Biorobotics Laboratory BioRob

Adam - A Modular Robot Simulation and Evolution Tool

This undergraduate project is sometimes confused with my Master Thesis, which is on the same topic but obviously more recent and elaborate. Please visit the Master Thesis project page first if you haven't yet seen it.

What is Adam?

Adam is a system to evolve and simulate modular robots. The simulation accurately models rigid body dynamics (kinematics, gravity, friction, collisions, etc) in a world that consists simply of an infinite plane.

Adam robots are built with a number of different, well-defined module types. Modules are like LEGO blocks: Each type has certain qualities and positions where it is possible to put another building stone. Adam is built to make it easy to add new types of modules. One could define modules that correspond to real hardware and then actually build the evolved robots but for now, we test the system only with simple hinge modules.

The hinge module consists of two cubes, fixed together with a hinge. Other elements can be attached at each one of the ten free faces. The hinge can be rigid or powered by a motor. Each powered hinge has a PD controller on its angle. Additionally, the joint may be elastic or not.

The user can build Adam robots with a simple script. Furthermore, this allows evolved robots to be saved in plain text format. They can then be inspected and edited by the user with the text editor of his choice.

A simple quadruped robot that was built with the script.

Locomotion strategies of evolved robots included walking, rolling, jumping, ratcheting, caterpillar like locomotion, etc.

Sample movies of evolved robot

I removed these movies because they were bad quality. The movies available on my Master Thesis project page are much more interesting.

Peer-reviewed Proceedings

More recent publications and the pdf of the following paper are available on my Master Thesis project page.

D. Marbach and A.J. Ijspeert. Co-evolution of configuration and control for homogenous modular robots. In F. Groen et al., editor, Proceedings of the Eighth Conference on Intelligent Autonomous Systems (IAS8), pages 712-719. IOS Press, 2004.

Documentation

Project report: report.pdf

Slides of the presentation: slides.pdf