中文摘要：

针对水果采摘机器人关节伺服存在不确定性及其上界值无法测量等问题,提出了一种基于RBF网络的滑模变结构控制策略,通过RBF神经网络实现对不确定性的上界进行自适应预测,进一步改进了控制器的效果。在Mat Lab中进行仿真试验,结果表明：与传统的基于上界已知的滑模控制器相比,本文所提出的基于RBF网络的滑模变结构控制策略具有位置控制精度高、收敛速度快及抖动抑制能力强等特点,控制效果得到了进一步的提高。

英文摘要：

Pointing on the problem that there existed uncertainties in fruit harvesting robot joint servo and its upper bound value can not be measures easily, this paper proposes a sliding mode variable structure control strategy based on RBF net- work, in the proposed strategy the RBF neural network is used to predict the upper bound value of the uncertainties, the effect of the controller is further improved. In order to verify the effectiveness of the RBF-based adaptive sliding mode method, simulation experiment was carried out in MATLAB, the simulation results show that compared with the tradition- al sliding mode controller whose upper bound value is known, the proposed the sliding mode variable structure control strategy based on RBF network has high position control accuracy, fast convergence speed, good chattering reducing, also the control effect has been further improved.