Tactile-Based Human Intent Recognition for Robot Assistive Navigation

2025-09-19

AI Summary

Existing robot navigation interfaces fail to replicate the natural, efficient physical interactions observed between humans and caregivers, thereby limiting assistive effectiveness. To address this, we propose Tac-Nav: a tactile navigation system integrating a cylindrical tactile skin onto the Stretch 3 mobile manipulator to infer navigation intent directly from hand-hold postures. Crucially, we introduce the first explicit geometric modeling of cylindrical tactile sensors and design a Cylinder-Kernel Support Vector Machine (CK-SVM) classifier to suppress rotational ambiguity, significantly enhancing robustness. Experiments demonstrate CK-SVM achieves 97.1% and 90.8% accuracy in simulation and real-world settings, respectively—outperforming four baseline methods. A user study confirms Tac-Nav’s superior intuitiveness and user preference over conventional joystick and voice-based control. This work establishes a novel, natural, and reliable haptic human–robot navigation paradigm for individuals with mobility impairments.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Interaction Techniques and Devices
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

Robot assistive navigation (RAN) is critical for enhancing the mobility and independence of the growing population of mobility-impaired individuals. However, existing systems often rely on interfaces that fail to replicate the intuitive and efficient physical communication observed between a person and a human caregiver, limiting their effectiveness. In this paper, we introduce Tac-Nav, a RAN system that leverages a cylindrical tactile skin mounted on a Stretch 3 mobile manipulator to provide a more natural and efficient interface for human navigational intent recognition. To robustly classify the tactile data, we developed the Cylindrical Kernel Support Vector Machine (CK-SVM), an algorithm that explicitly models the sensor's cylindrical geometry and is consequently robust to the natural rotational shifts present in a user's grasp. Comprehensive experiments were conducted to demonstrate the effectiveness of our classification algorithm and the overall system. Results show that CK-SVM achieved superior classification accuracy on both simulated (97.1%) and real-world (90.8%) datasets compared to four baseline models. Furthermore, a pilot study confirmed that users more preferred the Tac-Nav tactile interface over conventional joystick and voice-based controls.

Problem

• Developing tactile-based intent recognition for robot assistive navigation systems
• Creating robust classification algorithm for cylindrical tactile sensor data
• Providing more natural interface compared to joystick and voice controls

Innovation

• Cylindrical tactile skin for natural interface
• Cylindrical Kernel SVM for robust classification
• Algorithm models sensor geometry for accuracy