摘要
针对倒立摆系统,提出了在结构上可生长的神经网络控制方案。网络利用细胞生长结构算法,在工作域中实现对状态变量的模式分类,并通过新神经元的插入实现网络规模的生长演化。在输出域中针对倒立摆控制任务采用强化Hebb学习机制,实现不同的神经元以最佳方式响应不同性质的信号刺激。仿真表明,通过神经网络自身的发育,该方案有效控制了倒立摆系统。
A new approach of controlling the inverted pendulum by a growing neural network is presented. The network adopts growing algorithm from reference to Growing Cell Structures to perform the pattern classification in work field. This growing mechanism can he evolved through the continuous growing of the new nerve cell. And reinforcement Hebb Synaptic Modification is used as the self-learning method to make the neurons in difference fields to respond to the different stimulus in the best way. Finally, the experimental results show that the neural network scheme can interact autonomously with the environment and control the inverted pendulum effectively by growing manners of neural system itself.
出处
《微计算机信息》
北大核心
2005年第11S期91-93,共3页
Control & Automation
基金
国家自然科学基金(63075017)资助课题
关键词
细胞生长结构
Hebb学习
倒立摆
Growing Cell Structures
Hebb Learning
Inverted Pendulum
