Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BUDD-e: an autonomous robotic guide for visually impaired users

Jinyang Li, Marcello Farina, Luca Mozzarelli, Luca Cattaneo, Panita Rattamasanaprapai, Eleonora A. Tagarelli, Matteo Corno, Paolo Perego, Giuseppe Andreoni, Emanuele Lettieri

TL;DR

Visually impaired users face safety and accessibility challenges navigating indoor and outdoor public spaces, including healthcare facilities. The authors introduce BUDD-e, an autonomous robotic guide built on the Yape TWIP platform and a Smart Tether System, with an H2-based robust longitudinal controller and an autonomous driving stack for safe following and obstacle avoidance. The paper details mechanical design, target-tracking distance estimation, and a multi-layer control framework combining distance control, tether guidance, global/local planning, and a supervisory FSM, plus an extensive user-centered evaluation in a hospital setting and a final demonstration. Results indicate high usability (SUS) scores, positive acceptance (TAM), and safe, responsive guidance across indoor/outdoor scenarios, highlighting practical impact and avenues for future refinements.

Abstract

This paper describes the design and the realization of a prototype of the novel guide robot BUDD-e for visually impaired users. The robot has been tested in a real scenario with the help of visually disabled volunteers at ASST Grande Ospedale Metropolitano Niguarda, in Milan. The results of the experimental campaign are throughly described in the paper, displaying its remarkable performance and user-acceptance.

BUDD-e: an autonomous robotic guide for visually impaired users

TL;DR

Visually impaired users face safety and accessibility challenges navigating indoor and outdoor public spaces, including healthcare facilities. The authors introduce BUDD-e, an autonomous robotic guide built on the Yape TWIP platform and a Smart Tether System, with an H2-based robust longitudinal controller and an autonomous driving stack for safe following and obstacle avoidance. The paper details mechanical design, target-tracking distance estimation, and a multi-layer control framework combining distance control, tether guidance, global/local planning, and a supervisory FSM, plus an extensive user-centered evaluation in a hospital setting and a final demonstration. Results indicate high usability (SUS) scores, positive acceptance (TAM), and safe, responsive guidance across indoor/outdoor scenarios, highlighting practical impact and avenues for future refinements.

Abstract

This paper describes the design and the realization of a prototype of the novel guide robot BUDD-e for visually impaired users. The robot has been tested in a real scenario with the help of visually disabled volunteers at ASST Grande Ospedale Metropolitano Niguarda, in Milan. The results of the experimental campaign are throughly described in the paper, displaying its remarkable performance and user-acceptance.

Paper Structure

This paper contains 27 sections, 19 equations, 21 figures, 4 tables.

Table of Contents

  1. Introduction
  2. The BUDD-e robotic guide
  3. The mechanical setup
  4. The control system
  5. The Yape longitudinal speed model identification
  6. The Yape longitudinal speed discrete-time switching model
  7. $\mathcal{H}_2$ design
  8. Distance controller design
  9. Integral action
  10. Experimental validation
  11. Robot-user distance estimation
  12. Autonomous Navigation
  13. Localization
  14. Global Planner
  15. Local Planner and Collision Avoidance
  16. ...and 12 more sections

Figures (21)

  • Figure 1: Scheme of the BUDD-e robotic guide
  • Figure 2: The Yape robot
  • Figure 3: The Smart Tether System: 1. Box with electronic components; 2. Motor; 3. Shaft; 4. Polyurethane tube; 5. Proximity sensor; 6. Rigid cable; 7. Load cell; 8. Elastic cable; 9. Handle.
  • Figure 4: Example of the data used to identify the longitudinal dynamical model of Yape. Black dashed line: $v_{\rm ref}$; gray solid line: measured Yape velocity $v$; blue solid line: simulated Yape velocity $v$.
  • Figure 5: Response (i.e., measured user-robot distance $d$) to step-wise inputs of the distance control systems designed: \ref{['eq:control1']} (blue line) and \ref{['eq:control2']} (red line). Dashed lines: reference distances $d^o$.
  • ...and 16 more figures