基于自适应滤波的无人机编队防碰撞系统研究

Research on Anti-collision System of UAV Formation Based on Adaptive Filtering

针对无人机双机近距编队飞行时,易发生碰撞事故的问题,文中提出了有效控制的防碰撞算法,并建立了半物理仿真系统。采用自适应Kalman滤波为防碰撞控制算法,使编队的距离保持在安全范围之内。该算法通过预测僚机的状态,实时控制僚机机动,当编队达到警戒距离时,僚机迅速执行规避动作,双机分离。搭建了模拟编队飞行的半物理仿真平台,用Zigbee实行无线通讯,以超声波实现机间测距。实验结果表明,该系统能防止编队发生碰撞,并且有效抑制滤波发散,具有较好的鲁棒性和实时性。

Aiming to the problem of collision when two unmanned aerial vehicles are in close formation flight, a new algorithm of collision avoidance combined with the HIL （Hardware-In-loop） simulation platform is constructed. Use the adaptive Kalman filter algorithm to make the separation between two airplanes can be held in safety range. The action of the trailing aircraft is controlled by forecasting its condition through the algorithm, and then the evasive action of two aircrafts will be carried out to move away if the separation between two airplanes approaches the warning distance. The HIL platform is constructed to do the formation flight simulation and the wireless communication is achieved by Zigbee to use ultrasonic wave to measure the distance. Results from the experiment show that the collision avoidance can be effectively achieved by the system and filter divergent can be inhibited. Also the system has good robust and real time effect.