Search

Uricotelic species, such as squamate reptiles, birds and insects, effectively eliminate nitrogen as uric acid in a solid form commonly called urates. Observations made over a decade suggested that the voided urates produced by colubroids (modern snake species) exhibit remarkable differences from those of boids and pythons (ancient snake species). Here, we compare the urates generated by eight captive snake species fed the same diet. Although all fresh urates were wet at the time of excretion, those produced by modern snakes dried to a powdery solid, whereas those of ancient species dried to a rock-hard mass that was tightly adherent to surfaces. Powder X-ray diffraction and infrared spectroscopy analyses performed on voided urates produced by five modern and three ancient snakes confirmed their underlying chemical and structural differences. Urates excreted by ancient snakes were amorphous uric acid, whereas urates from modern snakes consisted primarily of ammonium acid urate, with some uric acid dihydrate. These compositional differences indicate that snakes have more than one mechanism to manage nitrogenous waste. Why different species use different nitrogen-handling pathways is not yet known, but the answer might be related to key differences in metabolism, physiology or, in the case of ancient snakes, the potential use of urates in social communication.

SUMMARYThis paper addresses the motion planning and control problem of a system of 1-trailer robots navigating a dynamic environment cluttered with obstacles including a swarm of boids. A set of nonlinear continuous control laws is proposed via the Lyapunov-based Control Scheme for collision, obstacle, and swarm avoidances. Additionally, a leader–follower strategy is utilized to allow the flock to split and rejoin when approaching obstacles. The effectiveness of the control laws is demonstrated through numerical simulations, which show the split and rejoin maneuvers by the flock when avoiding obstacles while the swarm exhibits emergent behaviors.

Z razvojem računalnikov so se razvijale tudi računalniške simulacij, ki so pomemben del našega vsakdana. Kot posledica na razvoj računalnikov in računalniških simulacij so se v zgodovini na najrazličnejših področjih testi in eksperimenti poenostavili, pocenili in izboljšali. To je neposredno vplivalo na razvoj najrazličnejših sistemov, ki jih uporabljamo še danes. Čeprav je bilo izdelanih že mnogo simulacij na vseh področjih, pa so nekatere že zastarele oziroma bi se jih dalo z uporabo današnjih programskih jezikov in programske opreme veliko lažje izdelati. Ena izmed pomembnih simulacij je bila simulacija ptičjega leta iz leta 1986 poimenovana Boids. Od začetka programa Boids pa do danes, se je v spremenjenih različicah pojavil že v mnogih računalniških igrah, filmih ter raziskavah. Namen diplomskega dela je bil izdelati program v programskem jeziku Java, ki bo simuliral ptičji let. Za zgled pa je bil program Boids iz leta 1986. Simulacija je iz dveh delov: osnovnega programa, ki vsebuje tri glavne principe iz programa Boids, ter optimizacija in izboljšava izgleda, ki simulacijo izdelajo bolj razumljivo za vsakega uporabnika. Program smo izdelali s programsko opremo Processing ter jezikom Java. Končni rezultat je simulacija ptičjega leta po zgledu programa Boids v novejšem programskem jeziku in z novejšo programsko opremo. With the development of computers, computer simulations have also evolved, which are an important part of our daily lives. As a result of the computer development and computer simulations, tests and experiments in various fields have been simplified, made cheaper and improved, which has directly affected the development of various systems that we still use today. Although many simulations have already been made in all areas, some are already outdated or would be much easier to do using today's programming languages and software. One of the important simulations was a 1986 bird flight simulation called Boids. Since the beginning of the Boids program, it has appeared in modified versions in many computer games, movies and researches. The purpose of the diploma thesis was to create a program in the Java programming language that will simulate bird flight, and the example was the Boids program from 1986. The simulation consists of two parts, a basic program containing three main principles from the Boids program and optimization an visual improvements, which make the simulation easier to understand for every user. The program was created with the Processing software and the Java programming language. The end result is a bird flight simulation modeled on Boids in a newer programming language and with newer software.

We propose a model that combines existing walking and crowd models (the social-force, following behavior, and Boids models) to simulate pedestrian flow under realistic states of congestion. We recorded pedestrians walking in a passage inside Shinagawa Station to capture the actual conditions and compared this with our simulation results to verify the reliability of the existing and proposed models. This confirmed that the proposed model could reproduce actual pedestrian flow better than the existing models. Using the verified proposed model, we examined the number of rows deep in which pedestrians can comfortably walk down a crowded passage. We then calculated the magnitude of the average number of people leaving as well as their walking speed depending on the number of rows deep in which the pedestrians walked. This analysis confirmed that pedestrians could comfortably walk down a passage that is 16-m wide by 64-m long when there are only two or three rows of pedestrian flow.

