Evolutionary robotics

Evolutionary Robotics (ER) is a methodology that uses evolutionary algorithms to develop controllers for autonomous robots. It allows to automatically create robot controllers from an initially random population by selecting according to a predefined fitness function. For a number of reasons, evolution is usually first done in computer simulation and only working controllers are transferred to real robots.
Robot controllers typically consist of artificial neural networks, and evolution typically creates the controllers by modifying the strengths of connections between the neurons in the neural network.

ER relates to work done in the domains of Robotics, Artificial Life and Evolutionary Computation.

History

The foundation of ER was laid with work at the national research council in Rome in the 90s, but the initial idea of encoding a robot control system into a genome and have artificial evolution improve on it dates back to the late 80s.

The term evolutionary robotics has been introduced in 1993 by Cliff, Harvey and Husbands at the University of Sussex. In 1992 and 1993 two teams, a team surrounding Floreano and Mondada at the EPFL in Lausanne and a research group at the COGS at the University of Sussex reported the first experiments on artificial evolution of autonomous robots. The success of this early research triggered a wave of activity in labs around the world trying to harness the potential of the approach.

Lately, the difficulty in "scaling up" the complexity of the robot tasks has shifted attention somewhat towards the theoretical end of the field rather than the engineering end.

Evolutionary Robotics

Evolutionary robotics is done with many different objectives, often at the same time. These include creating useful controllers for real-world robot tasks, exploring the intricacies of evolutionary theory (such as the Baldwin effect), reproducing psychological phenomena, and finding out about biological neural networks by studying artificial ones.
Creating controllers via artificial evolution requires a large number of evaluations of a large population. This is very time consuming, which is one of the reasons why controller evolution is usually done in software. Also, initial random controllers may exhibit potentially harmful behaviour, such as repeatedly crashing into a wall, which may damage the robot.
Transferring controllers evolved in simulation to physical robots is very difficult and a major challenge in using the ER approach. The reason is that evolution is free to explore all possibilities to obtain a high fitness, including any inaccuracies of the simulation.
This need for a large number of evaluations, requiring fast yet accurate computer simulations, is one of the limiting factors of the ER approach.