不同模糊神经网络配置在自动驾驶车辆跟随算法中的比较研究

2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE) Pub Date : 2016-01-01 DOI:10.1109/ICCSCE.2016.7893609

John Paolo A. Ramoso, M. Ramos

{"title":"不同模糊神经网络配置在自动驾驶车辆跟随算法中的比较研究","authors":"John Paolo A. Ramoso, M. Ramos","doi":"10.1109/ICCSCE.2016.7893609","DOIUrl":null,"url":null,"abstract":"This paper investigates modelling vehicle following using Fuzzy-Neural Network (FNN). Architecture, training sets, and learning rate are manipulated to create 24 combinations of FNN. Generating two sets of weights per combination yields 48 simulations. Acceleration and deceleration profiles from seven electrical tricycles are observed while navigating through the University of the Philippines. A force equation has been applied to simulate vehicular dynamics. Each combination is then subjected to test run simulations to examine vehicular reactions to distance maintenance, velocity matching, and change in applied force to the vehicle. Results show that two 2 hidden layer FN and two NN allow a vehicle to successfully follow a lead vehicle.","PeriodicalId":6540,"journal":{"name":"2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE)","volume":"19 1","pages":"413-418"},"PeriodicalIF":0.0000,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Comparative study of different Fuzzy-Neural configurations for autonomous vehicle following algorithm\",\"authors\":\"John Paolo A. Ramoso, M. Ramos\",\"doi\":\"10.1109/ICCSCE.2016.7893609\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper investigates modelling vehicle following using Fuzzy-Neural Network (FNN). Architecture, training sets, and learning rate are manipulated to create 24 combinations of FNN. Generating two sets of weights per combination yields 48 simulations. Acceleration and deceleration profiles from seven electrical tricycles are observed while navigating through the University of the Philippines. A force equation has been applied to simulate vehicular dynamics. Each combination is then subjected to test run simulations to examine vehicular reactions to distance maintenance, velocity matching, and change in applied force to the vehicle. Results show that two 2 hidden layer FN and two NN allow a vehicle to successfully follow a lead vehicle.\",\"PeriodicalId\":6540,\"journal\":{\"name\":\"2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE)\",\"volume\":\"19 1\",\"pages\":\"413-418\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCSCE.2016.7893609\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCSCE.2016.7893609","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 3

摘要

本文研究了用模糊神经网络(FNN)建立车辆跟随模型。结构、训练集和学习率被操纵来创建24种FNN的组合。每个组合生成两组权重可以产生48次模拟。在菲律宾大学导航时，观察了七辆电动三轮车的加速和减速曲线。应用力方程对车辆动力学进行了模拟。然后对每种组合进行测试运行模拟，以检查车辆对距离维护，速度匹配和施加在车辆上的力的变化的反应。结果表明，两个2隐层FN和两个NN可以使车辆成功地跟随前导车辆。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Comparative study of different Fuzzy-Neural configurations for autonomous vehicle following algorithm

This paper investigates modelling vehicle following using Fuzzy-Neural Network (FNN). Architecture, training sets, and learning rate are manipulated to create 24 combinations of FNN. Generating two sets of weights per combination yields 48 simulations. Acceleration and deceleration profiles from seven electrical tricycles are observed while navigating through the University of the Philippines. A force equation has been applied to simulate vehicular dynamics. Each combination is then subjected to test run simulations to examine vehicular reactions to distance maintenance, velocity matching, and change in applied force to the vehicle. Results show that two 2 hidden layer FN and two NN allow a vehicle to successfully follow a lead vehicle.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2016 6th IEEE International Conference on Control System, Computing and Engineering (ICCSCE)

自引率

0.00%

发文量