Abstract—Path planning in unstructured area while dealing with narrow spaces is an area of research which is receiving extensive interest. Many existing algorithms are able to produce safe paths but the presented concepts are either not adapted to narrow spaces or they are unable to learn from the past experience to improve repeated movements from the same agent or followed trajectories by other agents. This paper introduces an original concept based on Ant-Air phenomenon for safe path planning in a cluttered environment where narrow passages are treated. The algorithm presented is able to learn from the past experience and hence improve the already generated trajectory further by using some lessons learned from the past experience. The concept is applicable in various domains such as mobile robot path planning, manipulator trajectory generation and part movement in narrow passages in real or virtual assembly/disassembly process.
Index Terms—Path planning, collision detection and avoidance, self-learning algorithm, assembling or disassembling, narrow spaces.
The authors are with University of Luxembourg, L-1359 Luxembourg, Pakistan (e-mail: rafiq.ahmad@uni.lu, peter.plapper@uni.lu).
[PDF]
Cite: Rafiq Ahmad and Peter Plapper, "Ant-Air Self-learning Algorithm for Path Planning in a Cluttered Environment," International Journal of Materials, Mechanics and Manufacturing vol. 4, no. 2, pp. 127-130, 2016.