| 引用本文: | 刘暾东,陆蒙,邵桂芳,王若宇.机器人减速器传动误差建模与优化[J].控制理论与应用,2020,37(1):215~221.[点击复制] |
| LIU Tun-dong,LU Meng,SHAO Gui-fang,WANG Ruoyu.Modeling and optimization of rotate vector reducer transmission error[J].Control Theory and Technology,2020,37(1):215~221.[点击复制] |
|
| 机器人减速器传动误差建模与优化 |
| Modeling and optimization of rotate vector reducer transmission error |
| 摘要点击 1845 全文点击 740 投稿时间：2018-07-25 修订日期：2019-04-03 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2019.80553 |
| 2020,37(1):215-221 |
| 中文关键词 RV减速器 等价模型 传动误差 参数辨识 粒子群算法 |
| 英文关键词 RV reducer Equivalent model transmission error parameter identification particle swarm optimization |
| 基金项目 厦门大学校长基金项目,福建省产学合作项目,厦门市科技项目 |
|
| 中文摘要 |
| 为提高工业机器人RV减速器的传动精度，合理分配各零件的加工和装配公差，本文提出一种基于等价模型的RV减速器传动误差建模与优化方法。该方法根据RV减速器的传动结构，构建17自由度的等价误差模型，利用传统经验参数进行求解，获得减速器仿真传动误差；同时，将仿真传动误差与实际测量传动误差进行对比，运用最小二乘法建立经验参数辨识模型；在此基础上通过粒子群算法优化辨识模型中的经验参数,将该参数运用到实际RV减速器生产中，结果显示：与传统经验参数建立的误差模型相比，本文提出的方法使得传动精度的仿真误差平均缩小9.99%，大幅度提高了等价误差模型的准确性。 |
| 英文摘要 |
| In order to improve the transmission accuracy of RV reducer used in industrial robot and assign the machining and assembly tolerance reasonably, this paper proposes a method for modeling and optimizing the transmission error of a RV reducer based on an equivalent model. According to the transmission structure of the RV reducer, this method constructs an equivalent error model of 17 degrees of freedom, which is solved by using traditional empirical parameters to obtain the simulation transmission error of the reducer. Then, we use the least squares method to build the parameter identification model by comparing the simulation transmission errors and experimental transmission errors. On this basis, the particle swarm optimization algorithm is used to optimize the empirical parameters in the identification model, and the parameters are applied to the actual RV reducer production. The result shows that compared with the error model established by the traditional empirical parameters, the method proposed in this paper reduces the simulation error of transmission accuracy by 9.99%, which greatly improves the accuracy of the equivalent model. |
|
|
|
|
|