Development of an EKF-based localization algorithm using compass sensor and LRF
This work addresses localization for mobile robots in applications like telecontrol and autonomy, but it is incremental as it applies an existing EKF method to standard sensors.
The paper tackled mobile robot localization by developing a sensor fusion model combining odometry, compass, and laser range data using an extended Kalman filter, resulting in enhanced localization sufficient for navigation tasks.
This paper presents the implementation of a perceptual system for a mobile robot. The system is designed and installed with modern sensors and multi-point communication channels. The goal is to equip the robot with a high level of perception to support a wide range of navigating applications including Internet-based telecontrol, semi-autonomy, and autonomy. Due to uncertainties of acquiring data, a sensor fusion model is developed, in which heterogeneous measured data including odometry, compass heading and laser range is combined to get an optimal estimate in a statistical sense. The combination is carried out by an extended Kalman filter. Experimental results indicate that based on the system, the robot localization is enhanced and sufficient for navigation tasks.