Peer-to-Peer Ultra-Wideband Localization for Hands-Free Control of a Human-Guided Smart Stroller

Sensors (Basel). 2024 Jul 25;24(15):4828. doi: 10.3390/s24154828.

Abstract

We propose a hands-free control system for a human-guided smart stroller. The proposed method uses real-time peer-to-peer localization technology of the human and stroller to realize an intuitive hands-free control system based on the relative position between the human and the stroller. The control method is also based on functional and mechanical safety to ensure the safety of the stroller's occupant (child) and the pilot (parent) during locomotion. In this paper, first, we present a preliminary investigation of the humans' preference for the relative position in the context of hands-free guided strollers. Then, we present the control method and a prototype implemented with an electric wheelchair and UWB sensors for localization. We present an experimental evaluation of the proposed method with 14 persons walking with the developed prototype to investigate the usability and soundness of the proposed method compared to a remote joystick and manual operation. The evaluation experiments were conducted in an indoor environment and revealed that the proposed method matches the performance of joystick control but does not perform as well as manual operation. Notably, for female participants, the proposed method significantly surpasses joystick performance and achieves parity with manual operation, which shows its efficacy and potential for a smart stroller. Also, the results revealed that the proposed method significantly decreased the user's physical load compared to the manual operation. We present discussions on the controllability, usability, task load, and safety features of the proposed method, and conclude this work with a summary assessment.

Keywords: hands-free control; human-centered design; human–machine interface; motorized stroller; peer-to-peer localization.

MeSH terms

  • Adult
  • Equipment Design
  • Female
  • Humans
  • Male
  • User-Computer Interface
  • Walking / physiology
  • Wheelchairs*

Grants and funding

This work was supported by the Japan Science and Technology Agency (JST) for the CREST research project on Social Signaling (JPMJCR19A2) and the Japan Society for the Promotion of Science KAKENHI [grant number 23H00485].