Automatic driving software debugging and testing system and method, medium and equipment

Abstract

The invention discloses an automatic driving software debugging and testing system and method, a medium and equipment. The system comprises a human-computer interaction interface module, an automatic driving data acquisition module, an automatic driving data archiving module, an automatic driving scene display module, an automatic driving data generation module, an automatic driving data write-back module and an automatic driving scene simulation display module. The method mainly comprises scene data recording and scene data playback. According to the invention, the recorded scene data is written into a message queue and a shared memory, so that the automatic driving software performs simulation operation based on the scene data, a user does not need to perform debugging in a real vehicle actually, the software debugging in a limited environment is avoided, and the situation that the scene data evans instantly and cannot be reproduced during the operation of the real vehicle is avoided; by recording the scene data and replaying the scene data afterwards, the scene with the problem at the moment can be reproduced infinitely theoretically, and the debugging and testing efficiency of the single module and the whole module of the automatic driving software can be effectively improved.

Description

technical field [0001] The present invention relates to automatic driving, in particular to an automatic driving software commissioning system, method, medium and equipment. Background technique [0002] The key technologies of self-driving cars include basic technologies such as the car's own hardware, software platform, system security platform, vehicle communication platform technology, and core algorithms. Among them, the upper layer of the autonomous driving software is the application layer, which runs the "core control algorithm" and "safety management". The core control algorithm of intelligent driving, including environment perception and positioning, path decision planning and vehicle control. At the same time, according to the requirements of intelligent driving, it realizes functional safety and information security management, including error monitoring, information security strategy, processor monitoring and implementation of safety correction, and safety warn...

AI Technical Summary

Problems solved by technology

If there is no sensor data that matches the current scene, algorithm engineers cannot efficiently locate and troubleshoot problems

[0004] The limitation of the existing technology is that the perception module of the automatic driving software can only be commissioned based on the scene environment image data, but other modules such as decision-making, planning, positioning, etc. can't do anything

Images