| Su Hongsheng (苏红升),Ding Wei,Lei Jingtao.[J].高技术通讯(英文),2021,27(2):193~199 |
|
| Joint position control of bionic jumping leg driven by pneumatic artificial muscle |
| |
| DOI：10.3772/j.issn.1006-6748.2021.02.010 |
| 中文关键词: |
| 英文关键词: pneumatic artificial muscle (PAM), bionic leg, radial basis neural network, position control |
| 基金项目: |
| Author Name | Affiliation | | Su Hongsheng (苏红升) | (School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, P.R.China) | | Ding Wei | (School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, P.R.China) | | Lei Jingtao | (School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, P.R.China) |
|
| Hits: 1433 |
| Download times: 1156 |
| 中文摘要: |
| |
| 英文摘要: |
| The bionic legs are generally driven by motors which have the disadvantages of large size and heavy weight. In contrast, the bionic legs driven by pneumatic artificial muscles (PAMs) have the advantages of light weight, good bionics and flexibility. A kind of bionic leg driven by PAMs is designed. The proportional-integral-derivative (PID) algorithm and radial basis function neural network (RBFNN) algorithm are combined to design RBFNN-PID controller, and a low-pass filter is added to the control system, which can effectively improve the jitter phenomenon of the joint during the experiment. It is verified by simulation that the RBFNN-PID algorithm is better than traditional PID algorithm, the response time of joint is improved from 0.15s to 0.07s, and the precision of joint position control is improved from 0.75° to 0.001°. The experimental results show that the amplitude of the change in error is reduced from 0.5° to 0.2°. It is verified by jumping experiment that the mechanism can realize jumping action under control, and can achieve the horizontal displacement of 500mm and the vertical displacement of 250mm. |
|
View Full Text
View/Add Comment Download reader |
| Close |
|
|
|
