Simulation scenario generation method and unmanned driving system test method

Abstract

This application relates to a simulation scenario generation method and an unmanned driving system test method, and the simulation scenario generation method is characterized by comprising the following steps of acquiring and processing scene description sentences to obtain a word vector sequence, wherein the scene description sentences include environment description required for generating a simulation scene and/or own vehicle behavior description; inputting the word vector sequence into scene element selection models to obtain a simulation scene element set output by the scene element selection models, wherein the scene element selection models include an encoder model and a decoder model; and generating a simulation scene based on the simulation scene element set. The application can output the scene element set corresponding to the word vector sequence by the scene element selection models according to the generalized scene description sentences, with no need to input accurate anddetailed scene descriptions to improve the applicability, and can generate a simulation scene based on the simulation scene element set to obtain a simulation scene with large randomness, thereby improving the complexity and comprehensiveness of the simulation scene.

Description

technical field [0001] The present application relates to the field of computer application technology, in particular to a method for generating a simulation scene and a method for testing an unmanned driving system. Background technique [0002] With the development of vehicle control technology, unmanned driving technology has emerged, that is, through the sensor system installed on the unmanned vehicle to perceive the road environment around the vehicle, and use the unmanned driving algorithm to detect the perceived road, vehicle position and obstacles. By processing the object information, the driving route can be automatically planned, and the driverless car can be controlled based on the driving route, so that the driverless car can reach the predetermined target location. [0003] To improve the safety of self-driving algorithms and self-driving cars, multiple tests of self-driving algorithms or self-driving cars are required. At the same time, in order to increase t...

AI Technical Summary

Problems solved by technology

[0004] However, in the process of implementation, the inventors found that there are at least the following problems in the traditional technology: the current simulation scene generation method requires human

Images