Data acquisition method and device for unmanned vehicle

Abstract

The present application discloses a data acquisition method and device for an unmanned vehicle. A specific implementation of the method includes: selecting the laser radar with the highest importance level as the first laser radar among at least one laser radar of the unmanned vehicle; obtaining the start time of the current time window and performing the following data processing steps: The point cloud data packets collected by each laser radar in the at least one laser radar after the start time of the current time window perform real-time acquisition and storage operations; determine whether any condition in the condition group is satisfied; in response to determining If any condition in the condition group is satisfied, the start time of the current time window is set as the current time, and the data processing step is continued. This embodiment realizes aligned storage of point cloud data packets collected by at least one lidar of the unmanned vehicle.

Description

technical field [0001] The present application relates to the technical field of motor vehicles, specifically to the technical field of unmanned vehicles, and in particular to a data acquisition method and device for unmanned vehicles. Background technique [0002] The control system of an unmanned vehicle relies on sensors to perceive the surrounding environment. In order to obtain more and more comprehensive information, unmanned vehicles are usually equipped with a variety of sensor devices. Among all sensors, lidar carries the most important functions such as vehicle positioning and obstacle detection. At present, different lidars are installed in different positions of unmanned vehicles, and multiple lidars have different coverage angles due to different positions. In order to achieve a better control effect, the control system of the unmanned vehicle will fuse the data collected by multiple lidars. [0003] At present, the industry mainly adopts the following two fu...

AI Technical Summary

Problems solved by technology

However, due to the different ranges and angles collected by different lidars, the most complete information can only be obtained by merging them together, and processing them separately will cause some point cloud data output by lidars to contain too little eigenvalue information, which greatly reduces the The accuracy of the algorithm recognition

[0006] Therefore, the existing multi-channel lidar data fusion technology has the problem of ghosting caused by data out-of-sync or too little eigenvalue information and low algorithm recognition accuracy.

Images