Neuro-Optimizer: A New Artificial Intelligent Optimization Tool and Its Application for Robot Optimal Controller Design

The main objective of this paper is to introduce a new intelligent optimization technique that uses a predictioncorrection strategy supported by a recurrent neural network for finding a near optimal solution of a given objective function. Recently there have been attempts for using artificial neural networks (ANNs) in optimization problems and some types of ANNs such as Hopfield network and Boltzmann machine have been applied in combinatorial optimization problems. However, ANNs cannot optimize continuous functions and discrete problems should be mapped into the neural networks architecture. To overcome these shortages, we introduce a new procedure for stochastic optimization by a recurrent artificial neural network. The introduced neurooptimizer (NO) starts with an initial solution and adjusts its weights by a new heuristic and unsupervised rule to compute the best solution. Therefore, in each iteration, NO generates a new solution to reach the optimal or near optimal solutions. For comparison and detailed description, the introduced NO is compared to genetic algorithm and particle swarm optimization methods. Then, the proposed method is used to design the optimal controller parameters for a five bar linkage manipulator robot. The important characteristics of NO are: convergence to optimal or near optimal solutions, escaping from local minima, less function evaluation, high convergence rate and easy to implement.