Lidar rotation control method for target detection

Abstract

The invention discloses a laser radar rotation control method for target detection. The laser radar rotation control method comprises the steps that S1, a laser radar is installed on an operating platform to form a detection system, the change trends acquired at different operating speeds of the operating platform that point cloud distributions change with the scanning speed of the laser radar areobtained; S2, when the detection system performs target detection, the operating speed of the operating platform is acquired in real time, the change trend of corresponding point distribution is obtained according to the operating speed acquired in real time, and the current required laser radar scanning speed is determined according to the change trend of the point distribution; S3, the laser radar is controlled to rotate according to the scanning speed of the laser radar determined in the step S2 so as to complete target detection. The laser radar rotation control method has the advantagesof being simple to achieve, small in working blind zone and large in effective detection distance and making scanned point cloud data rich.

Description

technical field [0001] The invention relates to the technical field of target detection in unmanned systems, in particular to a laser radar rotation control method for target detection. Background technique [0002] As a new type of active ranging sensor, lidar has high measurement accuracy, long detection distance, and is not affected by changes in ambient light. It can quickly obtain target information complementary to images, and is especially suitable for obstacle detection in unmanned systems. . Pedestrian and vehicle detection has always been a hot topic in the research fields of intelligent transportation, unmanned vehicles, and automobile safety assisted driving. Pedestrian detection based on visual information is easily affected by illumination, target attitude changes, and occlusion, which reduces its detection accuracy; In the realistic and complex traffic environment, the higher the accuracy of target detection and the farther the effective detection distance, t...

AI Technical Summary

Problems solved by technology

[0003] For the target detection of laser radar, it is usually devoted to the research on how to use the point cloud information obtained by laser radar to realize the accurate detection of obstacles. There is no research on the quality of the obtained point cloud information. However, in the case of unmanned vehicles In the actual application of the laser radar, there is a working blind spot and the farther the target is from the laser radar, the sparser the surface point cloud obtained is. The sparse point cloud will further reduce the effective detection distance of the system, and the low-quality point cloud data is also the cause The system effectively detects the root cause of the limited distance

[0004] In order to make up for the disadvantage of the weak geometric shape information of the target acquired by single-line lidar and improve the quality of point cloud data acquired by lidar, one method is to use 2D lidar or 3D laser scanning to achieve uniform scanning, where 3D laser scanning is usually a combination of 2D laser The radar and IMU (inertial measurement unit) are fixed at one end of the spring, relying on the 2D lidar and IMU at the head of the spring to generate vibrations, and then obtain the 3D point cloud by solving the pose recorded by the IMU in real time, but whether it is 2D scanning or The 3D scanning method is to directly set the scanning angle and speed to control the lidar scanning, without considering the running speed of the carrier platform, but the moving speed of the carrier platform will affect the swing scanning of the lidar, so the above control of the lidar scanning The quality of the obtained point cloud is still not high, which affects the effective detection distance of the system; another method is to use the RBF interpolation algorithm to interpolate the sparse point cloud data in the horizontal direction, and then use the resampling algorithm to perform the densification of the densified 3D point cloud. Adjustment, although high-quality point cloud data can be obtained, the effect of this type of interpolation method is easily affected by the projection direction of the target point cloud, and there are also requirements for the number of scan lines covering the target

Images