4. Lidar Calibration¶
In multi-sensor intelligent body navigation, multiple sensors are usually installed for data fusion and calculation. In mobile mapping and environment perception, it is usually necessary to arrange multiple Lidars to ensure a larger FOV and provide a stable data source for more robust attitude estimation algorithms and rich environment details. However, no matter what kind of sensor: camera, lidar, millimeter wave radar, they all have their own coordinate system, that is, the data generated by all sensors are based on the sensor’s own coordinate system. In order to facilitate algorithm research and testing, it is necessary to convert the data obtained by the respective sensors to the same coordinate system, which is the external parameter calibration of the sensor.
4.1. Lidar-Lidar Calibration¶
The Livox_Viewer integrates the function of manually calibrating Livox LiDARs, which can easily and accurately calibrate multiple Livox LiDARs with overlapping FOVs. For details, please refer to: Livox Viewer Manuel
4.2. Lidar-Camera Calibration¶
Livox provides a method to manually calibrate the external parameters between Livox lidar and camera, and it has been verified on Mid-40, Horizon and Tele-15. It contains code for calculating internal camera parameters, obtaining calibration data, optimizing external parameters and lidar camera fusion applications. In this scheme, the corners of the calibration board are used as the calibration targets. Due to the non-repetitive scanning feature of Livox lidar, the density of the point cloud is relatively high, and it is easier to find the accurate position of the corners in the lidar point cloud. The calibration and fusion of camera lidar can also get good results. For details, please refer to: Camera-LiDAR-Calibration Manual
For the multi-sensor fusion scanning scheme, please refer to the Livox open source algorithm livox high precision mapping which shows in detail how to synchronize and calibrate the Mid-40 lidar and apx-15 inertial navigation module, and use this component for real-time high-precision mapping.