Training method, evaluation method, control method and device of automatic driving model

Abstract

The invention discloses a training method, an evaluation method, a control method and a device for an automatic driving model. The driving track and driving behavior of each vehicle in an actual traffic scene can be mapped through a simulation system, further the driving information of a large number of vehicles in the actual traffic scene can be obtained, and the training data size of the model is increased. In the process of training the automatic driving model, the average deviation value between each target simulation vehicle and each automatic driving vehicle is continuously calculated, the target automatic driving model can be determined in time when the average deviation value is smaller than the first threshold value, and a training environment does not need to be redeployed when the average deviation value is not smaller than the first threshold value. The method improves the training efficiency, reduces the calculation amount, greatly reduces the cost, and can be widely applied to the technical field of automatic driving.

Description

technical field [0001] The invention relates to the technical field of automatic driving, in particular to a training method, an evaluation method, a control method and a device for an automatic driving model. Background technique [0002] Autonomous driving technology usually includes technologies such as high-precision maps, environmental perception, path planning, and path tracking control. In path planning technology, the training of driving models is an important part of automatic driving, and the safe driving of automatic driving depends on the accuracy and stability of driving strategies. The effect of the driving strategy depends on the training quality of the automatic driving model. Therefore, how to improve the training quality of the automatic driving model is particularly important. [0003] In related technologies, when training an automatic driving model, it is necessary to drive autonomously to collect relevant driving data. This method results in high labor...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, when training an automatic driving model, it is necessary to drive autonomously to collect relevant driving data. This method results in high labor costs and a small amount of data. For example, only the driving data of a certain driver’s driving habits can be collected, and only a small amount of driving data can be collected. , which in turn affects the training quality of the autonomous driving model

[0004] In addition, when evaluating the training quality of the autonomous driving model in related technologies, the trained model is generally tested through test cases to verify whether the model is qualified. When the test case confirms that the model is unqualified, new training data needs to be collected for redeployment In the training environment, to train the model again, the workload is heavy and the work efficiency is low; moreover, the relevant technology cannot grasp the timing of the model test in time, and the test process is often started before the model is trained, and the numerous test processes further increase the work. quantity

Images