Abdel-Illah Mouaddib

Integrating Human Recommendations in the Decision Process of Autonomous Agents

In order to help agents in a difficult situation, we integrate the human in the agents decision process. We develop a new model called Human Help Provider in a Markov Decision Process (HHP-MDP). We define HHP-MDP as a middle ground between an autonomous agent and a teleoperated agent called adjustable autonomy. The global approach of HHP-MDP is based on (1) a recommendation is translated into a partial policy, (2)then, this partial policy is translated into transition and reward functions associated to the agent Markov Decision Process. This new model is quickly integrated into the agents decision process.

Decentralized Markov Decision Processes for Handling Temporal and Resource constraints in a Multiple Robot System

We consider in this paper a multi-robot planning system where robots realize a common mission with the following characteristics: the mission is an acyclic graph of tasks with dependencies and temporal window validity. Tasks are distributed among robots which have uncertain durations and resource consumptions to achieve tasks. This class of problems can be solved by using decision-theoretic planning techniques that are able to handle local temporal constraints and dependencies between robots allowing them to synchronize their processing. A specific decision model and a value function allow robots to coordinate their actions at runtime to maximize the overall value of the mission realization. For that, we design in this paper a cooperative multi-robot planning system using distributed Markov Decision Processes (MDPs) without communicating. Robots take uncertainty on temporal intervals and dependencies into consideration and use a distributed value function to coordinate the actions of robots.

GREAT: a model of progressive reasoning for real-time systems

The problem of producing timely responses when faced with deadlines is an important issue for real-time systems. To deal with this problem different kinds of scheduling algorithms have been developed within the real-time community. Unfortunately, these approaches fail when applied to knowledge based systems, mainly because AI techniques rely on time-consuming algorithms with unpredictable, or highly variable performances. To face this problem researchers in AI have introduced deliberative techniques that enable to adapt the way a task is working in function of the available time. For this purpose approximation algorithms of two kinds have been developed: iterative refinement and multiple methods. The model we propose in this paper, belongs to iterative refinement methods. The main advantage of our approach is the capability of designing iterative refinement methods using a general rule language.<<ETX>>

Utilisation des processus décisionnels de Markov pour l’aide à la maîtrise d’une maladie animale

La maitrise dune maladie animale non reglementee est a linitiative des eleveurs et est parfois incitee par des organisations professionnelles pour ameliorer la situation sanitaire ou economique dune zone. Une organisation peut donc proposer a ses membres des recommandations pour maitriser la maladie, ces recommandations pouvant varier au cours du temps selon levolution de la situation epidemiologique dans la zone. Lenjeu est donc de pouvoir proposer des outils daide a la prise de decision et devaluer a priori limpact des decisions proposees sur la propagation dun agent pathogene en termes epidemiologiques (prevalence de la maladie) et economiques. Dans cet article, nous evaluons lapport des processus decisionnels de Markov (MDP). Nous proposons un modele de propagation intertroupeaux ou une action de maitrise est recommandee par un decideur collectif pour optimiser le cout de la maladie et de sa maitrise au niveau du groupe. Nous supposons que le decideur collectif connait la proportion deleveurs qui vont suivre sa recommandation. Lutilisation dun MDP integrant un modele epidemiologique permet dindiquer a chaque pas de temps sil faut faire une recommandation ou non selon la situation epidemiologique. La strategie obtenue consiste en des recommandations non systematiques. Bien que lobjectif soit doptimiser les couts, la prevalence dans la zone est aussi diminuee. La definition dune strategie adaptative est un avantage de notre approche qui permet de proposer des strategies non classiquement proposees et etudiees.