RECOLTE (2021/2024)

Tue, 24 Aug 2021 00:00:00 +0000

This work is carried out in the framework of a collaboration between the International Unit of Mathematical Modeling and Complex-Systems (UMMISCO) - attached to Sorbonne University and the Research Institute for Development (IRD), the Gastin Berger University from Saint-Louis in Senegal and the Multi-Agent System (MAS) group of LIP6.

The objective of this project is - building on previous work - to propose, design and implement a decisional architecture able to adapt the degree of autonomy of an artificial entity over time. In addition to the development of an architecture supporting the different modes mentioned, particular attention will be paid to the form that the transition from one mode (autonomous, semi-autonomous, remotely operated) to the other takes.

The platform, equipped with a set of embedded sensors (GPS, inertial unit, altimeter,..), a camera, a high performance camera and an M2M communication system, will have to evolve in a spatialized and geometrically constrained context in order to detect, identify, characterize and follow periodically and semi-automatically the targets present in the concerned area. The platform will have to be able to adapt dynamically to the elements encountered from an initial route provided off-line and to be able to switch, on-line, from an autonomous mode to a semi-autonomous or remotely operated mode (and vice versa) in a context of shared authority with a low-skilled operator.

Dem@t-Factory (2010-2012)

Tue, 27 Apr 2010 00:00:00 +0000

Supervising a system consists in the detection, isolation, identification and then repair of any fault that may occur in it. Nowadays, supervised systems are often distributed and communications are asynchronous.When communica- tions are disrupted, information takes an indefinite time to reach its destination, which slows down or stops the su- pervision process. The question is thus how to efficiently supervise a distributed system with unreliable communica- tions ? We claim that the answer to this question requires a distributed supervision system.In this document, we present ADS2, a multi-agent supervision architecture that is robust in a context of unreliable communications. ADS2 has four main characteristics :

Each agent can perform both di- agnosis and repair,
Each agent can find a balance be- tween a quick local diagnosis and repair, and a delayed, sys- temic one, based on the respective costs of misdiagnosis and communication,
Agents may form a coalition to restore a consistent view of the system state in case some had to act locally with incomplete information at an earlier stage.
Each agent is able to untertwine the diangostis and repair steps in order to reduce the size of the diagnosis search space and thus the time of malfunction.

The document develops our proposal along these four characteristics, and evaluates ADS2 using an industrial case-study.

Partners : Safig, A2iA, BancTec, Jouve, le laboratoire CEDRIC CNAM, le laboratoire A2iA ESIEE, LIP6, Temis

Autonomy | Cédric Herpson

RECOLTE (2021/2024)

Dem@t-Factory (2010-2012)