In real-time positioning problems, some unstable situations often occur, such as GPS signal loss and map drifting. In order to have better localization and map buildings, this chapter proposes a coarse-and-fine hybrid positioning system, which integrates the global information and feature-based simultaneous localization and mapping (GF-SLAM). The system can operate adaptively when external signals are unstable and avoid the cumulative error from local methods. Generally, it applies the feature-based SLAM (F-SLAM) for coarse positioning with particle filter. If available, it fuses the accurate global information by extended Kalman filter (EKF) for precise positioning and revises the deviation in mapping, thereby achieving an effective combination of two positioning modes. In addition, we introduce semantic mapping and lifelong localization approaches to recognize semi-dynamic objects in non-static environments. We also propose a generic framework that can integrate mainstream object detection algorithms with mapping and localization algorithms. The mapping method combines an object detection algorithm and a SLAM algorithm to detect semi-dynamic objects and constructs a semantic map that only contains semi-dynamic objects in the environment. In summary, this chapter proposes methods to resolve self-positioning and map drift when positioning fails.
Multi-Sensor Fusion Localization
Multi-sensor Fusion for Autonomous Driving ; Kapitel : 6 ; 119-150
2023-05-11
32 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Sensor fusion for AUV localization
Tema Archiv | 1990
|Sensor fusion for localization of automated vehicles
TIBKAT | 2018
|