Exploration and Navigation for Mobile Robots With Perceptual Limitations

To learn a map of an environment a mobile robot has to explore its workspace using its sensors. Sensors are noisy and have perceptual limitations that must be considered while learning a map. This paper considers a mobile robot with sensor perceptual limitations and introduces a new method for exploring and navigating autonomously in indoor environments. To minimize the risk of collisions as well as to not exceed the range of sensors, we introduce the concept of a travel space as a way to associate costs to grid cells of the map, based on distances to obstacles. During exploration the mobile robot minimizes its movements, including rotations, to reach the nearest unexplored region of the environment, using a dynamic programming algorithm. Once the exploration ends, the travel space is used to form a roadmap, a net of safe roads that the mobile robot can use for navigation. These exploration and navigation method are tested using a simulated and a real mobile robot with promising results.