Using robots to investigate the evolution of adaptive behavior

- Robots can be evolved using supercomputers to perform a range of desired tasks.
- Analysis of evolved robots can produce hypotheses about why traits evolved in nature.
- I describe the evolution of modularity, morphological computation, and prospection.

In the field of evolutionary robotics, investigators evolve populations of autonomous machines to exhibit some desired behavior. The neurology, morphology, or both may be placed under evolutionary control, and different behaviors can be selected for. Results from this approach can generate unique and surprising hypotheses about why certain behaviors evolve, regardless of whether they emerge in organisms or machines. To illustrate this approach, I describe recent work on the evolution of modularity, morphological computation, and prospection.