A CPG-driven Autonomous Robot

Jörg Conradt, Paulina Varshavskaya
conradt@ini.phys.ethz.ch, paulina@csail.mit.edu

We have built an autonomous mobile robotic worm [ref 1] which facilitates exploring motion principles based on neural Central Pattern Generator (CPG) circuits [ref 2] in a truly distributed system.
The main aim of the project is to demonstrate elegant motion on a robot with a large number of degrees of freedom under the control of a simple distributed neural system as found in many animals' spinal cord. Currently, the robot consists of up to 60 individual segments that all run a local CPG. Sparse adjustable short- and long-range coupling between these CPGs synchronizes all segments, thus generating overall motion. A wireless connection between a host computer and the robot allows changing parameters during operation (eg. individual coupling coefficients, traveling speed, and motion amplitude). Additionally, users can modify the CPG algorithm and reprogram the segments during operation.
The robot can demonstrate various motion patterns based on extremely simple neural algorithms. We are currently implementing more advanced neural CPGs [eg. refs 3-5] and will compare them in terms of motion robustness and traveling speed.
This work originated during the 2002 Telluride Workshop of Neuromorphic Engineering and was partially funded by the Institute of Neuromorphic Engineering and the Institute of Neuroinformatics.

More details in our NIPS2003 poster or our ICANN 2003 paper.


#1: Conradt, J., and Varshavskaya, P. (2003). Distributed Central Pattern Generator Control for a Serpentine Robot. Proceedings of the Joint International Conference on Artificial Neural Networks and Neural Information Processing (ICANN/ICONIP), Istanbul
#2: Cohen, A. H., Holmes, P. J., and Rand, R. H. (1982). The Nature of the Couplings Between Segmental Oscillators of the Lamprey Spinal Generator for Locomotion: A Mathematical Model. Journal of Mathematical Biology, Vol.13, pp.345-369.
#3: Ijspeert, A.J., Hallam, J. and Willshaw, D.(1999). Evolving swimming controllers for a simulated lamprey with inspiration from neurobiology. Adaptive Behavior 7:2, 1999, pp 151-172.
#4: Grillner, S., Deliagina, T, Ekeberg, Ö, Manira, A. El, Hill, R.H., Lansner, A., Orlovsky, G.N., and P. Wallén (1995). Neural networks that co-ordinate locomotion and body orientation in lamprey. Trends in Neuroscience 18(6): 270-279.
#5: Cohen, A. H., and D.L. Boothe (2003). Sensorimotor interactions: Principles derived from central pattern generators. In Handbook of Brain Theory and Neural Networks, ed. M. Arbib. MIT Press.

Short demonstration videos of WormBot motion (more videos to come soon!)  

Video showing a variety of motion patterns:    MPG (3MB)      (most interesting)

The robot in planar "snake like" configuration:    MPG (10MB)

The robot in vertical "worm like" configuration:   MPG ( 9MB)

Here is a zoom-in onto one of the robot's segments

And here is a photograph of the current Worm Robot prototype

myspace profile visitors