Authors:
H. Schäfer
;
M. Proetzsch
and
K. Berns
Affiliation:
Robotics Research Lab, University of Kaiserslautern, Germany
Keyword(s):
Obstacle Detection, Obstacle Avoidance, Behaviour-based Navigation, Mobile Robotics, Short-term Memory, Distributed Minimal World Model, Algorithms, Biologically Inspired Robotics.
Related
Ontology
Subjects/Areas/Topics:
Autonomous Agents
;
Informatics in Control, Automation and Robotics
;
Mobile Robots and Autonomous Systems
;
Robotics and Automation
;
Vehicle Control Applications
Abstract:
In this paper a biologically inspired approach for compensating the limited angle of vision in obstacle detection systems of mobile robots is presented.
Most of the time it is not feasible to exhaustively monitor the environment of a mobile robot. In order to nonetheless achieve safe navigation obstacle detection mechanisms need to keep in mind certain aspects of the environment. In mammals this task is carried out by the creature’s short-term memory. Inspired by this concept an absolute local map storing obstacles in terms of representatives has been introduced in the obstacle detection and avoidance system of the outdoor robot RAVON. That way the gap between the fields of vision of two laser range finders can be monitored which prevents the vehicle from colliding with obstacles seen some time ago.