Al-Dakheeli Muhammed, Hadeer Essam, Beshoy Alber, Kirolos Samuel, Hagar Muhammed, M. Wagdy, Nouran Khaled, Hadeer Fawzy, Aya Tarek, Mohamed Abdel Salam, M. El-Kharashi
{"title":"Developing AI Agent with Functional Mockup Units for Car Autonomous Navigation","authors":"Al-Dakheeli Muhammed, Hadeer Essam, Beshoy Alber, Kirolos Samuel, Hagar Muhammed, M. Wagdy, Nouran Khaled, Hadeer Fawzy, Aya Tarek, Mohamed Abdel Salam, M. El-Kharashi","doi":"10.1109/icecs53924.2021.9665639","DOIUrl":null,"url":null,"abstract":"In this paper we present our implementation of a Deep Queue Network (DQN) AI Agent model for car autonomous navigation. The agent is capable of lane keeping without making any collisions with the surrounding vehicle and has learnt to move fast and safe in intersections. The model has been trained using two front camera sensors (depth and segmentation) and a collision detector. We also demonstrate how to connect this agent to functional mockup units (FMUs) to simulate the mechatronics part of the car. The deployment of our model has been demonstrated in a CARLA car simulator environment.","PeriodicalId":448558,"journal":{"name":"2021 28th IEEE International Conference on Electronics, Circuits, and Systems (ICECS)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 28th IEEE International Conference on Electronics, Circuits, and Systems (ICECS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/icecs53924.2021.9665639","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In this paper we present our implementation of a Deep Queue Network (DQN) AI Agent model for car autonomous navigation. The agent is capable of lane keeping without making any collisions with the surrounding vehicle and has learnt to move fast and safe in intersections. The model has been trained using two front camera sensors (depth and segmentation) and a collision detector. We also demonstrate how to connect this agent to functional mockup units (FMUs) to simulate the mechatronics part of the car. The deployment of our model has been demonstrated in a CARLA car simulator environment.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
