摘要
为了增强移动机器人在动态环境中的学习和适应能力,提出一种基于GA-BP算法优化的神经网络的具有学习的机器人行为控制方法。单纯的BP算法有易陷入局部极小、收敛速度慢的缺点,根据遗传算法具有全局寻优的特点,将二者结合起来形成一种训练神经网络的混合GA-BP算法。实际的实验结果显示,提出的方法对机器人的学习和适应能力有很大的增强,并且提高了机器人行为的准确性和快速性,可以有效、可靠地运用于机器人地面控制,并可以方便地应用于其他方面。
The problems of learning and evolutionary ability of robots to adapt to dynamic environments are discussed. A robot control algorithm with neural network training is formed by combining BP algorithm and genetic algorithm, according to the advantage of the globe optimal searching of genetic algorithm, in order to overcome the shortcomings that BP algorithm is usually trapped to a local optimum and has a low convergence speed. This algorithm can effectively and reliably be used in the robot control. The simulation results show that the algorithm has the advantages of time series prediction capability and can also effectively be used in other fields.
出处
《控制工程》
CSCD
北大核心
2009年第1期91-94,共4页
Control Engineering of China
基金
江苏省高校自然科学研究指导性计划基金资助项目(KK0410182)
