Think Small, Act Big: Primitive Prompt Learning for Lifelong Robot Manipulation

2025-04-01

AI Summary

To address catastrophic forgetting and poor cross-skill knowledge transfer in lifelong robotic continual learning, this paper proposes a two-stage learning framework grounded in reusable and extensible primitive prompts. The core contribution is the first-ever motion-aware primitive prompting mechanism, which explicitly models semantic and kinematic commonalities across skills, enabling effective knowledge sharing. Our method supports incremental learning via frozen pretrained prompts alongside dynamically instantiated new prompts. It integrates multi-skill joint pretraining, parameter-efficient prompt optimization, and motion-prior embedding. Evaluated under simulation-to-real transfer settings, the framework demonstrates significant improvements on a newly constructed large-scale robotic skill dataset and real-world robot platforms: 37% faster convergence for novel skills and a 22% gain in cross-task generalization accuracy—substantially outperforming current state-of-the-art approaches.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Machine Learning: Life-Long and Continual Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Building a lifelong robot that can effectively leverage prior knowledge for continuous skill acquisition remains significantly challenging. Despite the success of experience replay and parameter-efficient methods in alleviating catastrophic forgetting problem, naively applying these methods causes a failure to leverage the shared primitives between skills. To tackle these issues, we propose Primitive Prompt Learning (PPL), to achieve lifelong robot manipulation via reusable and extensible primitives. Within our two stage learning scheme, we first learn a set of primitive prompts to represent shared primitives through multi-skills pre-training stage, where motion-aware prompts are learned to capture semantic and motion shared primitives across different skills. Secondly, when acquiring new skills in lifelong span, new prompts are appended and optimized with frozen pretrained prompts, boosting the learning via knowledge transfer from old skills to new ones. For evaluation, we construct a large-scale skill dataset and conduct extensive experiments in both simulation and real-world tasks, demonstrating PPL's superior performance over state-of-the-art methods.

Problem

• Addressing catastrophic forgetting in lifelong robot learning
• Leveraging shared primitives across multiple robot skills
• Enhancing new skill acquisition via knowledge transfer

Innovation

• Primitive Prompt Learning for lifelong manipulation
• Two-stage learning with multi-skills pre-training
• Motion-aware prompts capture shared primitives