DGPS-BASED LOCALIZATION AND PATH FOLLOWING APPROACH FOR OUTDOOR WHEELED MOBILE ROBOTS

Differential Global Positioning System (DGPS) sensors provide a way for an outdoor wheeled mobile robot to achieve better localization sensing that results in better navigation and control. In this paper, we propose a new practical path following approach with successful applications to the outdoor wheeled mobile robots (WMR). The primary focus of the approach is to use differential GPS (dGPS) technology as the only sensing device for localization while the odometry-based localization method for outdoor environment is not desirable. In using a dGPS sensing device for measurement, filter estimation techniques are also considered to account for measurement inaccuracies. Of the several techniques, the extended Kalman filter (EKF) is chosen and implemented as part of this approach for improved results. A secondary focus of the approach is to incorporate path planning as an integral part of the proposed technique, in which the boundary influence to the interested object is well integrated in addition to the conventional path following approaches. This allows for a better consideration of the inaccuracies inherent in GPS measurements by planning for them in order to ensure the robot achieves its objective. The dGPS-based proposed path following approach is firstly simulated in the simulation environment. The proposed approach is then applied and tested on a WMR experimental platform for a set of experimental cases with comparisons.