Genetic programming for automatic design of self-adaptive robots

Autor:	Stéphane Calderoni, Pierre Marcenac
Rok vydání:	1998
Předmět:	business.industry Computer science Autonomous agent Foraging Evolutionary algorithm Genetic programming Tree (data structure) Factor (programming language) Robot Reinforcement learning Design process Artificial intelligence business Reinforcement computer computer.programming_language
Zdroj:	Lecture Notes in Computer Science ISBN: 9783540643609 EuroGP
DOI:	10.1007/bfb0055936
Popis:	The general framework tackled in this paper is the automatic generation of intelligent collective behaviors using genetic programming and reinforcement learning. We define a behavior-based system relying on automatic design process using artificial evolution to synthesize high level behaviors for autonomous agents. Behavioral strategies are described by tree-based structures, and manipulated by genetic evolving processes. Each strategy is dynamically evaluated during simulation, and weighted by an adaptative value. This value is a quality factor that reflects the relevance of a strategy as a good solution for the learning task. It is computed using heterogeneous reinforcement techniques associating immediate and delayed reinforcements as dynamic progress estimators. This work has been tested upon a canonical experimentation framework: the foraging robots problem. Simulations have been conducted and have produced some promising results.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::5935e1fe6d2cd2878661b430184f0360 https://doi.org/10.1007/bfb0055936 Zobrazit plný text záznamu