摘要
通过对漂浮基空间机械臂冲击动力学的建模,计算出完成抓取操作后空间机械臂各关节的运动速度,从而为数值仿真计算提供初始条件.针对目标物体及空间机械臂系统惯性参数存在未知的情况,设计了冲击后空间机械臂及目标物体组合系统的载体姿态及关节运动滑模神经网络控制,并对平面自由漂浮空间机械臂系统抓取目标物体的操作及冲击后控制进行了数值仿真.仿真结果表明,空间机械臂抓取目标物体时的冲击对系统有重大的影响,所设计的控制方案可保证系统对期望运动的追踪.
To simulate the post-impact dynamics, the velocities of base's attitude and each joint after impact are derived from the impact model. Based on the results and considering the uncertain parameters, the NN control using sliding mode is designed for the free-floating space manipulator. In particular, it doesn't require position measurement, velocity nor acceleration of the base because of an effective exploitation of the particular property of the system dynamics during the derivation of the system dynamic model in the post-impact phase. The numerical simulation carried out confirms that the controller proposed is feasible and effective.
出处
《厦门理工学院学报》
2013年第2期29-33,共5页
Journal of Xiamen University of Technology
基金
福建省教育厅科技项目(JA10227)
厦门理工学院高层次引进人才科技项目(YKJ11011R)
关键词
空间机械臂
冲击动力学
滑模控制
神经网络
space manipulator
impact dynamics
sliding mode control
neural network