Search

Magistrska naloga zajema analizo kolesnega ležaja avtomobila. Ležaji so osnovni strojni elementi, ki omogočajo vodenje vrtečih strojnih delov in zagotavljajo najmanjše torno trenje in obrabo tornih površin. Kolesni ležaji, ki smo jih preverjali, spadajo v skupino radialnih kotalnih ležajev s krogličnimi ali stožčastimi kotalnimi telesi. Sedaj se v praksi preverja stanje kolesnih ležajev, tako, da mehanik, ko je vozilo na dvigalu, zavrti kolo in pozorno posluša, kakšne zvoke pri tem kolesni ležaj oddaja. Če presodi, da je ležaj obrabljen, se ga zamenja, v primeru pa, da ni popolnoma prepričan, se opravi še testna vožnja, kjer med vožnjo posluša, če nastajajo zvoki, ki so znak obrabljenega ležaja. Trenutna metoda je precej subjektivna, saj je odvisna od mehanikove sposobnosti poslušanja in njegove presoje, ki temelji na njegovih izkušnjah. Naloga je bila zastavljena tako, da smo del, kjer mehanik zavrti kolo in posluša zvok, zamenjali z meritvijo, ki smo jo opravili z mikrofonom, med testno vožnjo pa smo opravili še dodatno merjenje s pospeškometrom. Meritev s pospeškometrom je predstavljala referenčno metodo. Takšna metoda je tudi v skladu z metodo, ki jo podaja mednarodni standard za obrabo ležajev v industrijskem okolju. Merjenje z mikrofonom poteka na podoben način, kot poteka ocena stanja ležaja sedaj. Kolo še vedno zavrti mehanik, na mesto, kjer pa mehanik postavi uho in posluša zvok, ki ga oddaja ležaj, pa sedaj postavimo mikrofon. Zvok posnamemo in ga nato v programskem okolju DewesoftX dodatno obdelamo, da lahko ocenimo stanje ležaja. Če po obdelavi še vedno nismo popolnoma prepričani o stanju kolesnega ležaja, opravimo še meritev s pospeškometrom. Merjenje s pospeškometrom poteka med vožnjo. V bližino kolesnega ležaja, za katerega sumimo, da je obrabljen, namestimo pospeškometer. Nato med testno vožnjo opravimo meritev. Tako kot pri oceni stanja s pomočjo mikrofona tudi tukaj podatke meritev obdelamo v programskem okolju DewesoftX. Po opravljeni obdelavi lahko natančno analiziramo in ocenimo stanje ležaja. Pri obrabljenem ležaju lahko z uporabo kinematičnega modela ležaja določimo tudi, katera komponenta ležaja je uničena. Z mikrofonom smo opravili meritve na več kot 70 kolesnih ležajih. V večini primerov se je mehanikova ocena stanja kolesnega ležaja ujemala z oceno stanja na podlagi meritev opravljenih z mikrofonom. S pospeškometrom pa smo opravili več kot 15 meritev. V vseh primerih smo lahko jasno videli in ocenili stanje ležaja. S primerjavo ocen, ki smo jih podali na podlagi meritev, opravljenih z mikrofonom, in ocen, ki smo jih podali na podlagi meritev opravljenih s pospeškometrom, smo ugotovili, da ni bilo primera, da bi na podlagi meritve z mikrofonom ocenili stanje ležaja kot sprejemljivo, nato pa bi na podlagi meritve s pospeškometrom ocenili, da je ležaj obrabljen. Smo pa imeli primer, ko smo na podlagi meritve z mikrofonom ocenili kot obrabljen ležaj, nato pa na podlagi meritve s pospeškometrom ocenili ležaj kot sprejemljiv. Metoda nam je omogočila podrobno analizo okvare ležajev. Ugotovili smo, da bi v mehaničnih delavnicah in servisih lahko za oceno stanja kolesnega ležaja uporabljali mikrofon in pospeškometer. Ocenjevanje stanja kolesnega ležaja na podlagi meritve mikrofona in pospeškometra ima prednost, ker za ocenitev ne potrebujemo mehanika. Stanje ležaja lahko oceni nekdo brez izkušenj v mehanični delavnici. Slabost uvedbe mikrofona in pospeškometra je velika začetna investicija in časovna potratnost namestitve merilnega sistema. Kot možnost nadgradnje vidim preventivni merilni sistem integriran serijsko v vozilo. S takšnim merilnim sistemom bi bilo mogoče zaznati že zgodnjo fazo obrabe ležaja, poslati opozorilno obvestilo v nadzorni sistem vozila in se izogniti morebitnim nadaljnjim poškodbam na vozilu. The master's thesis covers the analysis of a car wheel bearing. Bearings are basic machine elements that enable the guidance of rotating machine parts and ensure minimum friction and wear of friction surfaces. The wheel bearings we analyzed were ball bearing and tapered roller bearings. In practice the condition of the wheel bearings is checked by lifting the vehicle, then mechanic turns the wheel and listens carefully to what sounds the wheel bearing makes. If he judges that the bearing is worn, it is replaced, but if he is not completely sure, a test drive is carried out, where he listens while driving for sounds that are a sign of a worn bearing. The current method is quite subjective as it depends on the mechanic's ability to listen and his judgment based on his experience. The task was set in such a way that we replaced the part where the mechanic turns the wheel and listens to the sound with a measurement made with a microphone, and during the test drive we made an additional measurement with an accelerometer. The accelerometer measurement was the reference method. Such a method is also in accordance with the method given by the international standard for mechanical vibration in an industrial environment. The measurement with a microphone is carried out in a similar way to the assessment of the condition of the bearing now. The wheel is still turned by the mechanic, but in the place where mechanic places his ear and listens to the sound emitted by the bearing, we now place a microphone. We record the sound and then further process it in the DewesoftX software so that we can assess the state of the bearing. If, after analysis, we are still not completely sure about the condition of the wheel bearing, we also measure it with an accelerometer. Accelerometer measurement takes place while driving. We place an accelerometer near the wheel bearing that we suspect is worn. Then we take a measurement during the test drive. Data is analyzed in the same software environment as measurement we made with a microphone. After processing, we can accurately analyze and assess the condition of the bearing. In the case of a worn bearing, the kinematic model of the bearing can also be used to determine which component of the bearing is destroyed. We performed measurements on more than 70 wheel bearings with a microphone. In most cases, the mechanic's assessment of the condition of the wheel bearing matched the assessment of the condition based on the measurements made with the microphone. We made more than 15 measurements with the accelerometer. In all cases, we could clearly see and assess the condition of the bearing. By comparing the ratings given based on the measurements made with the microphone and the ratings given based on the measurements made with the accelerometer, we found that there was no case in which the condition of the bearing was assessed as acceptable based on the measurements made with the microphone, and then, based on the measurement with the accelerometer, we would estimate that the bearing is worn. However, we had a case where, based on a measurement with a microphone, we assessed the bearing as worn, and then, based on a measurement with an accelerometer, the bearing was assessed as acceptable. We found that a microphone and an accelerometer could be used in mechanical workshops and repair shops to assess the condition of the wheel bearing. Evaluating the condition of the wheel bearing based on the microphone and accelerometer measurements has the advantage that a mechanic is not needed for the evaluation, as the condition of the bearing can be evaluated by someone without experience in a mechanic's workshop. The disadvantage of introducing a microphone and accelerometer is the large initial investment and the time-consuming installation of the measurement system. As an upgrade option, I see a preventive measurement system integrated as standard in the vehicle. With such a measuring system, it would be possible to detect the early stage of bearing wear, send a warning message to the vehicle's control system and avoid possible further damage to the vehicle.

