Hybridizing evolutionary computation and reinforcement learning for the design of almost universal controllers for autonomous robots.
Darío MaravallJavier de Lope AsiaínJosé Antonio Martín H.Published in: Neurocomputing (2009)
Keyphrases
- evolutionary computation
- reinforcement learning
- autonomous robots
- evolutionary algorithm
- genetic programming
- computational intelligence
- machine learning and data mining
- evolutionary design
- evolvable hardware
- evolutionary learning
- robot control
- mobile robot
- genetic algorithm
- fuzzy systems
- learning classifier systems
- optimisation problems
- artificial life
- robot navigation
- learning algorithm
- swarm intelligence
- control system
- evolutionary strategy
- fitness function
- optimal policy
- evolutionary search
- optimization algorithm
- fuzzy logic
- evolutionary approaches
- temporal difference methods