Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm
This paper proposes a robot calibration method that uses an extended Kalman filter (EKF) and a neural network based on Levenberg–Marquardt combined accelerated particle swarm optimization (LMAPSO) to improve the accuracy of the robot’s absolute position. After the EKF optimizes...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9438679/ |
| _version_ | 1818671377442406400 |
|---|---|
| author | Ha Xuan Nguyen Hung Quang Cao Ty Trung Nguyen Thuong Ngoc-Cong Tran Hoang Ngoc Tran Jae Wook Jeon |
| author_facet | Ha Xuan Nguyen Hung Quang Cao Ty Trung Nguyen Thuong Ngoc-Cong Tran Hoang Ngoc Tran Jae Wook Jeon |
| author_sort | Ha Xuan Nguyen |
| collection | DOAJ |
| description | This paper proposes a robot calibration method that uses an extended Kalman filter (EKF) and a neural network based on Levenberg–Marquardt combined accelerated particle swarm optimization (LMAPSO) to improve the accuracy of the robot’s absolute position. After the EKF optimizes all geometric parameters, the robot position still contains non-geometric errors due to joint clearance, gear backlash, and link deflection that are impossible to model. Therefore, an artificial neural network model (ANN) is designed to compensate for these un-modeled errors. The Levenberg–Marquardt combined accelerated particle swarm optimization (LMAPSO) provides a robust optimization search algorithm to optimize the weight and bias of the neural network based on the training set. An experiment on a five-bar parallel robot shows that geometric and non-geometric calibration reduced the maximum absolute position error from (1.548 to 0.045) mm. The experimental results demonstrate the proposed calibration method’s effectiveness with the robot’s absolute position accuracy improving by 98%. |
| first_indexed | 2024-12-17T07:23:02Z |
| format | Article |
| id | doaj.art-e3aa288310a34289815105105f5c2372 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T07:23:02Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-e3aa288310a34289815105105f5c23722022-12-21T21:58:43ZengIEEEIEEE Access2169-35362021-01-019754157542510.1109/ACCESS.2021.30825349438679Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO AlgorithmHa Xuan Nguyen0https://orcid.org/0000-0002-0678-0808Hung Quang Cao1https://orcid.org/0000-0003-0689-3113Ty Trung Nguyen2Thuong Ngoc-Cong Tran3https://orcid.org/0000-0001-6211-7499Hoang Ngoc Tran4https://orcid.org/0000-0002-1401-3668Jae Wook Jeon5https://orcid.org/0000-0003-0037-112XDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaDepartment of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, South KoreaThis paper proposes a robot calibration method that uses an extended Kalman filter (EKF) and a neural network based on Levenberg–Marquardt combined accelerated particle swarm optimization (LMAPSO) to improve the accuracy of the robot’s absolute position. After the EKF optimizes all geometric parameters, the robot position still contains non-geometric errors due to joint clearance, gear backlash, and link deflection that are impossible to model. Therefore, an artificial neural network model (ANN) is designed to compensate for these un-modeled errors. The Levenberg–Marquardt combined accelerated particle swarm optimization (LMAPSO) provides a robust optimization search algorithm to optimize the weight and bias of the neural network based on the training set. An experiment on a five-bar parallel robot shows that geometric and non-geometric calibration reduced the maximum absolute position error from (1.548 to 0.045) mm. The experimental results demonstrate the proposed calibration method’s effectiveness with the robot’s absolute position accuracy improving by 98%.https://ieeexplore.ieee.org/document/9438679/Accelerated particle swarm optimizationrobot calibrationLevenberg-Marquardtneural network |
| spellingShingle | Ha Xuan Nguyen Hung Quang Cao Ty Trung Nguyen Thuong Ngoc-Cong Tran Hoang Ngoc Tran Jae Wook Jeon Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm IEEE Access Accelerated particle swarm optimization robot calibration Levenberg-Marquardt neural network |
| title | Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm |
| title_full | Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm |
| title_fullStr | Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm |
| title_full_unstemmed | Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm |
| title_short | Improving Robot Precision Positioning Using a Neural Network Based on Levenberg Marquardt–APSO Algorithm |
| title_sort | improving robot precision positioning using a neural network based on levenberg marquardt x2013 apso algorithm |
| topic | Accelerated particle swarm optimization robot calibration Levenberg-Marquardt neural network |
| url | https://ieeexplore.ieee.org/document/9438679/ |
| work_keys_str_mv | AT haxuannguyen improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm AT hungquangcao improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm AT tytrungnguyen improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm AT thuongngoccongtran improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm AT hoangngoctran improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm AT jaewookjeon improvingrobotprecisionpositioningusinganeuralnetworkbasedonlevenbergmarquardtx2013apsoalgorithm |