针对目前路径跟踪控制检测方法精度低、实时性能差的问题,提出一种基于深度机器学习的构建树形神经网络CTNN(Constructing Tree shaped Neural Net)的深度学习算法.该算法通过深度机器学习,构建针对性强的学习集,同时在模型车中实现.将传统机器学习算法与文章所提出的算法在相同行驶条件下的实时响应进行比较,仿真结果表明,CTNN算法在恶劣的行驶环境中,实时性、鲁棒性均得到一定程度的提高.
Research on intelligent vehicle path tracking control based on CTNN algorithm
In order to solve the problems of low precision and poor real-time performance of current path track-ing control detection methods,a deep learning algorithm for constructing tree shaped neural network(CTNN)based on deep machine learning is proposed.Through the deep machine learning,a more targeted learning set are constructed in this algorithm,and the algorithm is implemented in the model car.The real-time response results of the traditional machine learning algorithm and the algorithm proposed in this paper under the same driving conditions are compared.The simulation results show that the real-time performance and robustness of the CTNN algorithm are improved to a certain extent in the harsh driving environment.
machine visionpath following controltree neural networkmodel car simulation
NETL
NSTL
万方数据
