Towards Optimal Dynamic Localization for Autonomous Mobile Robot via Integrating Sensors Fusion

When it comes to optimal dynamic localization, high accuracy and robustness localization is the main challenge for the autonomous mobile robot. In this paper, an optimal dynamic localization framework with integrating sensors fusion is considered. The global point map is utilized to provide absolute pose observation information, and the multi-sensor information is applied to realize robust localization in complex outdoor environments. The multi-sensor technique, including 3D-Lidar, global positioning system (GPS), and inertial measurement unit (IMU), is adopted to construct the global point map by pose optimization so that the absolute position and attitude observation information can still be provided when the outdoor GPS signal fails. Meanwhile, in the case of optimal localization, the system kinematics equation is constructed by the IMU error model, and the map pose is matched by map scanning. Moreover, the GPS position information participates in multi-source fusion when the GPS signal is reliable. Finally, the experimental results show that the average localization error is within 0.05 meters, reflecting the flexibility of dynamic localization.