Arnaud Doniec

Distributed Constraint Reasoning Applied to Multi-robot Exploration

Exploration of an unknown environment is one of the major applications of Multi-Robot Systems. Many works have proposed multi-robot coordination algorithms to accomplish exploration missions based on multi-agent techniques. Some of these works focus on multi-robot exploration under communication constraints. In this paper, we propose an original way to formalize and solve this issue. Our proposal relies on distributed constraint satisfaction problems (disCSP) which are an extension of classical constraint satisfaction problems (CSP). Compared to other works, our proposal is fully distributed and guaranties the exploration of an unknown environment with maintenance of connectivity between all the members of a robots’ team.

Non-normative Behaviour in Multi-agent System: Some Experiments in Traffic Simulation

Most of the works related to norms and multi-agent systems focus on the design of normative agents systems making the assumption that agents always respect norms. Our aims in this article are (i) to discuss the relevance of this assumption in some specific contexts and to highlight some benefits of designing non-normative behaviour agents, (ii) to expound the methodology followed in a concrete application which consists in traffic simulation at junction. In particular, based on statistical traffic results, we show how non-normative behaviours contribute to improving the realism of simulation.

Metrics for the Evaluation of DisCSP: Some Experiments on Multi-robot Exploration

Many algorithms to solve Distributed Constraint Satisfaction Problems (DisCSP) have been introduced in the literature. In this paper, we propose to compare three different algorithms to solve DisCSP. Contrary to algorithms of the literature which are evaluated on graph coloring problems or uniform random binary DisCSPs, we use a multi-robot exploration problem. We show that, for this real world application, the comparison of algorithms may be improved by using additional metrics than those used in the literature. We will define other metrics that can be used for measuring different aspects of the multi-robot exploration problem. The aim of our attempt for defining metrics is to analyze and compare different aspects of complexity of this multi-robot problem. We will observe that using both classical and real world metrics is interesting to obtain a better and more precise comparison.

Dealing with Large MDPs, case study of waterway networks supervision

Inland waterway networks are likely to go through heavy changes due to a will in increasing the boat traffic and to the effects of climate change. Those changes would lead to a greater need of an automatic and intelligent planning for an adaptive and resilient water management. A representative model is proposed and tested using MDPs with promising results on the water management optimization. The proposed model permits to coordinate multiple entities over multiple time steps in order to avoid a flood in the waterway network. However, the proposed model suffers a lack of scalability and is unable to represent a real case application. The advantages and limitations of several approaches of the literature are discussed according to our case study.

Multi-agent Simulation Design Driven by Real Observations and Clustering Techniques

The multi-agent simulation consists in using a set of interacting agents to reproduce the dynamics and the evolution of the phenomena that we seek to simulate. It is considered now as an alternative to classical simulations based on analytical models. But, its implementation remains difficult, particularly in terms of behaviors extraction and agents modelling. This task is usually performed by the designer who has some expertise and available observation data on the process. In this paper, we propose a novel way to make use of the observations of real world agents to model simulated agents. The modelling is based on clustering techniques. Our approach is illustrated through an example in which the behaviors of agents are extracted as trajectories and destinations from video sequences analysis. This methodology is investigated with the aim to apply it, in particular, in a retail space simulation for the evaluation of marketing strategies. This paper presents experiments of our methodology in the context of a public area modelling.

Comportements anticipatifs dans les systèmes multi-agents : Application à la simulation de trafic routier

Multi-agent systems allow the simulation of complex phenomena which are not easily describable in an analytical way. This approach is often based on the coordination of agents whose actions and interactions involve the emergence of the phenomenon to be simulated. In this article, we expose a road traffic simulation issue. To answer this problem, we propose an anticipatory multi-agent behavioral model based on constraints networks processing. Our proposition is validated by simulating a real traffic situation: simulated data are compared to real data collected on the road.

Purchase Intention Based Model for a Behavioural Simulation of Sale Space

Simulation of retail space is a growing topic in the multi-agent systems community. Those systems vary depending on many issues, such as the type of store, or type of behaviour. The kind of issue that is wished to be simulated, or the type of data used to build the simulation, are other subjects of variations. Our ambition is to develop a simulator using a generic model, based on real data easy to collect. In this paper, we focus on the agent model. We develop it and make some experiments to test it.

Large Markov Decision Processes Based Management Strategy of Inland Waterways in Uncertain Context

Inland waterways management is likely to go through heavy changes due to an expected traffic increase in a context of climate change. Those changes are going to require adaptive and resilient management of the water resource. A representative model of the inland waterway has been proposed, using Markov decision processes to model the dynamic and uncertainties of the waterway. It is used to obtain an optimal plan for the distribution of the water on the network that takes into account the uncertainties arising for the operation of such networks. A subnetwork of the Hauts-de-France is modeled using this approach based on real data of traffic and water levels. The produced plans are tested on different scenarios under expected and unexpected conditions of traffic and climate to observe the quality and resilience of the generated plan during its execution. Simulations will show the advantages and limitations of such a modeling of the inland waterway network.

Towards Robots-Assisted Ambient Intelligence

An integrated network of mobile robots, personal smart devices, and smart spaces called “Robots-Assisted Ambient Intelligence” (RAmI) can provide for a more effective user assistance than if the former resources are used individually. Additionally, with the application of distributed network optimization, not only can we improve the assistance of an individual user, but we can also minimize conflict or congestion created when multiple users in large installations use the limited resources of RAmI that are spatially and temporally constrained. The emphasis of RAmI is on the efficiency and effectiveness of multiple and simultaneous user assistance and on the influence of an individual’s actions on the desired system’s performance. In this paper, we model RAmI as a multi-agent system with AmI, user, and robot agents. Moreover, we propose a modular three-layer architecture for each robot agent and discuss its application and communication requirements to facilitate efficient usage of limited RAmI resources. Our approach is showcased by means of a case study where we focus on meal and medicine delivery to patients in large hospitals.