Že odkar sem začel s treniranjem dvigovanja uteži, sem razmišljal o tem, kako bi lahko izboljšal proces učenja, kako bi natančneje meril napredek ter poglobil razumevanje fizikalnih pojavov ob dvigovanju. Večina že obstoječih sistemov za vrednotenje fizikalnih veličin pri dvigovanju, je bilo cenovno neugodnih, sploh za nekoga, ki ne pozna, kako bi vse te zajete podatke uporabil pri treningu. V okviru projekta diplomske naloge, sem razvil zadovoljivo natančen ter prenosljiv asistenčni sistem, za merjenje fizikalnih veličin pri dvigovanju uteži. Asistenčni sistem v glavnem sestavljajo Boschov orientacijski senzor BNO055 in Adafruitov mikrokrmilnik Feather M0 WiFi, ki na začetku nastavi parametre senzorja, potem pa bere senzorski signal in ga preko WiFi modula pošilja na osebni računalnik. Za verifikacijo asistenčnega sistema, smo uporabili referenčni sistem Qualisys za optični zajem gibanja. Vse podatke senzorskega dela in referenčnega sistema smo sprejeli, obdelali, prikazali in posneli. Shranjene signale smo potem obdelali z različnimi orodji, med glavnimi z orodjem MathCAD. V diplomski nalogi sem opisal tudi fizikalne osnove dvigovanja uteži, kar omogoča lažje razumevanje pravilne tehnike dvigovanja. Za nazornejši prikaz koristnosti asistenčnega sistema smo izvedli analizo celotnega dviga s pomočjo videoposnetka. V zadnjem delu aktivnosti smo preizkusili nekaj praktičnih primerov uporabe sistema, ki nam pomagajo pri načrtovanju in izvedbi vsakodnevnih treningov. Ever since I started training in weightlifting, I have been looking for ways to improve my learning process, how to measure the progress more accurately and deepen my understanding of physical phenomena while weightlifting. Most of the existing systems for physical performance evaluation in weightlifting are too expensive, especially for someone who doesn’t know how to use the captured data during training. As part of the project thesis, I developed an adequately accurate and transferable assistance system for measuring physical perfomance in weightlifting. The assistance system consists mainly of Bosch orientation sensor BNO055 and Adafruit microcontroller Feather M0 WiFi, which initially sets the sensors parameters, then continues with reading the sensors signal and sending it to the PC via the WiFi module. We used the reference Qualisys system for optical motion capture to verify the assistance system. The captured data from the developed sensor system and reference system was received, processed, displayed and recorded. The stored signals were then processed using various tools, mainly with the tool MathCAD. In my thesis, I also described the physical basics of weightlifting, which makes it easier to understand the correct lifting technique. For a more ilustrated demonstration of the usefulness of the assistance system, we had performed an analysis of the overall lift with the help of a video recording. During the last activity, we tested some practical examples of the use of the system, which are helping us with planning and performing our daily weightlifting training.

Ciljno nalaganje kot eno izmed bolj dostopnih aditivnih tehnologij je že bilo uporabljeno za izdelavo piezoupornih statičnih/kvazistatičnih zaznaval deformacij, zračnih tokov, sile. Nedavne raziskave so pokazale, da bi bilo mogoče ciljno nalaganje uporabiti za izdelavo dinamičnih zaznaval. Zaradi vpliva številnih procesnih parametrov, nehomogenosti struktur, izdelanih s ciljnim nalaganjem, anizotropije in nizke ločljivosti tehnologije, je izdelava dinamičnih zaznaval trenutno omejena. V doktorskem delu je obravnavana dinamična piezoupornost s ciljnim nalaganjem izdelanih struktur z vidika kontinuuma. Zasnovana je eksperimentalna metoda za določitev dinamičnih koeficientov piezoupornosti, koeficienti pa so nato eksperimentalno določeni. Na podlagi pridobljenih informacij o piezoupornosti so nato definirani konstrukcijski principi za izdelavo piezoupornega pospeškomera izdelanega s ciljnim nalaganjem. S pomočjo konstrukcijskih principov je izdelan prototipni pospeškomer z majhnimi gabaritnimi merami, nizko križno občutljivostjo in 0-425,Hz uporabnim frekvenčnim področjem. Fused filament fabrication, one of the most accessible additive manufacturing technology, has already been used to manufacture piezoresistive static/quasistatic strain sensors, force sensors and wind sensors. Recent research showed that fused filament fabrication could also be used for the manufacturing of dynamic sensors. However, the manufacturing of the dynamic sensors is currently limited due to the anisotropic mechanical and piezoresistive behaviour, the heterogeneity, the large number of process parameters and poor resolution of the fused filament fabrication technology. In the thesis dynamic piezoresistivity of fused filament fabricated structures is researched. Fused filament fabricated structures are treated as a continuum. An experimental method for the identification of dynamic piezoresistive coefficients is proposed. Piezoresistive coefficients are then determined using the proposed method. Acquiered knowledge about the piezoresistive behaviour of fused filament fabricated structures is adopted to define design principles for piezoresistive accelerometer. Prototype accelerometer based on the defined principles is manufactured. Prototype accelerometer has low cross-axis sensitivity, 0-425,Hz useful frequency range and is low in size.

V zaključni nalogi so predstavljene meritve vibracij s piezoelektričnim merilnikom v prostoru, kjer je vrstični tunelski mikroskop (VTM). Pri vrstični tunelski mikroskopiji mora biti mikroskop mehansko čim bolj ločen od okolice, saj za proučevanje površine uporablja konico, ki se nahaja tik ob površini, razdalja med njima pa mora biti konstantna. Najprej smo preverili delovanje na vseh frekvenčnih območjih, nato pa smo merili vpliv vibracij na mizi vrstičnega tunelskega mikroskopa. Ker so zunanje motnje pomembne za delovanje VTM, smo jih merili pri različnih pogojih. Vibracije iz okolja smo dušili s pnevmatskimi nogami, za katere smo ugotovili, da zmanjšajo nivo določenih vrhov za približno 10 dB. Ko smo prižgali rotacijsko in nato še turbo črpalko, se je intenziteta vibracij pri vklopu posamezne črpalke povečala za približno 10 dB. In this thesis piezoelectric measurements of vibrations of a room with a scanning tunneling microscope (STM) are presented. When performing scanning tunneling microscopy, a sharp conductive tip that must be kept at a constant distance from the sample and the microscope has to be mechanically as separate from the surroundings as much as possible in order to achieve maximum resolution. First, functionality check over a broad frequency range was performed. Next, the influence of forced vibrations to the STM table under various conditions was investigated. In order to dampen externally induced vibrations, pneumatic table-legs were used. Our experiments have shown that the use of pneumatic legs decreases vibrations up to SI{10}{dB}. In addition, an increase of SI{10}{dB} in vibration intensity was detected after using a rotary or turbo pump.

V diplomski nalogi smo obravnavali metode, ki se uporabljajo za merjenje vozne dinamike dvokolesnih vozil. Izdelali smo merilno napravo za merjenje vozne dinamike dvokolesnih vozil (posebej brez motorja). Predstavili smo komponente, iz katerih smo sestavili merilno napravo, in programsko kodo, ki je krmilila napravo. Na koncu smo opravili različne preizkuse za merjenje vseh zadanih fizikalnih količin ter analizirali dobljene rezultate. In the thesis we addressed methods that are used to measure the driving dynamics of two wheeled vehicles. We made a measuring device for measuring the driving dynamics of two-wheeled vehicles (especially the human-powered ones). We presented the components from which we assembled the measuring device and the program code that controlled the device. In the end, we preformed several tests to check how the device works and then analyzed the obtained results.

