Safe coordination control policy for multiple input constrained nonholonomic vehicles

In this paper we propose a decentralized control policy for coordinating a group of nonholonomic autonomous vehicles (non necessarily homogeneous) with input constraints moving in ℝ
²
, aiming to satisfy individual (trajectory tracking) and common objectives (guaranteed vehicles safety). The proposed policy models the discrete operation modes of each participating vehicle as a hybrid automaton, while obtaining control inputs for each discrete mode from a particular model predictive control formulation or from a preselected constant control input. The proposed control policy results in partial fulfillment of the trajectory tracking problem and safety guarantees for all involved vehicles while preserving the overall system stability. The policy incorporates elements of model predictive control and hybrid system stability theory, more specifically stability results related to multiple Lyapunov functions.