Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning
This paper describes the motion control of hyper redundant robots using a learning control scheme based on linear combination of error history. The learning control scheme is formulated with three elements: general solution of inverse kinematics with pseudo inverse of Jacobian matrix to achieve main...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Japan Society of Mechanical Engineers
2008-11-01
|
| Series: | Journal of Advanced Mechanical Design, Systems, and Manufacturing |
| Subjects: | |
| Online Access: | https://www.jstage.jst.go.jp/article/jamdsm/2/6/2_6_1011/_pdf/-char/en |
| _version_ | 1818107926238199808 |
|---|---|
| author | Daisuke MATSUURA Nobuyuki IWATSUKI |
| author_facet | Daisuke MATSUURA Nobuyuki IWATSUKI |
| author_sort | Daisuke MATSUURA |
| collection | DOAJ |
| description | This paper describes the motion control of hyper redundant robots using a learning control scheme based on linear combination of error history. The learning control scheme is formulated with three elements: general solution of inverse kinematics with pseudo inverse of Jacobian matrix to achieve main task, condition to achieve several subtasks and compensation by linear combination of obtained time history of output error. In order to make planar serial manipulator with hyper redundancy achieving main task, tracking of desired trajectory by its output link while performing obstacle avoidance as subtask, several subtask setting schemes to prevent from partially singular configuration caused by interference between main task and subtasks are proposed. The backward learning scheme is also proposed to obtain optimum initial configuration for the proposed learning control scheme. Several simulations and experiments with a planar 10R serial manipulator demonstrate the effectiveness of proposed control scheme. |
| first_indexed | 2024-12-11T02:07:13Z |
| format | Article |
| id | doaj.art-104352c10b8d48669e1ed8f105f4ab04 |
| institution | Directory Open Access Journal |
| issn | 1881-3054 |
| language | English |
| last_indexed | 2024-12-11T02:07:13Z |
| publishDate | 2008-11-01 |
| publisher | The Japan Society of Mechanical Engineers |
| record_format | Article |
| series | Journal of Advanced Mechanical Design, Systems, and Manufacturing |
| spelling | doaj.art-104352c10b8d48669e1ed8f105f4ab042022-12-22T01:24:21ZengThe Japan Society of Mechanical EngineersJournal of Advanced Mechanical Design, Systems, and Manufacturing1881-30542008-11-01261011102010.1299/jamdsm.2.1011jamdsmMotion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward LearningDaisuke MATSUURA0Nobuyuki IWATSUKI1Dept. of Mechanical Sciences and Engineering, Tokyo Institute of TechnologyDept. of Mechanical Sciences and Engineering, Tokyo Institute of TechnologyThis paper describes the motion control of hyper redundant robots using a learning control scheme based on linear combination of error history. The learning control scheme is formulated with three elements: general solution of inverse kinematics with pseudo inverse of Jacobian matrix to achieve main task, condition to achieve several subtasks and compensation by linear combination of obtained time history of output error. In order to make planar serial manipulator with hyper redundancy achieving main task, tracking of desired trajectory by its output link while performing obstacle avoidance as subtask, several subtask setting schemes to prevent from partially singular configuration caused by interference between main task and subtasks are proposed. The backward learning scheme is also proposed to obtain optimum initial configuration for the proposed learning control scheme. Several simulations and experiments with a planar 10R serial manipulator demonstrate the effectiveness of proposed control scheme.https://www.jstage.jst.go.jp/article/jamdsm/2/6/2_6_1011/_pdf/-char/enlearning controlhyper redundant roboterror historyvortex fieldobstacle avoidance |
| spellingShingle | Daisuke MATSUURA Nobuyuki IWATSUKI Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning Journal of Advanced Mechanical Design, Systems, and Manufacturing learning control hyper redundant robot error history vortex field obstacle avoidance |
| title | Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning |
| title_full | Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning |
| title_fullStr | Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning |
| title_full_unstemmed | Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning |
| title_short | Motion Control of Hyper Redundant Robots by Learning Control Based on Linear Combination of Error History and Initial Configuration Optimization with Backward Learning |
| title_sort | motion control of hyper redundant robots by learning control based on linear combination of error history and initial configuration optimization with backward learning |
| topic | learning control hyper redundant robot error history vortex field obstacle avoidance |
| url | https://www.jstage.jst.go.jp/article/jamdsm/2/6/2_6_1011/_pdf/-char/en |
| work_keys_str_mv | AT daisukematsuura motioncontrolofhyperredundantrobotsbylearningcontrolbasedonlinearcombinationoferrorhistoryandinitialconfigurationoptimizationwithbackwardlearning AT nobuyukiiwatsuki motioncontrolofhyperredundantrobotsbylearningcontrolbasedonlinearcombinationoferrorhistoryandinitialconfigurationoptimizationwithbackwardlearning |