Formation tracking control for unmanned vehicles using enhanced sequential convex programming-based model predictive control

This paper proposes an enhanced sequential convex programming-based model predictive control (ESCPMPC) scheme for formation tracking control problems. Considering coupled input constraints, a tracking error dynamic equation is established based on the position error between the leader and the follower, and a model predictive controller (MPC) is formulated for formation tracking. To improve the real-time control capability, we integrate MPC with sequential convex programming (SCP) by linearizing kinematics and convexifying obstacle avoidance constraints, thereby transforming the nonconvex optimization into a series of convex subproblems. While this approach efficiently approximates the solution to the original nonconvex problem, the linearization errors introduced during each SCP iteration can accumulate and potentially make the optimization problem infeasible. To address this issue, we propose an enhanced SCP (ESCP) method, which corrects these linearization errors. To ensure system stability, a terminal controller and a corresponding terminal set are computed. The recursive feasibility and stability of the proposed method are theoretically demonstrated. Finally, numerical simulations validate the effectiveness and computational efficiency of the proposed method in achieving formation tracking control for unmanned vehicles.