Na področju števcev za rolke je malo kakovostnih produktov. Druge rešitve, kot so aplikacije na mobilnih napravah, niso dovolj natančne zaradi drugačnega stila vožnje z rolko, kot na primer pri hoji ali pa vožnji z avtomobilom. Radi bi razvili števec, ki bo deloval na načelu Hallovega magnetnega senzorja. Dodatno pa bo števec vseboval natančne meritve nagiba, naklona in smeri neba. Tako bomo lahko na računalniku prikazali grafe in zemljevide na podlagi posnetih podatkov. The area of speedometers for skateboards has little quality products. Other solutions, like applications on mobile devices, simply aren't accurate enough, because of the different driving style with a skateboard than when you walk or drive in a car. We would like to develop a speedometer, which will work using a Hall magnetic sensor. Speedometer will also include accurate readings of roll, pitch and directional angle. That way we can show graphs and maps on the computer based on the recorded data.

Telesno dejavnost lahko ocenimo s subjektivnimi metodami, kot so vprašalniki, ankete, intervjuji itd., ali jo izmerimo z objektivnimi metodami, kot so merilniki srčnega utripa, pospeškometri, pedometri itd. Raziskave s subjektivno metodo so veliko cenejše in lažje izvedljive, kot raziskave z objektivno metodo, vendar je veljavnost podatkov, pridobljenih s subjektivno metodo, vprašljiva. Zato smo se odločili, da preverimo, kakšne so razlike med subjektivno in objektivno metodo ugotavljanja telesne dejavnosti, pri čemer smo se še posebej osredotočili, ali obstajajo tovrstne razlike med otroki glede na delež maščevja v telesu. V raziskavi smo uporabili podatke 100 otrok, starih 11 let, ki so sodelovali v študiji ARTOS. Kot subjektivno metodo ocenjevanja telesne dejavnosti smo uporabili vprašalnik CLASS, ki so ga za otroke izpolnjevali njihovi starši oziroma skrbniki. Kot objektivno metodo pa smo uporabili napravo BodyMedia SenseWear Armband PRO 3, ki vsebuje dvoosni pospeškometer in več drugih senzorjev. Izkazalo se je, da starši močno precenjujejo čas, ki ga njihovi otroci preživijo zmerno do visoko intenzivno telesno dejavni, pri čemer smo mejo za to dejavnost postavili na 4 MET. Pri dečkih so starši ta čas v povprečju precenjevali za 42 %, pri deklicah pa za 40 %. Za ugotavljanje razlik med subjektivno in objektivno metodo glede na delež maščevja v telesu smo vzorec razdelili na kvartile. V prvem kvartilu so bili merjenci z najnižjim deležem maščevja v telesu, v četrtem pa merjenci z najvišjim deležem maščevja v telesu. Starši so v povprečju čas, ki ga njihovi otroci preživijo zmerno do visoko intenzivno telesno dejavni, z vsakim kvartilom bolj precenjevali. V prvem kvartilu za 27 %, v drugem za 34 %, v tretjem za 41 % in v četrtem za 69 %. Naši izsledki so pomembni za kritičen razmislek o preučevanju in spremljanju telesne dejavnosti otrok z vprašalniki, ki temeljijo na poročanju staršev. Še bolj pa o tem, da se izsledki tovrstnih študij uporabljajo za postavljanje in vrednotenje politik na tem področju. Physical activity can be assessed by subjective methods such as questionnaires, surveys, interviews, etc., or by objective methods such as heart rate monitors, accelerometers, pedometers, etc. Subjective methods are much cheaper and more affordable than objective methods, but the validity of their output is questionable. Therefore, we decided to examine the differences between the subjective and objective methods of assessing physical activity, with particular focus on whether there are such differences between children in terms of the body fat mass. The study used data from 100 children aged 11 who participated in the ARTOS study. As subjective method of physical activity assessment, we used the proxy-report CLASS questionnaire that was completed by parents. As objective method we used the BodyMedia SenseWear Armband PRO 3 device, which contains a two-axis accelerometer and several other sensors. The analysis showed that parents greatly overestimated the time their children spend in moderate to vigorous physical activity - setting the limit fort this activity at 4 MET. On the average, parents overestimated this time by 42% in boys and 40% in girls. To determine the differences between the subjective and objective methods in regard of body fat mass, the sample was divided into quartiles. The first quartile included the children with the lowest body fat mass and the fourth quartile the children with the highest body fat mass. With each quartile parents increasingly overestimated the time their children spend in moderate to vigorous physical activity. In the first quartile the parents overestimated that time by 27%, in the second quartile by 34%, in the third by 41% and in the fourth by 69%. Our findings are important for critical reflection on the study and monitoring of children's physical activity with questionnaires based on parental reporting. What is more, the findings of such studies are used to set and evaluate policies in this area.

Globalno sprejeta priporočila za telesno dejavnost otrok in mladine priporočajo minimalno 60 minut zmerno do visoko intenzivne telesne dejavnosti nad dan. Po različnih podatkih je v Evropi manj kot 50 % otrok telesno dejavnih v skladu s priporočili, po podatkih Svetovne zdravstvene organizacije pa globalno več kot 80 % otrok naj ne bi dosegalo priporočil. Raziskovalci iz različnih delov sveta ugotavljajo, da raven telesne dejavnosti upada z odraščanjem in da so dekleta telesno manj dejavna kakor fantje. Namen našega raziskovanja je ugotoviti, kako se spreminja telesna dejavnost slovenskih otrok v obdobju 11 – 14 let in ugotoviti, ali obstajajo razlike v telesni dejavnosti fantov in deklet. Spremljali smo telesno dejavnost otrok v starostnem obdobju 11 – 14 let na vzorcu ljubljanskih osnovnošolcev. V prvem letu meritev je sodelovalo 114 preizkušancev, v drugem 90, tretjem 88 in četrtem 50. Devet preizkušancev je sodelovalo v vseh štirih časovnih točkah. Za merjenje telesne dejavnosti smo uporabili večsenzorne merilnike SenseWear Pro3, ki so podatke beležili v minutnih povprečjih. Zmerno intenzivnost telesne dejavnosti smo opredelili s 4 MET, visoko intenzivnost pa s 7 MET. Za statistično obdelavo smo izbrali metodo analize variance za odvisne vzorce. Razlike v celotni telesni dejavnosti so se izkazale za statistično značilne med starostnima skupinama 11 in 14 let (p = 0,006 - 2,34 h) in starostnima skupinama 12 ter 14 let (p = 0,016 -0,62 h). Razlike v povprečni intenzivnosti (MET) celotne telesne dejavnosti pri različnih starostih so bile statistično neznačilne F(3, 21) = 1,576, p = 0,225. Prav tako razlike v količini zmerno do visoko intenzivne telesne dejavnosti F(3, 21) = 1,232, p = 0,323. Analiza sedečega časa je pokazala, da ni statistično značilnih razlik v tej dejavnosti pri različnih starostih F(3, 21) = 1,207, p = 0,323. Prav tako ni razlik med spoloma v količini sedečih dejavnosti F(1, 7) = 0,318, p = 0,590. Zaradi okrnjenega vzorca ugotovitev ne moremo posploševati. Vendar pa bodo lahko uporabne tako za spodbude telesne dejavnosti v šolskem in domačem okolju kot tudi za primerjavo rezultatov prihodnjih tovrstnih raziskav v Sloveniji. Globally accepted recommendations for physical activity of children and youth state that children should engage in moderate to vigorous physical activity at least for 60 minutes daily. Referring to different reports in Europe less than 50 % of children and youth is sufficiently active. According to World Health Organization less than 80 % of children achieve the recommendations. There are generally accepted ideas among researchers that level of physical activity is decreasing with age and that boys are more active than girls. The aim of our study is to investigate longitudinal changes in physical activity in children between the ages of 11 and 14 and observe gender differences. Total 114 children took part of our research at the age of 11. There were 90, 88 and 50 children at age 12, 13 and 14 respectively. There were nine participants who were measured at all of the ages. We used SenseWear Pro3, multi-sensor accelerometer for measuring physical activity. It collected data at 60 s epoch. We defined moderate activity with 4 MET and vigorous with 7 MET. For statistical analysis, we used two- way mixed ANOVA as a test for repeated measures. We observed significant differences in total physical activity (TPA) between age groups of 11 and 14 (p = 0,006 -2,34 h) and between age groups of 12 and 14 (p = 0,016 - 0,62 h). There were no significant differences observed in average intensity of TPA between different age groups F(3, 21) = 1,576, p = 0,225. The same goes for differences in moderate to vigorous activity F(3, 21) = 1,232, p = 0,323 and sedentary activity F(3, 21) = 1,207, p = 0,323. Also, there were no significant gender differences in sedentary activity F(1, 7) = 0,318, p = 0,590. Due to considerable drop out of measured children and small final sample avaliable for longitudinal analysis, we can not generalize our findings. Nevertheless findings can be useful for school or home-based physical activity initiatives. Additionally, we see our study as good basis for further research of physical activity in Slovenian children.

