How to Peel with a Knife: Aligning Fine-Grained Manipulation with Human Preference

TL;DR

This paper introduces a two-stage learning framework for fine-grained manipulation tasks like peeling with a knife, combining force-aware imitation learning and human preference-based fine-tuning to achieve high success rates and generalization.

Contribution

It presents a novel approach that integrates quantitative and qualitative feedback for policy refinement in contact-rich, subjective tasks, demonstrating effective generalization and high success rates.

Findings

Achieves over 90% success rate on various produce

Improves performance by up to 40% through preference-based finetuning

Strong zero-shot generalization to unseen and out-of-distribution produce

Abstract

Many essential manipulation tasks - such as food preparation, surgery, and craftsmanship - remain intractable for autonomous robots. These tasks are characterized not only by contact-rich, force-sensitive dynamics, but also by their "implicit" success criteria: unlike pick-and-place, task quality in these domains is continuous and subjective (e.g. how well a potato is peeled), making quantitative evaluation and reward engineering difficult. We present a learning framework for such tasks, using peeling with a knife as a representative example. Our approach follows a two-stage pipeline: first, we learn a robust initial policy via force-aware data collection and imitation learning, enabling generalization across object variations; second, we refine the policy through preference-based finetuning using a learned reward model that combines quantitative task metrics with qualitative human feedback, aligning policy behavior with human notions of task quality. Using only 50-200 peeling trajectories, our system achieves over 90% average success rates on challenging produce including cucumbers, apples, and potatoes, with performance improving by up to 40% through preference-based finetuning. Remarkably, policies trained on a single produce category exhibit strong zero-shot generalization to unseen in-category instances and to out-of-distribution produce from different categories while maintaining over 90% success rates.