基于模型预测控制的多无人机避碰航迹规划研究
Study on Collision Avoidance Path Planning for Multi-UAVs Based on Model Predictive Control

针对无人机编队内避碰问题，提出一种可实时机间避碰以及多碰撞冲突管理的多无人机协同航迹规划方案。通过对UAV编队避碰问题进行分析，采用分布式模型预测控制（decentralized model predictive control，DMPC）方法，将其转换为滚动在线优化问题。设计了避碰管理单元，采用交互图更新机制解决多碰撞管理问题，运用基于角度变化的协同避碰规则，解决了分布式避碰问题的动作一致性问题。仿真结果表明，该方案可以有效解决4架无人机间多碰撞冲突问题，同时代价更小。
To solve the collision avoidance problem of multiple Unmanned Aerial Vehicles(UAVs), a cooperative path planning scheme for UAVs with collision avoidance and multi-collision avoidance management is proposed. Based on the analysis of the collision avoidance problem, decentralized model predictive control(DMPC) is used to transform the collision avoidance problem into receding horizon control. To manage multiple collisions of each UAV, collision management unit is designed and reaction graph is presented. Rule for cooperative collision avoidance based on angle changing are put forward to solve the action consistency problem. The simulation results show that the proposed scheme can solve the multiple collisions problem of UAVs with lower costs