Pomembno področje razvoja v robotiki so novi pristopi k učenju robotov, ki bi njihovo uporabo približali ljudem brez strokovnega znanja iz robotike. V Projektnem delu je predstavljen princip vodenja robotske roke na podlagi kretenj rok uporabnika. Za zajem kretenj rok smo izbrali senzorski zapestnici Myo Armband, ki nam nudita meritve pospeška, kotne hitrosti in smeri magnetnega polja, iz katerih algoritem senzorske fuzije izračuna orientacijo v prostoru. V projektu smo predstavili možno teoretično razlago algoritma združevanja meritev za pridobivanje orientacije. Na podlagi informacije o orientaciji obeh zapestnic smo razvili kinematični model človeške roke in ga simulirali v simulacijskem programu V-REP. Algoritem vodenja s pomočjo kretenj in kinematičnega modela človeške roke smo preizkušali na simulaciji robotske roke Kinova Mico. Rezultate simulacije vodenja robotske roke s kretnjami smo predstavili v obliki videa. In this project we present a method for intuitive control of a collaborative robotic arm using hand and arm gestures performed by the user. The proposed system uses two Myo Armband units to capture the motion of the arm and a personal computer to process the measurements and simulate the motion of the Kinova Mico manipulator in the V-REP simulation environment. Arm motion is captured by recording orientation of both Myo Armband units. A mathematical model of the human arm is built in order to compute the target position for the robot to reach. The resulting simulation is presented in video form as well.

V zaključnem delu sta opisana delovanje in razvoj varnostnega vezja za osebne predmete. V delu smo se dotaknili problematike varovanja osebnih predmetov in razvili elektronsko napravo za varovanje le teh. Opisane so nekatere funkcije mikrokrmilnika, SPI in UART komunikacija, ter bluetooth modul HC-05 in inercijski MEMS modul LSM6DS0. Napravo smo razvili z namenom varovanja osebnih predmetov in preprečitev kraj. Pri tem smo upoštevali kriterije za čim preprostejšo, zanesljivo in robustno napravo. In the final work we described the use and development of safety circuit for personal items. We touched the problem of protecting the personal belongings and developed an electronic device to protect them. Some of the functions of microcontroller, SPI and UART communication and Bluetooth modul HC-05 and inertial MEMS module LSM6DS0 are described in more detail. We developed the device, with the intention of protection of the personal belongings and to prevent alienation. For this, we complied with the criteria for a simpler and robust device.

Namen diplomske naloge je predstaviti principe delovanja, natančnost in uporabnost inercialnih navigacijskih sistemov. Predstavljeni so postopki za izboljšanje kakovosti koordinat, posebej v kombinaciji z GNSS. Na praktičnem primeru sta preizkušeni in predstavljeni uporabnosti in natančnosti določitve položaja z uporabo nizkocenovnega inercialnega senzorja v pametnem telefonu. Na osnovi analize smo potrdili uporabnost nizkocenovnih senzorjev za potrebe orientacije v prostoru in uporabo v panogah geodezije, kjer je potrebna manjša natančnost določitve položaja. Purpose of this graduation thesis is presentation of principles of operation, accuracy and usefulness of inertial navigation systems. Presented are also procedures for improving the quality of coordinates, specifically in connection with GNSS. With the use of low-cost inertial sensors in a smart phone, usability and precision are tested and presented on practical cases. We showed that low-cost sensors are suitable for providing orientation data but can not be used in surveying and related disciplines.

Nevrologi trenutno merijo in štejejo tremorje pri obolelih osebah tako, da opazujejo osebo v različnih pozicijah. Opazujejo predvsem roke. Nimajo na voljo naprave, s katero bi lahko izmerili tresenje rok obolelih oseb, ki bi jim te meritve prikazala in naredila preprosto obdelavo podatkov za lažjo analizo. Narejena je bila mobilna aplikacija, ki z uporabo žiroskopa in senzorja pospeška izmeri, hrani in vizualizira podatke tremorjev obolelih oseb. Aplikacija omogoča nevrologom pridobiti natančne podatke o stanju obolele osebe, prav tako omogoča primerjavo s starimi meritvami in tako nudi vpogled v napredovanje bolezni. Aplikacija je preprosta za uporabo in dostopna osebam s slabim vidom in slabim nadzorom prstov. Neurologists currently measure and count tremors of patients, by observing them in different positions. They usually observe the patients arms. They do not have a device which would allow them to measure tremors of patients, would visualize tremors and would process the measured data for easier analysis. The developed mobile application solves these problems. It uses a gyroscope and an accelerometer to measure tremors. The measured data is saved and visualised. The application allows neurologists to get accurate data about the state of their patients, compare measured data with previous measurements and so it also allows to track the progress of a patients disease. The application is simple to use and accessible to people with bad vision and bad finger control.

V diplomskem delu smo analizirali in primerjali več ogrodij za navigacijo v zaprtem prostoru in z uporabo izbranega ogrodja razvili prototip aplikacije za iskanje izdelkov v zaprtem prostoru. Preizkusili smo ga v prostorih Fakultete za računalništvo in informatiko na različnih mobilnih napravah z operacijskim sistemom Android. V delu smo analizirali razvojna ogrodja MapsIndoors SDK, IndoorAtlas SDK, Indoo.rs SDK in jih med seboj primerjali. Pri izdelavi prototipa je bilo uporabljeno razvojno ogrodje IndoorAtlas SDK, ki vključuje tudi spletni vmesnik za nadzor aplikacij in lokacij, ter mobilna aplikacija MapCreator 2 za ustvarjanje podatkov o prostoru na podlagi geomagnetnih zapisov, akustičnih signalov in signalov oddajnikov WiFi ter Bluetooth. Poleg tega smo razvili tudi lastno programsko logiko za izračun kota med smerjo hoje uporabnika in njegove interesne točke z uporabo pospeškomera in giroskopa. This work presents the analysis and comparison of different development kits that provide solutions for navigation indoors. The work also provides a more in-depth overview of the development kit chosen for the development of a prototype application for locating products indoors. The prototype was tested on the premises of Faculty of Computer and Information science, on several different mobile devices with Android mobile operating system. We analysed and compared MapsIndoors SDK, IndoorAtlas SDK and indoo.rs SDK development kits. The development kit used for the prototype was IndoorAtlas SDK, which includes a web interface, to control applications and locations, and a mobile application MapCreator 2, for creating indoor data based on geomagnetic fields, acoustic signals, Bluetooth signals and WiFi signals. We have also developed our own program logic that calculates the angle between the users path and his point of interest. This was achieved with the use of accelerometer and gyroscope.

