Odor-source localization in the clean room by an autonomous mobile sensing system

Abstract Experiments on odor-source localization using an autonomous mobile system were performed in a clean room. A new exploratory algorithm was developed since the algorithm of upwind searching previously developed by our group was found to be applicable only on a limited situation. The switchover of two strategies, moving upwind and along the gas-concentration gradient, was introduced to extend the applicable area. A threshold value of the gas sensor output used for exchanging two strategies was automatically adjusted during localization movements. As the flexibility of the system was enhanced, an ethanol source was successfully localized from almost anywhere in the prepared measurement space.