Muñiz, F.Zalama, E.Gaudiano, P.Coronado-Lóez, J.2011-11-142011-11-141995-02https://hdl.handle.net/2144/2183Recently it has been introduced a neural controller for a mobile robot that learns both forward and inverse odometry of a differential-drive robot through an unsupervised learning-by-doing cycle. This article introduces an obstacle avoidance module that is integrated into the neural controller. This module makes use of sensory information to determine at each instant a desired angle and distance that causes the robot to navigate around obstacles on the way to a final target. Obstacle avoidance is performed in a reactive manner by representing the objects and target in the robot's environment as Gaussian functions. However, the influence of the Gaussians is modulated dynamically on the basis of the robot's behavior in a way that avoids problems with local minima. The proposed module enables the robot to operate successfully with different obstacle configurations, such as corridors, mazes, doors and even concave obstacles.en-USCopyright 1995 Boston University. Permission to copy without fee all or part of this material is granted provided that: 1. The copies are not made or distributed for direct commercial advantage; 2. the report title, author, document number, and release date appear, and notice is given that copying is by permission of BOSTON UNIVERSITY TRUSTEES. To copy otherwise, or to republish, requires a fee and / or special permission.Mobile robotObstacle avoidanceLearning controlNeural networksRobot navigationReactive behaviorTechnical ReportBoston University Trustees