Search

Predicting smartphone users location with WiFi fingerprints has been a popular research topic recently. In this work, we propose two novel deep learning-based models, the convolutional mixture density recurrent neural network and the VAE-based semi-supervised learning model. The convolutional mixture density recurrent neural network is designed for path prediction, in which the advantages of convolutional neural networks, recurrent neural networks and mixture density networks are combined. Further, since most of real-world datasets are not labeled, we devise the VAE-based model for the semi-supervised learning tasks. In order to test the proposed models, we conduct the validation experiments on the real-world datasets. The final results verify the effectiveness of our approaches and show the superiority over other existing methods., Comment: 11 pages, 10 figures

Vital records are rich of meaningful historical data concerning city as well as countryside inhabitants that can be used, among others, to study former populations and then reveal the social, economic and demographic characteristics of those populations. However, these studies encounter a main difficulty for collecting the data needed since most of these records are scanned documents that need a manual transcription step in order to gather all the data and start exploiting it from a historical point of view. This step consequently slows down the historical research and is an obstacle to a better knowledge of the population habits depending on their social conditions. Therefore in this paper, we present a modular and self-sufficient analysis pipeline using state-of-the-art algorithms mostly regardless of the document layout that aims to automate this data extraction process.

Recognizing user location with WiFi fingerprints is a popular approach for accurate indoor positioning problems. In this work, our goal is to interpret WiFi fingerprints into actual user locations. However, WiFi fingerprint data can be very high dimensional in some cases, we need to find a good representation of the input data for the learning task first. Otherwise, using neural networks will suffer from severe overfitting. In this work, we solve this issue by combining the Information Bottleneck method and Variational Inference. Based on these two approaches, we propose a Variational Information Bottleneck model for accurate indoor positioning. The proposed model consists of an encoder structure and a predictor structure. The encoder is to find a good representation in the input data for the learning task. The predictor is to use the latent representation to predict the final output. To enhance the generalization of our model, we also adopt the Dropout technique for each hidden layer of the decoder. We conduct the validation experiments on a real-world dataset. We also compare the proposed model to other existing methods so as to quantify the performances of our method.

Ant colony optimization (ACO) algorithms are a bio inspired solutions which have been very successful in combinatorial problem solving, also known as NP-hard problems, including transportation system optimization. As opposed to exact methods, which could give the best results of a tested problem, this meta-heuristics is based on the stochastic logic but not on theoretical mathematics demonstration (or only on certain well defined applications). According to this, the weak point of this meta-heuristics is his convergence, its termination condition. We can finds many different termination criteria in the scientific literature, yet most of them are costly in resources and unsuitable for practical problems. On the other hand, given the fact that the ACO is a stochastic approach, it seems difficult to decide whether to stop the algorithms in order to have the optimal result of the tested problems. Therefore, the thesis of this paper is to propose an approach based on the environment in order to determine the best termination criteria of the ACO, for an optimized solution.

International audience; Hardware-in-the-loop (HIL) test provides a time saving and safe environment for testing high-power electronic systems. But the main difficulty for the HIL test of power electronic system lies on the modeling of the complex and highpower system. This paper proposes a modeling method of the electrical system of high-speed train. With this method, the HIL test platform using field-programmable gate array boards is built. Besides, in order to meet the computing power requirement of the system modeling, we simulate the whole system using two dSPACE simulators and inside each dSPACE simulator, a multiprocessor system is achieved through Gigalink connection. The whole HIL system can be used to evaluate both the hardware and software performance of traction control unit and auxiliary control unit. The HIL results under steady state and transient conditions demonstrate modeling accuracy and provide a detailed insight into its development.

