OHP-RL: Online Human Preference as Guidance in Reinforcement Learning for Robot Manipulation

TL;DR

OHP-RL is a novel reinforcement learning framework that uses human preferences as guidance, enabling safer, more efficient robot manipulation with less human effort and improved stability.

Contribution

The paper introduces a state-dependent preference gate in RL that effectively incorporates intermittent human feedback for robot manipulation tasks.

Findings

Achieves higher success rates across tasks

Converges faster than prior methods

Requires less human intervention

Abstract

While reinforcement learning (RL) enables robots to acquire skills autonomously, its real-world deployment is severely limited by inefficient and unsafe exploration. Human-in-the-loop interventions offer a practical solution, yet existing methods typically exploit these interventions as auxiliary training signals, without fully capturing the richer information they provide about when and how autonomy should be guided. Human interventions often encode relative preferences over behavior under safety and task constraints, rather than prescribing exact actions to imitate. Motivated by this perspective, we propose Online Human Preference as Guidance in Reinforcement Learning (OHP-RL), a framework that leverages human interventions as preference information to guide policy learning. OHP-RL introduces a state-dependent preference gate that adaptively regulates when and to what extent human interventions should shape policy learning. This design enables the agent to benefit from intermittent and imperfect human feedback while preserving autonomous exploration and stable policy optimization. We evaluate OHP-RL on three challenging real-world contact-rich manipulation tasks on a Franka robot. Across all tasks, OHP-RL consistently achieves strong success rates, faster convergence, and substantially lower human intervention effort than prior approaches. Moreover, the learned policies exhibit more stable and human-aligned behavior throughout training.