{"title":"融合DQN算法的无人救援车导航","authors":"Wenguan Cao, Xiaoci Huang, Fanglin Shu","doi":"10.1145/3366194.3366293","DOIUrl":null,"url":null,"abstract":"When unmanned rescue vehicle (URV) performing rescue missions in the disaster area, URV will not be feasible to continue to use the built-in map for path planning, and even cause more serious consequences. Therefore, when a rescue vehicle works with a human being in a dynamic environment such as disaster recovery, it is necessary to quickly complete the task of adapting to the scene and learning to perform its duties. In this paper, the search and rescue robot first collects environmental information according to the camera sensor installed by itself, and then constructs the intelligent vehicle behavior decision model from the vehicle driving efficiency and optimal path. Secondly, the search and rescue robot estimates through the improved DQN network structure value function. And update the network parameters to get the corresponding Q value through the training network. Finally, the experimental results show that the algorithm can quickly generate a safe and smooth path that satisfies the kinematic constraints of the vehicle.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Unmanned rescue vehicle navigation with fused DQN algorithm\",\"authors\":\"Wenguan Cao, Xiaoci Huang, Fanglin Shu\",\"doi\":\"10.1145/3366194.3366293\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"When unmanned rescue vehicle (URV) performing rescue missions in the disaster area, URV will not be feasible to continue to use the built-in map for path planning, and even cause more serious consequences. Therefore, when a rescue vehicle works with a human being in a dynamic environment such as disaster recovery, it is necessary to quickly complete the task of adapting to the scene and learning to perform its duties. In this paper, the search and rescue robot first collects environmental information according to the camera sensor installed by itself, and then constructs the intelligent vehicle behavior decision model from the vehicle driving efficiency and optimal path. Secondly, the search and rescue robot estimates through the improved DQN network structure value function. And update the network parameters to get the corresponding Q value through the training network. Finally, the experimental results show that the algorithm can quickly generate a safe and smooth path that satisfies the kinematic constraints of the vehicle.\",\"PeriodicalId\":105852,\"journal\":{\"name\":\"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-09-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3366194.3366293\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3366194.3366293","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Unmanned rescue vehicle navigation with fused DQN algorithm
When unmanned rescue vehicle (URV) performing rescue missions in the disaster area, URV will not be feasible to continue to use the built-in map for path planning, and even cause more serious consequences. Therefore, when a rescue vehicle works with a human being in a dynamic environment such as disaster recovery, it is necessary to quickly complete the task of adapting to the scene and learning to perform its duties. In this paper, the search and rescue robot first collects environmental information according to the camera sensor installed by itself, and then constructs the intelligent vehicle behavior decision model from the vehicle driving efficiency and optimal path. Secondly, the search and rescue robot estimates through the improved DQN network structure value function. And update the network parameters to get the corresponding Q value through the training network. Finally, the experimental results show that the algorithm can quickly generate a safe and smooth path that satisfies the kinematic constraints of the vehicle.