V diplomski nalogi smo predstavili principe delovanja, natančnost in uporabnost inercialnih navigacijskih sistemov. Na praktičnem primeru smo preizkusili uporabnost in natančnost določitve položaja z nizkocenovnimi inercialnimi senzorji. Na osnovi ugotovitev smo določili področja morebitne uporabe te tehnologije v geodeziji in sorodnih vedah. In the thesis the principles of operation, accuracy and usefulness of inertial navigation systems were presented. In practical cases, we tested the usability and precision of positioning with low-cost inertial sensors. On the basis of the findings, we determined the areas of potential use of this technology in geodesy and related disciplines.

"Wearable Health Technology" je eno od področij, ki za delovanje izkorišča različne senzorje in se zaradi napredka le-teh vse bolj uveljavlja. V diplomski nalogi je predstavljena mobilna rešitev za sisteme Android, ki posega na omenjeno področje in za delovanje uspešno vključuje uporabne senzorske tehnologije. Cilj obravnavane tematike je izvedba avtomatizirane mobilne rešitve na pametnem telefonu za obvešcanje ob srčnih zastojih in padcih oseb v primerih, ko si te ne morejo same pomagati. Za razvoj omenjene aplikacije so bile uporabljene napredne tehnologije, kot so senzorji srčnega utripa BLE (angl. Bluetooth Low Energy), pospeškometri in sistem GPS (angl. Global Positioning System), ter nekaj sorodnih tehnologij, ki so vključene kot predmet raziskave. V osrednjem delu sta predstavljena implementacija in opis načrtovanih algoritmov za ugotavljanje padca in načinov gibanja, ki skrbijo za pravilno integracijo in komunikacijo uporabljenih tehnologij. Predstavitve ključnih aktivnosti, strukturne zasnove, delovanje mobilne rešitve in funkcionalnosti so podane v zadnjem delu, ki poleg omenjenega vsebuje še rezultate testiranj in analizo izboljšav na posameznih modulih. Wearable Health Technology is one of the fields that uses different sensors for operation. Due to development of sensors technology this field is more and more popular. In the thesis is presented a mobile solution for Android system, which takes part of this field and uses different sensors technologies. The main goal is to achieve an automated solution for mobile device that can inform others in case of falling or heart attack of a person. In the development of this solution were used technologies like heart rate BLE (Bluetooth Low Energy) sensors, accelerometers, GPS (Global Positioning System) and some others that are part of research. Focus in the second part is put on individual implementation of technologies and used algorithms for detect falling and different movements. Presentation of main activities, structural composition, functionality and app working are described in third part, which additionally includes the results of testing and future improvements of individual modules.

Namen študije je bil ugotoviti, koliko izvenšolska organizirana vadba atletike pripomore k dvigu količine in intenzivnosti gibalne/športne aktivnosti otrok med samo uro atletike in izven nje. Z merilnikom pospeška smo izmerili tedensko gibalno/športno aktivnost 22 otrok (11 dečkov, starih 6,6 ± 1,2 leta), vključujoč dve uri organizirane vadbe atletike. Med uro atletike je skupna gibalna/športna aktivnost 1787 ± 620 sunkov/minuto (brez atletike 615,5 ± 116,6), prevladuje gibalna neaktivnost (29 ± 11,7 minute) ter srednja in visoka intenzivna gibalna/športna aktivnost (19,0 ± 8,9 minute). Dve uri atletike na teden povečata povprečno dnevno srednjo in visoko intenzivno gibalno/športno aktivnost (16 %, P < 0,001) ter zmanjšata čas gibalne neaktivnosti (–10 %, P < 0,001). Atletika sicer prispeva k doseganju priporočil na področju gibanja, toda stremeti bi bilo treba k še večji intenzivnosti izvedbe ur atletike. The purpose of our study was to investigate if organised physical activity of athletics increase the overal amount and intensity of physical activity during and after an athletics session. Using an accelereometer we measured weekly physical activity in 22 children (11 boys, aged 6.6 ± 1.2 years), including two hours of organised athletics. During athletics physical activity reached 1787 ± 620 (without athletics 615,5 ± 116,6) counts/minute, with most prevalent physical inactivity (29 ± 11,7 minutes) and moderate to vigorous physical activity (19,0 ± 8,9 minutes). Two weekly hours of athletics increased average daily time of moderate to vigorous physical activity (16 % P < 0,001) and decreased the time of physical inactivity (–10 % P < 0,001). Athletics contrbutes to achieving physical activity norms however, more focus should be placed on increasing the intensity of physical activity during organised hours of athletics.

Merjenje vibracij ter njihova obdelava in prikaz je pogost način iskanja napak ter nenormalnega delovanja naprav v industriji. V ta namen smo v zaključni nalogi predstavili osnove vibracij ter opisali način obdelave in grafičnega prikaza izmerkov vibracij. Razložili smo pomen digitalizacije in pretvorbe signala. Razvili smo cenen merilni sistem za meritev vibracij ter obdelavo in prikaz izmerkov v realnem času. Sistem je zasnovan na Arduino platformi, obdelava podatkov pa poteka na računalniku v programskem paketu MATLAB. Merilni sistem smo testirali v kontroliranih pogojih v laboratoriju ter na realnem primeru. Dobljeni rezultati so presenetljivo natančni glede na zmogljivosti postavljenega merilnega sistema. Na koncu smo izvedli raziskavo dosegljive frekvence vzorčenja ter razložili vpliv neenakomernega intervala vzorčenja. Measuring vibration is a common way of error detection or abnormal device behaviour in industry systems. For the purposes of this assignment we presented the basics of vibrations and described the method of vibration data processing and their graphic display on the computer. We presented the importance of digitalization and signal conversion. A cheap measuring system was developed for measuring, proccessing and displaying vibration data in real time. The data acquisition was performed with Arduino microcontroller. Proccessing and display of the vibration data was performed on a personal computer with MATLAB software package. The measuring system was initially tested in a controlled environment in the laboratory. After successful tests in the laboratory the measuring system was used on a working machine. The results obtained were suprisingly accurate with regard to the capabilities of the measuring system components. Finally, we identified the maximum sampling frequency and analyzed the impact of the uneven sampling interval.

V diplomskem delu je predstavljena izvedba samo-ravnotežnega robota. Cilj diplomskega dela je bila izdelava robota, ki bo lovil pokončen položaj z možnostjo upravljanja preko brezžične komunikacije. Robot je opremljen s krmilnim sklopom, senzorji za ravnotežje in brezžičnim modulom. V delu so podrobno opisani posamezni sklopi, vse uporabljene komponente, delovanje programske opreme in izvedba prototipa. Prototipna zasnova omogoča nadgradnjo sistema z novimi funkcionalnostmi. V sklopu podamo predloge za izboljšave samo-ravnotežnega robota. This thesis focuses on execution of a self-balancing robot. The major objective of this study was to create a robot, which would be able to maintain an upright position by itself with the possibility of using wireless communication to manage it. The robot is equipped with control assembly, balance sensors and a wireless module. This study describes separate assemblies, all components, the software operation and the prototype implementation in details. The prototype design allows the system to be upgraded with new functionalities. Suggestions for the improvement of the above-mentioned robot are given in the conclusion.