International audience; Proton exchange membrane fuel cell (PEMFC) has been considered as one of the promising renewable power technologies for the vehicular application. This paper proposes a fuel cell electric vehicle (FCEV) power system model that can be implemented in the hardware-in-the-loop (HIL) emulation for real-time execution on field-programmable gate arrays (FPGAs). The FCEV model comprises three parts: a PEMFC, a Z-source inverter and a squirrel cage motor. To achieve an accurate and efficient FPGA resources’ utilization, the PEMFC model is implemented by CPU whereas the models of the DC-AC inverter and the electrical motor are built in FPGA. For the validation of the proposed power system, the real-time simulation tests are conducted with a high accuracy. The developed hybrid system model can reach a simulation time step of 100 ns for FPGA and 500 μs for CPU under the co-simulation mode. Moreover, the simulation under various system operating conditions indicates that the high performance can be reached by the hybrid system computed in real-time. The proposed real-time model can be used to design the on-line diagnostic and model predictive control method, which can help to test the FCEV before the commercial applications.

Autonomous vehicles that are able to communicate together allowed the emergence of a new way of controlling intersections. Recently an active research community stakes on cooperative intersection management. Vehicles and intersection are able to communicate together in order to improve traffic condition at intersections. Many simulations have shown that the cooperative intersection management outperforms traffic lights because there are no limited stages as within traffic light and authorized movements are adapted to the current situation. This paper focusses on speed synchronization at intersections. More precisely, a virtual platoon is formed by vehicles coming from different lanes. When traffic rate is low this allows avoiding useless stops and thus the speed synchronization saves energy and increases the average speed. However, experiments showed that some parameters need to be considered. Hence, as the traffic rate grows, vehicles need to stop and speed synchronization is no more efficient. The main reason of this drawback is that the studied concepts are limited to First Come First Served or to First Out First-Served. This paper extends the virtual platooning, where new rules form new virtual platoon able to adjust dynamically the whole behavior according to the traffic flow growth under safety conditions. Simulations based on real-world experiences show that the proposed approach is very efficient by increasing the average speed while keeping a high capacity of the intersection.

International audience; Conceiving artefacts and tools is one of the major key characteristics of the human species. Since the beginning of automation and even more since the spreading of computer science, this activity has become more and more challenging especially through the analysis of the different interaction modes between the artifacts and the human. Nowadays, developing an artifact that must interact with a human is requiring an integrative point of view which must include biological, physical and cybernetical concerns. The current norms aimed at characterizing the quality and the relevance of human machine interface are generally limited to standard interfaces and/or to specific domain. The goal of this paper is to propose an integrative conceptual framework and ergonomic design criteria for developing human systems.

Analyzing human mobility with geo-location data collected from smartphones has been a hot research topic in recent years. In this paper, we attempt to discover daily mobile patterns using the GPS data. In particular, we view this problem from a probabilistic perspective. A non-parametric Bayesian modeling method, the Infinite Gaussian Mixture Model (IGMM) is used to estimate the probability density of the daily mobility. We also utilize the Kullback-Leibler (KL) divergence as the metrics to measure the similarity of different probability distributions. Combining the IGMM and the KL divergence, we propose an automatic clustering algorithm to discover mobility patterns for each individual user. Finally, the effectiveness of our method is validated on the real user data collected from different real users.

International audience; Renewable energy sources like wind or hydrogen have been worked as a reliable option to traditional power sources. But for distributed power generation systems (DPGSs), the operation of the grid-connected converter under the weak ac grid condition has been a challenge for the stability and power quality issues. This paper shows the effect of the weak grid on the grid voltage-controlled inverters with LCL- filter and L- filter. The impedance-based analytical approach is employed to study the harmonic stability caused by the interaction between the power grid and the grid converter. A method of deriving the impedance ratios for different inverters is utilized to predict the harmonic stability of the power system through Nyquist diagrams and Bode plot. Time-domain simulations based on Matlab/Simulink are presented to validate the effectiveness of the theoretical approach and confirm conclusions from analytical studies. This study provides insight into the design of the controller under weak grid networks.

