论文部分内容阅读
针对机器人动力学的非线性、强耦合、时变特点，为了高速、高精度地控制机器人，提出了基于神经网络的机器人智能控制策略．讨论了机器人动力学模型的特殊性，指出模型中结构和非结构不确定性因素；研究了一种新型串级变结构神经网络（ＣＶＳＮＮ）；提出了一种基于新型ＣＶＳＮＮ的机器人控制策略．以精密Ｉ号机器人为仿真对象，证明该算法能够较好地补偿结构与非结构不确定因素，收敛速度快，能够自适应地调节自身的结构．
Aiming at the nonlinear, strong coupling and time-varying characteristics of robot dynamics, in order to control the robot with high speed and high precision, a robot intelligent control strategy based on neural network is proposed. The particularity of the robot kinematics model is discussed. The structural and unstructured uncertainties in the model are pointed out. A novel cascade variable structure neural network (CVSNN) is studied. A novel control strategy based on the new CVSNN is proposed. The precision robot I is used as the simulation object, which proves that the algorithm can compensate the structural and unstructured uncertainties well, converges fast and adjusts its structure adaptively.