Obstacle detection method and device, equipment and computer storage medium

Abstract

The invention provides an obstacle detection method and device, equipment and a computer storage medium. The obstacle detection method comprises the steps: data collected by a plurality of sensors atthe same moment are obtained, wherein the multiple sensors are mounted on a vehicle; the data collected at the same moment are combined, and a data set corresponding to the moment is obtained; and obstacle detection is conducted according to the data set, and obstacle information is obtained. The financial cost can be reduced, the information amount of the data can be enriched, the obstacle detection step can be simplified, the problem that the data in different formats are fused unsuccessfully is avoided, and thus safety of vehicle traveling is improved.

Description

【Technical field】 [0001] The present invention relates to the technical field of automatic driving, in particular to an obstacle detection method, device, equipment and computer storage medium. 【Background technique】 [0002] In the existing self-driving vehicles, in order to avoid the existence of blind spots in the self-driving vehicles and the failure to detect obstacles in the blind spots, multiple sensors are usually arranged on the self-driving vehicles to completely perceive the surrounding environment of the vehicle. all obstacles. [0003] However, when using multiple sensors for obstacle detection in the prior art, there will be the following problems: when multiple sensors of the same type are used for obstacle detection, if a relatively expensive sensor is used, although it can obtain a wider range and a larger amount of information Larger data, but it will consume huge financial costs. If relatively cheap sensors are used, although the consumption of financial ...

AI Technical Summary

Problems solved by technology

[0003] However, when using multiple sensors for obstacle detection in the prior art, there will be the following problems: when multiple sensors of the same type are used for obstacle detection, if a relatively expensive sensor is used, although it can obtain a wider range and a larger amount of information Larger data, but it will consume huge financial costs. If relatively cheap sensors are used, although the consumption of financial resources is less, it is impossible to obtain data with a wider range and more information; when using different types of sensors When performing obstacle detection, the output data of multiple sensors are usually fused. Therefore, the obstacle detection step is relatively cumbersome, and when the number of sensors is too large, it is easy to cause fusion failure. Once the data fusion fails, the vehicle cannot be detected. The surrounding obstacles will greatly affect the driving safety of the vehicle

Images