工业机器人误差分级解耦补偿与精度性能退化机理研究

以工业机器人构建智能制造系统已成为先进制造技术的主要趋势之一，但因现有误差补偿技术的限制，工业机器人仍无法满足高端制造领域的精度要求。为进一步提升工业机器人的精度性能，本项目重点研究工业机器人定位误差分级解耦补偿方法、关节位置敏感类误差补偿方法、以及机器人精度性能在线评估与预测性维护方法。首先研究基于自适应蒙特卡洛的参数辨识误差测量不确定度评价方法，根据评价结果优化各层级的参数标定策略，实现工业机器人误差分级解耦补偿；其次研究基于切比雪夫多项式的关节位置敏感类误差的建模方法，并探索性提出隐闭环标定方法实现该误差模型系数的高精度辨识方法；最后基于机器人运行状态参数分析研究工业机器人精度性能的退化机理，并研究基于深度学习的工业机器人精度性能在线评估与预测性维护方法。本项目研究成果有望提高工业机器人的精度性能，并促进工业机器人在高端制造领域中的应用发展。

The construction of intelligent manufacturing systems with industrial robots has become one of the main trends of advanced manufacturing technology. However, due to the limitations of existing error compensation technologies, the industrial robots can not satisfy the accuracy requirements of high-end manufacturing industries. In order to improve the accuracy performance of industrial robots, the main research of this project focuses on the positional error hierarchical-decoupling compensation method, the joint-position sensitive error compensation method, and the robot accuracy performance online assessment and predictive maintenance method. Firstly, the measurement uncertainty evaluation method of parameter identification error based on adaptive Monte Carlo is studied. Based on the evaluation results, the parameters calibration strategy of each level is optimized to realize the error hierarchical-decoupling compensation. Secondly, the model of joint-position sensitive error based on Chebyshev polynomial is researched. The implicit loop calibration method is explored and proposed to realize the high-precision coefficients identification of the error model. Finally, the mechanism of accuracy performance degradation of industrial robots is anlyzed based on the operating state information analysis. The online assessment based on deep learning and predictive maintenance methods of accuracy performance of industrial robots is studied. The resulting achievements of this project are expected to improve the accuracy performance of industrial robots effectively and promote the application and development of industrial robots in the high-end manufacturing industries.