Increasing Feasibility with Dynamic Priority Assignment in Distributed Trajectory Planning for Road Vehicles

摘要

Nonconvex, nonlinear optimal control problems for large-scale networked control systems (NCSs) can be distributed to accelerate computation time. One distribution strategy is priority-based non-cooperative distributed model predictive control. A key problem in this strategy is the priority assignment to agents, as the priorities highly impact the solution quality and even determine if a feasible solution for all agents exists. This problem has been investigated in the domain of robotics for many years, and has recently been picked up in the domain of road vehicles. We propose a distributed reprioritization process for priority-based non-cooperative distributed model predictive control and prove recursive feasibility for the NCS. We develop a dynamic priority assignment algorithm for road vehicles, which we use in our reprioritization process. In our algorithm, each vehicle determines its priority in a distributed fashion. Inspired by an approach from the domain of robotics, we increase the priority of a vehicle with the number of potential collisions with other vehicles on its planned trajectory. We evaluate our distributed reprioritization process and our dynamic priority assignment algorithm in experiments. We compare the algorithm's performance to dynamic random priorities and to static priorities.