A method for automatic acquisition of the relative position of a vehicle and an obstacle

Abstract

The invention provides a method for automatically obtaining the relative position of a vehicle and an obstacle. The camera is placed above the middle position of the vehicle-mounted laser radar calibration field, and spherical characteristic calibration objects are respectively arranged at two opposite corners of the calibration field to detect the calibration. The world coordinates of the sphere center of the sphere feature calibration object in the field, and detect the outer contour of the vehicle to be calibrated, calculate the circumscribing rectangle of the vehicle’s outer contour and obtain the world coordinates of the center of the circumscribing rectangle, and use the circumscribing rectangle of the vehicle’s outer contour The center is the origin of the car body coordinate system, and the X-axis and Y-axis of the car body coordinate system are set to be parallel to the two sides of the circumscribed rectangle of the vehicle's outer contour, and the rotation and translation of the car body coordinate system and the world coordinate system are transferred to the system , and finally calculate the coordinates of the sphere center of the sphere feature calibration object in the car body coordinate system. The invention can automatically acquire the position and posture of the vehicle body and the relative position relationship with obstacles, reduces the requirement on the position and posture of the vehicle when calibrating the laser radar, and improves the calibration efficiency and accuracy of the laser radar.

Description

technical field [0001] The invention relates to the technical field of automatic driving and intelligent networked vehicles, in particular to the field of sensor calibration, and in particular to a method for automatically acquiring the relative positions of a vehicle and an obstacle. Background technique [0002] The on-board lidar calibration of intelligent networked vehicles or autonomous vehicles is often completed in specific scenarios. Calibrating on-board sensors is a pre-procedure step in the development and testing of intelligent networked vehicles or autonomous vehicles. Currently, when calibrating the on-board lidar of intelligent networked vehicles, Usually, characteristic calibration objects are arranged in the calibration scene for calibration. Since the position of the coordinate system of the vehicle body and the origin of the coordinate system are fixed, there are strict requirements on the pose of the vehicle parked, which makes it difficult to park the cali...

AI Technical Summary

Problems solved by technology

[0002] The on-board lidar calibration of intelligent networked vehicles or autonomous vehicles is often completed in specific scenarios. Calibrating on-board sensors is a pre-procedure step in the development and testing of intelligent networked vehicles or autonomous vehicles. Currently, when calibrating the on-board lidar of intelligent networked vehicles, Usually, characteristic calibration objects are arranged in the calibration scene for calibration. Since the position of the coordinate system of the vehicle body and the origin of the coordinate system are fixed, there are strict requirements on the pose of the vehicle parked, which makes it difficult to park the calibration vehicle and reduces the The calibration accuracy is reduced, and even the calibration fails, which greatly affects the calibration efficiency of the vehicle-mounted lidar.

Since the calibration scene and the position of the characteristic markers are fixed, it is necessary to ensure that the lidar is parked according to the specified pose during the calibration, so as to ensure the validity of the relative positional relationship between the specific calibration object and the car body stored in advance; the current practice is When placing the vehicle, it is usually operated manually. There is no standardized and automated way to obtain the relative positional relationship between the vehicle body and the characteristic calibration object. Manual adjustment of the vehicle pose is inefficient and difficult, and the accuracy of the calibration results has certain inconsistencies. Deterministic, even lead to calibration failure

Images