基于改进BP 神经网络的移动机器人寻线控制神经网络系统具有自学习和自适应的能力，同时有很强的容错性和鲁棒性，适用于处理难于语言化的模式信息。为使移动机器人沿地面标志线自主运动，采用CCD 图像传感器与PC/104总线相结合的硬件系统，运用神经网络的模式识别功能，实现了机器人的寻线控制，实验结果表明该方法是可行的，能有效地提高移动机器人对环境的适应性和其智能化水平。- Neural Network has the ability of self-learning and self-adaptation, also has strong fault-tolerance and robustness, as well as adapts to deal with pattern information which is difficult to demonstrate by language. In order to let the robot move along guide lines, the CCD image sensor and PC/104 are used to compose hardware system, and applies pattern recognition function of neural network, then realizes line-tracking control of mobile robot. The experiment results indicate that the proposed method is practicable and can improve the adaptability and intelligence of mobile robot.