Search

In this paper several astrodynamical formation flying models are assessed against the experimental results derived from the SPACE-SI formation flying experiments performed in September 2011 with the OHB Sweden developed Prisma satellites Mango and Tango. In these formation flying experiments critical manoeuvres for three types of missions were investigated with respect to in-orbit performances. The experiments included parallel flying with in track displacement demonstrating high-resolution optical dual satellite imaging and radar interferometric constellation, circumvolution as well as encircling of the target demonstrating debris observation and parallel flying with the radial displacement demonstrating fractionated spacecraft and accurate pointing of the formation. The astrodynamic data of the experiment are used to verify several formation flying models including a nonlinear model, a linear Hill-Clohessy-Wiltshire model, STK models with four propagators (Earth mass point, J2, default HPOP and HPOP with all disturbances) and the hereby originally proposed extension to the Hill-Clohessy-Wiltshire model, a linear model for orbits with small eccentricities.

Among applications of formation flying, several case scenarios for High Resolution Remote Sensing Satellite Constellations were proposed in the literature. For a radar interferometric system a pair of satellites has to be at two different positions that are separated by a distance of several hundred meters during measurement sequence. The satellites can be either in the same orbit or in a part of approximately parallel orbits. During imaging the relative separation of the satellites has to be stable and precisely known. In the case of an optical payload, one satellite can hold the optical lens system and the other the imaging sensors. The satellites must fly one over the other or one behind the other at a close range. In the paper, several manoeuvres for Satellite Constellations are analysed and simulated with the respect to fuel consumption. A linear model based on Hill-Clohessy-Wiltshire equations is solved analytically for the fuel consumption analysis. Linear models are optimised serving an approximate solution with respect to optimal fuel consumption respecting constraints, such as maximal disposable time and the instant of required formation position. Better results are obtained when orbit eccentricity is taken into account, as shown in the simulated examples.

This paper presents the design of a mobile robots simulator which is applicable for robot soccer or for general use. The simulator consists of a group of mobile robots, the ball and includes knowledge of their dynamic behavior modeling, collisions modeling and visualization. Two different approaches to the design of this simulator are given and compared. By the first approach the simulator physics background was completely developed by our team, which enabled us to get a better insight into the problem domain and gave us the possibility to efficiently solve some simulator specifics. First the model of ball and robot motion was derived and then complex approximate collisions models, where the real robot shape is taken into consideration. Some new ideas of collision formulation, realization and real robot shape inclusion are used. By the second approach the simulator was developed using freely available physics engine ODE ? Open Dynamics Engine. This engine already includes physical background for rigid bodies dynamic ; it is up to the user to define mechanical and physical parameters of the objects to simulate e.g. dimensions, masses, friction, joints and the like. The implementation of both simulators are described and a comparisons of their reality description, computational efficiency, effort and knowledge needed to built the simulator, advantages and disadvantages are given.

A reliable tracking algorithm of multiple mobile robots is addressed in this paper. A central vision system is used to find robots in the environment where robots are marked with identical colour patches mounted on each of them. To distinguish robots a robust tracking algorithm considering robot kinematics is implemented. However, mistakes such as swaps of robots identifications can happen and are unrecoverable, resulting in wrong control. Therefore swaps of robots identifications and initialisation problems are tackled using a supervision algorithm which is based on the correlation of command, and estimated robots velocities.

In this paper a problem of tracking multiple robots is addressed. A central vision system is used to find possible robots in the environment using identical colour marks mounted on each of the robots. To distinguish robots a robust tracking algorithm considering robot kinematics is implemented. However, swaps of robots identifications can still happen and are usually unrecoverable. Therefore swaps of robots identifications and initial initialisation problems are tackled using a supervision, which is based on correlation of command, and estimated robots velocity is implemented.

In the paper a global vision scheme applied to a fast dynamic game - robot soccer is presented. The process of robots positions and orientations estimation is divided into two steps. In the first step, the Bayer format image is acquired from camera, then the RGB image is interpolated and pixels are classified into a finite number of classes. At the same time, a segmentation algorithm is used to find corresponding regions belonging to one of the classes. In the second step, all the regions are examined. Selection of the ones that are parts of the observed object is made by means of simple logic procedures. A data filtering is used to improve identified noisy data. The novelty is focused on the optimization of the image acquisition algorithm as well as the processing time needed to finish the estimation of possible object positions.

In this paper description and comparison of two trajectory planning methods for mobile robots with differential drives is given. The trajectory was planned in the way to minimize the time of reaching end point. Described algorithms are suitable in cases when desired robot configuration is defined not only with robot pose, but also with its velocity in the desired pose. This is the case, for example, in robot soccer where robot must kick the ball with desired orientation and velocity.

V sledeči magistrski nalogi smo predstavili postopek razvoja in izdelave ple- zajočega robota, ki ima konfiguracijo multipoda s petimi nogami. Naloga ro- bota je, da med plezanjem po zidu z vodo oskrbuje rastline. Osredotočili smo se predvsem na tiste elemente, ki so bolj zanimivi iz vidika robotike in vodenja sistemov. V času razvoja smo najprej razvili kinematični in dinamični model ro- bota in njegovih nog in algoritme za načrtovanja posameznih gibov in celotnega premika robota. Nato smo zasnovali regulator, ki je sposoben vodenja vseh pe- tih nog naenkrat, a neodvisno druga od druge in lahko deluje v treh različnih načinih: pozicijskemu, hitrostnemu in v načinu vodenja po sili. Na koncu smo vse elemente povezali v aplikacijo, ki smo jo implementirali v programskem jeziku Python. The present Master thesis introduces the design and production of a climbing robot with configuration of a five-legged multipod. This robot has been developed for watering plants while climbing the wall where plants were placed. Our focus has been mainly dedicated to those parts of the project, which in our belief are of significant importance in the fields of robotics and control theory. We have developed kinematics and dynamics models of the robot and its legs, algorithms for planning individual movements and movement of the robot in general, and a regulator for control of independent movements of all five of the robot’s legs simultaneously. The regulator enables manipulation over the robot’s motion in three different modes: via position control, via velocity control, and via force control. Last but not least, we have developed an application for controlling the robot’s motion based on aforementioned approaches, written in computer programming language Python.

Magistrsko delo se osredotoča na predstavitev potrebnih teoretičnih znanj za uspešno implementacijo računalniških sistemov v farmaciji. Ključni del je razumevanje programskega okolja DeltaV, ki predstavlja enega izmed najpogosteje uporabljenih procesnih računalniških sistemov v industriji (ang. Distributed control system, s kratico DCS). V prvi del magistrskega dela je vključena teorija. Začnemo s kratko predstavitvijo teorije sistemov DCS, nato sledi opis ključnih zadev standarda ISA S88.01, ki definira terminologijo in module šaržnih sistemov. Zadnji del pa predstavlja dobre avtomatizirane proizvodne prakse (GAMP), ki so ključne pri življenjskem ciklu vsakega računalniškega sistema v farmaciji. Drugi del predstavlja predstavitev programskega okolje DeltaV. V njem predstavimo šest ključnih programov v tem programskem okolju, ki so ključni za uspešno implementacijo in nato validacijo vsakega računalniškega sistema. Zadnji del magistrske naloge pa predstavlja končno implementacijo celotnega farmacevtskega proizvodnega obrata. V tem delu predstavimo ključne module vodenja in faze, ki smo jih implementirali. Na koncu pa še predstavimo uporabniški vmesnik in končni recepturi. The master thesis focuses on presenting the necessary theoretical knowledge for the successful implementation of computer systems in pharmacy. The key part is understanding the DeltaV software environment, which is one of the most commonly used distributed control systems (DCS) in the industry. The first part of the master thesis includes theory. We begin with a brief introduction to DCS systems theory, followed by a description of key aspects of the ISA S88.01 standard, which defines the terminology and modules of batch systems. The final part presents good automated manufacturing practices (GAMP), which are crucial for the life cycle of any computer system in pharmacy. The seconds part presents the DeltaV software environment. We present six key programs in the software environment that are essential for successful implementation and subsequent validation of each computer system. The last part of the master thesis presents the final implementation of the entire pharmaceutical production plant. In this part, we present the key control modules and phases that we have implemented. Finally, we present the user interface and final recipes.

Z razvojom avtomatizirane proizvodnje so v transportu avtonomna vozila pridobila pomembno vlogo v logistični industriji. Za učinkovito delovanje robotskih vozil je potrebno dobro načrtovati poti, da zagotovimo njihovo optimalno delovanje in s tem posledično preprečimo morebitne konflikte. V delu predlagamo pristop planiranja poti, ki temelji na algoritmu SIPP. Algoritem SIPP je hitra oblika algoritma A* za načrtovanje v dinamičnih okoljih, s konceptom varnega intervala in najzgodnejših časov prihoda v vozlišče. Algoritem pri načrtovanju poti upošteva prioriteto vozil. Algoritem na začetku določi pot vozilu z najvišjo prioriteto, nato sledijo ostala vozila. Vsa nižje prioritetna vozila morajo pri svojem načrtovanju upoštevati časovne zasedenosti vozlišč in cest prehodnih vozil. V primeru, da algoritem vozilu določi čakanje, ta čakajo v vozlišču. V posodobljenem algoritmu vozila namesto čakanja v vozlišču čakajo na cesti pred vozliščem. Vozila s čakanjem na cesti omogočijo prost prehod ostalim vozilom mimo vozlišča in tako lahko hitreje dosežejo svoj cilj. V primeru, da na trenutni cesti ni dovolj prostora, da bi vsa vozila imela zagotovljeno varno čakanje, potem posodobljeni algoritem predlaga čakanje na predhodni cesti. Pristop planiranja poti vozil je prikazan na simulacijskih primerih. Automated guided vehicles have gained an important role in the logistic industry with the development of automated production. For the efficient operation of robotics vehicles, it is necessary to plan routes well in order to ensure their optimal operation and thereby prevent potential conflicts. Is this work, we propose a route planning approach based on the SIPP algorithm. The SIPP algorithm is a fast form of the A* algorithm for scheduling in dynamic environments, with the concept of safe interval and earliest arrival times at a node. The algorithm takes into account the priority of vehicles when planning the route. The algorithm initially determines the path of the vehicle with the highest priority, then other vehicles follow. All lower priority vehicles must take into account the time occupation of node and roads of transit vehicles in their planning. If the algorithm assigns a waiting time to the vehicle, it waits in the node. In the updated algorithm, instead of waiting in the node, vehicles wait on the road in front of the node. By waiting on the road, vehicles allow free passage for other vehicles past the node and thus can reach their destination faster. In the event that there is not enough space on the current road for all vehicles to have a guaranteed safe waiting, then the updated algorithm suggests waiting on the previous road. The vehicle path planning approach is shown on simulation examples.

