Posts tagged "système"

IRIT - CLLE, Toulouse University

Keywords

Autonomous vehicles, self-adaptive multi-agent systems, driving control recovery.

Connected autonomous vehicles of level 3, called "conditioned automation", are vehicles in which the human driver delegates driving control in specific situations. During these situations, it may be necessary for the human to regain control of the driving activity. The main objective of this thesis is to develop a supervision system adapted to each driver, by integrating human factors, to allow a safe and efficient transition of two-way control between the human and the autonomous vehicle. For this, the system must identify, by self-observation and in real-time, situations in which the current driver will no longer be able to ensure driving. He must also provide a context for assessing the criticality of the situation as quickly as possible to anticipate and react to it as best as possible. The driving context is composed of indicators that characterize the elements that describe part of the driving process: human, vehicle, and environment. The system is based on self-adaptive multi-agentlearning systems.

Figure 1: Dynamic Learning using self-adaptive multi-agent system

Scientific goals

- Dynamic learning using multi-agent systems

- Generic approach to supervise the activity of a system

- Study the impact of the factors describing the different elements present in the system context on the ability of the system to converge towards a solution.

- Insure the acceptability of the system by human driver

Contacts

kristell.aguilar-alarcon_at_irit.fr, marie-pierre.gleizes_at_irit.fr, loic.caroux_at_univ-tlse2.fr

Context Presentation

The purpose of this project is to propose a cooperative agent model, based on the self-adaptive multi-agent system theory (AMAS), allowing an efficient and fast exploration of the parameter space, autonomously and automatically. This exploration should allow a continuous readjustment of the simulation until convergence, improving the control of the macro-level over the micro-level.

On an application standpoint, the purpose of this project is to produce a realistic traffic that satisfies the best a set of objectives and constraints at both micro and macro levels. This traffic should also allow interaction with humans and adapt to events that could occur in the virtual environment. 

Keywords

Self-adaptive Multi-agent Systems, Self-Calibration, Multi-Agent Simulation

Scientific goals

1. Enrich the AMAS theory with general learning mechanisms andstrengthen the coupling between micro and macro levels.

2. Propose a new generic calibration method of models.

3. Enrich GAMA tools

Contacts

damien.vergnet_at_irit.fr, frederic.amblard_at_ut-capitole.fr, elsy.kaddoum_at_irit.fr, nicolas.verstaevel_at_irit.fr

IRIT – Université Toulouse 1 – Capitole

 Keywords

calibration, simulation, multi-agent system, AMAS    

In many fields of science and engineering, simulation is a key part in understanding phenomena or predicting their future evolution. It is also a useful tool for planners and decision-makers in order to guide them in their decisions. A simulation is a computer model of a real-world system that contains entities in interaction and that is used to understand and/or predict the evolution of the system it represents.  

 In order to be as close as possible to real phenomena, simulation models have to be calibrated. Several different calibration methods exist, from classical optimization methods to multi-agent systems and data assimilation. As most simulation models are complex systems, usual calibration methods are not really appropriate to account for dynamics changes. One way to handle those changes is to tune parameters values while the simulation is running, using data observed on the real system. One additional constraint is that models are seen as black boxes, i.e. the calibration system has no insight of the inner workings of the model. This means that the calibration system has to learn the influence of each parameter on each model observation.    

This work proposes a new online calibration method, CALICOBA, based on adaptive multi-agent systems, that aims to solve these problems. It features two kinds of agents: parameters and objectives. The role of objective agents is to estimate the distance of observations from the real data. Parameter agents take these distances to evaluate the best value the corresponding model parameter should take in order to minimize the distance computed by objective agents. Parameter agents have to learn their influence on each objective agent in order to compute the best value.    

For instance, in the case of a simple traffic model with two parameters (maximum vehicle speed and reaction time) and two observables (traffic density and mean vehicle speed). Lets assume that there is a data source for the actual roads represented by this model. The role of CALICOBA would be to take in the observed data and compare it to each observable in order to compute a distance for each. With this information, the parameter agents have to determine in which direction to modify their values in order to decrease these distances.    

As this is still a work in progress, the system has yet to be tested on traffic models, but it is planned for the near future.

