Locating a time-varying contaminant source in naturally ventilated indoor environments: An experimental study to find effective multi-robot olfaction methods. (15th May 2022)

• Record Type: Journal Article
• Authors: Zhou, Suwan; Zhang, Canxin; Cai, Hao; Zhang, Biao; Feng, Qilin; Feng, Lihang; Li, Fei; Zhou, Bin
• Is Part Of: Building and environment. Volume 216(2022)
• Publication Date: 2022-05-15
• Subjects: Natural ventilation -- Time-varying source -- Multi-robot olfaction -- Source localization -- Experimental validation; Buildings -- Environmental engineering -- Periodicals; Building -- Research -- Periodicals; Constructions -- Technique de l'environnement -- Périodiques; Electronic journals; 696
• DOI: 10.1016/j.buildenv.2022.108954
• Languages: English
• ISSNs: 0360-1323

Abstract

Abstract: The multi-robot olfaction method is critical to ensure the health and safety of indoor environments. This research aims to find effective multi-robot olfaction methods for locating time-varying sources in naturally ventilated indoor environments. A source positioning system composed of three mobile robots was developed for locating two types of time-varying sources (periodic source and decaying source). A total of 120 validation experiments were first conducted in real natural ventilation environments. In these experiments, the average success rates of the proposed IPSO and IWOA methods reached 81.68% and 80.00%, respectively. In addition, these two methods only required approximately 30% of the source localization time compared with the two typical scanning methods. For a fair comparison of different methods, another 120 experiments were then conducted in imitated natural ventilation environments. In these experiments, the success rates of the IPSO and IWOA methods varied between 80% and 86.7%, while those of the SPSO and WUII methods ranged between 20% and 33.3%. The experimental results indicated that the IPSO and IWOA methods were far superior to the SPSO and WUII methods in locating a time-varying source in naturally ventilated indoor environments due to their strong robustness and high efficiency. Highlights: The coupling of time-varying source and natural ventilation was considered. Three mobile robots were employed to locate periodic and decaying sources. The experiments include real and imitated natural ventilation environments. A total of 240 validation and comparison experiments were conducted. Two multi-robot olfaction methods showed strong robustness and high efficiency.