Test analysis method for automatic driving system

A technology for automatic driving and system testing, which is applied in vehicle testing, neural learning methods, machine/structural component testing, etc., and can solve problems such as large amount of test data

CN113074959AActive Publication Date: 2021-07-06BEIJING INSTITUTE OF TECHNOLOGYGY
9 Cites 6 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2021-07-06

Smart Images

  • Figure 1
    Figure 1
Patent Text Reader

Abstract

The invention discloses a test analysis method for an automatic driving system, and the method comprises the steps: loading to-be-evaluated data, and appointing an evaluation index and an evaluation scene which need to be evaluated; loading standard scene data as truth value data according to the evaluation scene; obtaining the characteristic quantity of the to-be-evaluated data; pairing the to-be-evaluated data with the truth value data according to the characteristic quantity consistency and / or the position consistency to form a data pair, wherein the position consistency refers to the consistency of positioning data in the to-be-evaluated data, the characteristic quantity consistency means that a characteristic quantity error is smaller than a threshold value; determining the matching degree of the data pair matching the evaluation scene and the evaluation index, and taking the data pair of which the matching degree exceeds a threshold value as an analysis object; and carrying out evaluation index analysis on the analysis object to obtain an evaluation result. The method can automatically process and analyze massive automatic driving data, and has the advantages of high efficiency, large data processing amount and accurate statistical indexes.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to a test and analysis method for data required by an automatic driving vehicle, and belongs to the technical field of automatic driving vehicle testing. Background technique

[0002] The testing technology of self-driving cars is one of its core technologies, which is of great significance to the performance improvement of self-driving cars. For complex autonomous vehicle application scenarios, the amount of test data involved is huge, and traditional automotive test analysis methods are difficult to meet its needs. Therefore, methods for automatic processing and analysis of large amounts of data in different scenarios of autonomous driving are current research hotspots. The automated testing and analysis of self-driving cars mainly focuses on two aspects: the first is how to efficiently obtain data of different working conditions and the establishment of a scene database; the second is how to automatically process and analyze ma...

Examples

Embodiment Construction

[0033] The method of the present invention will be further described in detail below in conjunction with the accompanying drawings. figure 1 For the overall process of the present invention, as shown in the figure, the main steps of the automated testing method provided by this system are as follows:

[0034] Step S1: Load the data to be evaluated, specify the evaluation indicators and evaluation scenarios required for evaluation; load the standard scene data according to the evaluation scenarios as the real value data.

[0035] This step obtains the evaluation indicators and evaluation scenarios to be evaluated from the input terminal, and loads the data to be evaluated.

[0036] Among them, the evaluation indicators include, for example, the position accuracy and speed accuracy of the detection target, the position accuracy of the vehicle body, the speed accuracy, and the rate of change of horizontal and vertical acceleration;

[0037] The data to be evaluated defines a uni...