Entity

Time filter

Source Type


Yang J.,Hefei University of Technology | Yang J.,Changhzou Key Laboratory of Software Technology and Applications | Zhuang Y.,Changhzou Key Laboratory of Software Technology and Applications | Zhuang Y.,Changzhou Institute of Technology
Applied Soft Computing Journal | Year: 2010

This paper presents an improved ant colony optimization algorithm (IACO) for solving mobile agent routing problem. The ants cooperate using an indirect form of communication mediated by pheromone trails of scent and find the best solution to their tasks guided by both information (exploitation) which has been acquired and search (exploration) of the new route. Therefore the premature convergence probability of the system is lower. The IACO can solve successfully the mobile agent routing problem, and this method has some excellent properties of robustness, self-adaptation, parallelism, and positive feedback process owing to introducing the genetic operator into this algorithm and modifying the global updating rules. The experimental results have demonstrated that IACO has much higher convergence speed than that of genetic algorithm (GA), simulated annealing (SA), and basic ant colony algorithm, and can jump over the region of the local minimum, and escape from the trap of a local minimum successfully and achieve the best solutions. Therefore the quality of the solution is improved, and the whole system robustness is enhanced. The algorithm has been successfully integrated into our simulated humanoid robot system which won the fourth place of RoboCup2008 World Competition. The results of the proposed algorithm are found to be satisfactory. © 2009 Elsevier B.V. All rights reserved.


Yang J.,Changhzou Key Laboratory of Software Technology and Applications | Yang J.,Hefei University of Technology | Zhuang Y.,Changhzou Key Laboratory of Software Technology and Applications | Zhuang Y.,Changzhou Institute of Technology | Li C.,Changzhou Institute of Technology
International Journal of Advanced Robotic Systems | Year: 2013

This paper proposes a behavior-switching control strategy of an evolutionary robot based on Artificial Neural Network (ANN) and genetic algorithm (GA). This method is able not only to construct the reinforcement learning models for autonomous robots and evolutionary robot modules that control behaviors and reinforcement learning environments, and but also to perform the behavior-switching control and obstacle avoidance of an evolutionary robot in the unpredictable environments with the static and moving obstacles by combining ANN and GA. The experimental results have demonstrated that our method can perform the decision-making strategy and parameter setup optimization of ANN and GA by learning and can effectively escape from a trap of local minima, avoid motion deadlock status of humanoid soccer robotic agents, and reduce the oscillation of the planned trajectory among the multiple obstacles by crossover and mutation. We have successfully applied some results of the proposed algorithm to our simulation humanoid robotic soccer team CIT3D which won the 1st prize of RoboCup Championship and ChinaOpen2010 and the 2 nd place of the official RoboCup World Championship on 5-11, July 2011 in Istanbul, Turkey. In comparison with the conventional behavior network and the adaptive behavior method, our algorithm simplified the genetic encoding complexity, improved the convergence rate ρ and the network performance. © 2013 Yang et al.; licensee InTech.


Yang J.,Changhzou Key Laboratory of Software Technology and Applications | Yang J.,Hefei University of Technology | Zhuang Y.,Changhzou Key Laboratory of Software Technology and Applications | Zhuang Y.,Changzhou Institute of Technology | And 2 more authors.
Computers and Geosciences | Year: 2012

Inspired by Krige' variogram and the multi-channel filtering theory for human vision information processing, this paper proposes a novel algorithm for segmenting the textures based on experimental semi-variogram function (ESVF), which can simultaneously describe structural property and statistical property of textures. The single variogram function value (SVFV) and the variance distance obtained by ESVF are used as texture feature description for segmenting textures. The feasibility and effectiveness of the proposed method are demonstrated by testing on some texture images. The computational complexity of the proposed approach depends neither on the number of the textures nor on the number of the gray levels, and only on the size of the image blocks. We have proved theoretically that the algorithm has the advantages of direction invariability and a higher sensitivity to different textures and can detect almost all kinds of the boundaries of the shape textures. Experimental results on the Brodatz texture databases show that the performance of this algorithm is superior to the traditional techniques such as texture spectrum, SIFT, k-mean method, and Gabor filters. The proposed approach is found to be robust, efficient, and satisfactory. © 2012.

Discover hidden collaborations