An accurate identification method of the position of a broken tooth in the joint steering gear of service robot. (January 2023)

• Record Type: Journal Article
• Authors: Shi, Zhaoyao; Zhang, Pan; Yue, Huijun; Lin, Jinyu; Ding, Hongyu
• Is Part Of: Measurement and control. Volume 56:Number 1/2(2023)
• Page Start: 311
• Page End: 326
• Publication Date: 2023-01
• Subjects: Joint steering gear -- transmission error -- positioning broken tooth -- measurement test bench -- wavelet-Kalman filtering; Automatic control -- Periodicals; Engineering instruments -- Periodicals; Production engineering -- Periodicals; 629.8
• DOI: 10.1177/00202940221075286
• Languages: English
• ISSNs: 0020-2940

Abstract

The gearbox of the small joint steering gear of a service robot is characterized by a compact structure, a large reduction ratio, and a special manufacturing and assembling technology. However, when a tooth in the gear is broken, the gear transmission can be still maintained, so it is difficult to identify the broken tooth accurately. In order to provide a theoretical basis for the identification of a broken tooth in the study, the influence of gear tooth fracture on the time-varying meshing stiffness in the transmission process was firstly analyzed theoretically and its influencing mechanisms on the time domain and frequency domain characteristics of transmission error (TE) were further analyzed. Then, the TE measurement test bench of the joint steering gear was designed. The signal-to-noise ratios and root-mean-square errors of the TE data processed by wavelet noise reduction, Kalman filtering and combined wavelet-Kalman filtering algorithms were compared and analyzed. The analysis results showed that the joint algorithm had the best performance. Next, the TE data measured under multiple working conditions were processed with the joint filtering algorithm to obtain the time domain results and then the frequency domain results were obtained by Fourier transform. Whether there was a broken tooth in the joint steering gear could be judged based on the difference between the maximum and minimum values of the TE in the time domain. The broken-tooth gear shaft could be positioned based on the frequency of TE mutation waveform. Furthermore, combined with the frequency domain, the accurate identification of the position of the broken gear can be realized. Finally, the accurate positioning method of broken teeth was applied in the product inspection process. Broken teeth were successfully identified and located, thus proving the effectiveness of the identification method.