电光与控制, 2018, 25 (11): 36, 网络出版: 2018-11-25
改进APF与Bezier相结合的多无人机协同避碰航路规划
Multi-UAV Cooperative Collision Avoidance Route Planning Based on Improved Artificial Potential Field and Bezier Curve
多无人机 协同追踪 机间防碰 人工势场法 Bezier曲线 multi-UAV collaborative tracking inter-UAV collision avoidance artificial potential field Bezier curve
摘要
针对多无人机协同避碰航路规划问题, 提出了一种增加无人机自身斥力场的改进人工势场法(APF)与Bezier曲线相结合的方法。首先, 在给出了对无人机产生虚拟影响的一般人工势场后, 为了防止多无人机之间的碰撞, 定义了无人机自身产生的斥力势场, 由于该斥力势场产生的斥力函数具有分段连续的特点, 使无人机在防止机间碰撞时能有效规避频繁大转角和急促转弯现象的发生; 其次, 针对APF易陷入局部最优问题, 设计了一种基于虚拟障碍物的逃离局部最优方法; 最后, 提出分段Bezier曲线平滑算法对无人机的飞行轨迹进行在线实时平滑优化, 在两个分段Bezier曲线连接处保证了曲线的光滑性, 消除了航路规划中的路径振荡现象。仿真结果表明了所提方法的有效性。
Abstract
To address the issue of cooperative route planning for multiple Unmanned Aerial Vehicles (UAVs) to avoid collision, an improved Artificial Potential Field (APF) method combined with Bezier curve is proposed to increase the self-repulsion field of UAVs.Firstly, the general artificial APF that has virtual impact on the UAV is given.In order to prevent inter-UAV collision, the repulsive potential field generated by the UAV itself is defined.The repulsion function produced by the repulsion field is sectionallycontinuous, and thus the occurrence of frequent large angle flight and sharp turning can be effectively avoided while preventing inter-UAV collision.Secondly, to address the problem of APF local optimum, a method based on virtual obstacle is designed to avoid local optimum.Finally, a sectional Bezier curve smoothing algorithm is proposed to smooth and optimize the UAV flight path online in real time.The smoothness of the curve is guaranteed at the joint of the two sectional Bezier curves, and the path oscillation in the route planning is eliminated.The simulation results have proved the effectiveness of the proposed method.
杨勇, 丁勇, 黄鑫城. 改进APF与Bezier相结合的多无人机协同避碰航路规划[J]. 电光与控制, 2018, 25(11): 36. YANG Yong, DING Yong, HUANG Xin-cheng. Multi-UAV Cooperative Collision Avoidance Route Planning Based on Improved Artificial Potential Field and Bezier Curve[J]. Electronics Optics & Control, 2018, 25(11): 36.