Test Methods for Autonomous Vehicles

Abstract

The present application discloses a test method for an unmanned vehicle, including: the centralized control server performs test scene initialization according to a preset test case; after the test initialization is completed, the centralized control server runs the test case, and at the same time Trigger the unmanned vehicle in the test scene to perform path planning according to the preset starting point and end point, travel according to the planned route and report the real-time status information of the vehicle; during the operation of the test case, for the test case For each test condition, the centralized control server performs corresponding scheduling and control on the simulation equipment according to the current real-time status information of the unmanned vehicle and the real-time status information of the simulation equipment in the test scene, and completes the test condition. Construct. By adopting the invention, the automatic test requirement of complex scenes can be met, and the realization cost of the simulation equipment is low.

Description

technical field [0001] The invention relates to mobile communication technology, in particular to a test method for unmanned vehicles. Background technique [0002] With the accelerated popularization of 5G technology, the rapid promotion of Internet giants, and the maturity of automotive intelligent technology, the pace of the autonomous vehicle industry will continue to accelerate, and it is expected to achieve commercialization in 2020. [0003] Unmanned vehicles must pass closed field test verification, and only after comprehensive, systematic and objective testing and evaluation of vehicle functions and performance can they drive autonomously on open roads. [0004] In the existing unmanned vehicle field test technology, the simulation equipment mainly simulates people, bicycles, cars and other interference objects, and simulates sudden jumping, overtaking, slowing down, sudden braking, merging, turning and other operations to test unmanned vehicles. Whether the drivin...

AI Technical Summary

Problems solved by technology

In this method, the intelligent simulation device needs to realize some functions of the unmanned vehicle, for example, it can sense the external environment and obtain some operating conditions of the external equipment (including unmanned vehicles and other simulation equipment) (such as the speed of both, relative distance, etc.), and adjust its own behavior accordingly, the function is very powerful, but the design is too complicated, which leads to the abnormally high cost of intelligent simulation equipment

In addition, during the testing process of this method, the intelligent simulation device needs to decide its own behavior according to the real-time interaction with the external environment, while the intelligent simulation device cannot accurately collect some information related to the unmanned vehicle in real time during the test process. data for the intelligent simulation device to control its own behavior based on these data, so that the various simulation behaviors of the intelligent simulation device during the test cannot be pre-set

Furthermore, for the test case in the same field, the behavior of the intelligent simulation device cannot be kept consistent in different test processes, which will lead to deviation or failure of the test results, and the test success rate is very low, which cannot meet the requirements of the unmanned vehicle field test. automation needs

[0007] It can be seen that the existing test solutions for unmanned vehicles have problems such as only simple scene tests, high implementation costs, low test success rates, and inability to meet the automation requirements of unmanned vehicle field tests.

Images