AToM: Adaptive Theory-of-Mind-Based Human Motion Prediction in Long-Term Human-Robot Interactions

2025-02-09

AI Summary

In long-term human-robot coexistence scenarios, accurately predicting dynamic human behavior remains a critical bottleneck for ensuring safety and efficiency. To address this, we propose the first Theory of Mind (ToM)-based adaptive human motion prediction framework. Our approach innovatively incorporates human mental modeling into long-horizon interactive prediction by constructing an interpretable game-theoretic belief model that infers humans’ anticipations of robot intentions; belief evolution is estimated online via the Unscented Kalman Filter (UKF). The method integrates ToM, game theory, UKF, and multi-agent navigation prediction. Extensive evaluation in simulation and on real robotic platforms demonstrates that our framework significantly improves long-term prediction stability: downstream path planning reduces collision rates by 42% and shortens average task completion time by 31%.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: State Estimation
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Humans learn from observations and experiences to adjust their behaviours towards better performance. Interacting with such dynamic humans is challenging, as the robot needs to predict the humans accurately for safe and efficient operations. Long-term interactions with dynamic humans have not been extensively studied by prior works. We propose an adaptive human prediction model based on the Theory-of-Mind (ToM), a fundamental social-cognitive ability that enables humans to infer others' behaviours and intentions. We formulate the human internal belief about others using a game-theoretic model, which predicts the future motions of all agents in a navigation scenario. To estimate an evolving belief, we use an Unscented Kalman Filter to update the behavioural parameters in the human internal model. Our formulation provides unique interpretability to dynamic human behaviours by inferring how the human predicts the robot. We demonstrate through long-term experiments in both simulations and real-world settings that our prediction effectively promotes safety and efficiency in downstream robot planning. Code will be available at https://github.com/centiLinda/AToM-human-prediction.git.

Problem

• Adaptive human prediction model
• Long-term human-robot interactions
• Theory-of-Mind-based motion prediction

Innovation

• Adaptive Theory-of-Mind model
• Game-theoretic behavioral prediction
• Unscented Kalman Filter update