基于深度学习的竞速型智能车转向控制系统设计

Design of steering control system of racing smart car based on deep learning

转向控制系统是智能车自主行驶的关键。依托第十五届全国大学生智能汽车竞赛,针对自主设计的竞速型智能车进行转向控制系统的设计。该控制系统由数据采集车完成数据采集的过程,包括采集电磁强度、图像、转向角等数据创建用于深度学习的数据集,在此基础上训练和测试神经网络模型,模型测试车采用经训练和测试的神经网络模型,结合实时电磁强度数据,实现对其转向的实时控制。经过实测,该控制系统能够有效控制转向,使智能车完成复杂路况下的竞速任务。

The steering control system is the key to autonomous driving of smart cars.Relying on the 15th National Students’smart car race,steering control system is designed for racing intelligent autonomous vehicle design.The vehicle control system by a data acquisition process of data acquisition is completed,including the strength of an electromagnetic data acquisition,image,created a depth steering angle learning data sets,on the basis of training and testing the neural network model is trained using test vehicle model and a neural network model testing,real-time binding strength of the electromagnetic data,real-time control of its steering.After measurement,the control system can effectively control the steering,the smart car racing to complete the tasks under complicated road conditions.