Multi-sensor calibration method, multi-sensor calibration device, computer device, medium and vehicle
Embodiments of the present disclosure provide a multi-sensor calibration method, a multi-sensor calibration device, a computer device, a medium and a vehicle. The method includes: acquiring data acquired by each of at least three sensors in a same time period in a traveling process of a vehicle; determining a trajectory of each of the at least three sensors according to the data acquired by each of at least three sensors; and performing a joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors.
1. A multi-sensor calibration method, comprising:
acquiring data acquired by each of at least three sensors in a same time period in a traveling process of a vehicle;
determining a trajectory of each of the at least three sensors according to the data acquired by each of at least three sensors; and
performing a joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors,
wherein performing the joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors comprises:
selecting one from the trajectories of the at least three sensors as a target trajectory, and taking remaining trajectories as object trajectories; and
performing trajectory alignment of each of the object trajectories and the target trajectory, and determining an extrinsic parameter of each of the at least three sensors in response to determining that each of the object trajectories is aligned with the target trajectory.
2. The multi-sensor calibration method according to claim 1 , wherein the at least three sensors comprise at least one lidar, at least one positioning device, and at least one camera;
the trajectory of the lidar is determined based on a position relationship between a plurality of point clouds acquired by the lidar at different time points;
the trajectory of the positioning device is determined based on a plurality of positioning points acquired by the positioning device at different time points;
the trajectory of the camera is determined based on a position relationship between a plurality of images acquired by the camera at different time points.
3. The multi-sensor calibration method according to claim 2 , wherein determining the trajectory of the lidar comprises:
analyzing the plurality of point clouds acquired by the lidar at different time points;
performing feature extraction on the plurality of point clouds;
determining a position relationship between every two of the plurality of point clouds by feature alignment; and
determining the trajectory of the lidar based on the position relationship.
4. The multi-sensor calibration method according to claim 2 , wherein determining the trajectory of the camera comprises:
analyzing the plurality of images acquired by the camera at different time points, and obtaining a plurality of feature points on each of the plurality of images;
determining a coordinate of each of the plurality of feature points in a camera coordinate system;
determining a position relationship between every two of the plurality of images by feature point alignment; and
determining the trajectory of the camera according to the position relationship.
5. A multi-sensor calibration device, comprising:
one or more processors;
a storage device, configured to store one or more programs,
wherein when the one or more programs are executed by the one or more processors, the one or more processors are caused to:
acquire data acquired by each of at least three sensors in a same time period in a traveling process of a vehicle;
determine a trajectory of each of the at least three sensors according to the data acquired by each of at least three sensors; and
perform a joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors,
wherein in performing the joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors, the one or more processors are configured to:
select one from the trajectories of the at least three sensors as a target trajectory, and take remaining trajectories as object trajectories; and
perform trajectory alignment of each of the object trajectories and the target trajectory, and determine an extrinsic parameter of each of the at least three sensors in response to determining that each of the object trajectories is aligned with the target trajectory.
6. The multi-sensor calibration device according to claim 5 , wherein the at least three sensors comprise at least one lidar, at least one positioning device, and at least one camera;
the one or more processors are configured to:
determine the trajectory of the lidar based on a position relationship between a plurality of point clouds acquired by the lidar at different time points;
determine the trajectory of the positioning device based on a plurality of positioning points acquired by the positioning device at different time points; and
determine the trajectory of the camera based on a position relationship between a plurality of images acquired by the camera at different time points.
7. The multi-sensor calibration device according to claim 6 , wherein in determining the trajectory of the lidar, the one or more processors are configured to:
analyze the plurality of point clouds acquired by the lidar at different time points;
perform feature extraction on the plurality of point clouds;
determine a position relationship between every two of the plurality of point clouds by feature alignment; and
determine the trajectory of the lidar based on the position relationship.
8. The multi-sensor calibration device according to claim 6 , wherein in determining the trajectory of the camera, the one or more processors are configured to:
analyze the plurality of images acquired by the camera at different time points, and obtain a plurality of feature points on each of the plurality of images;
determine a coordinate of each of the plurality of feature points in a camera coordinate system;
determine a position relationship between every two of the plurality of images by feature point alignment; and
determine the trajectory of the camera according to the position relationship.
9. A non-transitory computer readable storage medium having stored thereon a computer program that, when executed by a processor, causes a multi-sensor calibration method to be implemented, the method comprising:
acquiring data acquired by each of at least three sensors in a same time period in a traveling process of a vehicle;
determining a trajectory of each of the at least three sensors according to the data acquired by each of at least three sensors; and
performing a joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors,
wherein performing the joint calibration on the at least three sensors by performing trajectory alignment on the trajectories of the at least three sensors comprises:
selecting one from the trajectories of the at least three sensors as a target trajectory, and taking remaining trajectories as object trajectories; and
performing trajectory alignment of each of the object trajectories and the target trajectory, and determining an extrinsic parameter of each of the at least three sensors in response to determining that each of the object trajectories is aligned with the target trajectory.
10. The non-transitory computer readable storage medium according to claim 9 , wherein the at least three sensors comprise at least one lidar, at least one positioning device, and at least one camera;
the trajectory of the lidar is determined based on a position relationship between a plurality of point clouds acquired by the lidar at different time points;
the trajectory of the positioning device is determined based on a plurality of positioning points acquired by the positioning device at different time points;
the trajectory of the camera is determined based on a position relationship between a plurality of images acquired by the camera at different time points.
11. The non-transitory computer readable storage medium according to claim 10 , wherein determining the trajectory of the lidar comprises:
analyzing the plurality of point clouds acquired by the lidar at different time points;
performing feature extraction on the plurality of point clouds;
determining a position relationship between every two of the plurality of point clouds by feature alignment; and
determining the trajectory of the lidar based on the position relationship.
12. The non-transitory computer readable storage medium according to claim 10 , wherein determining the trajectory of the camera comprises:
analyzing the plurality of images acquired by the camera at different time points, and obtaining a plurality of feature points on each of the plurality of images;
determining a coordinate of each of the plurality of feature points in a camera coordinate system;
determining a position relationship between every two of the plurality of images by feature point alignment; and
determining the trajectory of the camera according to the position relationship.