期刊
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
卷 18, 期 8, 页码 5338-5347出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TII.2021.3125449
关键词
Manipulators; Kinematics; End effectors; Force; Jacobian matrices; Robots; Force control; Data driven; kinematic control; motion-force control; recurrent neural network (RNN); redundant manipulators
类别
资金
- National Natural Science Foundation of China [62176109]
- CIE-Tencent Robotics X Rhino-Bird Focused Research Program [2021-01]
- Natural Science Foundation of Chongqing, China [cstc2020jcyjzdxmX0028]
- Chinese Academy of Sciences Light of West China Program
- Natural Science Foundation of Gansu Province [21JR7RA531, 20JR10RA639]
- CAAI-Huawei MindSpore Open Fund, China [CAAIXSJLJJ-2020-012 A]
- Gansu Provincial Youth Doctoral Fund of Colleges and Universities [2021QB-003]
- Fundamental Research Funds for the Central Universities [lzujbky-2021-65]
- Education Department of Gansu Province: Excellent Graduate student Innovation Star project [2021CXZX-120]
- Supercomputing Center of Lanzhou University
This article proposes a new data-driven motion-force control scheme to address the problem of redundant manipulator control. The scheme uses a recurrent neural network to estimate the structure information and demonstrates excellent performance and practicality.
Redundant manipulators play a critical role in industry and academia, which can be controlled from the kinematic or dynamic perspective. The motion-force control of redundant manipulators is a core problem in robot control, especially for the task requiring keeping contact with objectives, such as cutting, polishing, deburring, etc. However, when a manipulator's model structure is unknown, it is challenging to take motion-force control of redundant manipulators. This article proposes a data-driven-based motion-force control scheme, which solves the motion-force control problem from the kinematic perspective. The scheme can take effect and estimate the structure information, i.e., the model parameters involved in the forward kinematics when the structure of the manipulator is incomplete or unknown. A recurrent neural network is devised to find the solution to the scheme. Besides, the theoretical analysis is presented to prove the correctness of the scheme. Simulations and physical experiments running on seven degrees of freedom redundant manipulators illustrate the superb performance and practicability of the scheme intuitively. The key contribution of this article is that, for the first time, a motion-force control scheme aided with data-driven technology is proposed from a kinematic perspective for the redundant manipulators.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据