With the rapid development of wireless communication technology and intelligent mobile devices, unmanned aerial vehicle (UAV) cluster is becoming increasingly popular in both civilian and military applications. Recently, a swarm intelligence-based UAV cluster study, aiming to enable efficient and autonomous collaboration, has drawn lots of interest. However, new security problems may be introduced with such swarm intelligence. In this work, we perform the first detailed security analysis to a kind of flocking-based UAV cluster with 5 policies, an upgrade version of the well-known Boids model. Targeting a realistic threat in a source-to-destination flying task, we design a data spoofing strategy and further perform complete vulnerability analysis. We reveal that such design and implementation are highly vulnerable. After breaking through the authentication of ad hoc on-demand distance vector (AODV) routing protocol by rushing attack, an attacker can masquerade as the first-arrival UAV within a specific scope of destination and generate data spoofing of arrival status to the following UAVs, so as to interfere with their normal flying paths of destination arrival and cause unexpected arrival delays amid urgent tasks. Experiments with detailed analysis from the 5-UAV cluster to the 10-UAV cluster are conducted to show specific feature composition-based attack effect and corresponding average delay. We also discuss promising defense suggestions leveraging the insights from our analysis.

The flock structures produced by individuals, e.g., animals, self-organize and change their complexity over time. Although flock structures are often characterized by the spatial alignment of each element, this study focuses on their dynamic and hierarchical nature, temporal variations, and meta-structures. In hierarchical systems, sometimes, the upper structure is unchanged, whereas the lower components change constantly over time. Current clustering methods aim to capture the static and mono-layer features of complex patterning. To detect and track dynamic and hierarchical objects, a new clustering technique is required. Hence, in this study, we improve the generative topographic mapping (GTM) method to visualize such dynamic hierarchical structures as they continuously change over time. Using examples from our recent studies on the large-scale Boids model, we confirm that the newly developed method can capture the complex flocking objects as well as track the merging and collapsing events of objects.

Cities are working to make the lives of their citizens as safe as possible. This desire is accompanied by the availability of new devices such as video surveillance cameras in quantity. However, too many of these devices can be counterproductive. Too much information can prevent the identification of important information among the totality of available information. The objective of this thesis is first to find solutions to indicate concretely the type of abnormal situation taking place in order to focus the operator’s attention on this situation. This part is based on static analyses extracted from video streams of video surveillance cameras. The second objective of this thesis is to improve the first one thanks to an individualized monitoring allowing to keep the anonymity of the monitored persons. This solution is based on the prediction of displacement using Kalman filters. Finally, the last objective of this thesis is directly related to the lack of information available to test the efficiency of our different methods. This objective is based on the implementation of a simulation system allowing to get closer to a real situation by using autonomous virtual entities., Les villes oeuvrent de sorte à pouvoir rendre la vie de ses citoyens la plus sécurisée possible. Cette volonté s’accompagne de la mise à disposition de nouveaux dispositifs tels que des caméras de vidéosurveillance en quantité. Toutefois, le trop grand nombre de ces dispositifs peut être contre-productif.Une trop grande quantité d’informations peut empêcher d’identifier l’information importante parmi la totalité des informations disponibles.L’objectif de cette thèse est d’abord de trouver des solutions permettant d’indiquer concrètement le type de situation anormale se déroulant afin de focaliser l’attention de l’opérateur sur cette situation. Cette partie se base sur des analyses statiques extraites des flux vidéo des caméras de vidéosurveillance. Le second objectif de cette thèse a pour but d’améliorer le premier grâce à un suivi individualisé permettant de conserver l’anonymat des personnes suivies. Cette solution se base sur de la prédiction de déplacement à l’aide de filtres de Kalman. Enfin, le dernier objectif de cette thèse est directement lié au manque d’informations disponibles pour tester l’efficacité de nos différentes méthodes. Cet objectif repose donc sur la mise en place d’un système de simulation permettant de se rapprocher au mieux d’une situation réelle en utilisant des entités virtuelles autonomes.

