| 引用本文: | 陈迪赛,高健,罗于恒,张揽宇.五轴混联机构时间最优速度规划及精准插补[J].哈尔滨工业大学学报,2024,56(8):112.DOI:10.11918/202312064 |
| CHEN Disai,GAO Jian,LUO Yuheng,ZHANG Lanyu.Time-optimal velocity planning and accurate interpolation for five-axis hybrid mechanism[J].Journal of Harbin Institute of Technology,2024,56(8):112.DOI:10.11918/202312064 |
|
| 摘要: |
| 为提高五轴混联机构对复杂曲面及其特征的加工质量,提出一种准确识别曲率点的时间最优速度规划及精准插补方法。首先,采用向心法生成路径节点参数,作密集化处理生成速度节点参数；其次,用B样条描述速度曲线,通过伪加加速度将三阶约束模型转变为线性模型；然后,以速度节点参数为分段点,构造各段路径的时间积分函数,采用自适应辛普森积分法求解；最后,将所得离散点信息输入到运动控制卡进行轨迹加工。实验结果表明:本方法可快速收敛到全局最优,降低曲率点速度并迅速回升其邻域速度；计算所得路径总时间仅有3.5 μs的波动,与常规插补方法收敛后的精度一致,且收敛时间更短；位置和角度的最大插补误差分别为0.76×10-3 mm和0.9×10-3°,小于编码器分辨率,符合加工需求,特别是极大曲率处相比一般方法降低95.37%。因此,在路径曲线基础上生成的速度节点参数具备更好的曲率识别性,通过构造各段路径的时间积分函数求解路径总时间具备更高的鲁棒性与效率。 |
| 关键词: 五轴混联机构 时间最优速度规划 曲率 B样条 插补 |
| DOI：10.11918/202312064 |
| 分类号:TP273 |
| 文献标识码:A |
| 基金项目:国家自然科学基金 (U20A4,6) |
|
| Time-optimal velocity planning and accurate interpolation for five-axis hybrid mechanism |
|
CHEN Disai,GAO Jian,LUO Yuheng,ZHANG Lanyu
|
|
(State Key Laboratory of Precision Electronic Manufacturing Technology and Equipment (Guangdong University of Technology), Guangzhou 510006, China)
|
| Abstract: |
| To improve the processing quality of complex curved surfaces and their features in a five-axis hybrid mechanism, a time-optimal speed planning and precise interpolation method for accurately identifying curvature points is proposed. First, the centripetal method is used to generate path node parameters, and intensive processing is performed to generate speed node parameters. Secondly, a B-spline is used to describe the velocity curve, and the third-order constraint model is transformed into a linear model through pseudo-jerk. Then, the speed node parameters are used as segmentation points to construct the time integral function for each segment path, and an adaptive Simpson’s integration method is used to solve the function. Finally, the obtained discrete point information is input to the motion control card for trajectory processing. The experimental results show that the proposed research model in this paper can quickly converge to the global optimum, effectively reduce the curvature point velocity and quickly recover its neighborhood velocity. The total path time calculated only fluctuates by 3.5 μs, comparable to the accuracy achieved by conventional interpolation methods but with a shorter convergence time. The maximum interpolation errors of position and angle are 0.76×10-3 mm and 0.9×10-3° respectively, which are smaller than the encoder resolution and meet the processing requirements. Specifically, at the extreme curvature, the proposed method reduces the interpolation error by 95.37% compared to conventional approaches. Therefore, the speed node parameters generated based on the path curve have better curvature identification, and the total path time calculated by constructing the time integral function for each segment path has higher robustness and efficiency. |
| Key words: five-axis hybrid mechanism time-optimal velocity planning curvature B-spline curve interpolator |