V okviru magistrskega dela smo ustvarili orodje za simulacijo kolesnih mobilnih robotov v skladiščnih okoljih. Orodje sestavljata dva ključna gradnika: simulator mobilnega robota in sistem vodenja. Simulacija je realizirana v razvojnem okolju Unity, vodenje pa v okolju Matlab-Simulink. Med seboj komunicirata preko internetnega komunikacijskega protokola UDP. Simulacija je namenjena predvsem, da služi kot učni pripomoček študentom robotike. Dodatno smo implementirali še nekaj pogostih algoritmov vodenja, ki služijo kot primeri uporabe razvite rešitve. V uvodnih dveh poglavjih magistrske naloge je predstavljena tematika kolesnih mobilnih robotov: njihova umestitev v robotiki in industriji, njihova tipična okolica in senzorika. Poudarek je na razumevanju elementov mobilnih robotov, ki bodo relevantni v kasnejših poglavjih kot del simulacije. V naslednjem poglavju je predstavljena simulacija, ki smo jo ustvarili v okviru tega dela. Za pomoč pri razumevanju so na kratko predstavljene glavne uporabljene komponente in pomembni pojmi okolja Unity. Razloženi so vsi ključni elementi programa: kontrolnik kamere, kontrolnik grajenja, sistem shranjevanja in nalaganja, simulacija samega robota in implementacija UDP komunikacije z okoljem Matlab-Simulink. Predstavljene so tudi implementacije simuliranih senzorjev: svetlobni merilnik razdalje Lidar, senzor za sledenje črti in senzor za zaznavo NFC značk. V nadaljevanju je predstavljenih nekaj pogostih rešitev iskanja optimalne poti za mobilne robote: algoritem Dijkstra, algoritem A* ter iskanje v širino in iskanje v globino. Podrobno je predstavljena tudi izvedba algoritma A*, ki je bila v okviru tega magistrskega dela implementirana v okolju Matlab. V sklepnem delu je predstavljenih še nekaj najpogostejših algoritmov vodenja, ki smo jih kot primer uporabe za končnega uporabnika simulacije razvili v okolju Matlab-Simulink. Gre za algoritme vodenja v točko, vodenja po črti in izogibanja trkov. Within the scope of this thesis, we created a tool for simulating wheeled mobile robots in warehouse-like enviroments. It contains two key elements: a mobile robot simulator and its control system. The simulation was created in Unity development platform and the control system was made in Matlab-Simulink environment. The two elements communicate via the UDP internet communcication protocol. The simulation is intended to serve as a learning assist to robotics students. Additionally, we implemented a few common control algorithms, which serve as usage examples of the developed solution. In the opening two chapters of the thesis, the topic of wheeled mobile robotics is presented: its place in robotics and industrial applications, their typical environment and sensorics. An emphasis is placed on understanding those elements of mobile robots, that will be relevant in the upcoming chapters as parts of the simulation. Next chapter presents the simulation, that was created in the scope of this thesis. The most important used components and concepts of Unity environment are briefly explained to aid the reader's understanding. Key elements of the program are explained: a camera controller, a building controller, a save and load system, the robot simulation itself and the implementation of communication between the two used environments via UDP. Additionally, implementations of simulated sensors are also presented: Light ranging and detection sensor (Lidar), Line following sensor and NFC tag reader. Afterwards, a few common optimal path finding solutions are presented: Dijkstra's Algorithm, Algorithm A*, Breadth-first Search and Depth-first Search. The chapter also contains a detailed explanation of the implemented A* algorithm. A few of the most common control algorithms are presented in the closing chapter. We developed the following control algorithms in the Matlab-Simulink environment: control to point, line following control and obstacle avoidance. These serve as usage examples for the end user of the developed simulation.

Večanje človeške populacije in z njo potrebe po hrani predstavlja izziv za svetovno kmetijstvo. K večji proizvodnji pridelkov lahko veliko prispeva kmetijska robotika, ki bi v prihodnosti povečala donos, trajnost in dostop do pridelkov ter hkrati zmanjšala uporabo pesticidov. V tej nalogi smo se osredotočili na enega od mnogih izzivov preciznega kmetijstva in sicer na avtonomno navigacijo po kmetijski površini. Razvili smo navigacijski algoritem, ki je sposoben uspešno navigirati majhnega mobilnega robota po simuliranem polju koruze. Za načrtovanje poti, ki robota vodi po sredini vrst in iz ene vrste v drugo, smo uporabili meritve iz 2D LiDAR senzorja. Te podatke smo obdelali s postopkom rojenja DBSCAN in SLAM algoritmom. Robustnost navigacijskega algoritma smo preverili na poljih z različno razgibanostjo tal in gostoto posejanih rastlin. Pokazali smo, da kljub temu, da večja razgibanost tal vpliva na kvaliteto vodenja in delovanja SLAM algoritma, ima na kvaliteto navigacije bistveno večji vpliv gostota posejanosti rastlin. Naš algoritem je torej sposoben navigirati majhnega mobilnega robota po simuliranem polju koruze z razgibanim terenom, dokler je to dovolj gosto posejano. The increasing number of human population and its growing need for food presents a great challenge for global farming. In the future, agricultural robotics can greatly contribute to increased production of crops, which would increase yield, sustainability and access to crops and at the same time reduce the use of pesticides. In this work we focus on one of the many challenges of precision farming, namely autonomous navigation in an agricultural environment. We developed a navigation algorithm that is capable of successfully navigating a small agricultural robot in a simulated maize field. We used measurements from a 2D LiDAR sensor to plan a path that guides the robot through the field rows and from one row to another. These measurements were processed using the clustering algorithm DBSCAN and a SLAM algorithm. The robustness of the navigation algorithm was tested on simulated fields with different levels of terrain roughness and plant density. In our tests we show that higher levels of terrain roughness affect the quality of the robot driving and SLAM algorithm, however lowering the plant density has the largest effect on the quality of the navigation. We conclude that our algorithm is capable of navigating a small mobile robot in a simulated maize field as long as the field is sufficiently densely planted.

Iskanje poti za skupino agentov in koordinirano gibanje v skupnem okolju je osnovni problem večagentnih sistemov. Algoritme za večagentno planiranje poti lahko delimo v optimalne in podoptimalne. Iskanje optimalnih rešitev je zahtevno, saj prostor stanj raste eksponentno s številom agentov in zato problem ni rešljiv v polinomskem času. V primeru večjega števila agentov se uporabljajo podoptimalni algoritmi, ki hitro najdejo izvedljive rešitve. CBS (angl. Conflict-Based Search) algoritem je večagentni algoritem, ki deluje na osnovi iskanja trkov med agenti in deluje na dveh nivojih. Nižji nivo je namenjen iskanju optimalne poti posameznega agenta. Višji nivo pa preverja konflikte med agenti, gradi drevo konfliktov z detekcijo trkov med agenti in išče najboljšo rešitev brez trkov. Zaradi delovanja na dveh nivojih je veliko hitrejši od algoritma $A^{*}$, saj algoritem CBS ne upošteva opcij, kjer ne more priti do optimalne poti. Slabost algoritma se pokaže, ko imamo prostor, kjer je med računanjem do neke točke optimalnih poti več. Takrat mora algoritem preveriti vse poti in preveriti, če je katera izmed njih boljša. Namen diplomske naloge je pregled in spoznavanje delovanja algoritma CBS in njegovih nadgradenj, prav tako pa primerjati algoritem CBS z nekaterimi ostalimi algoritmi ter primerjanje algoritma CBS z njegovimi izboljšavami. Finding a path for a group of agents and coordinated movement in a common environment is a fundamental problem of multi-agent systems. Algorithms for multi-agent route planning can be divided into optimal and suboptimal. Finding optimal solutions is challenging as the state space grows exponentially with the number of agents and therefore the problem cannot be solved in polynomial time. Therefore, in the case of a larger number of agents suboptimal algorithms are used which quickly find viable solutions. The CBS (Conflict-Based Search) algorithm is a multi-agent algorithm that works on the basis of searching for conflicts between agents and works on two levels. The low level search is dedicated to find the optimal path of a single agent at a time. At the high level, search is performed on a tree based algorithm with a help of conflicts between agents. Because it is a two-level algorithm, it is much faster than the $A^{*}$ algorithm since the CBS algorithm examines fewer states while still maintaining optimality. A weakness of the algorithm occurs when we have an open space. There are a few optimal paths during this calculation. In this case, the algorithm must examine all of them to see if there is a better one. The purpose of the diploma thesis is to review and learn about the operation of the CBS algorithm and its improvements, as well as to compare the CBS algorithm with some of the other algorithms and to compare the CBS algorithm with its improvements.

