Flying Robot Project

(Latest update: 06/March/2001)


Project Description

This project is a part of biorobotics projects at AILab, Dept. of Information Technology, Univ. of Zurich, and aims at understanding mechanism of navigation in flying insect such as honeybees or flies. Flying insects mainly rely on visual information to navigate in noisy three dimentional space. Although it is well known that visual processing techniques in traditional approach require huge computational power, insects achieve navigation in complicated natural environment with tiny brain. Therefore, main point of this project focuses on the cheap mechanisms to realize vision based navigation. This project includes such topics as goal-directed navigation, landmark navigation, path integration, reute learning, landmark recognition and learning.

Recent result of goal-directed navigation experiments shows that vision based navigation could be achieved in significantly cheaper way in terms of visual processing cost, if the system employs the principles of the natural insects, namely sensory-motor coordination principle. This cheap mechanism is realized by using ego-motion of the agent induced by sensory-motor coordination, thus it could be applied for mobile agent in general. 

On the other hand, it is widely believed that bees can recognize the object, mainly as landmark in the context of landmark navigation. We are currently working on landmark navigation in the context of navigation using image motion detectors. This approach might lead to general concept of object recognition which takes sensory-motor coordination principle into account.


Hardware platform

       - Flying robot: Melissa


Publications
Fumiya Iida, Dimitrios Lambrinos, "Navigation in an autonomous flying
robot by using a biologically inspired visual odometer," Sensor Fusion
and Decentralized Control in Robotic System III, Photonics East,
Proceeding of SPIE, vol. 4196, pp.86-97 (2000)  pdf (2.1M)

Fumiya Iida,"Goal-Directed Navigation of an Autonomous Flying Robot
Using Biologically Inspired Cheap Vision," Proceedings of the 32nd
ISR(International Symposium on Robotics), 19-21 April 2001 (in press) pdf (415k)


Useful information
Other Flying Robot links
- Micro Air Vehicle "Entomopter" Project at Georgia Tech
- Autonomously Flying Helicopter at ETH
- MARVIN Project atTU Berlin
- Autonomous Flying Vehicle Project at Univ. of Southern California
- Autonomous Helicopter Project at Carnegie Mellon Univ.
Natural insect study and biorobotics 
- Visual Science Laboratory at National Australian Univ.
- CNRS

Companies
- Mobile Airships & Blimps
- Neuronics
- VTQ
- ces - complete electronic systems ag

Send questions and comments to Fumiya Iida
[Go to AILab]Go to AILab main page