The application discloses a kind of ultra-
wideband laser radar inertial navigation cooperative SLAM method and
system, it is applicable to the
robot positioning and mapping task under the environment without GPS.This method fuses the high-frequency motion prior of IMU, the relative
pose constraint of
LiDAR and the absolute
ranging information of UWB, constructs unified
factor graph optimization model, realizes multi-sensor cooperative positioning;
System initialization stage completes IMU bias calibration, UWB
base station coordinate configuration and
LiDAR initial attitude
estimation;In the running process, the
relative motion between key frames is obtained by using IMU pre-integration prediction
pose and
LiDAR point cloud registration, combined with UWB
ranging to construct residual term and introduce adaptive weight mechanism to suppress
ranging anomaly, improve fusion robustness;The
system of the application supports geometric feature loop detection, and combined with UWB to construct cross-frame constraint for
loop optimization, and the optimization result is used for real-time incremental update of
global map;The method has the advantages of high positioning accuracy, strong anti-interference ability, wide application scene, etc.