V številnih panogah si industrije brez podpore robotizacije v obliki manipulatorjev ne moremo več predstavljati. Manipulatorji so večinoma fiksno vpeti, saj lahko z delovnim dosegom učinkovito opravijo zahtevana opravila. V primeru zahtev po večjem dosegu pa je potrebno uporabiti mobilne manipulatorje, kjer je manipulator pritrjen na mobilno platformo. Slednje zahteva naprednejše algoritme za lokalizacijo in vodenje. V magistrskem delu smo rešili problem vodenja redundantnega neholonomičnega mobilnega manipulatorja. Uporabljali smo robotski sistem sestavljen iz mobilne platforme Pal Tiago Base, na kateri je bil nameščen robotski manipulator Panda Emika Franka. Za robotski sistem smo izdelali dva različna modela kinematike. Pri prvem modelu smo za vodenje sklepov mobilne platforme uporabili enačbe kinematike diferencialnega pogona. Pri drugem modelu pa smo mobilno platformo modelirali kot kombinacijo navideznega rotacijskega in navideznega translacijskega sklepa. Izdelali in preizkusili smo različne optimizacije za reševanje problema redundantnosti sestavljenega robotskega sistema. Za odpravo problema neholonomičnih omejitev pogona smo s pomočjo optimizacije kota med platformo in točko, mobilno platformo med vodenjem orientirali proti ciljni točki. Izdelali smo algoritem vodenja, ki smo ga preizkusili v simulacijskem in realnem okolju. Pri vodenju robota v realnem okolju smo meritve lege platforme in manipulatorja zajemali s pomočjo sistema OptiTrack. Za preizkus smo si zamislili dve različni poti točk, po katerih smo (v simulaciji) vodili našega robota. Ugotovili smo, da optimizacija manipulabilnosti vrha robota izboljša sledenje trajektoriji, vendar pa robot zahaja v limitne lege sklepnih vrednosti, ter posledično ne more nadaljevati svoje naloge. Z uporabo optimizacije za doseg želene lege sklepov smo rešili problem zahajanja sklepov v končne lege, poslabšali pa smo kakovost sledenja trajektoriji. In many industries, it is no longer possible to imagine factories without the support of robotization in the form of robotic manipulators. The manipulators are mostly fixed, as they can efficiently perform the required tasks within their working reach. In the case of requirements for greater reach, it is necessary to use mobile manipulators, where the manipulator is attached to a mobile platform. The latter requires a more advanced localization and guidance algorithms. In the master’s thesis, we solved the problem of controlling a redundant non-holonomic mobile manipulator. We used a robotic system consisting of a Pal Tiago Base mobile platform, on which Emiko Frank's Panda robotic manipulator was installed. We created two different kinematics models for the robot system. In the first model, we used differential drive kinematics equations to guide the joints of the mobile platform. In the second model, we modelled the mobile platform as a combination of virtual rotational and virtual translational joints. We developed and tested various optimizations to solve the redundancy problem of the composite robot system. We oriented the mobile platform towards the target point while guiding the robot to solve the problem of non-holonomic limitations of the drive, by optimizing the angle between the platform and the point. We developed a guidance algorithm which we tested in a simulation and real environment. Measurements of the position of the platform and the manipulator were captured using the OptiTrack system, when guiding the robot in a real environment. For the test, we imagined two different paths of points along which we (in the simulation) guided our robot. We found that optimizing the manipulability of the top of the robot improves trajectory tracking but the robot enters the limit positions of the joint values, and as a result cannot continue its task. By using optimization to reach the desired position of the joints, we solved the problem of the joints moving into their final positions but worsened the quality of trajectory tracking.

Skladišča po svetu postajajo vse večja in potrebujejo vse več vozil in ljudi, da jih upravljajo, zato je na tem področju velika želja po tem, da se jih čim bolj avtomatizira, saj jih s tem naredimo učinkovitejše. Fakulteta za elektrotehniko v Ljubljani razvija algoritem, ki simulira in planira poti večih vozil, pri čemer skrbi, da se vozila izogibajo trkom. Za planiranje poti skrbi znani algoritem A*, ki je pa nadgrajen, da upošteva in predlaga še čakanje pred vozlišči in pa umik na stranske ceste, če je to potrebno. Vendar pa bolj kot postaja nek program kompleksen in temeljit, kompleksnejša lahko postaja njegova uporaba, zato so za učinkovito uporabo nujno potrebne poenostavitve. Slabost tega programa za načrtovanje poti pa je, da je sama implementacija vsakega zemljevida zelo zamudna, saj je potrebno v kodo ročno vpisati ogromno parametrov, zaradi tega si je tak zemljevid tudi težko prestavljati in si ga je potrebno predhodno nekam načrtati. Poleg tega je pa izdelava tega zemljevida zelo specifična in se mora oseba, ki se tega loti, predhodno spoznati s pravili gradnje zemljevida. Zato sem napisal grafični vmesnik, ki to delo zelo olajša, saj lahko enostavno z miško premikamo ceste po zaslonu, jih povečujemo in obračamo s tipkami na tipkovnici ter v razna prazna polja vpisujemo parametre, ki jih potrebujemo. S tem uporabniku ni treba poznati same sestave in oblike kode, ki je potrebna za izdelavo zemljevida, saj za vse to poskrbi sam algoritem. Warehouses all around the world are becoming bigger and need more vehicles and people to manage them. Consequently we need to automate that as much as possible so the warehouses become more efficient. Faculty of electrical engineering in Ljubljana is developing an algorithm for planning the paths of multiple vehicles in a simulated warehouse. Planning the paths is taken care of by an A* algorithm, which is upgraded so it takes into account and suggests the waiting times before crossroads and retreats of the vehicles into the side roads if needed. However the more the software becomes thorough, the more complex its usage can become and to avoid that we need simplifications. The negative side of this algorithm for planning the paths is that the implementation of each map of roads for the vehicles is very time-consuming, because you need to write a lot of parameters one by one directly into the source code and because of the complexity of the maps you need to make additional sketches beforehand. Besides that, the making of this map is very specific and unintuitive so you would first need to learn how to make it. Therefore I programed a graphical user interface which makes the implementation of maps a lot easier, because you can move roads by mouse, make them larger or smaller or rotate them by pressing keys on keyboard and write in needed parameters simply by using input boxes. The user now doesn't need to know how to make the official complicated function that creates maps, because all that is done by my algorithm.

Zaradi vse večjega vključevanja ljudi v delovni prostor je narasla potreba po zanesljivem algoritmu, ki je sposoben detekcije in izogibanja ovir. V ta namen so se razvili prediktivni algoritmi vodenja, ki vnaprej predvidevajo trk z objektom in robota predhodno preusmerijo. V magistrski nalogi bom predstavil algoritem, ki rešuje problematiko vodenja mobilnih robotov v industrijskem okolju. Predstavljen algoritem se naslanja na bolj znan pristop dinamičnega okna in dinamičnega generiranja trajektorij ter skuša odpraviti njune pomanjkljivosti. Povečano je število možnih poti, med katerimi algoritem lahko izbira, ter njihova razvejanost. Dodal sem kompenzacijo časovnega zamika med zajemom lokalizacijskih podatkov ter njihovo uporabo v izračunu poti. Poleg osnovnih cen za odmik od cilja in poti sta na voljo še ceni za spremembo translatorne in kotne hitrosti, ki omogočata boljši nadzor nad dinamiko vozila. Algoritem sem testiral tako v simulacijskem okolju Gazebo kot v realnem okolju na robotu podjetja Epilog d. o. o. V simulacijskem okolju sem najprej preveril njegovo osnovno delovanje, nato pa še sposobnost izogibanja statičnim oviram. Na robotu pa sem opravil test gibanja iz točke A v B, izogibanje statični oviri ter izogibanje gibajoči se osebi. Rezultate vseh testov sem predstavil v obliki dveh grafov, ki prikazujeta razdaljo do končnega cilja ter napako sledenja globalni poti. V zaključku sem še opisal večje pomanjkljivosti algoritma ter predlagal možne izboljšave. Due to an ever increasing inclusion of people into the workspace of robots, the demand arose for a reliable algorithm that is capable of obstacle detection and avoidance. To fulfill that demand, a group of predictive control algorithms were developed that can predict collisions with objects and divert the robot in order to avoid them. The algorithm I present in this thesis is based on the more widely known dynamic window and trajectory rollout algorithms and tries to correct their flaws. The number of additional possible paths is increased and so are their branching options. I added a time delay compensation to compensate for the time difference between the capture of localization data and their usage in the algorithm. Besides the base cost parameters of distance to goal and path following error, two new were added. The costs of translational and angular velocity change, help increase the users control over the dynamics of the robot. I tested the algorithm both in the Gazebo simulation environment software and in a real environment with a mobile robot developed by the company Epilog d.o.o. The simulation was first used to test the basic functionalities of the algorithm and later its ability to avoid static obstacles. With the real robot, we performed tests that determined if it could travel from point A to B, avoid static obstacles and avoid a moving person. In the conclusion I describe all the major flaws of the algorithm and ways to improve it.

