Obstacle recognition method and device and electronic equipment

Abstract

The invention provides an obstacle recognition method and device and electronic equipment, and the method comprises the steps: firstly obtaining a three-dimensional point cloud data set of the current position of a vehicle, dividing the three-dimensional point cloud data set into a plurality of grids according to the coordinates of each piece of point cloud data, the grids with high similarity form a target grid cluster according to the distribution information corresponding to the point cloud data contained in each grid, and finally, obstacle information is determined according to the point cloud data contained in the target grid cluster. Compared with the prior art, ground segmentation is not carried out on the point cloud data, the three-dimensional point clouds are directly mapped into the grids, the grids are clustered according to the distribution information, the distribution characteristics of all the three-dimensional point clouds are reserved, and the final obstacle recognition result has good robustness.

Description

technical field [0001] The present application relates to the technical field of automatic driving, in particular to an obstacle identification method, device and electronic equipment. Background technique [0002] Environmental perception plays an important role as the eyes of self-driving cars. During the process of automatic driving, the vehicle must perceive the surrounding environment in real time to make its own coping strategies. Lidar stands out among many sensors with the advantage of high precision. With the widespread popularization of Lidar in the field of autonomous driving, research on Lidar point cloud algorithms is in full swing. [0003] At present, there are two types of lidar obstacle perception: deep learning network algorithm detection and traditional algorithm detection. On the one hand, the deep learning network algorithm detection requires a large amount of data collection and labeling. On the other hand, if the labeled data does not cover the scene,...

AI Technical Summary

Problems solved by technology

On the one hand, deep learning network algorithm detection requires a large amount of data collection and labeling. On the other hand, if there are no scenes covered in the labeling data, the deep learning network will not be able to detect obstacles.

The first goal is to ensure the safety of self-driving vehicles in the process of driving. In view of the disadvantages of deep learning network detection algorithms, traditional algorithm detection can achieve the goal of all inspections and ensure driving safety. However, the process of traditional algorithm detection is cumbersome and the algorithm Complexity is currently facing a big test

Since automatic driving has relatively high real-time requirements for the system, traditional algorithms cannot respond quickly in the face of emergencies, posing a major safety hazard

Images