International audience; The hardware-in-the-loop simulation plays a vital role in the test of high-power electronic system. Although the application of field-programmable gate array (FPGA) embedded system has enabled the real-time system simulating below 500 ns, the transient characteristic of high-voltage insulated-gate bipolar transistor (HVIGBT) is largely compromised. In this paper, a new piecewise HVIGBT model, considering its driver circuit effect and parasite parameter, is proposed for FPGA-based real-time simulation applications. With the attempt to reduce the simulation latency, we propose an FPGA solver with a parallel structural to divide the system into several layers. The model could not only provide accurate system-level performance of the power electronic converter but also give an insight into the transient behavior effect of high-power electronic system. Finally, a case study about emulation of traction system of high-speed train is also presented. Implementations are made on an FPGA Kintex-7 embedded in National Instruments FlexRIO PXIe-7975. The obtained results show that the proposed modeling algorithm can achieve both accuracy and efficiency within a fixed real-time simulation time step of 25 ns.

International audience; The application of field-programmable gate array (FPGA) in the hardware-in-the-loop simulation (HiLs) has enabled the time-step reaching hundreds of nanoseconds range. However, the time performance of FPGA-based real-time simulation remains to be improved. In this paper, a network analysis modeling method for FPGA-based real-time simulation of the power electronic converter is proposed. An ideal switching network unit, without requiring a large amount of memory, is used to determine the topology of the circuit and solve interface voltages/currents from the torn circuit. A parallel integration method is then implemented to obtain the status of the circuit element. At last, a diode-clamped three-level voltage source converter is used as a case study to demonstrate the effectiveness of the method. A 40 -ns simulation step is finally achieved on the National Instruments (NI) FlexRIO PXIe-7975 platform. The accuracy of the proposed method is also verified against the results from Matlab/Simulink.

Energy consumption and power management are key issues problem when using smartphone handsets or any other connected devices. This is linked to two opposite users’ expectations: having a better autonomy on one side and having devices with more computer power and bigger screens on the other side. In literature, many previous works focused on modeling the smartphone power drain at the hardware or software levels. As opposed to these approaches, we present, in this paper the construction of a handset energy model at user level based on coarse grain collected data. To that way, we adopt a crowd-sourcing approach, based on data collected by the Device Analyzer project. This data is covering all the activities/events of the user relatively to his/her smartphone. However, the data is collected within a fixed time period which involved imprecisions in the time stamp associated to each event. So as to model the energy drain and to improve events time location, we propose a top-down method based on reactive multi-agent systems paradigm. In this approach, each event is represented by an agent in a virtual world which can move along the time axis according to the behavior of its mates and which influences the battery discharging curve. As soon as the multi-agent system is stabilized, the behavioral parameters of the agents are used to build up the power model. The value of the multi-agent approach is tied to its ability to incorporate implicit information about battery drain from operating system events such as network usage. Finally, the proposed modeling method is tested on real users data and several coordination strategies are compared. Index Terms—Smartphone, energy model, multi-agent model, crowd-sourcing, user behavior

This paper presents enhanced parallel field-programmable gate array (FPGA) architecture for solving the model of diode-clamped three-level voltage source converter (VSC). The simulator is intended for the testing evaluation of digital control based on the hardware-in-the-loop (HIL) concept. The philosophies of the proposed implementation are: 1) it enables to the use of simulation as low as 40 nanoseconds, 2) it can provide a fixed timeline for the whole system execution; 3) it effectively utilize the parallel architecture in the FPGA. The implementation results in the National Instruments FIexRIO PXIe-7975 platform are compared with offline simulation in Matlab. The results show high accuracy of the modeling approach.

Augmented Reality (AR) is now more and more widespread in many application fields. Thanks to the new progresses in computer power, AR can now be used widely in education without any expensive additional devices. In this paper is presented the feedback of an experimental protocol using an AR application as an additional support for a History lesson in secondary schools. Even if the technical part has a lead role in the student experience, the most challenging issue is related to the choice of the teaching lesson as itself which must fit several, sometimes contradictory, requirements.

International audience; In real-time simulation applications, hardware-in-the-loop (HIL) within nanosecond time step of large size power electronic systems involving high-frequency control strategy is still a big challenge today. In this paper, we propose a new hybrid modeling approach based on a novel nodal analysis and solving method in order to achieve both fast and accurate real-time simulation. The decoupling method allows dealing with circuit element separately regardless the size of power electronic system. After system matrix is obtained, an innovative approach using Cholesky decomposition is proposed. At last, a comparative case study of traction system in the electrical locomotive train is also presented. For this case study, implementations are made on a Field Programmable Gate Arrays (FPGA) Kintex-7 embedded in National Instruments FlexRIO PXIe-7975. Results obtained show that the proposed modeling algorithm can achieve both accuracy and efficiency within a 50 ns fixed real-time simulation time step. Besides, the comparison with results obtained from Simpower system in Matlab allows evaluating the accuracy of our proposed modeling approach.