Za komunikacijo z računalnikom v vozilu se pojavlja vedno več aplikacij in naprav, ki zahtevajo branje podatkov preko standardiziranega priključka. Kadar je to edina dovoljena ali mogoča poveza za zbiranje telemetričnih podatkov, naletimo na problem, ko zajeti podatki niso v želeni resoluciji ali razpoložljivi, kot zahteva njihova nadaljnja uporaba. V diplomski nalogi smo predlagali, kako priti do večje ločljivosti podatka o hitrosti vozila, kot je ločljivost, ki ustreza standardu ISO 15031-5 ali SAE J1939-71. Razvili smo merilni modul s pospeškometrom, žiroskopom in magnetometrom. S Kalmanovim filtrom smo združili senzorske podatke in jih povezali s podatki, zajetimi iz avtomobilskega računalnika. Tako smo nekoliko izboljšali ločljivost. Rezultate smo statistično potrdili z 30 krajšimi in 31 daljšimi vožnjami, na katerih smo ocenjene hitrosti in prevožene poti primerjali z referenčnimi podatki. More and more applications and devices require communication with vehicle to retrieve data over standardized connector. When this is the only permissible or possible connection to collect telemetric data, we encounter a problem when data are not included in the desired resolution or available as required by their continued use. In this thesis, we propose how to get the higher resolution data on the vehicle velocity, such as resolution corresponding to ISO 15031-5 or SAE J1939-71. We developed a measuring module with accelerometer, gyroscope and magnetometer. With Kalman filter we fused sensor data and combined them with the data collected from the car computer, which slightly improved the accuracy of assessed velocity. Results were statistically confirmed by 30 short and 31 long trips, by comparing estimated velocity and driving distances with referential ones.

Diplomsko delo opisuje testiranje različnih umeritvenih postopkov inercijskih sistemov, s katerimi je možno odpraviti določene pomanjkljivosti, napake in posledično izboljšati meritev. Izvor in velikost napak je v posameznih senzorjih v večini primerov neznana, zato sem z metodo Allan varianca najprej določil karakteristiko naključnih procesov (beli šum, lezenje …). Napake senzorjev se v procesu umerjanja modelirajo z različnimi matematičnimi modeli, zato je tudi točnost in zanesljivost umeritvenih postopkov različna. V diplomskem delu sem z MIMS (SU6DOF) testiral natančnost šestih različnih umeritvenih metod pri znanih pogojih. Izmed vseh testiranih postopkov je bila najboljša in hkrati tudi najpreprostejša metoda Stančin, ki je imela za pospeškometer in žiroskop najmanjšo napako. Pri primerjavi različnih umeritvenih metod za magnetometer pa se je izkazalo, da različne metode dosegajo podobne rezultate. Izrazita je bila le razlika v velikosti napake med statično in dinamično umeritvijo magnetometra. The thesis describes testing of various calibration procedures of the magnetic inertial measurement system (MIMS), with which certain deficiencies, and errors can be eliminated, and consequently measurement can be improved. Errors of individual sensors are in most of the cases unknown, therefore, I first determined the characteristics of random processes (white noise, drift etc.) using the Allan variance method. Every sensor has its own errors that are modelled in the process of calibration with an appropriate numerical model. Therefore, the accuracy and reliability of calibration procedures is different. The performance of six different calibration procedures in known conditions was empirically evaluated using a commercially available MIMS (SU6DOF). A series of tests for accelerometer and gyroscopes have shown that the best method in terms of accuracy, simplicity, and errors was presented by Stančin. A comparison of various calibration methods for a magnetometer however, showed no difference in total average error. Only a difference in the size of the error among the static and dynamic calibration of magnetometer was distinct.

V diplomskem delu je opisana zasnova in izvedba vgrajenega sistema za beleženje linearnih pospeškov v treh prostostnih stopnjah. Izdelani vgrajeni sistem je mobilna naprava, ki odvisno od izbranega načina delovanja omogoča samodejno shranjevanje izmerjenih podatkov o pospešku na spominsko kartico. Podprta sta dva načina delovanja. Prvi način je neprekinjeno beleženje pospeškov z nastavljivo frekvenco vzorčenja, ki ga sprožimo ročno s pritiskom na tipko. Drugi način delovanja je prekinitveni način, kjer se beleži podatke o času, ko pospešek preseže v naprej določen prag. Obdelovanje in urejanje meritev je možno s klasičnim programskim orodjem v okolju Windows, kot je recimo MS Excel. The thesis describes the design and the construction of an embedded system for a linear acceleration logging in three degrees of freedom. The built embedded system is a mobile device which, regarding the selected operating mode, enables an automatic measured data storage. There are two different operating modes supported. The first one is a continuous acceleration logging with an adjustable sampling rate and could be turned on by pressing manually on a button. The second one is an interrupt mode where the time date is logged when the acceleration exceeds a predetermined threshold. It is also possible to process and edit measurements with a regular Windows environment software tool, like the MS Excel.

Sodobni pametni telefoni imajo vgrajene senzorje gibanja, ki se uporabljajo v številnih aplikacijah. Prouči lastnosti in delovanje senzorjev gibanja v pametnih telefonih. Načrtajte in izvedite sistem za zajem in obdelavo senzorskih signalov, ki bo omogočal analizo človeške hoje. Sistem naj prikaže parametre hoje, ki bodo omogočali izvedbo biološke povratne vezave. Delovanje sistema preizkusite pod različnimi testnimi scenariji v realnih okoliščinah. Up-to date smartphones have embedded inertial sensors that are used in numerous applications. Study the properties and operation of smartphone inertial sensors. Design and implement a system for acquisition and processing of sensor signals that would allow gait analysis. The system should present gait parameters that would allow implementation of biofeedback. Test the operation of the system under various scenarios and in realistic conditions.

Vsakodnevna aktivna pot v šolo oziroma iz šole je lahko pomemben vir gibalne/športne aktivnosti otrok in mladostnikov, nujno potrebne za njihov celostni razvoj in ohranjanje zdravja. Vendar pa je odnos med aktivno potjo v šolo oziroma iz šole ter gibalno/športno aktivnostjo otrok in mladostnikov še vedno nejasen. Namen članka je ugotoviti, ali aktivna pot v šolo oziroma iz šole vpliva na dvig gibalne/športne aktivnosti otrok. S sistematičnim pregledom elektronskih baz podatkov, kot so MEDLINE, Google Scholar, ScienceDirect in Springerlink, smo v maju 2012 poiskali potencialno primerne članke in jih 18 vključili v pregled. Študije zajemajo otroke in mladostnike, stare od 5 do 16 let, in kvantitativno analizirajo odnos med aktivno potjo in gibalno/športno aktivnostjo otrok. Kar 14 študij poroča, da so otroci in mladostniki, ki so na poti v šolo oziroma iz šole aktivni, bolj gibalno/športno aktivni kot tisti, ki se v šolo oziroma iz šole vozijo. Osem študij ugotavlja tudi pozitivne povezave med aktivno potjo in zmerno do visoko intenzivno gibalno/športno aktivnostjo otrok. Za podrobnejši vpogled na povezavo med aktivno potjo in gibalno/športno aktivnostjo otrok bi bilo potrebnih več longitudinalnih študij. Določitev enotnih kriterijev za spremljanje aktivne poti in gibalne/športne aktivnosti otrok ter spremljanje dejavnikov, ki vplivajo na odločitev za aktivno pot, bi pripomogla k medsebojni primerljivosti rezultatov posameznih študij in omogočila oblikovanje učinkovitih intervencijskih programov za spodbujanje aktivne poti med otroki in mladostniki. Daily active journey (AJ) to/from school can be an important source of physical activity (PA) of children and adolescents, necessary for their overall development and health preservation. Yet, the relation between AJ to/from school and PA of children and adolescents is still not clear. The purpose of this article is to find out whether the AJ to/from school influences the rise of PA. Systematic examination of electronic databases such as MEDLINE, Google Scholar, Science Direct and Springerlink, in May 2012 helped us find relevant articles of which 18 met the criteria for inclusion in our review. Studies include children and adolescents aged 5 to 16 years and quantitatively analyse the relationship between AJ and PA. Fourteen out of eighteen studies report that children and adolescents who practice AJ accumulate more PA compared to those who are driven to/from school. At the same time 8 studies report a positive correlation between AJ to/from school and moderate to vigorous physical activity (MVPA). More longitudinal studies would be necessary to get a deeper insight into the link between AJ and PA. Establishing uniformed criteria for monitoring AJ and PA and monitoring factors that influence the decision for the AJ would contribute to comparability of the results of individual studies and would at the same time make the development of effective intervention programmes for encouraging AJ among children and adolescents possible.

