Abstract
We introduce a hybrid algorithm for the self-semantic location and autonomous navigation of robots using entropy-based vision and visual topological maps. In visual topological maps the visual landmarks are considered as leave points for guiding the robot to reach a target point (robot homing) in indoor environments. These visual landmarks are defined from images of relevant objects or characteristic scenes in the environment. The entropy of an image is directly related to the presence of a unique object or the presence of several different objects inside it: the lower the entropy the higher the probability of containing a single object inside it and, conversely, the higher the entropy the higher the probability of containing several objects inside it. Consequently, we propose the use of the entropy of images captured by the robot not only for the landmark searching and detection but also for obstacle avoidance. If the detected object corresponds to a landmark, the robot uses the suggestions stored in the visual topological map to reach the next landmark or to finish the mission. Otherwise, the robot considers the object as an obstacle and starts a collision avoidance maneuver. In order to validate the proposal we have defined an experimental framework in which the visual bug algorithm is used by an Unmanned Aerial Vehicle (UAV) in typical indoor navigation tasks.
Highlights
Some of the most challenging behaviors of autonomous robots are related to navigation tasks
The aim of our work is to introduce an efficient method for autonomous robot navigation supported by visual topological maps (Maravall et al, 2013a), in which the coordinates of both the robot and the goal are not needed
After the theoretical foundations of the entropy-based search combined with the bug algorithm, we present the experimental work performed for its validation using an Unmanned Aerial Vehicle (UAV)
Summary
Some of the most challenging behaviors of autonomous robots are related to navigation tasks. According to a policy of safe navigation, the basic level of behaviors are devoted by strategies that allow the obstacle detection and the collision avoidance. Once these tasks have been conveniently solved, the level is route planning, i.e., the generation of routes or paths that allow the robot to reach specific places in the environment. Some information about the environment is needed for such kind of planning This information is managed by the robot control system and concerns the fundamental issue of environment mapping. The robot can store a sort of metric data related to the environment, topological information, including relationships between elements of the environment and features, maybe visual, associated to them, or any combination of both
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
