基于分布式模型预测控制的无人机编队控制

针对多四旋翼无人机编队在巡航飞行过程中队形形成和保持问题,采用分布式模型预测控制方法将该问题转化为在线滚动优化问题.建立线性时不变的编队运动模型,进而在考虑状态和输入约束,不考虑时延、外界干扰、噪声的情况下,利用领航跟随策略设计一种分布式模型预测控制器,通过引入自身和邻居的假设状态轨迹设计代价函数.其中邻居信息的交互是在有向、时不变通信拓扑结构下进行的.基于该控制器,无人机能够在跟踪目标轨迹的同时,快速形成预先设定的队形并保持队形飞行.通过引入终端等式约束保证系统稳定,进而将目标函数作为Lyapunov函数,给出编队系统渐近稳定的充分条件.最后,利用6架无人机仿真验证控制算法的有效性和优越性.

Formation control of multi-UAV based on distributed model predictive control algorithm

This paper presents a distributed model predictive control algorithm for the formation and maintenance of multi-quadrotor during the cruise fight. That is dealing with the formation control using the rolling optimization method. Firstly, a linear time-invariant formation motion model is established. Then using the leader-follower strategy, a distributed model predictive controller is designed by introducing the assumed state trajectory of itself and neighbors to the cost function, which is in the case of considering the state and input constraints, without considering the communication delay, external interference and noise. Unmanned aerial vehicles (UAVs) interact with local information based on a directional, time-invariant communication topology. Based on the controller, UAVs can quickly form a pre-set formation and maintain it while tracking the target trajectory. To ensure the stability of the system, the terminal equality constraint is introduced. Then taking the cost function as the Lyapunov function, the sufficient conditions for the asymptotic stability of the formation system are given. Finally, simulations with six UAVs demonstrate the effectiveness and superiority of the proposed algorithm.