International audience; In real-time Hardware-in-the-Loop (HIL) test applications for power electronic systems, the main hurdle is to tackle with the mathematical models of variable topology of complex and high frequency driven converter. The most widespread solution is to separate the whole system into subsystems. However, partitioning method usually introduces simulation time step latency between different subsystems, which causes numeric instabilities especially when stiff situation occurs. In this paper, we propose a novel parallel simulation approach which has no time step latency in the whole system division, from which a numerically stable system modeling can be realized. Its numerical accuracy of the solution, the architecture design, and the issue pertaining to the parallel calculation are discussed in detail in this paper. The pertinence of the developed solution is also tested using a case study relating to a traction system power electronic application. For this case study, implementations are made both on a 3 GHz Xeon CPU of RT LAB real-time simulator with a 2 μs simulation step and a Field Programmable Gate Arrays (FPGA) Kintex-7 embedded in National Instruments FlexRIO PXIe-7975 enabling a simulation step below 50 ns. Besides, comparison with results obtained from Simpower system in Matlab allows to evaluate the accuracy of our proposed modeling approach.

Among the innovations aimed at tackling the transportation issues in urban areas, one of the most promising solutions is the possibility of making virtual trains of vehicles so as to provide a new kind of transportation system. Even if this kind of solution is now widespread in literature, some difficulties still need to be resolved. For instance, one must find solutions to make the crossing of the train possible while maintaining train composition (trains must not be split) and safety conditions. This paper proposes a multi-level decision process aimed at dealing with this issue. This proposal is based on dynamic adaptation of train parameters which lead to trains crossing without stopping any of them. Results, obtained in simulations, make the comparison with a classical crossing strategy.

The use of technologies such as Virtual Reality (VR) and Augmented Reality (AR) is more and more widespread in our everyday life and is now mature enough to be used in critical medical or aerospace applications. The main issue is then to provide more adaptive abilities to VR/AR applications in developing them using human concerns as primary element. To that way, beyond the classical HITL concept this paper proposes an integrative point of view for virtual environment design based on the concept of Bio-CPS. The goal of this paper is to present Bio-CPS related concepts and to sketch some guidelines for future virtual environments design focusing on specific application fields.

The intelligent vehicle concept has been widely developed with the aim of solving problems caused by the increasing of vehicles in cities, such as congestion, air pollution and accident. With the increasing use of intelligent vehicles, traffic flow control methods have been improved as compared to the traditional methods, such as traffic light management. Among new possible methods, the one we want to explore in this paper is to combine several relative intelligent vehicles into one platoon with a reduced distance and allowing an increase of the road capacity. However, the use of platoons brings some issues at system level, for instance when several platoons encounter with each other at road nodes. In order to give answers to these issues, we define one organization strategies including defined interactions between vehicles and between platoons. In this paper, one multi-agent model is proposed to tackle with this issue. The proposed algorithm is implemented based on multi-level control method and is experimented in the scenario where one vehicles-train composed by three vehicles is reconfigured in two lane road.

International audience; Due to their properties such as high energy density, lithium-ion battery packs are more and more widespread for many kinds of applications. Although a classical Battery Management System (BMS) can improve the performance and safety of a battery pack, it cannot deal with issues related to aging and cell faults. This paper presents a fault-response approach for battery packs operating at the cell level. It uses a reconfigurable topology combined with agents, that can increase the resilience of a pack in the face of faults. A cosimulation technique based on Matlab/Simulink and the JADE platform is used to test and validate the proposed approach.

