An effective robotic processing errors prediction method considering temporal characteristics

Contribution

In this paper, a PSO-ARIMA robot milling errors prediction method is proposed. The method can effectively predict robot milling errors with high accuracy, enabling online prediction and perceive in advance. The main conclusions are as follows:

1. This paper proposes the PSO-ARIMA method that integrates statistical principles and machine learning to accurately predict positional errors while maintaining interpretability. In four validation tasks, the maximum MAE of the prediction results for position error and posture error are 0.021mm and 0.011°, which fulfills the requirements of practical applications.
2. The preprocessing module for data alignment was designed to minimize alignment errors present in the raw data. This effectively reduces the alignment errors associated with widespread communication time fluctuations, resulting in improved accuracy of the error calculation.
3. The prediction is enhanced by using conditional milling force data, meanwhile the PSO algorithm enables reliable and high-quality setting of the model’s unknown parameters. In the future, more in-depth research will be conducted on the robot milling errors prediction, and consider the integration of statistical prediction methods and machine learning prediction algorithms to further simplify the prediction model and improve the prediction accuracy and stability.

Recommended citation: R. DENG, X. TANG, T. ZHANG*, F. PENG, J. YUAN, R. YAN, An effective robotic processing errors prediction method considering temporal characteristics, Journal of Advanced Manufacturing Science and Technology, 0 (2024) 0, https://doi.org/10.51393/j.jamst.2024010. https://doi.org/10.51393/j.jamst.2024010