Automatic driving simulation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an automatic driving simulation method and device, electronic equipment and a storage medium. The method comprises the steps of building a simulation scene; constructing vehicle modeling information of the target vehicle, wherein the vehicle modeling information comprises simulation sensor information, vehicle dynamics information and three-dimensional digital-analog information; running automatic driving simulation of the target vehicle, and collecting simulation sensor sensing data of the target vehicle in an automatic driving simulation running process; inputting simulation sensor sensing data into a sensing algorithm module for learning iteration, and / or inputting simulation sensor sensing data of a target vehicle and true value sensor sensing data obtained based on a simulation scene into an evaluation module, thus evaluating a sensing algorithm corresponding to the simulation sensor according to an error between the sensing data of the simulation sensor and the sensing data of the true value sensor. Therefore, safe and efficient learning and evaluation of the automatic driving perception algorithm are realized.

Description

technical field [0001] The present application relates to the technical field of intelligent vehicles, in particular to an automatic driving simulation method, device, electronic equipment and storage medium. Background technique [0002] Autonomous driving technology is at the forefront of current vehicle intelligence technology. Autonomous driving can reduce traffic accidents, reduce fuel consumption and energy consumption, and can fundamentally change road traffic and cargo transportation, benefiting the entire society. The development and verification of autonomous driving Work is highly valued by the industry. [0003] For advanced automatic driving, the perception algorithm, as an important part, is the "eye" to ensure the safe operation of automatic driving. [0004] However, in the early development process of autonomous driving perception algorithms, a large number of iterative learning and testing were required. Using the data collected by real vehicles for learni...

AI Technical Summary

Problems solved by technology

[0004] However, in the early development process of autonomous driving perception algorithms, a large number of iterative learning and testing were required. Using the data collected by real vehicles for learning and testing, it was necessary to build a test scene and mark a large number of objects in the test scene, which consumed a lot of manpower. Material resources, and some typical scenarios are very difficult to reproduce, and it is impossible to effectively verify the autonomous driving perception algorithm

[0005] In addition, when the perception algorithm is still immature, deploying it directly on the real vehicle may easily lead to traffic accidents and pose a safety hazard

Images