作者: Edward W. Large , Henrik I. Christensen , Ruzena Bajcsy
DOI: 10.1177/027836499901800103
关键词:
摘要: The dynamic-systems approach to robot path planning defines a dynamics of behavior in which task constraints contribute independently nonlinear vector field that governs actions. authors address problems arise scaling this handle complex behavioral requirements. They propose operates the space constraints, determining relative contribution each constraint dynamics. Competition among is able deal with when combining contributions, making it possible specify tasks are more than simple navigation. To demonstrate utility approach, design system two agents perform cooperative navigation task. show how competition enables make decisions regarding execute given situation, resulting execution sequences behaviors satisfy discuss scalability competitive-dynamics autonomous systems.