First hand built robot. Uses Mark Tilden Nervous Net technology.

 
   

This robot crawls around the cage and backs up and turns when it hits the wall. The touch sensors (antennae) have carbonite tips that scratch the glass as they rub against the sides of the cage. A programmable microcontroller is used in the control circuitry.

 
   

Utilizing neural network and genetic algorithms, this robot evolves and learns to walk on its own. There are two microcontrollers on the robot, one that positions the servomotors, and one that reads sensory information from a number of touch sensors and an accelerometer. The second microcontroller communicates to a PC via a serial port.

  Video
 
Hums to itself while it walks around aimlessly.
More Info Sound 311k
 

Coming soon...Face Detection

 
Home | Sculpture | Robotics | Projects | Resume
All images and videos are property of Matthew Likens, please contact me if you would like to use them.
mattlikens@yahoo.com