• Title/Summary/Keyword: Aritificial Neural Network

Search Result 3, Processing Time 0.117 seconds

A variable PID controller for robots using evolution strategy and neural network (Evolution strategy와 신경회로망에 의한 로봇의 가변 PID제어기)

  • 최상구;김현식;최영규
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1997.10a
    • /
    • pp.1585-1588
    • /
    • 1997
  • In this paper, divide total workspace of robot manipulator into several subspaces and construct PID controller ineach subspace. Using EvolutionSTrategy we optimize the gains of PID controller in each subspace. But the gains may have a large difference on the boundary of subspaces, which can cause bad oscillatory performance. So we use Aritificial Neural Network to have continuous gain curves htrough the entire subspaces. Simualtion results show that the proposed method is quite useful.

  • PDF

Vehicle Load Analysis using Bridge-Weigh-in-Motion System in a Cable Stayed Bridge (BWIM 시스템을 사용한 사장교의 차량하중 분석)

  • Park, Min-Seok;Lee, Jung-Whee;Kim, Sung-Kon;Jo, Byung-Wan
    • Journal of the Earthquake Engineering Society of Korea
    • /
    • v.10 no.6 s.52
    • /
    • pp.1-8
    • /
    • 2006
  • This paper describes the procedures developing the algorithm for analyzing signals acquired from the Bridge Weigh-in-Motion (BWIM) system installed in Seohae Bridge as a part of the bridge monitoring system. Through the analysis procedure, information about heavy traffics such as weight, speed, and number of axles are attempted to be extracted from time domain strain data of the BWIM system. One of numerous pattern recognition techniques, artificial neural network (ANN) is employed since it can effectively include dynamic effects, bridge-vehicle interaction, etc. A number of vehicle running experiments with sufficient load cases are executed to acquire training and/or test set of ANN. Extracted traffic information can be utilized for developing quantitative database of loading effect. Also, it can contribute to estimate fatigue lift or current health condition, and design truck can be revised based on the database reflecting recent trend of traffic.

Tuning Backpropagation Networks for Analyzing NIR Data

  • M.A.Hana;W.F.McClure;T.B.Whitaker
    • Near Infrared Analysis
    • /
    • v.2 no.1
    • /
    • pp.15-23
    • /
    • 2001
  • Designing (specifying the number of nodes in each layer) and training (calibration and validation) back-propagation (BP) for analyzing NIR data can be an arduous and time-consuming task. Actually, training is somewhat trivial. A BP network may be trained by randomly dividing the data set (DS) into two parts, training the network with one part and checking its performance with the other part. However, this procedure is plagued with the lack of objective information about network characteristics - the required number of nodes in the hidden layer(s) and the number of epochs needed to train for optimal performance. Work reported in this paper compares a BP network tuning procedure with a conventional reference (training and testing) procedure. The tuning procedure, believed to have several novel attributes, involved randomly dividing a data set into five groups. Each of the five groups was randomly subdivided into two groups with 80% in a training set and 20% in a tuning set. Training was interrupted periodically after every 100 epochs. During each interruption, network performance was checked against the tuning set - each time recording the mean-squared error (MSE) and the number of epochs (K) needed to reach this point. This procedure continued until a plot of MSE vs total epochs identified a minimum MSE. The number of epochs required achieve minimum MSE was noted. Now optimized (or tuned), network performance was determined by testing the network with all available data. One nice feature of using the tuning method is that the entire process can be executed without user input - i.e., the whole process of developing and training a BP network becomes objective. Four different near infrared data sets (A, B, C and D) were used in this work. Tow of the data sets (A and B) were used to determine the concentration of nicotine in tobacco samples. The other two sets (C and D) were used as a basis for classifyign tobaccos. The optimum BP architecture for each of the four data sets were those consisting of 1, 5, 2 and 1 hidden units in the hidden layer, respectively. The suggested tuning method improved, though marginally in some cases, the true performances of all calibration models as well as their standard deviations. since this work was dependent upon the artificial neural network (ANN) literature, a glossary of terms is given at the end of this paper. Results indicate improved performance using the tuning procedure. In addition, BP network calibrations were better than multiple linear regression (MLR) calibrations on the same data.