Diplomsko delo zajema postopek seznanjanja z vsako od treh glavnih komponent za izvedbo avtinomnega mobilnega sistema – robotom (Roomba model 600), kamero (Intel RealSense T265) in vgradnim sistemom (Raspberry Pi 3). Delo obravnava vzpostavitev povezave med navedenimi napravami in na koncu še sestavo programa, ki s pomočjo lokalizacije na osnovi slike omogoča avtonomno delovanje mobilnega sistema. Za dosego cilja je potrebno dobro razumevanje vsake posamezne naprave in programskega vmesnika, ki ga le-ta uporablja. Poseben poudarek je narejen na opisu kamere, senzorjev in obdelavi signalov. Povezava kamere z drugimi napravami potrebuje namestitev ustreznih knjižnic, medtem ko povezava med robotom in vgradnim sistemom potrebuje poseben vmesnik. Za izvedbo vodenja mobilnega sistema so uporabljene osnovne enačbe za računanje s koti in hitrostmi, ki so skupaj povezane s proporcionalnim regulatorjem. Sam cilj naloge je izvedba avtonomnega mobilnega sistema, ki lahko potuje med želenimi točkami ter se na poti uspešno umika oviram. Glede na cenovni razpon naprav, je sistem sorazmerno stabilen, pogreški pa so relativno majhni. The thesis consists of the procedure of acquaintance with each of the three main devices that comprise an autonomous mobile system - robot (Roomba model 600), camera (Intel RealSense T265) and embedded system (Raspberry Pi 3). The work presents the coupling of the presented devices and writing of program code that uses vision-based localization for autonomous driving of the mobile system. To achieve this goal, a good understanding of each device and the programming environment it uses is required. In this thesis, particular emphasis is placed on the description of the camera, sensors and signal processing. The connection of the camera with other devices requires the installation of appropriate libraries, while the connection between the robot and the embedded system requires a special interface. For the implementation of autonomous driving, I have used some basic equations for the calculation of angle and velocity, which are combined in a P-type controller. The goal of the work is to build an autonomous mobile system that is able to drive autonomously between desired points while avoiding obstacles. Considering the price range of the devices used, the system is quite stable while the errors are relatively small.

Pri avtonomni vožnji robotskih vozil, je planiranje poti eden ključnih izzivov s katerim se soočamo. Za usklajeno delovanje skupine mobilnih robotskih vozil pri transportu materiala v sodobnih skladiščih in proizvodnih obratih, del planiranja poti predstavlja tudi izogibanje potencialnim konfliktom. V delu predlagamo nov pristop planiranja poti z algoritmom, ki temelji na znanem algoritmu A* in je nadgrajen za planiranje poti več robotskih vozil tako, da najde kompromisno rešitev brez trkov in nepotrebnih zastojev. Pristop upošteva prioritete transportnih nalogov, zemljevid v obliki uteženega usmerjenega grafa ter predvidena časovna okna zasedenosti segmentov zemljevida. Algoritem najprej poišče pot za vozila z višjimi prioritetami. Ob vsakem planiranju poti vozila, algoritem na koncu zabeleži predvidena časovna okna zasedenosti cest in vozlišč na zemljevidu za najdeno pot. Te zasedenosti se nato upoštevajo pri iskanju poti za vozilo z nižjo prioriteto tako, da ne ovira vožnje vozil z višjo prioriteto in se izogne konfliktom. Dve pomembni možnosti, ki jih algoritem upošteva in predlaga, sta čakanje pred vozliščem oz. na poti do vozlišča, da se prehod sprosti in pa možnost umika na stransko cesto v primeru onemogočenega čakanja. Algoritem upošteva tudi možnost hkratne vožnje več vozil po isti cesti ter ponovni zagon planiranja v poljubnem časovnem trenutku. Pristop je ovrednoten na simulacijskih primerih. With autonomous driving of robotic vehicles, path planning is one of the key challenges we face with. For coordinated operation of a group of automated guided vehicles (AGV) in the transport of material in modern warehouses and production facilities, part of path planning is also avoidance of potential conflicts. In this paper, we propose a new path planning approach with an algorithm which is based on the well-known algorithm A* and upgraded for path planning of multiple AGVs in such a way that it finds a solution without collisions and unnecessary congestions. The approach considers the priorities of transport orders, the map in the form of a weighted oriented graph, and the predicted time windows of the occupancy of the map segments. The algorithm first finds a path for AGVs with higher priorities. At the end of path planning for every AGV, the algorithm marks predicted time windows of occupancy of roads and nodes according to found path. These occupancies are then considered when planning a path for an AGV with a lower priority so that it does not obstruct the driving of AGVs with a higher priority and so that it avoids conflict. Two important options that the algorithm considers and suggests are waiting in front of the node or on the way to the node in order to clear the passage, and the possibility of r moving out of the way to a side road in case the waiting is disabled. The algorithm also takes into account the possibility of several vehicles driving on the same road at the same time and the option of restarting planning at any time. The approach is evaluated on simulation cases.

V diplomskem delu je opisan zagon in delovanje sistema žoga na plošči (angleško ball on plate system), ki je voden z mikrokrmilnikom Arduinom Uno in pametno kamero pixy. Namen diplomskega dela je predstavitev delovanja PID-regulatorja (proporcinalno-integrirni-diferencirni), ki se zelo pogosto pojavlja v avtomatizaciji, na učnem modelu kot je sistem žoge na plošči. Sistem žoga na plošči je sistem, ki z nagibanjem ploskve, na kateri je žogica, regulira položaj te žogice. Na začetku dela je predstavljeno teoretično ozadje delovanja PID-regulacije, sledi predstavitev uporabljene opreme in naprav. V osrednjem delu diplomske naloge so predstavljene dve različni programski kodi za izvedbo PID-regulacije z mikrokrmilnikom Arduino Uno, in sicer programska koda z že pripravljenimi funkcijami za realizacijo PID-regulacije in programska koda s funkcijo PID-regulatorja, ki sem jo napisal sam. Opisan je tudi postopek uglaševanja PID-regulatorja s pomočjo nastavitvenih pravil za integrirne sisteme. Predstavljeni so tudi odzivi sistema z regulacijo. V zaključku diplomskega dela so povzete lastnosti sistema in podane morebitne slabosti in izboljšave sistema. The diploma thesis describes the start-up and operation of the Ball on plate system with an Arduino Uno microcontroller and a smart pixy camera. The purpose of the diploma thesis is to present the operation of the PID-controller, which is often used in automation, on a learning model such as the Ball on plate system. Ball on plate system is a system that regulates the position of the ball by tipping the surface on which the ball is. At the beginning of the thesis, the theoretical background of PID-regulation is presented, followed by a presentation of all used equipment and devices. The main part of the diploma thesis presents two different program codes for the implementation of PID-regulation with Arduino Uno microcontroller, namely program code with already prepared functions for the implementation of PID-regulation and program code with PID-controller function, which I wrote myself. The process of specifying the parameters of PID-controller using the setup rules for integrated systems is also described. The responses of the system with regulation are also described. In the conclusion, the properties of the system are summarized and possible weaknesses and improvements of the system are given.

Dandanes so nadzorni sistemi vse pogosteje v uporabi. Uporabljajo se za preprečevanje materialne ali kakršne koli drugačne škode. Sistem nadzorovanja z razpoznavo obraza je idealna rešitev za marsikatero aplikacijo. Največkrat se razpoznava obraza uporablja za namene policije ali drugih organov pregona. V zadnjem času se je razpoznava obraza začela uporabljati tudi na širšem področju, pri čemer najbolj izstopata razpoznava obraza na pametnih telefonih in razpoznava obraza za dostopanje do bančnih storitev prek interneta. V diplomskem delu bo izdelana aplikacijo za nadzorni sistem s pomočjo kamere ESP32-CAM in laserskega senzorja razdalje VL53L1X. Kamera ima že vgrajeno funkcijo razpoznave obrazov. Naloga diplomskega dela je, da se kamera uporabi za razpoznavo obraza shranjenih oseb in vrne povratno informacijo o stanju (ali je oseba prepoznana pravilno ali ne). Senzor razdalje VL53L1X se uporablja kot pomoč pri nastavljanju pravilne razdalje med obrazom in kamero ESP32-CAM za najboljše rezultate razpoznave. Nowadays control systems are getting more and more popular each day. They are being used to prevent material and other types of damage. Face recognition control system is the ideal solution for many applications. Most commonly it is used for law enforcement purposes, but lately the use has expanded quite a bit. Most people come across it every day, by unlocking their phone with the face recognition feature that most of the smartphones have. It is also used for accessing banking information online. The diploma thesis describes the process of creating an application for a access control system with the help of an ESP32-CAM camera and the VL53L1X laser distance sensor. The camera already has a face recognition feature installed and my task is to use that feature for face recognition of the people, whoses faces I had to save onto the camera software, and camera can then return the feedback information about the status (meaning whether the person has been recognized correctly or not). The distance sensor VL53L1X is used just as a tool to help adjust the correct distance between the face and the ESP32-CAM camera for optimal results.

