成都大学学报（自然科学版）2026，Vol.45Issue(1)：58-65,8.DOI:10.3969/j.issn.1004-5422.2026.01.009
基于流形学习的多轴加工—机器人操作协同轨迹规划与奇异点规避算法
Multi-Axis Machining Robot Operation Collaborative Trajectory Planning and Singularity Avoidance Algorithm Based on Manifold Learning
摘要
Abstract
In multi-axis machining robot machining systems,the coupling between non-orthogonal motion chains and continuous rotation axes resulted in inherent kinematic redundancy and parameterized singu-larity in the configuration space,which cannot meet machining requirements.Therefore,a multi-axis ma-chining robot operation collaborative trajectory planning and singularity avoidance algorithm based on manifold learning was designed.A robot head coordinate system was established,and a multi-axis machi-ning robot operation trajectory kinematic model was constructed through robot rotation transformation and origin translation.Robot motion trajectory data were mapped to manifold space,an equivalent uncon-strained optimization objective function for trajectory planning within the manifold learning framework was constructed,and then the trajectory planning objectives under robot kinematic constraints were estab-lished.By using the matrix of the effect of robot operation speed on the end velocity,non-full-rank singu-larity in the machining robot operation trajectory could be avoided,thereby meeting the consistency re-quirement between the robot operation collaborative trajectory and the trajectory expectation.The final ex-perimental results showed that the linear velocities of the robot operating the X-axis,Y-axis,and Z-axis fluctuated within the ranges of 0～0.10 m/s,-0.15～0 m/s,and 0～0.5 m/s,respectively;the angular velocities of the X-axis,Y-axis,and Z-axis fluctuated within the ranges of 0～0.05 rad/s,-0.05～0 rad/s,and-0.03～0 rad/s.The planned values of linear velocities and angular velocities were close to the expected values,and the planning effect was good,which played an important role in high-quality ma-chining of robots.关键词
流形学习/多轴加工—机器人操作/协同轨迹规划/奇异点/规避算法/线速度/角速度Key words
manifold learning/multi-axis machining robot operation/collaborative trajectory planning/sin gularity/avoidance algorithm/linear velocities/angular velocities分类
信息技术与安全科学引用本文复制引用
陈永强,谷家坤..基于流形学习的多轴加工—机器人操作协同轨迹规划与奇异点规避算法[J].成都大学学报（自然科学版）,2026,45(1):58-65,8.基金项目
2023年度安徽省高校科学研究项目(2023AH052395) （2023AH052395）
芜湖市智能焊装工程技术中心项目(sgcjsyjzx06) （sgcjsyjzx06）
