VIRGO: Vision-based robot navigation research network

• Unité de recherche: PCRD EU
• Numéro de projet: 96.0148
• Titre du projet anglais: VIRGO: Vision-based robot navigation research network

Mots-clé (Anglais)

Robot navigation; distance estimation; flying robots; insect-inspired motion detection.

Autre Numéro de projet (Anglais)

EU project number: FMRXCT960049

Programme de recherche (Anglais)

EU-programme: 4. Frame Research Programme - 10.1 Stimulation of training and mobility

Description succincte (Anglais)

See abstract

Partenaires et organisations internationales (Anglais)

Coordinator: Foundation for Research & Technology (EL)

Résumé des résultats (Abstract) (Anglais)

The goal of the VIRGO Network is to coordinate European research and post-graduate training activities that address the development of intelligent robotic systems able to navigate in (partially) unknown and possibly dynamic environments. Navigation in unknown and dynamically changing environments has been associated with the evaluation of visual information. The research conducted by the AILab has primarily focused on biological models of flying insects' navigation, e.g. speed, orientation, altitude control. One problem for autonomous flying robot navigation is that, unlike terrestrial robots, flying robots cannot rely on proprioceptive cues (e.g. wheel encoders) for measuring the distance traveled.
We have developed a biologically inspired model of visual odometer based on Elementary Motion Detectors (EMDs). The model is inspired from the recent evidence that flying insects are using optical flow induced by egomotion to estimate distance traveled. We embedded this model into the autonomous flying robot and performed a series of goal-directed navigation experiments. The miniature panoramic vision system was used to provide input to the EMDs of the left and right visual fields. The outputs of the EMDs were in later stage spatially integrated by wide field motion detectors, and their accumulated response was directly used for the odometer. In a set of initial experiments, the robot moved through a corridor on a fixed route, and the outputs of EMDs, the odometer, were recorded. The results show that the proposed model can be used to provide an estimate of the distance traveled, as well as to achieve obstacle avoidance and course stabilization behaviours. Further investigation in the simulation have shown that goal-directed navigation can be potentially achieved by simple visual processing, and that the design flexibility of this approach leads to high adaptivity to the given task-environment. These interesting properties of the biologically inspired visual processing model have been summarized as a paradigm of 'cheap vision'.
The research activity has demonstrated at the Science Fair (Science et Cité) held in the large hall of Zurich main station, where 10 newly developed flying robots have been shown.
In addition, on the basis of the achievements of this project, a new European Union project is currently being organized together with some partners from the VIRGO network.

Références bases de données (Anglais)

Swiss Database: Euro-DB of the
State Secretariat for Education and Research
Hallwylstrasse 4
CH-3003 Berne, Switzerland
Tel. [phone]
Swiss Project-Number: 96.0148