针对四旋翼无人机队形保持、编队避障问题，在双向、时不变通信拓扑结构下，基于领航跟随策略，利用安全攸关模型预测控制(safety-critical model predictive control，S-MPC)和一致性理论，设计编队控制器并实现了具有避障能力的队形保持。采用分散式S-MPC算法，每架无人机在满足避碰条件无人机的可行区域内，仅规划自身运动来跟踪编队控制算法指定的轨迹。研究了各解耦后的无人机如何与其他无人机并行求解带有耦合约束的优化问题，从而保证了各无人机独立决策的一致性。同时，所提算法将控制障碍函数(control barrier func-tion，CBF)引入到MPC控制器的约束中，从而保证无人机飞行在远离障碍物的安全集合内，规划出的轨迹更为平滑，减小了系统能耗。最后，通过仿真实验验证了所提方法的有效性。
Obstacles avoidance for quadrotor formation based on consensus theory and S-MPC
Aiming at the problem of quadrotors unmanned aerial vehicle(UAV)formation maintenance and obstacle avoidance,the safety-critical model predictive control(S-MPC)and consistency theory is proposed to design the formation controller to achieve formation maintenance with obstacle avoidance ability.By using the decentralized S-MPC algorithm,each UAV only plans its own motion to track the trajectory specified by the formation control algorithm within the feasible area that meets the collision avoidance conditions.This paper studies how each decoupled UAV solves the optimization problem with coupling constraints in parallel with other UAVs,so as to ensure the consistency of independent decision-making of each UAV.At the same time,the proposed algorithm introduces the control barrier function(CBF)into the constraints of the MPC controller,so as to ensure that the UAV flies in a safe set far away from obstacles,the planned trajectory is smoother,and the system is reduced energy consumption.Finally,the effectiveness of the proposed method is verified by simulation experiments.
model predictive controlcontrol barrier function(CBF)formation obstacle avoidanceconsensus algorithm
万方数据
维普