V diplomskem delu so predstavljeni senzorji pametnih telefonov in njihova funkcionalnost. Še posebej smo se osredotočili na senzor pospeševanja, ki je sestavni del vsakega pametnega telefona Z implementacijo tega senzorja izmerimo podatke gibanja telefona, oziroma njegovega pospeška. Če senzor pospeševanja izmeri večja nihanja, skušamo ugotoviti za katero vrsto gibanja gre. V našem primeru poskušamo prepoznati padec. Prepoznavanje vrste gibanja zahteva posebno študijo, ki vključuje delo s klasifikacijskimi metodami. Zabeležena gibanja smo v okolju Weka z različnimi nastavitvami klasificirali in primerjali med seboj ter določili najprimernejše klasifikacijske metode za naš primer – prepoznavanje padca. Rezultati so pokazali razmeroma uspešno klasifikacijo primerkov. In the thesis are presented sensors of smart phones and their functionality. In particular, we focused on the accelerometer sensor, which are an integral part of every smart phone. Phone acceleration was measured with implementation of sensor accelerometer. If accelerometer determine greater variation, we try to determine what type of movement was involved in our case, we tried to identify fall. Identifying movements requires a special study involving work with classifier methods. Recorded movements were classified and compared with each other in Weka environment and determined the most appropriate classification method for our example – identifying the fall. The results showed us a relatively successful classification of specimens.

V diplomskem delu smo opisali operacijski sistem Android in sloje platforme Android. V naslednji fazi smo predstavili strojno opremo, ki smo jo uporabili za preizkus aplikacije. Nato je sledil opis razvoja mobilne aplikacije Android, katere namen je zajemanje in označevanje podatkov o hoji uporabnika. Predstavljeno je tudi delovanje senzorja žiroskopa in pospeškometra, ki sta bila ključna za delovanje aplikacije, ter senzorja GPS, ki je ključen za prikaz lokacije uporabnika. Aplikacija je bila testirana in uspešno deluje na Android različici 4.3 (telefon Samsung Galaxy SIII) in najnovejši Android različici 5.1.1 (telefon LG Nexus 5). In this thesis, we have described the Android operating system and the layers of Android platform. In the next phase, we have presented the hardware we have used for testing the application. Then the description of the development of Android mobile application aimed at capturing and tagging user’s walking data has been given. The operation of the gyro sensor and accelerometer has been described, which has been crucial for the functioning of the application and the GPS sensor, which is essential to detect the user's location. The application has been tested and has been successfully working on Android version 4.3 (phone Samsung Galaxy SIII) and the latest Android version 5.1.1 (phone LG Nexus 5).

V diplomskem delu smo implementirali mobilno aplikacijo za detekcijo in analizo patološkega in fiziološkega tremorja med vsakdanjo uporabo mobilnega telefona. Aplikacija je omejena na mobilne naprave z operacijskim sistemom android verzije 4.0 in vgrajenim pospeškometrom. Uporabniku omogočimo spremljanje pridobljenih podatkov iz pospeškometra v obliki grafa, ki se v tekom zajema podatkov samodejno posodablja. Aplikacija omogoča pregled obdelanih podatkov in prikaz frekvenčnega spektra pridobljenih signalov iz katerega ocenjujemo tip tremorja ter njegovo amplitudo in frekvenco. In this diploma, we implemented a mobile application for detection and analysis of pathological tremor during everyday use of a mobile phone. The application is limited to mobile devices with Android operating system version 4.0 (Ice Cream Sandwich) and a built-in accelerometer. The application allows the user to monitor the data obtained from the accelerometer in form of a graph, which is updated in real time. It provides an overview of processed data and displays the frequency spectrum of the obtained signals from which we estimate the amplitude and frequency of the tremors, as well as type of detected tremor.

V diplomskem delu smo s pomočjo ţiroskopa in pospeškometra, ki ju vsebuje vsaka mobilna naprava, izdelali mobilno aplikacijo za sprotno rotiranje navideznega 3D-objekta na zaslonu osebnega računalnika. Aplikacija je omejena na mobilne naprave z operacijskim sistemom Android ter omogoča hitro in intuitivno vzpostavitev komunikacije med njimi in streţniškim okoljem, ki skrbi za rotacijo izbranega 3D-objekta. Slednjega je mogoče poljubno spreminjati in s tem aplikacijo enostavno prilagoditi različnim poslovnim modelom, ki zahtevajo upravljanje in pregledovanje 3D-objektov (npr. ogled računalniških modelov, turističnih znamenitosti, oblek, avtomobilov, pričesk itd.). We created a mobile application for a real-time rotation of 3D virtual objects on the screen of a personal computer. This application uses gyroscope and accelerometer, embedded in every mobile device and is limited to the mobile devices with Android operating system. It allows a fast and intuitive establishment of communication between mobile device and the server’s environment, which rotates the selected 3D object on the screen of personal computer. Virtually any 3D object can be manipulated in this way. Thereby, the application can easily be adapted to the different business models, which require manipulation of the 3D virtual objects, such as review of computer models, sightseeing, review of clothes, cars, hair styles and so on.

V tem diplomskem delu je predstavljen primer sistema za pozicioniranje v zaprtih prostorih z uporabo NFC tehnologije. Znano je, da v zaprtih prostorih signal satelitov navigacijskim napravam ni viden, zato pozicioniranje, kot je npr. preko GPSa, ni mogoče. Delo obsega razvoj več aplikacij, vključno z Android aplikacijo na mobilnem telefonu. Uporabnik s pomočjo aplikacije na mobilnem telefonu prebere enolični identifikator iz NFC nalepke in iz strežnika pridobi trenutni položaj in seznam možnih ciljev. Nato lahko izbere cilj in aplikacija ga usmerja po najkrajši poti do cilja preko vmesnih NFC točk, ki jih na tej poti uporabnik lahko prebere. Na zaslonu telefona se prikazujejo informacije, kot so smer, oddaljenost do naslednje točke, celotna oddaljenost do cilja in trenutna hitrost. V delu so opisani matematični in fizikalni problemi in njihove rešitve. V zadnjem poglavju je bila izdelana statistična analiza, ki pripomore pri ugotavljanju anomalij vpliva močnega magnetnega polja. In this diploma thesis an example of the indoor positioning system using NFC technology is presented. The indoor signal of satellites, such as GPS, is not visible to the navigational devices, so the positioning there is not possible. The thesis covers the development of several applications including Android application for smartphones. The smartphone user reads ID from NFC label and gets his current position and the list of all possible destinations from the server. User can then select one destination and can receive the shortest path from the current position to the destination from the server. Mobile application navigates him between NFC labels on this shortest path until the destination is reached. The thesis also covers some mathematical and physical problems connected with the topic and possible solutions. In the last chapter the statistical analysis that can help find strong magnetic field anomalies is presented.

