Issues in evolutionary robotics
Proceedings of the second international conference on From animals to animats 2 : simulation of adaptive behavior: simulation of adaptive behavior
Automatic creation of an autonomous agent: genetic evolution of a neural-network driven robot
SAB94 Proceedings of the third international conference on Simulation of adaptive behavior : from animals to animats 3: from animals to animats 3
An introduction to genetic algorithms
An introduction to genetic algorithms
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Noise and the Reality Gap: The Use of Simulation in Evolutionary Robotics
Proceedings of the Third European Conference on Advances in Artificial Life
The artificial life roots of artificial intelligence
Artificial Life
Modeling adaptive autonomous agents
Artificial Life
Analyzing dynamic fitness landscapes of the targeting problem of chaotic systems
EvoApplications'12 Proceedings of the 2012t European conference on Applications of Evolutionary Computation
| Hi-index | 0.00 |
An autonomous robot “Khepera” was simulated with a sensory-motormodel, which evolves in the genetic algorithm (GA) framework, with thefitness evaluation in terms of the navigation performance in a mazecourse. The sensory-motor model is a developed neural network decodedfrom a graph-represented chromosome, which is evolved in the GAprocess with several genetic operators.It was found that the fitness landscape is very rugged when it isobserved at the starting point of the course. A hypothesis for thisruggedness is proposed, and is supported by the measurement of fractaldimension. It is also observed that the performance is sometimesplagued by “Loss of Robustness,” after the robot makes majorevolutionary jumps. Here, the robustness is quantitatively defined asa ratio of the averaged fitness of the evolved robot navigating inperturbed environments over the fitness of the evolved robot in thereferenced environment.Possible explanation of robustness loss is the over-adaptationoccurred in the environment where the evolution was taken place.Testing some other possibilities for this loss of robustness, manysimulation experiments were conducted which smooth out the discretefactors in the model and environment. It was found that smoothing thediscrete factors does not solve the loss of robustness. An effectivemethod for maintaining the robustness is the use of averaged fitnessover different navigation conditions.The evolved models in the simulated environment were tested bydown-loading the models into the real Khepera robot. It isdemonstrated that the tendency of fitness values observed in thesimulation were adequately regenerated.