International Conference
| Static Friction Torque Estimation for Robot Manipulators Using a Data-Driven Approach | |
|---|---|
| Year | 2023 |
| Month | August |
| Journal | 2023 IEEE International Symposium on Sensorless Control for Electrical Drives (SLED) |
| Author | Jaehoon Shim; Sangwon Lee; Daesung Jeon; Jung-Ik Ha |
| Link |
http://ieeexplore.ieee.org/abstract/document/10261391
26회 연결
|
In analyzing dynamic multi-joint robots, the accuracy of model-based robot dynamics is significantly affected by various forces that are not adequately modeled in the joints. Among these factors, joint friction torque plays a crucial role. Numerous studies have focused on modeling joint friction, and this paper proposes joint friction models for multi-joint robots based on a data-driven approach. The key ideas are summarized into two aspects. First, comprehensive data acquisition is essential to utilize the data-driven model. This study covers a method to obtain data that includes comprehensive information which is insufficiently addressed in previous research. Second, previous studies using neural networks have only considered partial joint information. This paper proposes a model that utilizes joint velocity, torque, and temperature information together to develop a comprehensive friction model. The proposed friction torque models are verified using a real-world 6-axis collaborative robot. |
|
Seoul National University, 1, Gwanak-ro, Gwanak-gu,
Seoul, Republic of Korea, 08826 Bldg. 301, Rm. 606