Explorations in evolutionary robotics
Adaptive Behavior
Evolutionary algorithms: from recombination to search distributions
Theoretical aspects of evolutionary computing
The Handbook of Brain Theory and Neural Networks
The Handbook of Brain Theory and Neural Networks
Modelling and Simulation of Robot Manipulators: A Parallel Processing Approach
Modelling and Simulation of Robot Manipulators: A Parallel Processing Approach
Neuro-fuzzy adaptive control based on dynamic inversion for robotic manipulators
Fuzzy Sets and Systems - Special issue: Fuzzy set techniques for intelligent robotic systems
Parameter control in evolutionary algorithms
IEEE Transactions on Evolutionary Computation
A comparison of predictive measures of problem difficulty inevolutionary algorithms
IEEE Transactions on Evolutionary Computation
Evolution of homing navigation in a real mobile robot
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hi-index | 0.00 |
This paper is concerned with the evolution development of the fuzzy neural networks. The learning procedures in such networks consists of two main development phases. First the network is evolutionary constructed through the use of evolutionary strategies and evolutionary programming. The second phase of the design of the network is aimed at its refinement and optimize the objective parameters to adapt to the nonlinear changes of the input space. There are two main optimization mechanisms working concurrently : structural development and paramertric refinement mechnisms using the integrated approach based on evolutionary programming and the evolution strategies.The evolutionary-based leaning has been compared with the gradientbased learning. The proposed technique is applied for controlling nonlinear dynamics of a multilink robot system. Extensive simulations have been performed, aimed at investigating the adaptive capability, tracking performance and the convergence properties of the proposed evolutionary model with respect to variety of situations. The results are remarkable compared with the fuzzy and neural networks optimized using gradient-based learning.