基于图像拟合的水面无人艇局部避障算法研究  

Study on Local Obstacle Avoidance Algorithm of Surface Unmanned Craft Based on Image Fitting

作　　者：徐小强[1] 陈涵 冒燕 XU Xiao-qiang;CHEN Han;MAO Yan(School of Automation,Wuhan University of Technology,Wuhan 430070,China)

机构地区：[1]武汉理工大学自动化学院,武汉430070

出　　处：《武汉理工大学学报》2022年第11期88-95,共8页Journal of Wuhan University of Technology

基　　金：军内计划科研项目(XK2019009);中央高校基本科研业务费专项资金(2019IVA045);山东省自然科学基金(ZR2020MF111).

摘　　要：针对水面无人艇全局避障算法建图速度缓慢而难以满足无人艇实时需求的现状,提出一种改进人工势场法进行局部路径规划。首先,针对电子海图的障碍物轮廓点稀疏且外轮廓不规则的问题,利用射线法和凹凸障碍物的转换思想处理点与障碍物之间的位置关系并进行环境建模;其次,根据转换的障碍物模型利用可行区域法进行可行区域与不可行区域的划分;然后,引入后视旋转避障的总体思想,结合可行区域法并利用直达法和一些典型改进思想进行无人艇避障算法的优化;最后,为满足无人艇制导算法直线化的需求,进行路径直线化来减少路径长度并再次削减路径点数量。仿真对比实验结果表明,算法的运行时间减少了45.6%,路径点数量减少了49.3%。Aiming at the current situation that the global obstacle avoidance algorithm of surface unmanned craft is slow and difficult to meet the real-time requirements of unmanned craft,an improved artificial potential field method is proposed for local path planning.Firstly,aiming at the problem that the contour points of obstacles in electronic chart are sparse and the outer contour is irregular,the position relationship between points and obstacles is processed and the environment is modeled by using the ray method and the conversion idea of concave convex obstacles;Secondly,according to the converted obstacle model,the feasible region method is used to divide the feasible region and the infeasible region;Then,the general idea of backward looking rotation obstacle avoidance is introduced,combined with the feasible region method,the direct method and some typical improved ideas are used to optimize the obstacle avoidance algorithm of unmanned craft;Finally,in order to meet the needs of linearization of UAV guidance algorithm,path linearization is carried out to reduce the path length and reduce the number of path points again.The simulation results show that the running time of the algorithm is reduced by 45.6%and the number of path points is reduced by 49.3%.

关 键 词：水面无人艇 局部避障算法 人工势场算法 可行区域法 图像拟合 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]