Biology Inspired Robot Behavior Selection Mechanism: Using Genetic Algorithm

• Published in: Bio-Inspired Computational Intelligence and Applications Vol. 4688; pp. 777 - 786
• Main Authors: Wang, Yiping, Li, Sheng, Chen, Qingwei, Hu, Weili
• Format: Book Chapter
• Language: English
• Published: Germany Springer Berlin / Heidelberg 2007; Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: Basal Ganglion; Behavior Selection; Genetic Algorithm; Mobile Robot Navigation; Salience Signal
• ISBN: 3540747680; 9783540747680
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-540-74769-7_82

Since behavior selection is a crucial issue not only in biology, but also in robotics, especially in behavior-based robotics, it is nature to consider the behavior selection problem both in biological view and robotic view. In recent years, accumulative evidences from neurobiology and anatomy have given rise to proposals that the basal ganglia-a group of subcortex nuclei in vertebrate brains- serve as a central selection mechanism. This paper introduces a robot behavior selection mechanism inspired by basal ganglia and makes explorations of applying genetic algorithm to the optimization of model parameters. The proposed method demonstrates its efficiency through a simulated robot foraging task and casts light on designing more intelligent and fluent behavior selection mechanism in the future.