Robust Constrained Receding Horizon Control for Trajectory Planning

This paper presents a receding horizon controller (RHC) that can be used to design trajectories for an aerial vehicle operating in an environment with disturbances. Various constraints are imposed in the problem, such as turning rate limits and bounds on the vehicle speed, and target regions and no-fly zones are included in the environment. The proposed algorithm modifies these constraints to ensure that the on-line RHC optimization remains feasible even when the vehicle is acted upon by unknown, but bounded, disturbances. The approach uses a robust control invariant admissible set as a terminal set that does not need to be a target set of the overall guidance problem. This result extends previous work in two ways: the vehicle is guaranteed to remain safe under the influence of disturbances; and much longer robust trajectories can be constructed on-line. The full algorithm is demonstrated in several numerical simulations.