1. Evolving-Controllers Versus Learning-Controllers for Morphologically Evolvable Robots
- Author
-
Miras, Karine, De Carlo, Matteo, Akhatou, Sayfeddine, Eiben, A. E., Castillo, Pedro A., Jiménez Laredo, Juan Luis, Fernández de Vega, Francisco, Artificial intelligence, Network Institute, Artificial Intelligence (section level), Computational Intelligence, Knowledge Representation and Reasoning, Castillo, Pedro A., Jiménez Laredo, Juan Luis, and Fernández de Vega, Francisco
- Subjects
Self-reconfiguring modular robot ,Modular robots ,Computer science ,Crossover ,Evolutionary robotics ,02 engineering and technology ,03 medical and health sciences ,0302 clinical medicine ,Control theory ,0202 electrical engineering, electronic engineering, information engineering ,Morphological evolution ,business.industry ,technology, industry, and agriculture ,body regions ,Life-time learning ,Robotic systems ,Mutation (genetic algorithm) ,Robot ,Learning methods ,020201 artificial intelligence & image processing ,Artificial intelligence ,Evolutionary Robotics ,business ,human activities ,030217 neurology & neurosurgery - Abstract
We investigate an evolutionary robot system where (simulated) modular robots can reproduce and create robot children that inherit the parents’ morphologies by crossover and mutation. Within this system we compare two approaches to creating good controllers, i.e., evolution only and evolution plus learning. In the first one the controller of a robot child is inherited, so that it is produced by applying crossover and mutation to the controllers of its parents. In the second one the controller of the child is also inherited, but additionally, it is enhanced by a learning method. The experiments show that the learning approach does not only lead to different fitness levels, but also to different (bigger) robots. This constitutes a quantitative demonstration that changes in brains, i.e., controllers, can induce changes in the bodies, i.e., morphologies.
- Published
- 2020
- Full Text
- View/download PDF