• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.434-439
• /
• 1994

Recently neural networks leave been proposed as new computational tools for solving constrained optimization problems because of its computational power. In this paper, the shortest path finding algorithm is proposed by rising a Hopfield type neural network. In order to design a Hopfield type neural network, an energy function must be defined at first. To obtain this energy function, the concept of a vector-represented network is introduced to describe the connected path. Through computer simulations, it will be shown that the proposed algorithm works very well in many cases. The local minima problem of a Hopfield type neural network is discussed.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.17 no.5
• /
• pp.29-35
• /
• 2003

In this paper, a new optimum design method is proposed for the linear induction motor(LIM). The Genetic Neural Network(GNN) is introduced in the optimum design of LIM and the simulation result is compared with the Genetic Algorithm(GA) and Neural Network(NN). The maximum thrust and trust/weight are selected as the object functions. The comparison showed that the proposed method is better than GA and NN.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.10
• /
• pp.25-34
• /
• 1999

In this paper, an algorithm is presented to recognize lane and obstacles based on highway road image. The road images obtained by a video camera undergoes a pre-processing that includes filtering, edge detection, and identification of lanes. After this pre-processing, a part of image is grouped into 27 sub-windows and fed into a three-layer feed-forward neural network. The neural network is trained to indicate the road direction and the presence of absence of an obstacle. The proposed algorithm has been tested with the images different from the training images, and demonstrated its efficacy for recognizing lane and obstacles. Based on the test results, it can be said that the algorithm successfully combines the traditional image processing and the neural network principles towards a simpler and more efficient driver warning of assistance system

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1998.04a
• /
• pp.216-225
• /
• 1998

Although artificial neural network based on backpropagation algorithm is an excellent system simulator, it has still unsolved problems of its structure-decision and learning method. That is, we cannot find a general approach to decide the structure of the neural network and cannot train it satisfactorily because of the local optimum point which it frequently falls into. In addition, although there are many successful applications using backpropagation learning algorithm, there are few efforts to improve the learning algorithm itself. In this study, we suggest a general way to construct the hidden layer of the neural network using binary genetic algorithm and also propose the various learning methods by which the global minimum value of the teaming error can be obtained. A XOR problem and line heating problems are investigated as examples.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.11
• /
• pp.192-199
• /
• 2002

This paper demonstrates that hybrid neural-genetic multimodel parameter estimation algorithm can be applied to structured system Identification of electro-hydraulic servo system. This algorithm are consist of a recurrent incremental credit assignment (ICRA) neural network and a genetic algorithm. The ICRA neural network evaluates each member of a generation of model and genetic algorithm produces new generation of model. We manufactured electro-hydraulic servo system and the hybrid neural-genetic multimodel parameter estimation algorithm is applied to the task to find the parameter values(mass, damping coefficient, bulk modulus, spring coefficient) which minimize total square error.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.6
• /
• pp.442-447
• /
• 2003

This paper demonstrates that a modified hybrid neural-genetic multimodel parameter estimation algorithm can be applied to structured system identification of an electro-hydraulic servo system. This algorithm is consists of a recurrent incremental credit assignment(ICRA) neural network and a genetic algorithm. The ICRA neural network evaluates each member of a generation of model and genetic algorithm produces new generation of model. The modified hybrid neural-genetic multimodel parameter estimation algorithm is applied to an electro-hydraulic servo system the task to find the parameter values such as mass, damping coefficient, bulk modulus, spring coefficient and disturbance, which minimizes the total square error.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.5
• /
• pp.196-203
• /
• 2003

This paper demonstrates that an improved hybrid neural-genetic multimodel parameter estimation algorithm can be applied to the structured system identification of an electro-hydraulic servo system. This algorithm is consists of a recurrent incremental credit assignment (ICRA) neural network and a genetic algorithm, The ICRA neural network evaluates each member of a generation of model and the genetic algorithm produces new generation of model. We manufactured an electro-hydraulic servo system and the improved hybrid neural-genetic multimodel parameter estimation algorithm is applied to the task to find the parameter values, such as mass, damping coefficient, bulk modulus, spring coefficient and disturbance, which minimize total square error.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.1
• /
• pp.76-82
• /
• 2005

Speech signal has various features of speakers. This feature is extracted from speech signal processing. The speaker is identified by the speaker identification system. In this paper, we propose the speaker identification system that uses the incremental learning based on neural network. Recorded speech signal through the microphone is blocked to the frame of 1024 speech samples. Energy is divided speech signal to voiced signal and unvoiced signal. The extracted 12 orders LPC cpestrum coefficients are used with input data for neural network. The speakers are identified with the speaker identification system using the neural network. The neural network has the structure of MLP which consists of 12 input nodes, 8 hidden nodes, and 4 output nodes. The number of output node means the identified speakers. The first output node is excited to the first speaker. Incremental learning begins when the new speaker is identified. Incremental learning is the learning algorithm that already learned weights are remembered and only the new weights that are created as adding new speaker are trained. It is learning algorithm that overcomes the fault of neural network. The neural network repeats the learning when the new speaker is entered to it. The architecture of neural network is extended with the number of speakers. Therefore, this system can learn without the restricted number of speakers.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.21 no.7
• /
• pp.103-108
• /
• 2007

Neural network is used in many fields of control systems currently. However, It is not easy to obtain input-output pattern when neural network is used for the system of a single feedback controller and it is difficult to get satisfied performance with neural network when load changes rapidly or disturbance is applied. To resolve these problems, this paper proposes a new mode to implement a neural network controller by installing a real object in place of activation function of Neural Network output node. As the Neural Network self adaptive control system is designed in simple structure neural network input-output pattern problem is solved naturally and real tin Loaming becomes possible through general back propagation algorithm. The effect of the proposed Neural Network self adaptive control algorithm was verified in a test of controlling the speed of a A.C. servo motor equipped with a high speed computing capable DSP (TMS320C32) on which the proposed algorithm was loaded.