Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators
Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2019-07-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fnbot.2019.00047/full |
| _version_ | 1818018854561906688 |
|---|---|
| author | Zhihao Xu Xuefeng Zhou Shuai Li |
| author_facet | Zhihao Xu Xuefeng Zhou Shuai Li |
| author_sort | Zhihao Xu |
| collection | DOAJ |
| description | Obstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By abstracting robots and obstacles into critical point sets respectively, the distance between the robot and obstacles can be described in a simpler way, then the obstacle avoidance strategy is established in form of inequality constraints by general class-K functions. Using minimal-velocity-norm (MVN) scheme, the control problem is formulated as a quadratic-programming case under multiple constraints. Then a deep recurrent neural network considering system models is established to solve the QP problem online. Theoretical conduction and numerical simulations show that the controller is capable of avoiding static or dynamic obstacles, while tracking the predefined trajectories under physical constraints. |
| first_indexed | 2024-04-14T07:45:13Z |
| format | Article |
| id | doaj.art-668635d5c6e5417ba00df79265951235 |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-04-14T07:45:13Z |
| publishDate | 2019-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-668635d5c6e5417ba00df792659512352022-12-22T02:05:22ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182019-07-011310.3389/fnbot.2019.00047466020Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant ManipulatorsZhihao Xu0Xuefeng Zhou1Shuai Li2Guangdong Key Laboratory of Modern Control Technology, Guangdong Institute of Intelligence Manufacturing, Guangzhou, ChinaGuangdong Key Laboratory of Modern Control Technology, Guangdong Institute of Intelligence Manufacturing, Guangzhou, ChinaSchool of Engineering, Swansea University, Swansea, United KingdomObstacle avoidance is an important subject in the control of robot manipulators, but is remains challenging for robots with redundant degrees of freedom, especially when there exist complex physical constraints. In this paper, we propose a novel controller based on deep recurrent neural networks. By abstracting robots and obstacles into critical point sets respectively, the distance between the robot and obstacles can be described in a simpler way, then the obstacle avoidance strategy is established in form of inequality constraints by general class-K functions. Using minimal-velocity-norm (MVN) scheme, the control problem is formulated as a quadratic-programming case under multiple constraints. Then a deep recurrent neural network considering system models is established to solve the QP problem online. Theoretical conduction and numerical simulations show that the controller is capable of avoiding static or dynamic obstacles, while tracking the predefined trajectories under physical constraints.https://www.frontiersin.org/article/10.3389/fnbot.2019.00047/fullrecurrent neural networkredundant manipulatorobstacle avoidancezeroing neural networkmotion plan |
| spellingShingle | Zhihao Xu Xuefeng Zhou Shuai Li Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators Frontiers in Neurorobotics recurrent neural network redundant manipulator obstacle avoidance zeroing neural network motion plan |
| title | Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators |
| title_full | Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators |
| title_fullStr | Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators |
| title_full_unstemmed | Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators |
| title_short | Deep Recurrent Neural Networks Based Obstacle Avoidance Control for Redundant Manipulators |
| title_sort | deep recurrent neural networks based obstacle avoidance control for redundant manipulators |
| topic | recurrent neural network redundant manipulator obstacle avoidance zeroing neural network motion plan |
| url | https://www.frontiersin.org/article/10.3389/fnbot.2019.00047/full |
| work_keys_str_mv | AT zhihaoxu deeprecurrentneuralnetworksbasedobstacleavoidancecontrolforredundantmanipulators AT xuefengzhou deeprecurrentneuralnetworksbasedobstacleavoidancecontrolforredundantmanipulators AT shuaili deeprecurrentneuralnetworksbasedobstacleavoidancecontrolforredundantmanipulators |