Title: An HVSM for improving the homing ability of visual robots

Authors: G.X. Peng, S.Q. Xie, Lei Cheng

Addresses: College of Computer Science, Huazhong University of Science and Technology, PR China. ' Department of Mechanical Engineering, University of Auckland, Private Bag 92019, Auckland, New Zealand. ' Department of Automation, Huazhong University of Science and Technology, PR China

Abstract: The ability of insects returning home after being displaced far away from home is remarkable while the homing ability of existing robot navigators is often limited. The snapshot model is a computational model developed by biologists for investigating insects| homing ability. It computes a given home vector of the current image of the insect|s surroundings only and a snapshot image is taken at target positions. This paper investigates the background of the snapshot model and the model is further developed and extended for robot visual homing. A new model, called horizontal and vertical snapshot model or HVSM, is proposed for developing effective control algorithms for robot visual homing. The experiments show that the HVSM has better performances by utilising information implicitly encoded in two dimensions of an image by a homing agent.

Keywords: robot visual homing; snapshot model; HVSM; vector mapping; robot navigation; robot vision; robot control; biomechatronics; biorobotics; homing agents.

DOI: 10.1504/IJISTA.2005.007305

International Journal of Intelligent Systems Technologies and Applications, 2005 Vol.1 No.1/2, pp.18 - 31

Published online: 05 Jul 2005 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article