Diplomska naloga opisuje postopek zasnove in izvedbe testne naprave za testiranje transportnih sistemov WA290 za pomivalno dezinfekcijske medicinske stroje, ki so produkt podjetja Belimed. Testno napravo sestavljata glavna elektro omara A02 in elektro omara A01, ki simulira stroj WD290/WD290 IQ in avtonomni transportni modul. Testna naprava je namenjena testiranju posameznih modulov avtomatiziranega sistema. Testna naprava se je izdelala v sklopu projekta v podjetju Belimed. Pri testni napravi sem sprva kreiral električno shemo v programu EPLAN Electric P8, namestil mehanske komponente, vgradil električne komponente in jih povezal ter napisal navodila za upravljanje s testno napravo. Na začetku diplomskega dela je opisno predstavljen avtonomni transportni sistem WA290 z vsemi implementiranimi moduli, sledi predstavitev vrste komunikacije, ki je v samem sistemu uporabljena. Nato sem opisal svoje prispevano delo v projektu v kronološkem zaporedju. Z ekipo smo začeli s samo zasnovo testne naprave, kjer smo določili njene osnovne sestavne komponente, postopek namestitve le-teh, zagotovitev potrebnih povezav samih elementov, določili potrebe izgleda in funkcionalnosti za čim bolj optimalno rabo naprave. S kreiranjem električne vezalne sheme sem elektrotehnične elemente ter njihove povezave prikazal v izvedenem projektu, ki služi uporabniku testne naprave za boljše razumevanje delovanja ter mu omogoča enostavnejše popravke ob morebitni napaki ali odpovedi določenega sestavnega dela. Opisan je tudi postopek mehanske in elektro montaže, pri čemer sem opisno izpostavil pomembnejše sestavne dele testne naprave. The diploma thesis describes a design and implementation process of a test device for testing WA290 transport systems for medical washer-disinfection systems, which are products of the company Belimed. The test device consists of the main electrical cabinet A02 and the electrical cabinet A01, which simulates the machine WD290 / WD290 IQ and autonomusautonomous transport module. The testing device is intended for testing individual modules of an automated system. The test device was manufactured as part of a project at Belimed. For the test device, I first created an electrical diagram in the EPLAN Electric P8 program, installed the mechanical components, installed the electrical components and connected them, and wrote instructions for operating the test device. At the beginning of the diploma work, the automated transport system WA290 with all implemented modules is descriptively presented, followed by a presentation of the type of communication used in the system itself. I then described my contribution to the project in chronological order, with the team starting with the design of the test device itself, where we determined its basic components, the process of installing them, providing the necessary connections of the elements themselves, determining the layout and functionality needs for the most optimal use of the device. By creating an electrical wiring diagram, I showed the electrical elements and their connections in a completed project, which serves the user of the testing device to understand the operation and allows him easier corrections in the event of a failure or failure of a particular component. The process of mechanical and electrical assembly is also described, where I have outlined the more important components of the test device.

V prvem delu magistrske naloge je predstavljena optimizacija obstoječega stroja, ki je namenjen za odlaganje zalitih statorjev na paleto na tekočem traku, po katerem se ti odpeljejo na strojno pregledovanje. Statorji so zloženi na pladnjih, ki jih naložene na vozičke zapeljemo v stroj. Servomotorji zalogovnikov s pomočjo prijemal dvignejo pladenj na prenos, ki ga premika dodaten servomotor. Robot pobira vsak stator posebej, ga odnese na mesto odčitavanja QR kode, nato na preverjanje odtenka njegove barve ter na koncu na paleto, ki se nahaja na tekočem traku. Najprej je predstavljenih nekaj mehanskih predelav, ki so bile potrebne za zanesljivejše delovanje stroja. Opisan je program, ki vključuje: izdelavo programskega bloka za premikanje servomotorjev, izdelavo celotnega avtomatskega cikla zalogovnikov in prenosa, optimizacijo ročnega režima stroja, kodo za izvedbo zajemanja meritev pri analizi merilnega sistema senzorja barve ter optimizacijo zaslonov na kontrolnem panelu. Programiranje je potekalo na Siemensovem krmilniku in panelu s pomočjo njihovega programskega okolja TIA Portal. Stroj zajema meritve barvnih odtenkov statorjev, s pomočjo katerih je izvedeno preverjanje njihove toplotne obdelave. Zato je v drugem delu magistrske naloge predstavljena analiza merilnih sistemov, ki jo moramo izvesti na našem senzorju. Merilni sistemi morajo zagotavljati, da pri opravljanju meritev na pride do prevelikih odstopanj od realne vrednosti. Analiza zato merilnemu sistemu preverja stabilnost, ki predstavlja spreminjanje pristranskosti skozi daljše časovno obdobje, pristranskost, ki predstavlja razliko med referenčno vrednostjo in povprečjem nekega števila meritev, linearnost, ki prikazuje razliko pristranskost skozi celotno merilno območje sistema, ponovljivost, ki nam predstavlja razliko med zaporednimi meritvami enega merilca in obnovljivost, ki nam predstavlja razliko med povprečnimi meritvami posameznih merilcev. Merilni sistem našim zahtevam ustreza takrat, ko so vse naštete lastnosti znotraj mejnih vrednosti. The first part of the master's thesis presents the optimization of the existing machine, which is intended for depositing cast stators on a pallet laying on a conveyor belt, after which they are taken for machine inspection. The stators are stacked on trays, which are loaded onto carts, and then together loaded into machine. The storage servomotors lift the tray to the transmission by means of grippers, which is moved by an additional servomotor. The robot picks up each stator separately, takes it to the location of the QR code reading, then to check the hue of its color, and finally to the palette located on the conveyor belt. First, some mechanical modifications that were necessary for more reliable operation of the machine are presented. The program is described with the following: production of a program block for moving servomotors, production of a fully automatic cycle of storage and transmission, optimization of manual machine mode, code for performing measurement capture in color sensor measurement system analysis and screen control panel optimization. Programming took place on a Siemens system using their TIA Portal programming environment. The machine also includes measurements of the color shades of the stators, with the help of which the verification of their heat treatment is performed. Therefore, in the second part of the master's thesis, the analysis of measuring systems is presented, which we must perform on our sensor. Measuring systems must ensure that excessive deviations from the real value do not occur when performing measurements. The analysis of the measuring system checks the stability, which represents the change of bias over time, bias, which represents the difference between the reference value and the average of some measurements, linearity, which shows the bias difference throughout the measuring range of the system, repeatability, which represents the difference between successive measurements of one appraiser and reproducibility, which represents the difference between the average measurements of individual appraisers. Repeatability and reproducibility can be considered together to obtain a variance equal to the sum of the variations within the measurement system. The measuring system meets our requirements when all the listed properties are within the limit values.

Problematika določevanja optimalne ali kvalitetne poti v prostoru je zahtevna naloga. Ima veliko možnih praktičnih aplikacij, pri čemer je pogosto zahtevan hiter izračun. Računska kompleksnost narašča v odvisnosti od povečevanja števila parametrov, njihovih vrednosti in zahtevnostjo postavitve ovir. Magistrsko delo temelji na uporabi RRT-metod z lastnostjo hitre predstavitve neoptimalnega gibanja po prostoru v obliki grafa prehajanja stanj. Znotraj katerega se določi najboljša pot z uporabo vzvratne metode iskanja poti. Z nadaljnjim reduciranjem odvečnih zavojev oziroma delov poti se pridobljeni poti izboljša kvaliteta. Z Bezierjevimi polinomi se v obliki zvezno gladke poti predstavi reducirana pot za možnost takojšnega sledenja. Izkoristi se hitro določanje neoptimalnega grafa prehajanja stanj z RRT-metodami, z nadaljnjo obdelavo se izboljša pomanjkljivost hitre gradnje s končnim produktom kvalitetne hitro pridobljene poti. V luči pohitritve gradnje RRT-dreves so implementirani nadgrajeni postopki posameznih korakov. S predhodnim definiranjem konfiguracijskega prostora vplivamo na hitrejše preverjanje veljavnosti točk oziroma trka z ovirami, na možnost uporabe iskanja najbližje točke le znotraj želenega dela okolice z namenom zmanjšanja števila preverjenih točk ter na upoštevanje oblike ovire pri ponovnem določanju nove točke, če prvotna točka ali njena povezava leži na oviri. Slednje omogoča učinkovitejše širjenje drevesa ob oviri ali v ozkih prehodih. Eksperimenti potrjujejo učinkovitost uporabe preprostih RRT-metod, brez naknadnega povezovanja bližnjih stanj, v kombinaciji z nadaljnjo obdelavo poti, kar omogoča hiter izračun kvalitetne poti. The problem of determining the optimal or suboptimal path in space is a demanding task with multiple applicability as well as necessity for fast calculation. The computational complexity increases depending on the increase in the number of parameters or their values and the complexity of obstacles arrangement. The master’s thesis is based on the use of RRT-methods with the fast presentation feature of possible non-optimal movement through space in the form of a state transition graph, within which the best path is determined by using the reverse path tracking method. The acquired path quality is further improved by reducing redundant turns and represented with Bezier polynomials in the continuous smooth path form for the possibility of immediate tracking. Therefore with the use of RRT-methods, a non-optimal state transition graph is rapidly yielded and in combination with further processing, of improving the disadvantage of fast construction, a high quality path outcome is thus quickly acquired. In the light of speeding up the construction of RRT-trees, upgraded procedures of individual steps are implemented, such as predefining the configuration space for faster tree node collision checking with a possible nearby obstacle, the choice of using the search for the nearest node only within the interesting part of the area in order to reduce the number of checked nodes and taking into account the obstacle shape when redefining a new node. The latter allows more efficient tree spreading along an obstacle or narrow passages. Experiments confirms the effective use of simple RRT methods, without subsequently linking of nearby states, in combination with further path processing, which allows fast calculation of quality path.

