{"title":"Backpropagation Neural Network for Tuning PID Pan-Tilt Face Tracking","authors":"D. Permatasari, D. Maharani","doi":"10.1109/ICITISEE.2018.8720968","DOIUrl":null,"url":null,"abstract":"This paper presents a method for solving tuning PID Pan-Tilt Face Tracking. PID conventional method is developed to self-tuning gain of PID using Backpropagation Neural Network (BPNN) during the process (online) then achieves the desired target of human face which has more robust and minimal error. This plant uses three input neuros (references input), five hidden neuros, and three output neuros (Kp, Ki, and Kd). For initialization learning rate (alpha) and momentum (gamma) using 0.1 and 0.3 with random initialization weight. The pan system result has a fast response with overshoot 0.68%, peak time 0.65s, and rise time 0.48s with Kp = 2.9416, Ki = 0.393, Kd = 8.647 and for tilt system with overshoot 1.59%, rise time 0.49 s, and peak time 0.7 s. PID controller by Backpropagation Neural Network, it is obtained better reference output results with faster and fewer responses overshoot.","PeriodicalId":180051,"journal":{"name":"2018 3rd International Conference on Information Technology, Information System and Electrical Engineering (ICITISEE)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 3rd International Conference on Information Technology, Information System and Electrical Engineering (ICITISEE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICITISEE.2018.8720968","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
This paper presents a method for solving tuning PID Pan-Tilt Face Tracking. PID conventional method is developed to self-tuning gain of PID using Backpropagation Neural Network (BPNN) during the process (online) then achieves the desired target of human face which has more robust and minimal error. This plant uses three input neuros (references input), five hidden neuros, and three output neuros (Kp, Ki, and Kd). For initialization learning rate (alpha) and momentum (gamma) using 0.1 and 0.3 with random initialization weight. The pan system result has a fast response with overshoot 0.68%, peak time 0.65s, and rise time 0.48s with Kp = 2.9416, Ki = 0.393, Kd = 8.647 and for tilt system with overshoot 1.59%, rise time 0.49 s, and peak time 0.7 s. PID controller by Backpropagation Neural Network, it is obtained better reference output results with faster and fewer responses overshoot.