计算机工程与应用2025，Vol.61Issue(24)：371-378,8.DOI:10.3778/j.issn.1002-8331.2501-0354
无地图环境下的四足机器人鲁棒运动规划方法
Robust Motion Planning Method for Mapless Environments of Quadruped Robots
摘要
Abstract
A hierarchical robust motion planning method is proposed to address the challenges of trajectory planning inef-ficiency and stability degradation arising from the intrinsic shape and anisotropic locomotion characteristics of quadruped robots in map-free complex environments.For target positioning in environments lacking complete maps,a polynomial MiniSnap algorithm combined with a weighted hybrid A*algorithm is employed at the front end to initialize the trajectory,effectively mitigating difficulties in trajectory tracking and reducing the tendency to become trapped in local optima.Meanwhile,to enhance both the efficiency and quality of local collision-free path searching,a weighting factor is integrated into the evaluation function,and constraints are imposed on yaw angle and lateral motion.Based on the initial trajectory obtained,the back-end adopts a strategy of optimizing B-spline control points by formulating an optimization model that considers constraints related to trajectory smoothness,collision avoidance,and dynamic feasibility,and iteratively refines the solution to provide a smooth,safe,and executable path for the quadruped robot.Experimental results show that the proposed method reduces planning time by 24%and execution time by 26.5%in obstacle-dense environments without prior maps,while demonstrating superior robustness compared to the mapless trajectory planning algorithm Ego-Planner.关键词
四足机器人/运动规划/轨迹优化/无地图自主导航Key words
quadruped robots/motion planning/trajectory optimization/mapless autonomous navigation分类
信息技术与安全科学引用本文复制引用
ZHANG Weiqi,ZHAO Yifan,ZHAO Liang,XIN Meiting..无地图环境下的四足机器人鲁棒运动规划方法[J].计算机工程与应用,2025,61(24):371-378,8.基金项目
国家自然科学基金(51209167,12002251) （51209167,12002251）
陕西省自然科学基金(2024JC-YBQN-0713,2025SYS-SYSZD-049). （2024JC-YBQN-0713,2025SYS-SYSZD-049）
