ExTraCT - explainable trajectory corrections for language-based human-robot interaction using textual feature descriptions

• Main Authors: Yow, J-Anne, Garg, Neha Priyadarshini, Ramanathan, Manoj, Ang, Wei Tech
• Other Authors: School of Mechanical and Aerospace Engineering
• Format: Article
• Language: English
• Published: 2025
• Subjects: Engineering; Human-robot interaction; Language in robotics
• Online Access: https://hdl.handle.net/10356/182191
• Institution: Nanyang Technological University

Introduction: In human-robot interaction (HRI), understanding human intent is crucial for robots to perform tasks that align with user preferences. Traditional methods that aim to modify robot trajectories based on language corrections often require extensive training to generalize across diverse ob...