Prevoz tovora se v logistiki in proizvodnji ponavadi opravlja z večjimi mobilnimi roboti, ki pa za infrastrukturo niso vedno najbolj primerni. So tudi težki, neokre- tni in zahtevajo veliko prostora, tudi ko ničesar ne prevažajo. V tem magistrskem delu je predstavljen algoritem za prevoz tovora z več manjšimi mobilnimi roboti, ki si tovor delijo. Manjši mobilni roboti so lahko lažji, hitrejši in bolj okretni, če je potrebno lahko učinkovito prevažajo tudi manjše tovore. V primeru večjega tovora pa se roboti s posebnim prijemalom pritrdijo na kolesa tovornega vozička, nase prevzamejo del teže tovora in voziček s tovorom odpeljejo na želeno mesto. Za prevoz tovora med točkami je bil razvit centraliziran algoritem, ki omogoča vodenje poljubnega števila robotov z diferencialnim pogonom. Algoritem je se- stavljen iz treh delov: sistema za izračun kinematike vožnje tovora, sistema za določanje orientacije robotov glede na tovor in dela, ki skrbi za vodenje robotov. Orientacija robotov glede na tovor se izračuna na podlagi treh metod, ki temeljijo na globalni legi robotov, odometriji robotov in kinematičnem modelu ter vizu- alnih markerjih. Za iskanje poti se uporablja algoritem A*, za sledenje poti pa prediktivni algoritem načrtovanja in sledenja trajektorije (Trajectory Rollout). Algoritem pretvori kontrolne ukaze za premik tovora iz algoritma za sledenje poti v kontrolne ukaze za posamezne mobilne robote. Implementacija algoritma temelji na okolju ROS (Robot Operating System). Algoritem je bil razvit in ovrednoten v simulacijskem okolju Gazebo. Nato se je prenesel in ovrednotil še na realnem sistemu – mobilnem robotu podjetja Epilog d.o.o. V simulaciji in na realnem sistemu so bili opravljeni eksperimenti z odprto- in zaprtozančnim vodenjem. Ustrezno delovanje algoritma se je določilo s tremi pokazatelji. Prvi pokazatelj je primerjava hitrostnega ukaza z dejansko hitrostjo tovora. Drugi pokazatelj so sile med roboti in tovorom, tretji pa bočno drsenje robotov. Algoritem je v vseh simuliranih in realnih eksperimentih deloval kar je pomenilo, da se je tovor premikal v skladu z želenimi ukazi brez bočnih drsenj robotov in brez prevelikih sil. Transport of load in production and logistics is usually done with bigger mobile robots that are not always the most suited for infrastructure. They are heavy, slow, and require lots of space even when not transporting anything. This mas- ter’s thesis presents the algorithm for transportation of load with multiple smaller mobile robots that share the load. Smaller robots are lighter, faster, and require less space. If needed they can more effectively transport smaller loads. In the case of bigger load robots use special gripper to attach themselves to the wheels of the transport cart and transfer some weight to themselves. A centralized algorithm was developed that is capable of controlling any num- ber of differential drive robots with the aim of transporting a load between points. The algorithm consists of three parts: a system for calculating kinematics of load transport, a system for determining the orientation of the robots with respect to load and robots control system. The orientation of robots with respect to load is calculated based on three methods that use the global pose of the robots, odom- etry of the robots and kinematic model, and visual markers. For path searching A* algorithm is used and for path following Trajectory Rollout algorithm is used. The algorithm is used to transform control orders for the load to controls for individual robots. The implementation of the algorithm is based on ROS (Robot Operating System) environment. The algorithm was developed and evaluated in Gazebo simulation environ- ment. It was then transferred and evaluated on a real system - on the mobile robot from company Epilog d.o.o. Open and closed-loop control experiments were done in simulation and on the real system. The proper working of the algorithm was determined using three indicators. First indicator in comparison of desired versus real velocity of the load. The second one is the forces between robots and the load. The third is sideways sliding of the robots. The algorithm worked in all simulated and real experiments which meant that the load was moving according to the commands without sideways sliding and without too big forces.

V diplomski nalogi je predstavljena izdelava avtonomnega zalivalnega sistema, ki je namenjen uporabi za manjše vrtove. Razviti sistem predvideva dva načina zalivanja, ki sta prilagojena specifikam tople grede in vrta. Algoritem se glede na večkrat dnevno izmerjene vrednosti temperature, vlage zemlje, vlage zraka in osvetljenosti odloči ali bo vrt v naslednjih jutranjih urah zaliv ali ne. Na začetku je predstavljen mikrokrmilnik Arduino in strojna oprema našega sistema. V osrednjem delu je opisan razvoj izdelka in koraki po katerih sem izdeloval avtonomni zalivalni sistem. Predstavljeni so tudi izračuni dnevne porabe električne energije razvitega sistema in meritve štirih spremenljivk, ki so ključne pri končnem algoritmu odločanja. Podrobneje je razloženo delovanje algoritma in izvedba s programsko kodo, ki sta ključna dela diplomske naloge. V zaključku so povzete lastnosti razvitega sistema in nakazane so možne izboljšave. In this thesis I present the development of autonomous watering system that can be used in small gardens. The developed system operates in two watering modes which are adapted to the requirements of a garden and a greenhouse application. The algorithm decides to water the garden on the next morning based on the measured values of temperature, soil moisture, air moisture and sunlight, captured several times in the previous day. In the beginning of my thesis I introduce the Arduino microcontroller and our system hardware. In the middle part I describe the development of the product and steps required to implement the autonomous watering system. I also present the calculation of the daily electrical energy requirement of the system and measurements of four main variables used in the decision making algorithm. I explain in more details the operation of the algorithm and the program code which is essential part of my thesis. In conclusion I summarise properties of the developed system and presented the potential improvements of the system.

Diplomska naloga je opisana izdelava sistema za regulacijo vlage in relativne temperature pri zorenju suhih mesnin. Sistem omogoča operaterju vpogled v veličine znotraj zorilne komore brez, da bi sam posegal vanjo, omogoča tudi spreminjanje teh veličin. Glavni cilj je bil razviti sistem, ki bo z nizko ceno opravljal večino funkcij profesionalnih sistemov za zorenje mesa brez prevelikega človeškega poseganja. Na krmilnik so povezani moduli za izvrševanje regulacije in za komunikacijo z uporabnikom. Glavni del naloge temelji na ustvarjanju programa za krmiljenje in za daljinsko upravljaje samega krmilnika. V prvem delu so predstavljeni elementi sistema in programska oprema povezana z njimi. V drugem delu pa je opisan sam programski del krmiljenja in daljinskega upravljanja ter nadzora krmilnika. Na koncu sledi še preizkus delovanja samega sistema. This thesis describes development of a system for automatic control of temperature and relative humidity with dry aging of meat. System enables user to observe and regulate key properties of the dry aging process. The main purpose of this thesis was to develop a system that enables successful application of the main properties of profesional systems intended for dry aging of meats on a budget and without too much human interference. Controller used in this thesis is connected to modules that actuate on the environment inside dry aging chamber and enable the use of remote controlling the system. The main part of the thesis is based on creating programs for automatic and remote controll of the system. First part consists of a presentation of a key hardware components and software that connects them together. Second part consists of programs descriptions used in the thesis. The final part is a test of the whole system.

Z razvojem avtomatsko vodenih vozil (AGV) in avtomatiziranih skladišč se je razvila potreba po razvoju algoritmov za avtonomno planiranje poti in tudi avtonomnem planiranju poti za parkiranje. Eden od ključnih problemov pri procesu parkiranja je planiranje poti. V tej študiji bo prikazano reševanje problema poti parkiranja štirikolesnega vozila AGV. V teoretičnem delu so predstavljeni trije algoritmi za planiranje poti: Dubinsove poti, RRT (ang. Rapidly-exploring random tree) algoritem in algoritem za iskanje poti s pomočjo krožnih lokov. V praktičnem delu je narejena študija algoritma za iskanje poti s pomočjo krožnih lokov, sam algoritem je implementiran in testiran v simulacijskem okolju. Glavni poudarek je na testiranju delovanja tega algoritma in predstavitvi rezultatov simulacije. Pot, ki povezuje začetno pozicijo AGV s končno pozicijo je sestavljena iz krožnih lokov. Algoritem je testiran v simulacijskem okolju, kjer so dimenzije parkirnega prostora in AGV-ja v naprej določene. Na začetku algoritem najde več kot eno končno krožnico, ki povezuje začetno in končno pozicijo AGV-ja. Krožnica, ki je izbrana kot končna izpolnjuje dva pogoja: radij končnega kroga mora bit večji od radija obračanja (Rturn) AGV-ja in tangentna točka med končno krožnico in krožnico obračanja AGV-ja mora bit nad X-osjo lokalnega koordinatnega sistema. Predlagani algoritem zagotavlja praktično metodo načrtovanja poti brez računsko zamudnega računanja, prav tako pa je sposoben najti končno pot iz katerekoli začetne pozicije AGV-ja. V tej študiji je bilo pokazano, da se dolžina poti spreminja s spreminjanjem naslednjih parametrov: pozicija in orientacija AGV-ja, krmilni kot, parkiranje naprej ali vzvratno. Narejeni so bili štirje optimizacijski algoritmi. Prvi dva algoritma optimirata pozicijo in orientacijo AGV-ja, če poznamo njegov krmilni kot theta. S to optimizacijo dobimo najboljše območje, kjer se lahko parkirni proces začne in tudi najboljšo orientacijo AGV-ja za parkiranje. Tretji algoritem optimira krmilni kot theta, kjer imamo podano začetno pozicijo in orientacijo AGV-ja. S to optimizacijo dobimo krmilni kot, s katerim pridemo do najbolj optimalne (najkrajše) poti. Pri zadnji optimizaciji se preveri način parkiranja (naprej ali vzvratno) s katerim pridemo do najbolj optimalne (najkrajše) poti. S pomočjo vseh optimizacij je bilo ugotovljeno, da je pot najkrajša, ko je začetna pozicija AGV-ja tudi tangenta točka med končno krožnico in krožnico obračanja AGV-ja. Dobra stran algoritma je, da vedno najde pot. Z uporabo različnih vrst optimizacije lahko pot prilagodimo glede na potrebe uporabnika (optimalno pot, ki pomeni najkrajšo pot, ali pot z določenim številom manevrov ali druge).Nadaljnji razvoj algoritma bi lahko vseboval testiranje na realnem primeru v realnem okolju. A path planning algorithm for parking an autonomous vehicle. Recently, with the development of self-driving, autonomous vehicles and automated warehouses there have been a growing interest in the development of autonomous path planning as well as parking algorithms. In this study, a solution for path planning in autonomous parking for 4-wheel-AGV(Automated Guided Vehicle) is presented. In the theoretic part of the study 3 path planning algorithms are exposed such as: Dubbin’s path, RRT (Rapidly-exploring random tree) algorithm and an arc path planning algorithm. In the practical part of the study the arc path parking algorithm is implemented and tasted. The main focus is given to algorithm testing and presentation of obtained test results in simulation environment. The path is circle arc that connects the vehicle initial position and goal position of the vehicle inside the parking space. The algorithm is tested in a simulation environment where parking space and dimension of the vehicle are known. At the beginning from the initial position of the vehicle to the final position the algorithm find more than one potential final circle that can be used to defending the final arc path. But the circle that is picked as final path has to have radius larger than the turning radius of the vehicle and the tangent point between the potential final circle and circles with turning radius of the car have to be above X axis of the local coordinate system. The proposed path planning algorithm provides a practical method of path planning without expensive computation and it was proved that if find a final path from any initial position. In the study it was proved that the length of the final path in changing depending on this parameters: position in orientation of the vehicle, steering angle and parking forward/backward. In the study 4 different optimizations to find the shortest path were done so the optimal path can be found. First one is optimization of the positions and orientations of the vehicle where we know the steering angle thata. With this optimization the best area where the vehicle should start with the parking procedure and path planning and the best orientation of the vehicle are found. Third is optimization of the steering angle theta, where we have the initial starting position and orientation of the vehicle. With this optimization we get what is the best steering angle of the vehicle so we get the optimal path length. And the last optimization is with parking forward/backward. With the optimization it was shown that the length of the path is shortest when vehicle park forward if it is turn with the front side of it to the parking spot. And if it park backward it is better if it is turned with the back towards the parking spot. Also while doing all the optimizations it was shown that the shortest path is found if the tangent point between final circle and circle with turning radius is same as the position point of the vehicle. The good side of the algorithm is that it always find a path, without expensive computations. But the path might not be always optimal, that is why it is good to know the optimization parameter that is important. If it is searching for optimal path meaning the shortest path, or the path with defined number of maneuvers or other optimization. The further work on the algorithm would be implementing and testing in on a real example in real environment.

