基于2维激光雷达的小型地面移动机器人自主回收方法

发布时间：2018-06-23 01:22

本文选题：移动机器人 + 自主回收　；参考：《机器人》2017年05期

【摘要】：为了解决小型地面移动机器人在野外环境中的回收问题,利用2维激光雷达进行环境感知与末端精确引导,设计并实现了一个自主回收系统.首先,提出一种自适应曲率滤波算法预先对雷达数据进行滤波处理.然后,在检测引导目标过程中,结合基于密度和最近邻度量方法对散乱的数据进行聚类,并且利用引导目标和辅助目标之间的几何结构约束进行目标检测与匹配.最后,采用基于影响层划分和候选方向评估的方法进行实时避障与最优行驶方向选择,对机器人进行精确引导,从而实现机器人的自主回收.在野外环境中对该自主回收系统进行验证,对自主回收系统中的若干环节分别进行实验分析.实验结果表明,该方法能够有效地实现小型地面移动机器人在野外环境中的自主回收.
[Abstract]:In order to solve the problem of recovery of small mobile robot in the field environment, an autonomous recovery system is designed and implemented by using 2-D lidar for environment perception and terminal accurate guidance. Firstly, an adaptive curvature filtering algorithm is proposed to filter radar data in advance. Then, in the process of detecting the guided targets, the scattered data are clustered based on density and nearest neighbor metric, and the geometric structure constraints between the guided targets and the auxiliary targets are used to detect and match the targets. Finally, real-time obstacle avoidance and optimal direction selection are implemented based on impact layer partitioning and candidate direction evaluation, and the robot is guided accurately to achieve autonomous recovery. In the field environment, the autonomous recovery system is verified, and some links in the autonomous recovery system are analyzed experimentally. The experimental results show that this method can effectively realize the autonomous recovery of small mobile robots in the field.
【作者单位】：南京理工大学计算机科学与工程学院;
【基金】：国家自然科学基金青年基金(61603184) 核高基国家重点专项(2015ZX01041101) 江苏省社会安全图像与视频理解重点实验室创新基金(30916014107) 江苏省自然科学基金青年基金(BK20140794)
【分类号】：TN958.98;TP242

【相似文献】