Na tržišču je vse več sistemov za oddaljeno nadzorovanje starejših ljudi. Večina teh sistemov je dragih in ljudem nedostopnih, saj uporabljajo zelo napredno opremo za nadzor. Osnovne funkcije nadzora bi lahko opravljali s cenovno dostopnimi senzorji, kot so pospeškomeri. Namen diplomskega dela je bila izdelava sistema za nadzor sprememb v hoji človeka. Pri delu smo uporabljali pospeške leve in desne noge v vseh treh koordinatnih smereh. Ugotovili smo, da je najboljši pristop k rešitvi problema delitev naše množice podatkov na fiksne časovne intervale. Brez te predpostavke bi težko določili pravilne atribute potrebne za strojno učenje. Da bi bolje razumeli podatke enega časovnega intervala, smo hojo razdelili na njene primitivne elemente - korake. S pomočjo splošne karakteristike hoje smo zasnovali dva algoritma. Skupaj nam razdelita podatke na posamezne korake. Izbrali smo osem različnih atributov značilnih za hojo. Nato smo poskusili ustvariti univerzalni model strojnega učenja, kateri bi prepoznal razlike med normalno hojo in šepanjem. Testiranje smo opravili na različnih časovnih intervalih. Izkazalo se je, da je podatke bolje razrezati na daljše množice. Tako dobimo bolj zanesljive informacije o hoji človeka na obravnavanem intervalu. Ugotovili smo, da je SVM najboljši model strojnega učenja za predstavljen klasifikacijski problem. On the market, there are more and more di®erent systems for monitoring elderly people. Most of the systems are expensive and inaccessible since they use very sophisticated monitoring equipment. Basic functions of monitoring could be done by reasonably priced sensors such as accelerometers. The object of this work was to develop the system for monitoring changes in human gait. We used data from accelerometers attached to the left and right ankle. Accelerations are represented in three-dimensional coordinate system. We realized the best solution to our problem was to divide data set into ¯xed time intervals. Without this assumption it would have been di±cult to deter- mine the right attributes for machine learning. To better understand the data in one interval, we divided walking into its primitive elements - steps. By using some gait characteristics we were able to conceive two algorithms. Together they enabled us to divide data into single steps. Eight di®erent attributes distinctive for walking were chosen. The next thing was an attempt to create universal training data set that could recognize the di®erence between normal walking and limping. Testing was done on di®erent time intervals. It proved successful to divide data into longer time intervals. By doing so we got better information about gait in one time interval. SVM proved to be the optimal machine learning model for the presented classi¯cation problem.

Današnja sodobna informacijska družba je prepletena z najrazličnejšimi senzorji in vgrajenimi sistemi, ki beležijo, nadzirajo in vrednotijo svet okrog nas. Preko spletnih storitev so ti podatki dostopni praktično povsod. Ob poplavi tovrstnih podatkov, ki nas obdajajo, smo se osredotočili na zajemanje gibanja in biometričnih podatkov subjekta z namenom kasnejše analize zajetih podatkov na osnovi iskanja korelacij med aktivnostjo subjekta v obliki gibanja in biometričnih podatkov, kot so temperatura, srčni utrip in koncentracija kisika v krvi. Cilj magistrskega dela je izdelati baterijsko napajano nosljivo napravo, ki bo omogočala digitalno zajemanje teh podatkov. Identificirali smo primerne elektronske komponente in senzorje za pretvorbo iskanih fizikalnih veličin v električne ter jih primerno prilagodili za pretvorbo v digitalno obliko. Pri tem ločimo senzorje za zajemanje gibanja, ki morajo omogočati hitro zajemanje vzorcev, od biometričnih senzorjev za zajemanje temperature, srčnega utripa in oksidacije krvi, pri katerih ni potrebe po visoki vzorčni hitrosti zajemanja podatkov. Pri zajemanju gibanja smo se osredotočili na 6 prostostnih stopenj, ki vključujejo 3 linearne (translacijske) osi pomikov in 3 kotne (rotacijske) osi pomikov. Zajete podatke smo lokalno shranili in implementirali njihov prenos na osebni računalnik ali mobilno napravo za kasnejšo analizo. Tovrstne podatke lahko uporabimo v športne ali rehabilitacijske namene, za pomoč starejšim v smislu poziva pravočasne pomoči in podobno. Nabor njihovih aplikacij pa bo nedvomno sovpadal z njihovim razvojem in dostopnostjo. Modern society is intertwined with numerous sensors and embedded systems that monitor and control the world that surrounds us. Web applications are bringing these data to our fingertips. Since our surrounding world is well parameterized we focused on the acquisition of motion and biometric data of each individual for a purpose of later analysis based on correlation between one’s activity, in the sense of movement, and biometric data in form of temperature, heart rate and blood oxygen concentration. The goal of the thesis was to build a wearable, battery powered device capable of digital acquisition of the mentioned data. We have identified the appropriate electronic components and sensors to transform the required physical quantities into electrical ones, suitable for digital conversion. Sensors for motion data acquisition need a higher acquisition rate than biometric sensors, since biometric data does not vary so rapidly over time. In regard to motion we decided to capture motion in 6 degrees of freedom, being 3 linear and 3 angular axes of rotation. Acquired data were locally stored until transferred on a personal computer for later analysis. They can be later used for sports or rehabilitation purposes, aid for elderly in the sense of in time call for help etc. With the development of such and similar embedded systems, the number of applications for the acquired data will definitely grow.

V diplomski nalogi je obravnavana posodobitev krmilja na lesnem skobeljnem stroju, pri čemer je bil glavni cilj olajšati delo na stroju. V ta namen je bilo potrebno na stroj dograditi ustrezne senzorje, ki bodo omogočali avtomatsko uravnavanje delovne mize. Hidravlični cilindri so na stroju že bili nameščeni, vendar niso bili v uporabi. Dodatno k temu so dograjeni tudi varnostni elementi, ki bodo zagotavljali varno delo navkljub težkim obratovalnim razmeram, kontrolni panel pa bo uporabniku nudil vpogled v vse potrebne informacije in ustrezno signalizacijo. Celoten sistem je voden preko programirljivega krmilnika Arduino. The diploma work presents upgrade of wood thickness planer feeder and its control unit. The main goal is to make work easier for the operator. For the modernisation of the feeder unit, appropriate sensors, which will allow auto-leveling of work table, need to be installed. Although hydraulic cylinders were already installed on the machine, they were never used. Furthermore, also the safety functions of the machine were upgraded, which will allow safe operation despite harsh working environment. The whole system is based on an Arduino programmable controller, whereas all information will be displayed on a control panel.

V diplomskem delu, ki se nanaša na praktično uporabo podatkov, pridobljenih iz mobilnih senzorjev, smo predstavili, kako delujejo senzorji, kateri senzorji so v naših napravah in kako se odzivajo na spremembe v okolju. Pri tem smo se osredotočili na nekaj izbranih senzorjev in preučili njihovo delovanje. Teoretične ugotovitve smo podkrepili z izdelavo mobilne aplikacije, v kateri smo podatke, pridobljene iz senzorjev, uporabili za zagotavljanje funkcionalnosti spremljanja osnovne statistike smučarja. V ta namen smo v aplikaciji s pomočjo senzorjev pridobivali nekatere koristne podatke, kot so čas, orientacija naprave, višina in spremembe višine v določenem času. Na osnovi teh pridobljenih podatkov smo lahko izračunali nekaj, za smučarja zanimivih in koristnih, informacij, npr. trajanje dejanske smuke, trajanje počitka, število zavojev ipd. Na osnovi podanih informacij lahko smučar ugotavlja, kakšna je bila njegova dnevna aktivnost in se na podlagi tega lažje odloča tudi za v prihodnje (npr. izbira smučišča). In diploma thesis we have introduced the workings of sensors that relate to practical application of data obtained from mobile sensors. We have stated all types of sensors, which are in our devices and described their response to change in environment. We have focused on few of the selected sensors and then examined their functioning. We have supported our theoretical findings with creation of a mobile application in which we used the obtained data from sensors, to ensure the functionality of monitoring our basic statistics of a skier. We obtained useful data such as time, device orientation, height and change of height in certain time frame, with the use of sensors managed by application. Based on the data gathered, we were able to calculate useful information for the skier, such as duration of actual skiing, duration of skier's rest, turn counts and so on. Based on the given information, the skier can determine, what was the quality of his daily activity and then make easier future decisions (ex. choice of ski resort).