卷积神经网络(CNN)在汽车无人驾驶中的应用与分析

Application and Analysis of Convolutional Neural Network (CNN) in Driverless Vehicle

汽车无人驾驶技术在实际应用过程中会通过高精度的感应器监测交通环境,应用科学算法准确规划合理的行车路径,其中卷积神经网络在无人汽车驾驶技术应用当中有明显的研究价值。文章对卷积神经网络在汽车无人驾驶中的应用进行研究,综合概述卷积神经网络的发展情况,研究分析无人驾驶双目3D感知模型,无人驾驶物体检测模型,并综合说明3D感知与物体检测之间的关系,以期对卷积神经网络能够在汽车无人驾驶中得到有效应用。

In the practical application of the driverless vehicle technology,the traffic environment will be monitored by the high-precision sensor,and the reasonable driving path will be accurately planned by the scientific algorithm.Among them,convolutional neural network has obvious research value in the application of driverless vehicle technology.This paper studies the application of convolutional neural networks in driverless vehicles,summarizes the development of convolutional neural network(CNN),and analyzes the self-driving binocular 3D perception model and the driverless object detection model.The relationship between 3D perception and object detection is explained synthetically so that convolutional neural network can be effectively applied in driverless vehicles.