A Cooperative Method for Multi Robots based on π-MaxQ
Zesu Cai · Intelligent Computer and Applications · 2013
Aimed to the problem in which there exist the little concurrency,low real time and efficiency,and the curse of dimensionality in multi robots' cooperation,a cooperative method for multi robots based on π-MaxQ is proposed in the paper.Firstly,the mental states such as common knowledge,belief,goal,intention,promise,trust,knowledge update,etc.,are defined.Secondly,the KQML communication primitives based on the mental states are expanded.Thirdly,the confederative rewarding function based on MaxQ is constructed.The experiments proves the validity of the proposed method.