L-ALLIANCE - TASK-ORIENTED MULTIROBOT LEARNING IN BEHAVIOR-BASED SYSTEMS

A large application domain for multi-robot teams involves task-oriente d missions, in which potentially heterogeneous robots must solve sever al distinct tasks. Previous research addressing this problem in multi- robot systems has largely focused on issues of efficiency, while ignor ing the real-world situated robot needs of fault tolerance and adaptiv ity. This paper addresses this problem by developing an architecture c alled L-ALLIANCE that incorporates task-oriented action selection mech anisms into a behavior-based system, thus increasing the efficiency of robot team performance while maintaining the desirable characteristic s of fault tolerance and adaptivity. We present our investigations of several competing control strategies and derive an approach that works well in a wide variety of multi-robot task-oriented mission scenarios . We provide a formal model of this technique to illustrate how it can be incorporated into any behavior-based system.