Cilj magistrskega dela je določiti lego mobilnega robota z uporabo algoritma filtra delcev in odometrije v času gibanja v znanem okolju s pomočjo meritev razdalj. Pri tem smo uporabili kit komplet Starter proizvajalca Makeblock in ga nadgradili z ultrazvočnimi senzorji ter optičnimi inkrementalnimi dajalniki. Merilniki na mobilnem robotu pa zaradi fizikalnih omejitev, občutljivosti, ločljivosti in šuma iz okolja vsebujejo negotovost, kar zmanjšuje natančnost ocenjene lege. V magistrskem delu so opisane zgradba mobilnega robota in tehnične lastnosti posameznih sestavnih delov. Podana je ocena meritve razdalje z uporabo ultrazvočnega senzorja. Meritve so prikazane v obliki histograma in normalne porazdelitve ter ovrednotene s primerjavo srednje vrednosti izmerjene razdalje in prave vrednosti pri oviri iz lesa ter plastike. Podani sta tudi meritev hitrosti optičnega inkrementalnega dajalnika in njegova ocena. V zadnjem delu magistrskega dela je opisano modeliranje gibanja mobilnega robota v ravnini, določeno s kinematiko in dinamiko. Opisan je način določevanja lege mobilnega robota, ki je težje določljiva zaradi nedeterminističnosti. Pri tem je opisana izvedba algoritma filtra delcev v znanem okolju. Na koncu smo preverili delovanje algoritma filtra delcev in odometrije ter opisali njuno razliko glede na opravljeno meritev. The objective of this master's thesis is to determine the position of a mobile robot during motion within a familiar environment by means of the particle filter algorithm and odometry, and with the help of distance measurements. To do so, we used the Starter Kit Package by Makeblock, which we upgraded with ultrasonic sensors and optical incremental encoders. On account of the physical limitations, sensitivity, definition, and environmental noise, there is an uncertainty factor to the mobile robot's meters, which inhibits the accuracy of the determined position. This thesis describes the structure of the mobile robot and the technical specifications of its individual components. Laid out are the estimates of the distance measurements made by using the ultrasonic sensor. The measurements are presented by means of a histogram and normal distribution, and are evaluated by comparing the median value of the measured distance and the actual value with a wood or plastic obstacle. Also provided are the measurement of the optical incremental encoder's speed and its assessment. The final part of this thesis presents mobile robot motion modelling on a plane, determined by cinematics and dynamics. The manner of determining the mobile robot's position, which is more difficult to determine due to being non-deterministic, is explained. Also described is the implementation of the particle filter algorithm in a familiar environment. In conclusion, we also checked the functioning of the particle filter algorithm and odometry, describing their differences in terms of the measurements taken.

V prvem delu diplomske naloge je opisanih več načinov merilnih metod merjenja notranje luknje rotorskega paketa. Deljene so na brezkontakntne in kontaktne metode. Pri brezkontaktnih metodah je naštetih nekaj glavnih lastnosti merilnika Micro – Epsilon boreCONTROL in MBvision Rotor optical system vision ter njuno delovanje v avtomatiziranem sistemu. Pri kontaktnih metodah merjenja so opisana delovanja merilnikov DIATEST BMD, Mitutoyo SERIES 568 in TESA-Hite 400. Naštete so njihove pomembnejše lastnosti ter njihovo delovanje v avtomatiziranem sistemu, če je ta mogoč. V drugem delu diplomske naloge je podrobno opisano delovanje sistema s presvetlitveno kamero Keyence TM-040 in njen krmilnik TM-3001P. Najprej je predstavljena kamera, katera nam s presvetlitvijo vrne dvodimenzionalno sliko predmeta. Nato so našteti vsi vhodi in izhodi krmilnika ter na kratko opisane njihove namembnosti. Natančno je opisano, kako se nastavi program TM-Navigator in predstavljene so vse meritve, katere lahko uporabljamo. Naštete so osnovne lastnosti krmilnika Siemens 1212C in podano je, na katere načine lahko komunicira s krmilnikom Keyence TM-3001P. Za konec je opisano delovanje sistema kot celota in eden od načinov proženja zajemanja slike s programom TIA Portal. In the first part of the thesis a few measuring methods are described of how to measure the inside diameter of a rotor stack. They are divided in contactless and contact measuring methods. At contactless methods some of the main features of Micro – Epsilon boreCONTROL and MBvision Rotor optical system vision and their operations in an automated system are listed. At contact measuring method operations of DIATEST BMD, Mitutoyo SERIES 568 and TESA-Hite 400 are described. Their main features are listed and the possibility of their operation in the automated system is listed too. In the second part of the thesis the operation of the system with lighting camera Keyence TM-040 and its controller TM-3001P are described in detail. At first a camera which gives us a two-dimensional image of a measuring object is introduced. Secondly all the inputs and outputs of the controller are listed and their purpose is briefly described. It is thoroughly explained how to set the settings of TM-Navigator program and there are presented all the measurements for which we can use the program for. Then some of the basic features of controller Siemens 1212C are listed and the ways in which the communication with the controller Keyence TM-3001P is possible are introduced. Lastly the system operation is described as a whole and it is explained how to capture an image with the TIA Portal program.

