Robot swarm motion control with artificial potential functions and sliding mode technique

In this thesis motion control of a swarm of robots was studied. The control strategy developed in the thesis is a combination of artificial potential functions and sliding mode control technique. The members of the swarm move in 2-dimensional space with non-holonomic dynamics. In the thesis, artificial potential functions were used for constructing a virtual attraction-repulsion space for the swarm members and sliding mode control technique was used for forcing the agents to move according to virtual attractions and repulsions. The behaviors that were studied are aggregation, formation acquisition, and tracking. The tracking behavior was studied both for a single agent and for a swarm of agents. The proposed control strategy was developed via constructive analytical analysis and its effectiveness was tested with various simulations.