基于鲁棒神经网络的水下机器人运动控制

针对水下机器人神经网络控制中系统响应速度慢及对噪声较敏感的问题,依据变结构控制理论,提出了一种基于鲁棒神经网络的水下机器人控制方法E利用指数趋近律,推导出神经网络参数的镇定算法,并采用标准误差反向传播(EBP)算法最小化目标函数,最后在水下综合探测机器人仿真平台上进行了试验研究E试验结果表明,该控制方法对神经网络学习率的改变和外界扰动有很强的鲁棒性,大大降低了机器人机械传动系统的磨损,且能够保证神经网络快速、稳定地学习,从而满足实时性控制的要求,具有较高的理论和实用价值.

Motion control of underwater vehicles based on robust neural network

Aiming at low response speed and sensitization to noises in control system of underwater vehicles by adopting neural network,a novel method of control based on the robust neural network is proposed according to the variable structure control theory.The parameter ballasting algorithm of neural network was deduced according to the exponent approach law,and the error back propagation(EBP) algorithm was adopted to minimize the object function.Finally,simulation experiments were carried out on general detection platform of underwater vehicle.Results show that the proposed method has good robustness to external disturbance and changing of learning-ratio.It can greatly reduce the abrasion of the mechanically-driven system,and keep learning of neural network fast and stably.It also meets the requirement of real-time control and has theoretical and practical value.