改进Rao-Blackwellized粒子滤波的定位与建图方法及优化  

Improved Positioning and Mapping Method and Optimization of Rao-Blackwellized Particle Filter

作　　者：支奕琛 谷玉海[1,2] 龙伊娜 徐小力[1,2] ZHI Yi-chen;GU Yu-hai;LONG Yi-na;XU Xiao-li(Key Laboratory of Measurement and Control Technology,Ministry of Education,Beijing Information Science and Technology,Beijing 100192,China;Mechanical and Electrical Engineering,Beijing Information Science and Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学现代测控技术教育部重点实验室,北京100192 [2]北京信息科技大学机电工程学院,北京100192

出　　处：《机械设计与制造》2024年第9期251-257,261,共8页Machinery Design & Manufacture

基　　金：促进高校内涵发展--学科建设专项资助项目(5112011015)。

摘　　要：针对复杂地面场景下无人系统自主定位建图的难题,从机器人的建图角度出发,提出一种在复杂环境下机器人实现自我定位的方案,方案采用遗传算法中的交叉变异思想,对改进Rao-Blackwellized粒子滤波的同步定位与地图构建(RBPF-SLAM)算法中的重采样过程进行优化。经过仿真和自行设计的履带式无人系统实验平台进行实验验证,采用该方法优化后的SLAM算法,使得无人系统所构建的栅格地图在消耗较少粒子的情况下,能够绘制精度较高的栅格地图,达到了较好的自主定位和建图效果。Aiming at the problem of autonomous positioning and mapping of unmanned systems in complex ground scenes,from the perspective of robot mapping,a scheme for robots to realize self-positioning in complex environments is proposed.The scheme adopts the cross-mutation idea in genetic algorithm,which improves Rao.The re-sampling process in the synchronization posi⁃tioning and map construction(RBPF-SLAM)algorithm of the Blackwellized particle filter is optimized.After simulation and self-designed crawler unmanned system experimental platform for experimental verification,the SLAM algorithm optimized by this method enables the grid map constructed by the unmanned system to be drawn with higher accuracy while consuming less par⁃ticles.The grid map achieves a better effect of autonomous positioning and mapping.

关 键 词：移动机器人 RBPF-SLAM 定位建图 

分 类 号：TH16[机械工程—机械制造及自动化] TP3[自动化与计算机技术—计算机科学与技术]