| 基于神经网络及综合电机特性的柔性关节空间机器人全阶滑模控制 |
| |
| 引用本文： | 朱安,陈力.基于神经网络及综合电机特性的柔性关节空间机器人全阶滑模控制[J].福州大学学报(自然科学版),2019,47(6). |
| |
| 作者姓名： | 朱安 陈力 |
| |
| 作者单位： | 福州大学 械工程及自动化学院,福州大学 械工程及自动化学院 |
| |
| 基金项目： | 国家自然科学基金(11372073,11072061);福建省工业机器人基础部件技术重大研发平台资助项目(2014H21010011). |
| |
| 摘 要： | 研究了载体位置、姿态均不受控的自由漂浮柔性关节空间机器人机械臂轨迹跟踪及关节柔性振动主动抑制问题,利用系统动量、动量矩守恒关系及第二类Lagrange法,并综合考虑关节驱动电机特性,导出了其全系统动力学模型；然后基于奇异摄动理论将该动力学模型分解为描述机械臂刚性运动的慢变子系统及描述关节柔性振动的快变子系统,并结合关节电机输出力矩与电枢电流的关系,将对控制力矩的设计转化为对电流控制的设计。之后,针对关节柔性振动快变子系统,采用速度差值反馈控制方案对其进行了振动主动抑制；针对机械臂刚性运动慢变子系统,则基于RBF神经网络提出了一种全阶滑模控制方案；其中RBF神经网络用于逼近由系统不确定参数带来的未知非线性项,全阶滑模控制方案的引入在于使控制系统在兼备传统滑模控制方案鲁棒性强特点的同时,还能克服抖振问题。最后,系统数值仿真结果说明了所提方案的有效性。
|
| 关 键 词： | 柔性关节空间机器人 关节电机特性 奇异摄动理论 RBF神经网络全阶滑模控制 柔性振动主动抑制 |
| 收稿时间： | 2019/1/4 0:00:00 |
| 修稿时间： | 2019/2/2 0:00:00 |
Full Order Sliding Mode Control Based on Neural Network for a FlexibleJoint Space Robot Combine with Joint Motor DynamicsS |
| |
| Zhu An and Chen Li.Full Order Sliding Mode Control Based on Neural Network for a FlexibleJoint Space Robot Combine with Joint Motor DynamicsS[J].Journal of Fuzhou University(Natural Science Edition),2019,47(6). |
| |
| Authors: | Zhu An and Chen Li |
| |
| Affiliation: | College of Mechanical Engineering and Automation,Fuzhou University,College of Mechanical Engineering and Automation,Fuzhou University |
| |
| Abstract: | Track tracking and active suppression of flexible vibration for the free-floating flexible joint space robot combine with joint motor dynamics is studied. Based on the momentum conservation and the angular momentum conservation of the system, utilize the second Lagrange Equation and combine with joint motor dynamics, the system dynamic model is derived. Then, the model is decomposed into slow subsystems that describing the rigid motion of the manipulator and fast subsystems that describing the flexible vibration of the joints by using the singular perturbation theory. Combined with the relationship between output torque and armature current of joint motor, the design of control torque is transformed into the design of current control. For the fast subsystems, the velocity difference feedback control strategy is adopted to suppress the vibration actively. Because the RBF neural network can approximate the nonlinear term caused by the uncertain parameters of the system, full order sliding mode control can effectively overcome the chattering problem of traditional sliding mode control. Therefore, for slow subsystems, the full order sliding mode control strategy based on RBF Neural Network is adopted to realize trajectory tracking. Finally, the numerical simulation results demonstrate the effectiveness of the proposed control strategy. |
| |
| Keywords: | Flexible joint space robot Joint motor dynamics Singular perturbation theory RBF neural network full order sliding mode control Active vibration suppression of flexible vibration |
|
| 点击此处可从《福州大学学报(自然科学版)》浏览原始摘要信息 |
|
点击此处可从《福州大学学报(自然科学版)》下载全文 |
