论文部分内容阅读
针对移动机器人未知环境下的趋光控制问题,模拟人或动物“感知-行动”认知机制,对具有趋光特性的移动机器人进行设计,提出一种基于Boltzmann机神经网络的趋光控制方法.该方法首先应用知识集对机器人趋光控制器的Boltzmann机神经网络进行趋光训练;然后应用Boltzmann机神经网络的运行机制实现趋光控制.仿真实验表明,该方法能够提高机器人学习的控制精度.
Aiming at the problem of photomodulation control in unknown environment of mobile robots, the cognitive mechanism of human-animal or animal perception-action is simulated and a mobile robot with phototactic characteristics is designed. Based on Boltzmann neural network, Method is proposed in this paper.At the beginning of this method, the Boltzmann machine neural network is used to train the photommuno-dynamics of the robot controller, and then the operation mechanism of the Boltzmann machine is used to realize the photometric control.The simulation results show that this method can improve the control of robot learning Accuracy.