文献类型：会议论文

英文题名：LoomoRescue: An Affordable Rescue Robot for Evacuation Situations

作者：Matthies, Denys J. C.[1];Schmidt, Sven Ole[1];He, Yuqi[1,2];Yu, Zhouyao[1,2];Hellbruck, Horst[1]

机构：[1]Tech Univ Appl Sci, Lubeck, Germany;[2]East China Univ Sci & Technol, Shanghai, Peoples R China

会议论文集：4th International Conference on Design, Operation and Evaluation of Mobile Communications (MOBILE) Held as Part of the 25th International Conference on Human-Computer Interaction (HCII)

会议日期：JUL 23-28, 2023

会议地点：Copenhagen, DENMARK

语种：英文

外文关键词：Rescue Robot; Computer Vision; Optical Vital Sign Detection; Posture Recognition; SLAM; Autonomous Navigation

摘要：Rescue robots can play an important role in disaster situations, such as locating people for evacuations. This paper demonstrates how to transform an affordable consumer robot, such as the Loomo Segway, into an intelligent rescue robot. Our proof-of-concept shows how LoomoRescue can autonomously browse offices to locate people and how to detect their vital signs through posture and heart rate detection in real-time. Our indoor localization is a SLAM approach based on an external UWB position system plus a movement correction with an ultrasound sensor in combination with an IMU. The accuracy of linear movement showed minor deviation with an average error of 1.65%, while the angular movement showed an error of 2.43%. We classify three types of critical postures with an average detection rate of 78.33% within a distance of 1-20 m. Our optical heart rate detection is 87.3% accurate to the ground truth. We envision that such an affordable robot can be used for evacuation purposes as it may be part of the standard inventory in the future.