International audience; Since a decade multi-agents became a widespread solution to tackle different kinds of issues in various application fields. Among the two main trends in multi-agent approaches (cognitive vs. reactive), the reactive one is particularly interesting for applications that require both fast response time, adaptability and robustness. Reactive Multi-agent Systems rely on simple interaction schemes between entities that can be inspired whether by Biology or Physics. The goal of this paper is to explore the possibility of using interference fringes and waves properties from a multi-agent standpoint so as to tackle a real problem, longitudinal platoon regulation, already explored by different strategies including particle agents approaches or standard PID controller.

Vehicle platoon approaches found in literature deal generally with column formations adapted to urban or highway transportation systems. This paper presents an approach in which each platoon vehicle follows a virtual vehicle, in order to cope with issues such as different platoon geometries. Those different types of formations can be encountered in a wide range of field such as the military or agriculture. A platoon formation is composed of a vehicle which assumes the platoon leader role (generally human driven) and other vehicles which play the follower role. A follower vehicle assigns a local leader role to one of the vehicles it perceives. The approach presented here bases on a predefined translation of position, by a follower vehicle, calculated from the perceived position of its local leader vehicle. This translation depends on the desired platoon geometry, expressed in terms of a predefined longitudinal and lateral distance of a follower vehicle relatively to its local leader position. Each vehicle is implemented as an agent which makes decisions depending only on its own perception.

The use of individual vehicles is increasing in inner cities involving a large number of unwanted side effects. However, the rise of intelligent vehicle technologies allows finding a solution to some of these problem bringing new vehicle usage. For traffic jam management, aside to widespread traffic light control, one of the promising solutions is the possibility to form virtual trains with vehicles so as to increase road capacities. This solution, however relevant it is, suffer from limitations when the road network has many interconnections where interactions between trains are hard to define. This requires thus a control process which can deal with train, vehicles, and hardware control issues. The goal of this paper is to propose a multi-level cooperative control system. This proposal relies on a dynamic adaptation of train parameters in runtime and allows crossroads and roundabout sharing without stopping any vehicle. The proposal is tested in simulation coping with the roundabout scenario.

We demonstrate how a multi-agent top-down approach can be used to interpolate between battery level measurements on a phone handset. This allows us to obtain a high fidelity trace whilst minimising the data collection overhead. We evaluate our approach using data collected by the Device Analyzer project which collects handset events and polled measurements from Android devices. The value of the multi-agent approach lies in the fact that it is able to incorporate implicit information about battery level from operating system events such as network usage. We compare our approach to interpolation using Bezier curves and show a 50 % improvement in mean error and variance.

Vehicle platoon systems can be considered as good alternative solutions to traffic problems encountered in urban environment. Indeed, they allow to propose new public transportation models based on size adaptable trains. In other fields, such as military operation theaters and agricultural fields, platoon systems can also solve specific problems. The goal of this paper is to propose an agent based decision process that allow to handle the control of vehicles platoon. The decision process is composed of 5 interconnected layers each dealing with one specific aspect of the decision process including the perception interpretation, the spatial configuration choice and the control of the vehicles. This model has been tested in simulation by taking into account several geometrical configurations.

International audience; This paper deals with intelligent autonomous navigation of a vehicle in cluttered environment. We present a control architecture for safe and smooth navigation of a Unmanned Ground Vehicles (UGV). This control architecture is designed to allow the use of a single control law for different vehicle contexts (attraction to the target, obstacle avoidance, etc.) [4]. The reactive obstacle avoidance strategy is based on the limit-cycle approach [2]. To manage the interaction between the controllers according to the context, the multi-agent system is proposed. Multi-agent systems are an efficient approach for problem solving and decision making. They can be applied to a wide range of applications thanks to their intrinsic properties such as self-organization/emergent phenomena. Merging approach between control laws is based on their properties to adapt the control to the environment. Different simulations on cluttered environments show the performance and the efficiency of our proposal, to obtain fully reactive and safe control strategy, for the navigation of a UGV.

