@inproceedings{oai:ipsj.ixsq.nii.ac.jp:02006385,
 author = {Narumon,Jadram and Yuri,Nishikawa and Midori,Sugaya and Narumon Jadram and Yuri Nishikawa and Midori Sugaya},
 book = {Proceedings of Asia Pacific Conference on Robot IoT System Development and Platform},
 month = {Dec},
 note = {As populations age, low-speed mobility devices (LMDs), such as electric wheelchairs, are becoming increasingly important for supporting the mobility of older adults. However, outdoor use of LMDs can be stressful due to environmental factors such as uneven road surfaces, congestion, noise, and weather conditions. These stressors may discourage continued use, underscoring the need to identify and reduce potential stress locations where users are likely to experience stress. Existing research on potential stress location during LMD rides remains limited, focusing mainly on physical hazards, which might overlook potential stress locations. This study proposes a method to identify and visualize potential stress locations for LMD users based on user-identified environmental factors using multimodal sensor data. A field experiment was conducted in Toyosu, Tokyo, involving 34 participants navigating sidewalks with electric wheelchairs. Multimodal sensor data (temperature, humidity, noise, brightness, and acceleration), video recordings, and post-ride surveys were collected. Survey analysis identified three main stress factors: poor road surface conditions, nearby pedestrians or cyclists, and narrow bridges or sidewalks. These factors were quantified, aggregated into stress scores, and classified into three levels for each 100-meter route segment. This visualization demonstrates how multimodal sensing data can be transformed into actionable units for urban planning and accessibility improvement., As populations age, low-speed mobility devices (LMDs), such as electric wheelchairs, are becoming increasingly important for supporting the mobility of older adults. However, outdoor use of LMDs can be stressful due to environmental factors such as uneven road surfaces, congestion, noise, and weather conditions. These stressors may discourage continued use, underscoring the need to identify and reduce potential stress locations where users are likely to experience stress. Existing research on potential stress location during LMD rides remains limited, focusing mainly on physical hazards, which might overlook potential stress locations. This study proposes a method to identify and visualize potential stress locations for LMD users based on user-identified environmental factors using multimodal sensor data. A field experiment was conducted in Toyosu, Tokyo, involving 34 participants navigating sidewalks with electric wheelchairs. Multimodal sensor data (temperature, humidity, noise, brightness, and acceleration), video recordings, and post-ride surveys were collected. Survey analysis identified three main stress factors: poor road surface conditions, nearby pedestrians or cyclists, and narrow bridges or sidewalks. These factors were quantified, aggregated into stress scores, and classified into three levels for each 100-meter route segment. This visualization demonstrates how multimodal sensing data can be transformed into actionable units for urban planning and accessibility improvement.},
 pages = {24--31},
 publisher = {情報処理学会},
 title = {Mapping Potential Stress Locations for Low-Speed Mobility Rides Based on User-Identified Environmental Factors Using Multimodal Sensor Data},
 volume = {2025},
 year = {2025}
}