Abstract
This paper investigates the time-varying formation tracking and collision avoidance problems for multi-unmanned ground vehicle (UGV) system which consists of omnidirectional vehicles on directed topology. To prevent the failure of collision avoidance caused by the lack of information interaction when the directed graph is not unilaterally connected, partial vehicles are elected as observers and their detection behaviors represented by sectors are realized by potential field and yaw control. Firstly, an improved repulsive potential field force is designed, and two composite potential field forces are proposed for single-to-single detection and single-to-multiple detection situations. Inspired by the equilibrium of forces, it is proven that the reconstructed potential field force which is not affected by the complexity of the collision situation can eradicate the local minima or deadlock states. Secondly, in order to generate detection behaviors according to a basic communication graph, an algorithm is put forward. Then, the protocol of position and velocity control, as well as the protocol of yaw control, are presented respectively. The former ensures the realization of formation tracking, while the latter, together with composite potential field forces, makes the detection behavior continuous. Moreover, according to Lyapunov-based theory, it is proven that the given method can achieve time-varying formation tracking and collision avoidance. Finally, simulations demonstrate the theoretical results.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call