Model of a predatory stealth behaviour camouflaging motion |
| |
Authors: | Anderson Andrew James McOwan Peter William |
| |
Institution: | Department of Computer Science, Queen Mary, University of London, London E1 4NS, UK. aja@dcs.qmul.ac.uk |
| |
Abstract: | A computational model of a stealth strategy inspired by the apparent mating tactics of male hoverflies is presented. The stealth strategy (motion camouflage) paradoxically allows a predator to approach a moving prey in such a way that it appears to be a stationary object. In the model, the predators are controlled by neural sensorimotor systems that base their decisions on realistic levels of input information. They are shown to be able to employ motion camouflage to approach prey that move along both real hoverfly flight paths and artificially generated flight paths. The camouflaged approaches made demonstrate that the control systems have an ability to predict future prey movements. This is illustrated using two- and three-dimensional simulations. |
| |
Keywords: | |
本文献已被 PubMed 等数据库收录! |
|