Scientific goals

- Online self-calibration

- Automatic learning of model input/output interactions

Contact

Damien Vergnet – damien.vergnet_at_irit.fr

Frédéric Amblard – frederic.amblard_at_irit.fr

Elsy Kaddoum – elsy.kaddoum_at_irit.fr

Nicolas Verstaevel – nicolas.verstaevel_at_irit.fr

Context Presentation

The development of sustainable smart cities requires the deployment of Information and Communication Technology (ICT) to ensure better services and available information at any time and everywhere. As IoT devices become more powerful and low-cost, the implementation of an extensive sensor network for an urban context can be expensive.

This thesis addresses the problem of estimating missing information in urban contexts. The objective is to estimate accurate environmental information where physical sensors are not available. The proposed solution, HybridIoT, uses the Adaptive Multi-Agent System (AMAS) to estimate accurate environmental information under conditions of uncertainty arising from the urban application context in which the project is applied, such as openness, heterogeneity and large-scale, which have not been explored by the state-of-the-art solutions.

Keywords

Smart city, Cooperative Multi-Agent Systems, Missing Information Estimation, Heterogeneous Data Integration

Scientific goals

- Limiting the number of ad hoc devices to be deployed in an urban environment

- The exploitation of heterogeneous data acquired from mobile, intermittent devices

- Real-time processing of information

- Self-calibration of the system

Contacts

davide.guastella@irit.fr, Marie-Pierre.Gleizes@irit.fr, valerie.camps@irit.fr

Context Presentation

The current digital transformation requires the creation of autonomous applications that can be adapted to complex, dynamic, heterogeneous, and unpredictable environments. These systems must be equipped with proactive learning capabilities.

To this end, Self-Adaptive Multi-Agent principle allows the decentralization and self-observation of the learning process. Each knowledge granule is an autonomous agent that cooperates with its neighbors to improve learning from exogenous and endogenous feedbacks. Detecting and solving concurrences, conflicts and incompetence leads to active endogenous learning.This work on an adaptive decentralized learning mechanism is applied to the learning of a robotic arm inverse kinematics.

 Keywords

Self-Adaptive Learning, Endogenous Learning, Adaptive Multi-Agent Systems, Robotics

Scientific goals

•    Design a Self-Learning System

•    Lifelong and Endogenous Learning

•    Genericity and Scalability

Contacts

bruno.dato@irit.fr, frederic.migeon@irit.fr, marie-pierre.gleizes@irit.fr

Context Presentation

This research project deals with energy efficiency in buildings to mitigate the climate change. Buildings are the highest source of energy consumption worldwide. However, a large part of this energy is wasted, mainly due to poor buildings management. Therefore, being accurately informed about consumptions and detecting anomalies are essential steps to overcome this problem. Currently, some software exists to record, store, archive, and visualize big data such as the ones of a building, a campus, or a city. Yet, they do not provide Artificial Intelligence (AI) able to automatically analyze the streaming data to detect anomalies and send alerts, as well as adapted reports to the different stakeholders.

The system designed in the SANDFOX project has for objective to fill this gap. To improve the energy management, an innovative system should aim at visualizing the streaming data, editing reports, and detecting anomalies, for different stakeholders, such as policy makers, energy man-agers, researchers, technical staff or end-users of these buildings.

The paper presents the User-Centred Design approach that was used to collect the required needs from different stakeholders. The developed AI system is called SANDMAN (semi-Supervised ANomaly Detection withMulti-AgeNt systems). It processes data in a time constrained manner to detect anomalies as early as possible. SANDMAN is based on the paradigm of self-adaptive multi-agent systems. The results show the robustness of the AI regarding the detection of noisy data, of different types of anomalies, and the scaling.

Keywords

Anomaly detection, dashboard, multi-agent system, smart buildings, energy management

Scientific goals

•    Anomalies detection in smart buildings streaming data by AI,

•    Restitution of the information to different stakeholders through an adapted dashboard.

Contacts

berangere.lartigue@univ-tlse3.fr, stephanie.combettes@irit.fr, marie-pierre.gleizes@irit.fr,  mathieu.raynal@irit.fr

Context Presentation

The theme addressed in this PhD concerns autonomous and connected vehicles. It is essential, after making a vehicle more reliable, to study how several connected autonomous vehicles will be able to interact to maximize the behavior of the collective (fluidity, fuel consumption and pollution). The Society of American Automotive Engineers (SAE International) has defined 5 levels of autonomy: level 1, in which the driver performs all maneuvers, at level 5, in which the vehicle is completely autonomous and can do without driver and / or passenger.This PhD concerns both level 4 autonomous vehicles in which the vehicle drives and the supervised driver can regain control of driving, as well as level 5 vehicles (total autonomy of the vehicle).

