论文部分内容阅读
针对智能车模型,提出了基于机器视觉的神经网络转向控制算法。该算法的输入为一张道路图的黑线位置值,输出为前轮转角。试验结果表明,这种算法能够很好地学习操作员给定的控制策略,具有较好的稳定性和鲁棒性。
For the smart car model, a neural network steering control algorithm based on machine vision is proposed. The input of this algorithm is the black line position value of a road map, the output is the front wheel corner. The experimental results show that this algorithm can well learn the control strategy given by the operator and has good stability and robustness.