A novel localization algorithm integrated with sensor fusion for position estimation of mobile robot in the road following and roundabout environment

A novel Artificial Intelligence algorithm based autonomous mobile robot localization in road environments is presented in this paper. The sensor fusion data, employing camera and odometry, are used to build the local maps for road environment. The localization of the robot is performed using a novel Artificial Intelligence algorithm, called Priori-Posteriori-Laser-Simulator, to find the position and heading-angle of robot along movement in two phases, namely priori and posteriori stages as 1st phase and Kalman filter in 2nd phase. In the priori stage, the path of the robot is planned in the built-map using Laser Simulator through generating a series of points as lines to drive the robot in the middle of the road environments. In the posteriori-stage, the position and robot heading angle are measured using sensor fusion system and the laser-simulator updates the position accordingly through a comparison between the planned and actual positions. The Kalman filter is used in 2nd phase to remove noise and enhance to enhance the localization accuracy. The proposed algorithm has been tested in indoor and outdoor road environments to show its effectiveness and performance for localizing robot in such environments. It is also compared with fuzzy-logic, adaptive H-infinity filter, fuzzy logic with Kalman filter and Oriented-Rotated Binary Simultaneous Localization and Mapping 2 algorithms on localizing robot at road environment. Results show the capability of the proposed algorithm to enable the robot to move effectively in road environments, recognize the features of road terrains and localize the robot during autonomous navigation in road environments.