A Multi-Layer Navigation System for Smart Wheelchairs with a Limited Input Interface

Piero Simonetto; Anna Polato; Francesco Pasti; Muhammad Rameez ur Rahman; Sebastiano Vascon; Nicola Bellotto; Emanuele Menegatti; Stefano Tortora

cs.ROOct 28, 2026

A Multi-Layer Navigation System for Smart Wheelchairs with a Limited Input Interface

Authors: Piero Simonetto, Anna Polato, Francesco Pasti, Muhammad Rameez ur Rahman, Sebastiano Vascon, Nicola Bellotto, Emanuele Menegatti, Stefano Tortora,

Organizations: Department of Information Engineering, University of Padova, Italy · Ca’ Foscari University of Venice · European Centre for Living Technology · Padua Neuroscience Center, University of Padova, Italy

Abstract

Smart wheelchairs are robotic devices that gather information from the surroundings and allow people to safely drive through various environments. They usually rely on the interpretation of high-level user's intentions, typically decoded through human-robot interface (HRI) or brain-machine interface (BMI). However, navigating in an unknown environment and accurately understanding user commands remain significant challenges. Misinterpretation of commands or the environment can lead to user frustration, distrust of the system, or even unsafe situations. In this paper, we present a novel Multi-Layer Navigation System (MLNS) for a smart wheelchair that enables navigation in unknown environments while interpreting and safely contextualizing user's intentions even in the case of limited and faulty commands. The stack relies on a Virtual Sensors-Driven Potential Fields (ViSPF) method for obstacle detection that accounts for the wheelchair's actual size, a Model Predictive Control (MPC) module for trajectory generation, and an open-vocabulary computer vision algorithm for target identification. Finally, a system based on Priority Queue (PQ) was implemented for target selection. We validated our system on a real smart wheelchair with experiments involving 10 participants without disabilities, including scenarios with faulty input interfaces. The results demonstrate the robustness and effectiveness of our system in real-world environments, proving its potential for practical applications in assistive technologies.

Figures & tables

Fig. 1: The proposed mlns ( mlns ) for a smart wheelchair semi-autonomous navigation. Given the user and OpenNav inputs, a pq combined with mpc and apf+vs controls the navigation of the wheelchair.

Fig. 2: Graphical representation of the proposed apf+vs . The wheelchair is represented with a rectangle surrounded by its four wheels. C represents the center of the wheelchair and $p_{\{1,2,3,4\}}$ the new vs . $\alpha_{\{C,p_{1},p_{2},p_{3},p_{4}\}}$ .

Fig. 3: Map of the laboratory where the experiment took place. The green point is the starting position, Target X and Y are desks, and Target Z is a person.

Fig. 4: A) Normalized heatmaps of the aggregated trajectories produced during the experiments (center). B-D) Normalized heatmaps of the trajectories divided per target and modality. White dashed lines represent the average Baseline path for each target. Heatmap represents a color gradient of the crossed positions, from the more (in yellow) to the less (in blue) common.

Fig. 5: Distribution of the paths length traveled for each target and navigation modality. Horizontal brackets indicate pairs with a statistically significant difference (Kruskal-Wallis with Dunn’s post-hoc test, $p<0.05$ )

Fig. 6: Distribution of the completion time (top) and number of commands sent by the user (bottom) of each target.

	b	wo	on
X	92.1 $\pm$ 17.3	108.4 $\pm$ 44.1	91.5 $\pm$ 31.5
Y	69.9 $\pm$ 15.5	74.0 $\pm$ 65.2	77.3 $\pm$ 38.3
Z	77.7 $\pm$ 13.8	81.7 $\pm$ 39.2	77.7 $\pm$ 23.9

TABLE I: Mean and standard deviation of completion time per target and method

	b	wo	on
X	4 $\pm$ 3.5	5.3 $\pm$ 5.4	3.1 $\pm$ 7.6
Y	3.2 $\pm$ 1.8	4.7 $\pm$ 9.4	5.8 $\pm$ 6.1
Z	3.1 $\pm$ 4.5	3.9 $\pm$ 6.1	2 $\pm$ 2.9