The topic of this work is the problem solving issue using simple interacting entities. The main difficulty of such a method stems from the paradox between the complexity of the problem, on one side, and the conceptual and functional simplicity of the entities on the other side. In order to overcome this paradox, we propose a method based on a reactive multi-agent system. This solving principle points out the importance of the interaction, and their integration in the agents' environment. As for the use of such a system to solve a real problem, the environment plays a key role. First, it represents the problem to solve on both topological and dynamical points of view. Second, it takes action into the solving process by determining the collective evolution of the agents thanks to the implementation of a system's Physics. In this paper, we describe the solving model we propose and its validation through experimental test performed both on simulation and with real targets.

Multi-agent systems are now wide spread in scientific works and in industrial applications. Few applications deal with the Human/Multi-agent system interaction. Multi-agent systems are characterized by individual entities, called agents, in interaction with each other and with their environment. Multi- agent systems are generally classified into complex systems categories since the global emerging phenomenon cannot be predicted even if every component is well known. The systems developed in this paper are named reactive because they behave using simple interaction models. In the reactive approach, the issue of Human/system interaction is hard to cope with and is scarcely exposed in literature. This paper presents Sphericall, an application aimed at studying Human/Complex System interactions and based on two physics inspired multi-agent systems interacting together. The Sphericall device is composed of a tactile screen and a spherical world where agents evolve. This paper presents both the technical background of Sphericall project and a feedback taken from the demonstration performed during OFFF Festival in La Villette (Paris).

Since a couple of years, transportation systems have been entering into mutation. The part of the autonomous abilities of vehicles increases significantly with their associate constraints and problematic. Among these abilities, trajectory following is one of the minimum competence required for vehicles or mobile robots autonomy. Generally, the computation of the path to follow is made “off-line” and then is transmitted to the vehicle. The main problem then encountered is the fact that the model of the evolution area of the vehicle may have changed between the computation of the plan and its execution. If unpredictible events occur during the vehicle moves, one must recompute a local plan taking into account these new events and respecting as far as it is possible the initial plan. Since the computation of the plan for a trajectory is costly and since the computation power of a vehicle is limited, it is hardly possible to embed this kind of ability directly into vehicles. This paper proposes a reactive multi-agent based solution to address this problem with low computation power requirement and adaptive abilities allowing obstacle avoidance.

For more than 20 years, researchers have designed models in order to describe swarm intelligence and apply the resulting techniques to complex problems. However, there is still a gap between these models and current MAS methodologies. The goal of this paper is to propose a principled and methodological approach for the engineering of systems based upon swarm intelligence. The constraints are, on the one hand, to enable the analysis, design and implementation of such systems; and, on the other hand, to formally analyze and verify properties of resulting systems. The principles of the approach are based, on the one hand, on requirement driven activities that produce goals to be fulfilled by the system of interest and, on the other, hand on an ontological modeling of the problem domain. This ontological modeling conceptualizes the phenomenon one seek to imitate and thus allows it understanding. The produced ontology is refined through the methodology activities down to organizational models. (C) 2012 Elsevier Ltd. All rights reserved.

Nowadays, automotive equipments tend to offer more and more drinving assistance. Among these, car park assist systems are now widespread on both commercial and research levels. This paper proposes a car park assistance based on the use of the multi-agent paradigm. This approach based on the transformation of vehicle perception into a virtual environment for agents and on the evaluation of agency properties. This evaluation is then turned into a decision for vehicle guidance. The proposal is evaluated in simulation by comparing it with industrial solutions.

This paper presents a reactive multi-agent system for echelon platoon organization. Platoon systems are sets of vehicles that move together while keeping a predefined geometrical configuration without any material coupling. Each vehicle represents an autonomous agent that behaves based only on its own perceptions. The distributed platoon organization problem consists in defining the algorithms to be executed by each vehicle's embedded software, in order to maintain the desired platoon configuration during displacements. Platoon systems found in literature deal generally with column formations adapted to urban or highway transportation systems. Other formations such as line, echelon,... can be encountered in fields like agriculture and the military. In this paper, we focus on the platoon echelon formation. An approach based on a virtual vehicle-to-vehicle interaction model composed of two springs is proposed. Those virtual springs attach a platoon's vehicle to its local leader, another platoon's vehicle. In this work, five different spring's attachment points are evaluated, to compare them and conclude about the more suitable one, depending on platoon's trajectory. Eventually, from the evaluation results, it can be conceived to make attachment points evolve during platoon operation.

This paper presents a reactive multi agent approach to the platoon control problem for the linear configuration. Platoon is a train of vehicles composed of a head vehicle and a variable number of followers. Vehicle-to vehicle coupling is virtual. Each follower vehicle controls its movement by interacting only with the preceding one. To this end, platoon control was designed as a reactive multi agent system where each follower vehicle is an agent. Each agent behavior is specified by a physics inspired interaction model, which allows to compute vehicle speed and direction from a single perception: the distance to the preceding vehicle. We present the physics inspired interaction model together with a safety verification case study and simulations results.

Since a couple of years, multi-agent systems have been used for a wide range of applications such as problem solving, modelling and simulation, Reactive agent based systems are characterized by their capability to solve complex problems, while maintaining functional and conceptual simplicity of each element. These approaches exhibit effectiveness whatever the considering fields (life simulation, robots and cooperation). One of the main problems of such method is the difficulty to characterize/measure emergent phenomena. Based on preceding work, this paper presents a reactive agent based vehicle platoon algorithm with an integrated obstacle avoidance ability. The vehicle decision process is then considered as a multi-agent system, the agents of which build collectively the decision taking into account both platoon and obstacle avoidance constraints. The decision is built through measurement of agency collective state and dynamics.

Nowadays, urban environments suffer from recurrent traffic jam with associated side effects. One reason of this is the social implicit priority given to personal cars, which are preferred to public transportation systems the main drawback of which is the time and path rigidity. Recently, some alternative transportation systems have been developed based on size adaptable trains of vehicles called platoon. These approaches still suffer from rigid path planning. The paper presents one possible solution to overcome this drawback. The proposal is based on the use of existing crossroads as hubs able to reconfigure vehicles train while crossing. Thanks to this solution, each train component could have its specific path in the public transportation grid. This paper presents also a comparative study of exposed algorithms relatively to a classical traffic light schedule.

Vehicle collision avoidance is a promising safety approach to new transportation systems, with innovative capabilities, such as obstacle detection, vehicle collision avoidance control strategy and adaptability to different obstacles. This paper presents a Reactive Multi-agent solution to the vehicle collision avoidance control problem with a linear configuration. In our case, vehicle collision avoidance is designed as a reactive multi-agent system in which agents interact with other agents and the obstacles situated in the environment by using physics inspired behaviors. Collision avoidance stability emerges as a global result of the individual interacted agents. Vehicle avoidance control strategy stems from calculating the trajectories of the vehicle based on the decision process of the reactive multi-agent system. Furthermore, the adaptation to different kind of obstacles is made by tuning model's physical parameters. In order to assert the transition from abstract to concrete, simulations experiments have been implemented and simulations results are analyzed.

Recently, many research efforts are directed towards co-evolutionary algorithms. The present work aims at the assessment of Hierarchical Cooperative CoEvolution (HCCE) being properly formulated to address hierarchical problems where simple components having separate design objectives, are parts of other more complex ones. HCCE is able to highlight the specialties of each component and additionally enforce their successful integration in a composite structure. Here we present HCCE describing also the internal dynamics that provide its effectiveness in addressing difficult distributed design problems. Additionally, the results described in the present work attest to its validity and superior performance against ordinary Unimodal evolution, and Enforced SubPopulation coevolution.

Vehicular platoon are a promising approach to new transportation systems, with innovative capabilities, such as vehicle sharing and adaptability to demand. This paper presents a multiagent solution to the platoon control problem with a linear configuration. In our case, a platoon is a vehicle train composed of a head vehicle and a variable number of followers. The head vehicle is human-driven or autonomous, whereas each follower vehicle controls its movement by interacting only with the preceding one. To this end, the platoon control system was designed as a reactive multiagent system where each follower vehicle is an agent. Each agent's behavior is specified by a physics inspired model, which allows to compute vehicle's speed and direction from a single perception: the distance to the preeding vehicle. Platoon stability emerges as a global result of individual interactions. Furthermore, adaptability to different kinds of vehicles is obtained by adjusting the model's physical parameters. Another important aspect is related to platoon's evolution, mainly by vehicle merging and splitting. To assert the transition from abstract to concrete, both simulations and experimentations have been implemented.

Recently, many research programs have investigated the concept of intelligent vehicles and their integration in the city of tomorrow. The aim is to develop an intelligent transportation system based on a fleet of fully automated cars designed for short trips at low speed in urban areas. This system will offer advantages of high flexibility, efficiency, safety, and thus, will improve the quality of life in our cities including but not restricted to protection of the environment, better management of parking areas, and others. One of the key functions that such transportation system must achieve concerns the notion of autonomous navigation of vehicles. To reach this goal, we are working on vehicle environment perception using passive and active sensor technologies. In this chapter, we address an artificial-vision based environment perception system for autonomous vehicle navigation. We are particularly interested in obstacle detection using stereo vision, image road line tracking for vehicle road, line following and landmarks recognition for local positioning. The developed techniques are implemented and have been tested using a fully automated vehicle platform for autonomous navigation in urban areas.

Even if the multi-agent paradigm has been evolving for fifteen years, the development of concrete methods for problem solving remains a major challenge. This paper focuses on reactive multi-agent systems because they provide interesting properties such as adaptability and robustness. In particular, the role of the environment, which is effectively where the system computes and communicates, is studied. From this analysis a methodology to design or engineer reactive systems is introduced. Our approach is based on the representation of the problem's constraints considered as perturbations to stabilize. Agents are then defined, in the second place, as a means of regulating the perturbations. Finally, the relevancy of our proposition is justified through the development of two solving models applied to real and complex problems.

Multi–agent systems can be considered as a part of distributed artificial intelligence approaches. They are particularly characterised by emerging global properties which stem from combinations of local interaction between individual pro–active elements called agents. Multi–agent systems have already been used in a wide range of applications but few applications integrate interaction with external devices and/or human beings. This paper presents a multi–agent–based application aimed, at producing emergent paintings based on both internal classical interaction and external triggering devices, such as music/sounds sources and motion captured by accelerometer sensors. The application results are computer generated paintings which can be classified as cyberart. The main challenge of this work is to obtain pertinent emergent paintings in relation with external trigger specificities.

Colloque avec actes et comité de lecture. internationale.; International audience; We present a vision based localization system for a mobile robot that uses a representative set of images obtained during an initial exploration of the environment. This set of images makes it possible to represent the environment as a partially markov decision process. The originality of this approach is the resulting data fusion process that uses both image matching and decision made by the robot in order to estimate the set of plausible position of the robot and the associated probabilities. Image matching or recognition is achieved using principal components analysis.

This work presents an approach to the obstacle avoidance problem, applicable in the frame of driver assistance. A decision, expressed as a proposed acceleration vector for the vehicle, is elaborated from the evaluation of a set of indicators characterizing the global state of a system of reactive agents (RMAS). Those agents evolve in a virtual environment produced on the base of vehicle’s perceptions of the material environment around it. Agent-to-agent and agent-to-environment interactions are defined in order to produce a distribution of agents over the virtual environment. This distribution, taken as the global state of the system, is analyzed by applying a set of indicators inspired from statistical physics, to calculate a new vehicle’s acceleration vector. This work presents the details of the RMAS model and its interaction laws, together with the global state evaluation functions. The approach has been applied to experimentation with a laboratory vehicle. Some experimental results are presented here.

Hardware in the loop (HIL) test provides a timesaving and safe environment for testing prototypes. However, the main difficulty for the real time simulation of power electronic system is the modeling method for the complex system. This paper proposes a modeling method for traction system in high speed train for transportation application. In this paper, it proposes hardware in the loop test setup for railway high-speed train with field-programmable gate array (FPGA) boards of dSPACE simulator. Besides, in order to meet the computing power requirement of the system modeling, a multi-processor system of dSPACE is achieved through Gigalink connection. The whole HIL system can be used to evaluate both the hardware and software performance of traction control unit (TCU). The real time simulation results under steady-state and transient conditions demonstrate modeling accuracy and provide detailed insight into the development of this vehicle.