This paper considers leader-following distributed control for nonholonomic mobile agents. The mobile agents consist of a group of wheeled mobile robots with differential drive configurations. With the aid of graph theory, various inter-agent communication scenarios are studied ranging from weakly connected to strongly connected graph. The distributed control scheme defined per agent consists of two levels; one level defines the joint-level velocity control which uses a gradient-based adaptive law to compensate for mobile robot agent dynamics and a sliding-mode control; another level defines the upper level which is responsible for Cartesian control. The consensus algorithm which executes the leader-following protocol resides within the Cartesian control. Several network topologies have been studied by applying the proposed Cooperative Control scheme.
展开▼