International audience; Complex systems and systems of systems (SoS) are systems characterized by the interconnection of a large number of components or subsystems. The complexity of such systems increases with the number of components and their manner of connectivity. The global behaviour of complex systems and SoS exhibits some properties that cannot be predicted by analysing components or subsystems in isolation. The verification of these properties called "emergent properties" is considered a crucial issue when engineering such systems. The purpose of this paper is to give an overview and verify emergent properties. In a first step, we have taken the blinker and the traffic light of the game of life as examples to verify emergence by using refinement techniques; in a second step, since the refinement is not straightforward, we have taken another example, the Boids model, and by using timed automata and UPPAAL model checking techniques, we have been able to simulate and verify emergent properties. Povzetek: Prispevek podaja pregled emergentnih lastnosti v kompleksnih sistemih.

Among Game AI, in macromanegement level, applying of deep learning algorithms like reinforcement learning has been proven to be useful. In this paper, we researched the development of Game AI and chose Starcraft II as our testing game. We compare various deep learning algorithms to normal algorithms like boids algorithms, potential field pathfinding etc., in the micromanegment level, to see their difference in run-time performance, training time and CPU utilization level.

This chapter presents a model for influencing and shaping that employs computational red teaming. The model is demonstrated and validated computationally using a simulation environment where social influencing and shaping are applied to an artificial society. Network topology for situation awareness and a trust factor on perceived information were added to the classic Boids model to enable the investigation of influence and shaping. The chapter builds on the difference between influence and shaping. First, this distinction is important because it implies that influencing is a sufficient condition for shaping. Second, it is important in computational social sciences to facilitate the creation of models that are not ambiguous about the socio‐psychological phenomena under investigation. Third, the distinction is important because it clarifies that influencing and shaping work on different time scales – influencing is effective in the short term, whereas shaping is more effective in the long term.

In this paper, we propose a multi-agent based art production framework. In existing artwork creation systems, images were generated using artificial life and evolutionary computation approaches. In artificial life, swarm intelligence or Boids model, and in evolutionary computation, genetic algorithm or genetic programming are commonly used to create images. These automated artwork creation systems make it easy to create artistic images even if the users are not professional artists. Despite the high possibility of these creation systems, however, much research has not been done so far. In this paper, we propose an art production framework that generates images using multi-agents with chaotic dynamics features. Agents act on the canvas following the three rules of Boids model. In addition, each agent possesses a chaotic neural network which trained by differential evolution algorithm, so that colors can be evolved to represent a better style. As a result, we propose an art production framework for generating processing artworks that contain highly complex dynamics. Finally, we created the glitch artworks using the proposed framework, which shows a new glitch style.

Nowcasting is vital for PV farms in order to match supply and demand on energy markets. The accuracy of existing forecasting methods relies on short term modelling of cloud movement and dynamics by combining satellite images with ground based allsky observations. Cloud movement is modelled based on complex nonlinear numerical models which on the short term are outperformed by image analysis techniques. In this research we propose a simplified nature inspired model for nowcasting which scales with the data. The model is based on forecasting wind direction extracted from motion vector fields by modelling wind using a modified Boids Flocking algorithm. We show that the model is accurate in predicting the cloud coverage up to several hours ahead and that it scales with the number of particles on shared memory systems suited for commodity machines available to PV farm operators.

We study swarms as dynamical systems for reservoir computing (RC). By example of a modified Reynolds boids model, the specific symmetries and dynamical properties of a swarm are explored with respect to a nonlinear time-series prediction task. Specifically, we seek to extract meaningful information about a predator-like driving signal from the swarm’s response to that signal. We find that the naive implementation of a swarm for computation is very inefficient, as permutation symmetry of the individual agents reduces the computational capacity. To circumvent this, we distinguish between the computational substrate of the swarm and a separate observation layer, in which the swarm’s response is measured for use in the task. We demonstrate the implementation of a radial basis-localized observation layer for this task. The behavior of the swarm is characterized by order parameters and measures of consistency and related to the performance of the swarm as a reservoir. The relationship between RC performance and swarm behavior demonstrates that optimal computational properties are obtained near a phase transition regime.

In this wide-ranging survey, Claus Emmeche outlines many of the challenges and controversies involved in the dynamic and curious science of artificial life. He describes the work being done by an international network of biologists, computer scientists and physicists who are using computers to study life as it could be, or as it might evolve under conditions different from those on Earth. Many artificial-life researchers believe that they can create new life in the computer by simulating the processes observed in traditional, biological life-forms. The flight of a flock of birds, for example, can be reproduced faithfully and in all its complexity by a relatively simple computer program that is designed to generate electronic "boids". Are these "boids" then alive? The central problem, Emmeche notes, lies in defining the salient differences between biological life and computer simulations of its processes. His study touches on every aspect of this complex and rapidly developing discipline, including its connections to artificial intelligence, chaos theory, computational theory, and studies of emergence.