In this PhD, the aim is to study how each vehicle communicates with its neighboring vehicles and how it behaves / reacts in response to the information provided by its neighborhood. A lock consists in determining the information that is relevant to communicate among all the data recovered from the numerous sensors scattered in the vehicle and effectors. 

Keywords

Intelligent Transport System, Connected and Autonomous Vehicles, Multi-Agent System

Scientific goal

•    Learn how to use any data and which ones are the best to communicate

Contact

guilhem.marcillaud@irit.fr

Context Presentation

Soon, Air Traffic Control (ATC) will have to cope with a radical change in the structure of air transport. Apart from the increase in traffic that will push the system to its limits, the insertion of new aerial vehicles such as UAVs into the airspace, with different flight performances, will increase its heterogeneity. Previous work investigated the collision avoidance management problem using a decentralized distributed approach. To do so, an autonomous and generic multi-agent system has been proposed to address this complex problem.

The aim of this work is to test the genericity of the proposed multi-agent system (CAAMAS), already tested in simulation with airplanes, using it with real drones in ENAC’s drone aviary. Work carried out during the semester has allowed 1) to interface CAAMAS with ENAC drone software, Paparazzi UAV, 2) ease scenarists work by creating a user friendly interface to generate flight plans, 3) make CAAMAS easier to use by creating some configuration interfaces.Further work will focus on the trajectories modification of the agents to take into account the specificity of the UAV’s.

 Keywords

Autonomous UAV; Multi-Agent System; Self-Avoidance ; Self-Adaptation ; Artificial Intelligence ;

Scientific goals

•    Self-Organization of an autonomous vehicles trafic

Contacts

augustin.degas@irit.fr, elsy.kaddoum@irit.fr

Context Presentation

Connected autonomous vehicles of level 3 are vehicles in which the human driver delegates driving control in specific situations. During these situations, it may be necessary for the human to regain control of the driving activity.

The main objective of this thesis is to develop a system for the safe and efficient transition of two-way control between the human and the autonomous vehicle.For this, the system must identify by self-observation and in real time situations in which the current driver will no longer be able to ensure driving. He must also provide a context for assessing the criticality of the situation as quickly as possible in order to anticipate and react to it as best as possible. The driving context is composed of indicators that characterize the elements that describes part of the driving process. The system is based on self-adaptive multi-agent learning systems.

Keywords

self-adaptive multi-agent systems, autonomous vehicle

Scientific goals

•    Dynamic learning using multi-agent systems

•    Generic approach to supervise the activity of a system

•    Insure the acceptability of the system by human driver

Contacts

kristell.aguilar-alarcon@irit.fr, marie-pierre.gleizes@irit.fr, loic.caroux@univ-tlse2.fr

Présentation du contexte

La réalité d'un campus intelligent ou plus généralement d'une ville intelligente passe par une observation régulière de l'environnement par des capteurs ad-hoc, afin d’agir dans l’environnement avec des dispositifs automatiques pour améliorer le bien-être des usagers. Ces capteurs permettent d’obtenir une connaissance des activités humaines et des conditions dans lesquelles ces activités sont menées, mais le déploiement d'un grand nombre de capteurs peut être coûteux. Les coûts sont principalement liés à l'installation, la maintenance et les infrastructures de capteurs dans les bâtiments existants. Pour ces raisons, l’objectif de cette thèse vise à réduire ces coûts en utilisant quotidiennement des milliers d’informations partielles et intermittentes provenant de smartphones des usagers du campus de l’Université Toulouse III Paul Sabatier. Ces traitements sont fondés sur une technologie d’Intelligence Artificielle par systèmes multi-agents coopératifs.

Figure 1 : «On utilise les informations des dispositifs intermittents et mobiles pour fournir des estimations précises»

Objectifs scientifiques

- Apprendre à partir de données brutes, imprécises et intermittentes sans feedback.

- Fournir les informations en continu, même en l’absence de données de smartphone des usagers.

- Utiliser une approche hybride de l’Internet des objets qui mixe capteurs réels et capteurs virtuels.

Mots clés

Systèmes multi-agents auto-adaptatifs, fusion de données, apprentissage, smart campus

Contacts

Davide Andrea Guastella, Valérie Camps, Marie-Pierre Gleizes, {davide.guastella, camps, gleizes}@irit.fr