V diplomski nalogi je predstavljen učni komplet za delo z razvojno tiskanino Arduino v okolju MATLAB. V glavnem delu je opisana pot do avtomatske regulacije žogice na gredi dolgi 40cm. Regulacija deluje preko programa MATLAB, kamor vpisujemo programsko kodo in programa Simulink, kamor rišemo blokovno shemo. Program MATLAB je preko USB kabla povezan na razvojno ploščo Arduino Uno, katera daje vhodno/izhodne signale sistemu. V prvem delu diplomske naloge se osredotočimo na zaprtozančno PI-regulacijo enosmernega motorja, ki je preko gredi povezan z drugim enosmernim motorjem oz. tahometrom. Posebej izvedemo P in I-regulacijo ter za vsako od regulacij navedemo njene glavne prednosti in slabosti. Simuliramo metodo nihajnega preizkusa in prek znane Nichols-Zieglerjeve metode določimo ojačenja in . Predstavimo model motorja, ki nam poda vse njegove lastnosti. Za lažje računanje modela si pomagamo z Laplace-ovo transformacijo, kjer sta glavni neznanki ojačenje K ter časovna konstanta T. Regulaciji izvedemo s pomočjo enotine stopnice ter preko metode določanja polov. V regulacijo je vključen tudi model sistema, ki prikazuje odziv sistema ob idealnih pogojih. V drugem delu diplomske naloge je predstavljen nadzor pozicije motorja in sicer pri močnostnem menjalniku, kjer je glavna lastnost prepoznavanje pozicije oz. kota gredi, ki je povezana z motorjem. Močnostni menjalnik lahko na vhodu krmilimo s potenciometrom ali s pulznim generatorjem, na izhodu je potenciometer, ki je povezan z motorjem menjalnika in daje podatke na analogni vhod. Podane so osnovne lastnosti sistema drugega reda, kjer uporabljamo PID-regulacijo ter na koncu izpeljanki PI-D in I-PD-regulacijo. Ojačenja regulatorjev so izračunana preko polov sistema, ki jih določimo z enačbo sistema drugega reda. V tretjem delu diplomske naloge uporabimo vse znanje, ki smo ga pridobili iz prvih dveh poglavij. Glavna naloga je regulacija žogice na točno določeni točki na gredi. Pozicijo žogice na gredi spremlja svetlobni senzor, ki oddaja in sprejema infrardeči žarek ter tako določa pozicijo žoge na gredi. Gred uravnava močnostni menjalnik, ki prav tako kot senzor daje podatke mikrokrmilniku na analogni vhod. V prvem delu izvedemo PID-regulacijo, v drugem delu pa uporabimo regulacijo s prostorom stanj. V prostoru stanj določimo spremenljivke stanj ter matrične enačbe, ki določajo relacije med vhodom in izhodom sistema. In the diploma thesis a training kit for working with the development circuit board Arduino in MATLAB environment is presented. The main part describes the development of automatic control of ball on 40cm beam. The control works through the MATLAB program, where we enter the program code and the Simulink program, where we draw a block diagram. The MATLAB program is connected via the USB cable to the Arduino Uno development board, which gives the input / output signals to the system. In the first part of the diploma thesis, we concentrate on the closed loop PI-control of a DC motor, which is connected via a shaft to another DC motor, tachometer. In particular, we perform the P and I-controller, and for each of the control we indicate its main strengths and weaknesses. We simulate the oscillation test method where, through the known Nichols-Ziegler method, we determine the reinforcements and introduce an engine model that gives us all its properties. To facilitate calculating the model, we are using Laplace transformation, where the main unknowns are gain K and the time constant T. The control is performed using the step signal and the pole method. A control system also includes a system model that shows the system's response under ideal conditions. In the second part of the diploma thesis, the control of the position of the engine is presented in the case of a power transmission, where the main characteristic is recognition of the position or angle of the beam that is connected to the engine. The power transmission can be controlled at the input with a potentiometer or with a pulse generator, at the output there is a potentiometer that is connected to the gearbox motor and supplies the data to the analog input. The basic properties of the second-order lag system are given, where we use the PID-control, and finally the derivatives of PI-D and I-PD-control. The controller gains are calculated over the poles of the system, which are determined using the second-order system equation. In the third part of the thesis, we use all the knowledge we have gained from the first two chapters. The main task is to control the ball at a specific point on the shaft. The position of the ball on the beams is accompanied by a light sensor that emits and receives an infrared ray, thus determining the position of the ball on the beam. The shaft controls the power transmission, which, like the sensor, gives the microcontroller data to the analog input. In the first part we carry out the PID-control, while in the second part we use the control in the state space area. In the state space we define the state variables and the matrix equations that define the relations between the input and the output of the system.

V diplomski nalogi je predstavljena uporaba sistema Keyence pri pregledovanju objektov. Predstavljen je celoten proces uporabe strojnega vida v avtomatiki. Opisano je, kako izberemo ustrezno kamero in objektiv ter za kakšne predmete in pogoje mora imeti kamera določene elemente in lastnosti. Podana je tudi enačba za pomoč pri izbiri objektiva. Navedeni so problemi, ki se pojavijo pri postavitvi kamere ali osvetlitve in rešitve, ki lahko pridejo v pomoč. Prikazanih je tudi nekaj načinov postavitve osvetlitve. Predstavljene so komponente kamere in objektiva, katere moramo poznati, ko ju nastavljamo. Na koncu so tudi praktični primeri, v katerih je bil uporabljen sistem Keyence. V njih je prikazana postavitev kamere in svetilke ter njihove nastavitve. Opisana je tudi uporaba Keyencovih orodji za pregledovanje objektov in nekaj filtrov za izboljšanje kvalitete slike. The disertation presents the use of the Keyence system for objects inspection. The entire process of using machine vision in automation is presented. How to choose the appropriate camera and lens, for which objects and conditions the camera must have certain characteristics and elements is described. An equation to help with the lens selection is given too. Some problems that can appear while setting up the camera and lighting are mentioned but in addition there are also some useful solutions. Ways to set the illumination are shown too. The components of the camera and the lens which we need to know when they are being set are presented. At the end there are practical examples in which the Keyence system is used. The placement and settings of cameras and lamps are also shown. The use of Keyence's tools for objects inspection and some filters in order to improve the quality of the image we captured with the camera are described too.

V diplomski nalogi je opisana izdelava krmilnega sistema za akvarij, ki omogoča popolno avtomatizacijo le tega. Sistem olajšuje vzdrževanje akvarija in ob enem tudi izboljša življenjske pogoje ribam ter rastlinam. Glavni cilj je bil razviti sistem, ki izpolnjuje vse glavne zahteve za samodejno vodenje akvarija brez človeške pomoči. Vse funkcije posameznih komponent sem povezal z mikrokrmilnikom, ki deluje vzajemno z dajalnikom realnega časa. Tako je zagotovljeno, da se sistem lahko tudi ob prekinitvi napajanja nato vrne v pravilno delovanje. Glavne naloge, ki jih sistem opravlja so: - hranjenje, - regulacija temperature vode, - krmiljenje čistilne naprave, luči ter črpalke za kisik, - prikaz časa in temperature preko LCD zaslona, - prekinitev delovanja filtra v času hranjenja in - zvezno prižiganje luči (simulacija naravnega okolja). V prvem delu naloge je podan opis vseh uporabljenih elementov, ter razlog za njihovo uporabo. V drugem delu je nato opisana njihova implementacija v celotni sistem in način reševanja posameznih problemov pri izgradnji in montaži sistema. Celotni krmilni sistem je trenutno prilagojen za namene uporabe v moji aplikaciji. V primeru bolj kompleksih zahtev je sistem seveda možno prilagoditi in razširiti. These diploma thesis describes the process of making a control system for the aquarium, which enables the complete automation of all functions. The system makes the maintenance of the aquarium easier and also improves the living conditions of both fish and plants. The main goal was to develop a system that fulfils all the main requirements for automatic control of the aquarium without human assistance. All functions of the individual components are linked to a microcontroller that interacts with the real time clock. This ensures that the system can return to the correct operation state even when the power supply is interrupted. The main tasks performed by this system are: - regulation of the water temperature, - feeding, - steering the cleaning system, lights and the Oxygen pump, - displaying the time and water temperature on the LCD screen, - interrupting the power supply for the filter unit between feeding and - continuous signal control of the lights (Simulation of the natural environment). In the first part of these diploma thesis is a description of all the given elements that I have used in my system and it also explains the reasons for their use. The second part then describes their implementation in the entire system. It also makes to show the way of solving individual problems in the construction and installation of the system. The entire control system is currently adapted for use in my specific application. The system can of course be adapted and extended in case of more complex requirements.

V diplomskem delu smo obravnavali delovanje kamere Pixy in pokazali, kako je iz podatkov, ki nam jih posreduje, mogoče izračunati razdaljo med kamero in objektom. Za programiranje robota smo v nadaljevanju v programskem okolju Lego Mindstorms EV3 Home Edition opisali funkcijske bloke, povezane s kamero. Opisali smo tudi funkcijski blok za PID regulator in pokazali, kako izračunamo njegove parametre za reguliranje motorjev Lego. V empiričnem delu smo preizkusili praktično delovanje kamere Pixy. Kamero smo pritrdili na robota, sestavljenega iz kompleta Lego Mindstorms, in simulirali lov plenilec–plen. Program smo naredili blokovno v programskem okolju Lego Mindstorms EV3 Home Edition, pokazali pa smo tudi možnost pisanja programske kode v jeziku Robotc. This thesis discusses the operation of Pixy digital camera, and demonstrates the possibility of calculating the distance between the camera and the object, based on the data the camera displays. The main chapters deal with robot programming where function blocks connected with the camera are described by using Lego Mindstorms EV3 Home Edition software environment. The function block for the PID controller has also been described, along with the calculation of its parameters that regulate the Lego motors. In the empirical part of the thesis, Pixy camera operation has been tested in practice. The camera was attached to a robot built from a Lego Mindstorms kit, and the predator—prey hunt has been simulated. The program was developed in Lego Mindstorms EV3 Home Edition software environment, although the possibility of a code-based solution written in Robotc has also been demonstrated.

Roboti, ki omogočajo gibanje v vse smeri, se imenujejo holonomični roboti. S svojimi štirimi večsmernimi kolesi (Omni wheels) se lahko robot premika ali vrti v vse smeri brez spremembe orientacije. Načrtan je bil tridimenzionalni nosilni okvir mobilnega robota, ki je bil natisnjen na 3D tiskalniku. Nanj so pritrjene komponente robota. Premikanje je bilo realiziramo s koračnimi motorji, ki omogočajo zelo natančno pozicioniranje brez povratne zanke, vse skupaj pa poteka pod nadzorom mikrokrmilnika Arduino Mega. Premik povzročimo z vnosom referenčnih vrednosti, izvrši pa se preko brezžične povezave Bluetooth. Robots that allow movement in all directions, are called holonomic robots. With its four multi–directional wheels (Omni wheels), the robot can move or rotate in any direction without changing its orientation. A supporting frame of the mobile robot was drawn and was then printed on a 3D printer. Robot components are attached to it. The moving was realized with stepper motors, which enable very accurate positioning without any feedback. The motors are controlled with a microcontroller Arduino Mega. The movement is caused with the input of new referent values and is executed via Bluetooth wireless connection.