This report investigates how the construction and query time of multiple spatial partitioning data structures is impacted by spatial distribution of and number of agents in a crowd simulation. In addition a method is investigated for updating the data structures less frequently at the cost of increasing the radius queried, without affecting the correctness of the queries. The data structures are tested in a simulation using a Boids model and update and query times are measured. It is found that the performance of the grid is better than the quad tree and the kd- tree for low number of agents, but deteriorates more quickly when the number of agents increase. It is also found that this approach can decrease the sum of time spent updating and the time spent querying in the simulation. The effectiveness of this method is highly dependent on the update of the data structure. Denna rapport undersöker hur konstruktion och grannsökning av flera datastrukturer för spatial partitionering påverkas av spatial fördelning av simuleringens agenter och antal agenter i simuleringen. Dessutom undersöks en metod för att uppdatera datastrukturerna mindre ofta, på bekostnad av att utöka grannsökningens radie, utan att påverka grannsökningens korrekthet. Datastrukturerna testas i en simulering baserad på Boids och uppdaterings- och frågetider för datastrukturerna mäts. Det visar sig att prestandan av grid är bättre än prestandan av quad tree och kd- tree för ett litet antal agenter, men att prestandan för grid försämras snabbare när antalet agenter ökar. Dessutom visar sig denna metod kunna ge en minskning i den totala tiden som går åt till att göra grannsökningar och uppdateringar av datastrukturen. Hur effektiv denna metod är beror i hög grad på hur lång uppdateringstiden är för den använda datastrukturen.

Boids (bird-oids) is a widely used model to mimic the behaviour of birds. Shoids (sheep-oids) rely on the same boids rules with the addition of a repulsive force away from a sheepdog (a herding agent). Previous work assumed homogeneous shoids. Real-world observations of sheep show non-homogeneous responses to the presence of a herding agent. We present a portfolio of information-theoretic and spatial indicators to track the footprints of shoids with different parameters within the shoid flock. The portfolio is named the Centre of Influence to indicate that the aim is to identify the influential shoids with the highest impact on flock dynamics. We use both synthetic simulation-driven data and measurements collected from live sheep herding trials by an unmanned aerial vehicle (UAV) to validate the proposed measures. The resultant measures will allow us in our future research to design more efficient control strategies for the UAV, by polarising the attention of the machine learning algorithm on those shoids with influence footprints, to drive the flock to improve the herding of sheep.

Izrada složene simulacije predstavlja veliki izazov, pogotovo ako je ta simulacija interaktivna. Unity već dugi niz godina nudi alate za izradu interaktivnih simulacija te se najčešće koristi za izradu video igara. Kako se zahtjevi tržišta video igara mijenjaju s dolaskom novih tehnologija tako se stvara potreba za novim pristupom za izradu video igara. Uređaji za prikaz virtualne stvarnosti i prijenosni uređaji zahtijevaju učinkovita rješenja koja štede bateriju i procesorske cikluse. U svrhu rješavanja ovih problema Unity je ponudi novu DOTS tehnologiju. Cilj ovog rada je proučiti razliku između novog, DOTS pristupa koji je vođen podatkovno orijentiranom paradigmom te klasičnog pristupa koji je vođen objektno orijentiranom programskom paradigmom. Za potrebe usporedbe izgrađene su dvije simulacije boida, jedna koristi klasični pristup, druga koristi DOTS. Designing a complex simulation represent a great challenge, especially if the simulation is interactive. Unity, for quite some time now, has been offering tools for creating interactive simulations and has been mostly used for creating video games. With new, emerging technologies needs of the video game industry are changing. Virtual reality headsets and mobile devices require solutions that save battery power and processor cycles. Trying to solve these issues Unity has offered their new DOTS technology. The goal of this thesis is to compare the new, data-oriented DOTS design with the classical, object-oriented design. For this test two simulations were constructed, one using the classical approach, the other using DOTS.

Personalized systems are required in many domains. However, gathering training data for personalization from individuals, as is necessary with deep learning, is a difficult and timeconsuming task. With our proposed method, less or no training data is required to adapt to individuals’ preferences, even when they shift over time. We introduce a potential field based method “Dark Side Ternary Stars” which has three components, GAGPL, Wordoids, and EGO. In this paper, we focus on two of them, ”Wordoids”, which adopt extends Boids algorithms to perform individualized classification of keywords by topic and improved our previous work ”GAGPL”, which calculates the individualized semantic orientation of sentences by using learned words per topic. As experimental results, we applied this method to news articles about Japanese professional baseball and we show that our method can obtain individualized semantic orientations and summaries of the article per individual.