基于深度学习的无人驾驶关键技术的研究

Research ofDriverless and Key Technology Based on Deep Learning

近年来,无人驾驶领域快速发展,具有广泛的应用价值。本文论述了无人驾驶技术中的关键部分,采用多种模型进行路况信息提取和处理,介绍了卷积神经网络和MobileNet模型,配合TensorFlow框架来实现无人驾驶的深度学习。通过实验证实,该系统具有较好的适应性,在一定程度上满足了无人驾驶汽车的要求。

In recent years, the rapid development of the driverless area has wide application value. This paper discusses the key parts of the technology of unmanned driving, uses various models to extract and process road condition information, introduces the convolutional neural network and MobileNet model, and cooperates with the TensorFlow framework to realize the in-depth learning of unmanned driving. Experiments show that the system has good adaptability and meets the requirements of driverless vehicles to a certain extent.