Towards Computational Models of Insect Motion Detectors for Robot Vision

摘要

In this essay, we provide a brief survey of computational models of insect motion detectors and bio-robotic solutions to build efficient and reliable motion-sensing systems. Vision is an important sensing modality for autonomous robots, since it can extract abundant motion features from a visually cluttered and dynamic environment. However, modelling a dynamic vision system for motion detection in both a cheap and robust manner is still an open challenge. In nature, animals, that have experienced millions of years of evolutionary development, can be prominent models to study motion-detecting strategies. This essay introduces a few state-of-the-art bio-inspired motion detectors and corresponding robotic applications, in order to demonstrate the effectiveness of mimicking insect motion perception strategies for robot vision and gather cross-disciplinary attention.