| 基于反步法的农业采摘机器人移动路径跟踪控制研究 |
| |
| 引用本文： | 张光磊,贾鹤鸣.基于反步法的农业采摘机器人移动路径跟踪控制研究[J].安徽农业科学,2014(23):8023-8026. |
| |
| 作者姓名： | 张光磊 贾鹤鸣 |
| |
| 作者单位： | [1]东北林业大学信息与计算机工程学院,黑龙江哈尔滨150040; [2]东北林业大学机电工程学院,黑龙江哈尔滨150040 |
| |
| 基金项目： | 中央高校基本科研业务费专项（DLl3BB04）. |
| |
| 摘 要： | 为实现农业采摘机器人路径跟踪控制,基于李雅普诺夫稳定性理论设计了一种反步法控制器。通过调节控制器参数抵消了机器人误差模型中的部分未知非线性项,简化了控制器的设计形式。利用神经网络对模型未知项和外界干扰进行估计,设计鲁棒自适应控制器在线补偿设计的神经网络的估计误差,提高了神经网络的学习精度。仿真试验结果表明,设计的控制器可实现采摘机器人在存在外界干扰作用下对期望路径的精确跟踪。
|
| 关 键 词： | 农业采摘机器人 路径跟踪 反步法 神经网络 |
Path-following Controller for Agricultural Picking Robot Based on Backstepping |
| |
| Institution: | ZHANG Guang-lei, JIA He-ming ( College of Information and Computer Engineering, Northeast Forestry University, Harbin, Hei- longjiang 150040; College of Mechanical and Electrical Engineering, Northeast Forestry University, Harbin , Heilongjiang 150040) |
| |
| Abstract: | In order to implement the path following control of agricultural picking robot, based on Lyapunov stability theorem, a backstepping method is designed. Through the tuning of the controller＇ s parameters, the unknown nonlinear terms in error dynamic robot model can be elim- inated. Thus the form of designed controller can be simpler. Neural network is adopted to compensate the model uncertainties and environmen- tal disturbances. An adaptive robust controller is designed to compensate the estimated error of neural network on-line, and the learning pre- cise of neural network can be improved. The simulation and experiment results illustrate that the controller can make agricultural picking robot track the desired arbitrary path precisely. |
| |
| Keywords: | Agricultural picking robot Path following Backstepping Neural network |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