TABLE II: Mean and standard deviation of number of commands per target and method

	b	wo	on
X	16.6766 $\pm$ 2.5074	19.8891 $\pm$ 7.1318	15.9831 $\pm$ 4.2086
Y	12.0068 $\pm$ 2.0267	12.7959 $\pm$ 10.3488	12.2596 $\pm$ 5.7333
Z	14.3877 $\pm$ 1.5002	14.8688 $\pm$ 6.6185	13.7696 $\pm$ 3.2384

TABLE III: Mean and standard deviation of length of the paths per target and method

Robot Navigation

Robot navigation studies how robots perceive, localize, plan, and move safely toward goals in static or changing environments.

See all 556 papers

Towards Real-World Applications with an Autonomous Powered Wheelchair
Jul 7, 2026Simone Arreghini, Alessandro Giusti, Alex Bordini +3Robot Navigation Assistive Robotics
Dalle Molle Institute for Artificial Intelligence (IDSIA), USI-SUPSI, Lugano, Switzerland. · Genny Factory, Sant’Antonino, Switzerland.
ChairNav: Cross-Embodiment Pre-training and Personalization for Long-Horizon Wheelchair Navigation
Oct 15, 2026Junhua Huang, Zhizheng Liu, Honglin He +2Robot Navigation Assistive Robotics
University of California, Los Angeles · Louisiana State University
SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs
Mar 14, 2026Yutong Zhang, Shaiv Y. Mehra, Bradley S. Duerstock +1Robot Navigation Social Robot Navigation
Edwardson School of Industrial Engineering, Purdue University, West Lafayette, IN 47907, USA · Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN 47907, USA

Jul 7, 2026J/K move · Enter open · S save

Towards Real-World Applications with an Autonomous Powered Wheelchair

Simone Arreghini, Alessandro Giusti, Alex Bordini, Enrico Ferrara+2

Dalle Molle Institute for Artificial Intelligence (IDSIA), USI-SUPSI, Lugano, Switzerland. · Genny Factory, Sant’Antonino, Switzerland.

Robot Navigation Assistive Robotics Human-Robot Interaction

Open paper AskPDF arXiv

Wheelchair users call for assistive mobility systems that provide active support, adapt to dynamic environments, and are intuitive and user-friendly. However, powered wheelchairs typically still provide limited autonomy and lack effective integration with advanced perception and navigation capabilities, particularly in complex real-world environments. This paper presents a preliminary study toward autonomous powered wheelchairs for real-world assistive mobility. We introduce a proof-of-concept prototype that integrates autonomous perception, gesture-based interaction, and navigation on a commercially available self-balancing powered wheelchair. The proposed system builds upon Genny Zero, a commercial self-balancing wheelchair that enables hands-free and intuitive operation through body-weight shifting. To extend its capabilities toward autonomous operation, we integrate an RGB-D camera for human-aware perception and interaction, together with a LiDAR sensor for localization and navigation. We demonstrate the integrated system in two assistive applications: (i) hailing, allowing users to call the wheelchair from a distance; and (ii) people-following, where the wheelchair follows a person using leader-follower strategies, including a constrained indoor navigation example. The results highlight the potential of combining autonomous robotics with assistive mobility platforms, while also showing the feasibility of the proposed integration and identifying the main technical challenges that must be addressed before moving toward user-ready, accessible, and intelligent mobility solutions. A video demonstrating the experimental setup and results is available at: https://youtu.be/LVAix_Qx7bM.

Abstract

Robot Navigation

Explore similar work

Towards Real-World Applications with an Autonomous Powered Wheelchair

ChairNav: Cross-Embodiment Pre-training and Personalization for Long-Horizon Wheelchair Navigation

SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs