Search papers, labs, and topics across Lattice.
This paper introduces a low-burden, offline preference learning framework for personalizing assistive robots, using natural language feedback from users with paralysis. The framework leverages LLMs grounded in the Occupational Therapy Practice Framework (OTPF) to translate unstructured language into decision trees representing user needs, followed by automated LLM-based safety verification. A simulated meal preparation study with 10 adults with paralysis demonstrated significant workload reduction and clinically-validated safety and preference alignment compared to traditional methods.
LLMs can translate free-form natural language feedback from users with paralysis into safe and personalized robot control policies, drastically reducing the burden of preference learning.
Physically Assistive Robots (PARs) require personalized behaviors to ensure user safety and comfort. However, traditional preference learning methods, like exhaustive pairwise comparisons, cause severe physical and cognitive fatigue for users with profound motor impairments. To solve this, we propose a low-burden, offline framework that translates unstructured natural language feedback directly into deterministic robotic control policies. To safely bridge the gap between ambiguous human speech and robotic code, our pipeline uses Large Language Models (LLMs) grounded in the Occupational Therapy Practice Framework (OTPF). This clinical reasoning decodes subjective user reactions into explicit physical and psychological needs, which are then mapped into transparent decision trees. Before deployment, an automated"LLM-as-a-Judge"verifies the code's structural safety. We validated this system in a simulated meal preparation study with 10 adults with paralysis. Results show our natural language approach significantly reduces user workload compared to traditional baselines. Additionally, independent clinical experts confirmed the generated policies are safe and accurately reflect user preferences.
