PECAN: Personalizing Robot Behaviors through a Learned Canonical Space

2024-07-22

arXiv.org

AI Summary

This study addresses the limitation in robotic behavior personalization—reliance on indirect feedback for inferring user preferences—by proposing a direct, cross-task style control paradigm. Methodologically, it constructs a low-dimensional, continuous normative space from human driving demonstrations, enabling users to intuitively select semantically interpretable behavior styles (e.g., “defensive driving”) via click-based interaction; a behavior decoding network then generalizes these selections to diverse task-specific actions. An interactive visualization interface supports real-time exploration and selection. The core contributions are: (1) the first framework enabling direct, semantic, and cross-task human control over robot behavior style; and (2) a normative space that jointly ensures interpretability and style consistency across tasks. User studies demonstrate that PECAN significantly improves preference satisfaction and interaction intuitiveness, while maintaining consistent stylistic generalization across multiple driving tasks.

Technology Category

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

Application Category

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

Abstract

Robots should personalize how they perform tasks to match the needs of individual human users. Today's robot achieve this personalization by asking for the human's feedback in the task space. For example, an autonomous car might show the human two different ways to decelerate at stoplights, and ask the human which of these motions they prefer. This current approach to personalization is indirect: based on the behaviors the human selects (e.g., decelerating slowly), the robot tries to infer their underlying preference (e.g., defensive driving). By contrast, our paper develops a learning and interface-based approach that enables humans to directly indicate their desired style. We do this by learning an abstract, low-dimensional, and continuous canonical space from human demonstration data. Each point in the canonical space corresponds to a different style (e.g., defensive or aggressive driving), and users can directly personalize the robot's behavior by simply clicking on a point. Given the human's selection, the robot then decodes this canonical style across each task in the dataset -- e.g., if the human selects a defensive style, the autonomous car personalizes its behavior to drive defensively when decelerating, passing other cars, or merging onto highways. We refer to our resulting approach as PECAN: Personalizing Robot Behaviors through a Learned Canonical Space. Our simulations and user studies suggest that humans prefer using PECAN to directly personalize robot behavior (particularly when those users become familiar with PECAN), and that users find the learned canonical space to be intuitive and consistent. See videos here: https://youtu.be/wRJpyr23PKI

Problem

• Robots need to personalize behaviors for individual human users
• Current methods infer preferences indirectly from selected behaviors
• PECAN enables direct style selection via a learned canonical space

Innovation

• Learns low-dimensional canonical space from demonstrations
• Enables direct style selection via interface clicks
• Decodes selected style across multiple tasks