A mixed analog-digital vision sensor for detecting objects approaching on a collision course

Authors:
Hirotsugu Okuno;Tetsuya Yagi
Affiliations:
Division of Electrical, Electronic and Information Engineering, Osaka University, Osaka, Japan;Division of Electrical, Electronic and Information Engineering, Osaka University, Osaka, Japan
Venue:
Robotics and Autonomous Systems
Year:
2009

Citing 4
Cited 0

A bio-inspired visual collision detection mechanism for cars: Optimisation of a model of a locust neuron to a novel environment

Neurocomputing
A Robot Vision System for Collision Avoidance Using a Bio-inspired Algorithm

Neural Information Processing
An analog VLSI chip emulating sustained and transient response channels of the vertebrate retina

IEEE Transactions on Neural Networks
Collision detection in complex dynamic scenes using an LGMD-based visual neural network with feature enhancement

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Bio-inspired vision system is a particularly good candidate for navigation of mobile robots and vehicles because of its computational advantages, e.g., low power dissipation and compact hardware. Previously, we had designed a mixed analog-digital integrated vision system for collision detection inspired by a locust neuronal circuit model. The response of the system was, however, susceptible to the luminance of approaching objects and the vibratory self-motion of a car when it was installed on a miniature mobile car. In the present study, we developed a new collision detection algorithm to overcome these problems based on robust image-motion detection and applied the algorithm to control a miniature mobile car.