Lukas Lichtensteiger, Ph.D.

*** I'm sorry but this page is very outdated, I will work on it asap. Meanwhile, please send me a mail if you need any information ***
E-mail:   llicht@ifi.uzh.ch

Address:  
          Artificial Intelligence Lab
          Dept. of Computer Science
          University of Zurich
          Andreasstrasse 15
          CH - 8050 Zurich, Switzerland

Office:   AND-2.14

Phone:    ++41 - 1 - 635 45 63
Fax:      ++41 - 1 - 635 45 07

Research Summary

My research focuses on the interdependence between morphology and control architecture of an agent with regard to its (intelligent) behavior. In order to develop a theoretical framework for the role of morphology I am performing experiments with real-world robots on two different levels of abstraction:
 
  • The "low-level, physical part": Evolving Sensor Morphologies using Adaptive Hardware. How can an agent's morphology, specifically the details of its sensor morphology, contribute to intelligent behavior in a given task environment? How can physical laws in the robot-environment interaction be exploited to solve problems at the lowest possible level? For a given task environment, which problems should be dealt with by morphology and which by the control architecture? What is the "optimal morphology" for an agent with regard to a given task environment, from a designer's point of view? What can we learn from biological systems?

 
  • The "high-level, information theoretic part": Reconfigurable Modular Robots.How does body morphology interdepend on neural morphology, i.e., an agent's control architecture? How should the ensemble of the two be organized? Modular units? What kind of communication between modules is optimal? Where should sensors/actuators be positioned, and how should they be integrated in the control architecture? Does ontogenetic development help the system "to get organised"?


Both projects take a lot of inspiration from biological systems. They are also part of the Embodied Artificial Evolution Group.
 

Robots

Flexible Robot Building Kit.

T-Bot.

Artificial Compound Eye (EyeBot).
 

Simulators

Simple Java Robot Simulator.
 

Publications

Lichtensteiger, L. (2000), Adaptive hardware for evolving sensor morphology of a compound eye
In: SPIE’s International Technical Group Newsletter (Robotics and Machine Perception), Vol. 9, No. 2 (November 2000), Society of Photo-Optical Instrumentation Engineers, Bellingham, WA, USA

Lichtensteiger, L. (2000), Towards optimal sensor morphology for specific tasks: Evolution of an artificial compound eye for estimating time to contact
In: Sensor Fusion and Decentralized Control in Robotic Systems III, Gerard T. McKee, Paul S. Schenker (eds.), Proceedings of SPIE Vol. 4196, pp. 138-146 (2000).

L. Lichtensteiger and R. Salomon (2000), The Evolution of an Artificial Compound Eye by Using Adaptive Hardware
In: Proceedings of the 2000 Congress on Evolutionary Computation, 16-19 July 2000, La Jolla Marriott, San Diego, CA, USA, pp. 1144-1151

Salomon, R. and Lichtensteiger, L. (2000), Exploring different Coding Schemes for the Evolution of an Artificial Insect Eye
Proceedings of the First IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, 11-12 May 2000, San Antonio, TX, USA

Lichtensteiger, L. and Eggenberger, P. (1999), Evolving the Morphology of a Compound Eye on a Robot
Proceedings of the Third European Workshop on Advanced Mobile Robots (Eurobot '99), (Cat. No.99EX355). IEEE, Piscataway, NJ, USA; 1999; p.127-34.

M. Sc. in Theoretical Physics

Lichtensteiger, L. and Durrer, R. (1999), Are there static textures?
Physical Review D. vol.59, no.12; 15 June 1999; p.125007/1-6.

Lichtensteiger, L. (1996), Spherically Symmetric Solutions for Sigma-Models in 3 and 4 Dimensions
Master's Thesis, Department of Theoretical Physics, University of Zurich, Switzerland
 


215accesses so far,

215including ifi