Editing Evolutionary robotics (section)

'''Evolutionary robotics''' is an [[Embodied cognition|embodied]] approach to [[Artificial Intelligence]] ('''AI''') in which [[robots]] are automatically designed using [[Darwinism|Darwinian]] principles of [[natural selection]].<ref name="ACM">{{cite journal |last1=Bongard |first1=Josh |title=Evolutionary Robotics |journal=Communications of the ACM |date=2013 |volume=56 |issue=8 |pages=74–83 |doi=10.1145/2493883 |s2cid=16097970 |language=English|doi-access=free }}</ref> The design of a robot, or a subsystem of a robot such as a [[Neural network|neural controller]], is optimized against a [[Fitness (biology)|behavioral goal]] (e.g. run as fast as possible). Usually, designs are evaluated in [[in silico|simulations]] as fabricating thousands or millions of designs and testing them in the real world is prohibitively expensive in terms of time, money, and safety.

An evolutionary robotics experiment starts with a population of randomly generated robot designs. The worst performing designs are discarded and replaced with [[mutations]] and/or [[crossover (genetic algorithm)|combinations]] of the better designs. This [[evolutionary algorithm]] continues until a prespecified amount of time elapses or some target performance metric is surpassed.

Evolutionary robotics methods are particularly useful for engineering machines that must operate in environments in which humans have limited intuition (nanoscale, space, etc.). Evolved simulated robots can also be used as scientific tools to generate new hypotheses in biology and cognitive science, and to test old hypothesis that require experiments that have proven difficult or impossible to carry out in reality.