Tutorial on Embodiment

3.2.2.1. Behavioral diversity - crazy bird*

 

To investigate how to gain controllability and increase the behavioral diversity of robots, a quadrupedal robot was built with the LEGO MINDSTORMS NXT kit (see Fig. 3.2.2.1.1).

It only has two motors and no sensors, the two motors are synchronized, turn at the same speed and in the same direction. The front legs are attached eccentrically to the wheels driven by the motors and are physically connected to the hind legs. Four experimental setups were tested with different phase delay between left and right legs (i.e., 0° , 90° , 180°, and 270°). Despite its simplicity, the robot could reach every point on a table by changing a single control parameter (i.e., the phase delay between its legs). The controller only has to induce a global bias rather than exactly controlling the joint angles or other parameters of the agents movement (Rinderknecht et al. 2007).


If we keep the controller of the quadrupedal robot as is, but modify the morphology by removing the hind legs and adding two loosely attached rubber feet, which can turn and move a bit, we obtain the "Crazy bird" (see Fig. 3.2.2.1.2. and video 3.2.2.1.1.), that gets its name due to its rich behavioral diversity.

Fig. 3.2.2.1.1. A quadrupedal LEGO robot.


Fig. 3.2.2.1.2. Crazy bird - a slightly modified version of the quadrupedal LEGO robot in fig. 3.2.2.1.1.

Video 3.2.2.1.1. crazy bird. The two motos running forward with constant speed. Due to the robot's body shape and the interaction with the environment behavioral diversity can emerge.

 

* This case study has previously appeared in Pfeifer & Gomez, 2009.

References:

Rinderknecht, M., Ruesch, J., Hadorn, M. The lagging legs exploiting body dynamics to steer a quadrupedal agent. In: International Conference on Morphological Computation, Venice, Italy (2007)
Pfeifer, R. & Gomez, G. (2009). Morphological computation - connecting brain, body, and environment. In B. Sendhoff, O. Sporns, E. Körner, H. Ritter, & K. Doya, K. (eds.), Creating Brain-like Intelligence: From Basic Principles to Complex Intelligent Systems (pp.66-83). Berlin: Springer.

 

<<PREVIOUSHOMENEXT>>