非平坦地形下移动机器人位姿精准估计研究  

Research on Accurate Estimation of Mobile Robot Posture in Unflat Terrain

作　　者：董作峰 叶玉刚 郑宪秋 秦冬冬 DONG Zuo-feng;YE Yu-gang;ZHENG Xian-qiu;QIN Dong-dong(School of Innovation and Entrepreneurship,Shanxi Institute of Technology,Shanxi Yangquan 045000,China;School of Computer Science and Information Security,Guilin University of Electronic Technology,Guangxi Guilin 541004,China)

机构地区：[1]山西工程技术学院创新创业学院,山西阳泉045000 [2]桂林电子科技大学计算机与信息安全学院,广西桂林541004

出　　处：《机械设计与制造》2024年第10期305-308,314,共5页Machinery Design & Manufacture

基　　金：国家自然科学基金青年基金项目—新疆缓斜巨厚煤层季节性地表水保护开采研究(52104140);阳泉市重点研发计划项目—煤矿掘进盾构机刀盘的设计与创新(2020YF047);阳泉市重点研发计划项目—基于ROS技术的四足机器人研发与应用(2020YF036)。

摘　　要：非平坦地形上移动机器人的姿态和运动变得更加复杂和多样化,导致机器人位姿估计精准度和效率降低,为此提出非平坦地形下移动机器人位姿精准估计方法。根据移动机器人的运动特性和约束条件获取运动学函数,将该函数输入Mahony算法,得到粗略的姿态估计值。将姿态估计值和里程计观测数据作为EKF算法的初始参数,通过滤波和融合处理进行位姿估计,结合PL-ICP点云匹配方法进一步更新位姿估计结果。将更新后的位姿估计作为EKF算法的先验估计,并与其他传感器数据进行融合,得到初步位姿估计结果。将初步位姿估计结果作为果蝇优化算法的优化目标,获取更加精准的移动机器人位姿估计结果。实验结果表明:所提方法能够在确保移动机器人稳定运行的基础上,有效提高其位姿估计的精准度和效率,取得了较好的应用效果。The pose and motion of mobile robots on uneven terrain have become more complex and diverse,leading to a decrease in the accuracy and efficiency of robot pose estimation.Therefore,an accurate estimation method of mobile robot posture in unflat terrain is proposed.The kinematic function is obtained according to the motion characteristics and constraints of the mobile robot,which is input into the Mahony algorithm to get a rough attitude estimate.The pose estimation and odometer observation data are taken as the initial parameters of the EKF algorithm,and the pose estimation is performed by filtering and fusion processing.The PL-ICP point cloud matching method is combined to further update the pose estimation results.The updated pose estimation is used as the prior estimation of EKF algorithm and fused with other sensor data to obtain the preliminary pose estimation results.The preliminary pose estimation results are used as the optimization target of fruit fly optimization algorithm to obtain more accurate pose estimation results of mobile robot.The experimental results show that the proposed method can effectively improve the accuracy and efficiency of pose estimation on the basis of ensuring the stable operation of the mobile robot,and obtain a good application effect.

关 键 词：移动机器人 运动学函数 位姿估计 EKF算法 PL-ICP点云匹配方法 果蝇优化算法 

分 类 号：TH16[机械工程—机械制造及自动化] TP242[自动化与计算机技术—检测技术与自动化装置]