Automatic driving method and system and vehicle

Abstract

The invention relates to an automatic driving method and system and a vehicle. The method comprises the following steps that a vehicle front image sequence and a vehicle speed sequence are obtained; processing the front image sequence by a pre-trained convolutional neural network to obtain a multi-frame image feature vector sequence, and connecting the multi-frame image feature vector sequence with low-dimensional features of the vehicle speed sequence to obtain and output a coding feature vector of each frame; a pre-trained long-term and short-term memory network processes the coding featurevector of each frame and a state vector obtained by processing the coding feature vector of the previous frame in sequence to obtain a driving instruction corresponding to the coding feature vector ofthe current frame; and controlling an executing mechanism of the vehicle to execute the driving instruction. The system is a carrier for implementing the method, and the vehicle comprises the system.According to the invention, the accuracy and the real-time performance of vehicle anthropomorphic automatic driving can be improved.

Description

technical field [0001] The present invention relates to the technical field of automatic driving, in particular to an automatic driving method, a system thereof, and a vehicle. Background technique [0002] In traditional autonomous driving, a modular approach based on a rule system is adopted, which is generally divided into several modules: perception, fusion, decision-making and control. Its advantage is that the tasks of each module are clear, and when there is a problem in the system, it can be quickly checked, and the reliability of the system is high. However, this solution relies on the fine design of each module, and the artificially designed system often cannot cover various driving scenarios, so its ability to deal with complex road conditions is limited. Moreover, the perception module of this scheme often requires a large amount of labeled data for modular deep neural network training, which requires a lot of manpower and material resources for data labeling. ...

AI Technical Summary

Problems solved by technology

Although the existing method 3 increases the speed prediction compared with the existing method 1, the existing method 3 can only realize the simple lane keeping function, and cannot cope with more complex driving environments

[0007] In summary, the existing end-to-end deep neural network technology for imitating driving behavior is not yet mature and needs further improvement

Images