SCOPE improves robot planning by fixing scene graph errors
SCOPE: Scene-graph Construction Optimized by Plan Error-correction for Robot Long-term Planning Datasets
Robotics
Summary
Robots need to plan many steps ahead to complete complicated tasks, but current methods sometimes struggle because the data describing the world isn’t diverse or accurate enough. The authors created SCOPE, a system that automatically builds more varied and physically possible scenes for training robot planners. It checks for mistakes by simulating plans and then fixes problems in the descriptions if errors occur, ensuring the data makes sense in the real world. This leads to better robot planning performance compared to earlier methods.
What this means in practice
- •For robotics engineers: Create more reliable training datasets for robot planners that reduce planning errors by correcting scene graph mistakes automatically.
- •For automated warehouse operators: Improve long-term robot task success by training on diverse and physically consistent data generated through error-correction feedback loops.
Authors
Chaoyue Li, Zeyu Gao, Yanfeng Lu, Fan Luo
Abstract
Task and Motion Planning (TAMP) stands as a fundamental challenge in robotics, essential for enabling autonomous agents to execute complex, long-horizon tasks in diverse environments. While Large Language Models (LLMs) introduce a new paradigm for TAMP by leveraging extensive common sense, their inherent reasoning limitations hinder direct application, and fine-tuning strategies to enhance these capabilities are currently bottlenecked by the scarcity of high-quality datasets. Existing automated generation methods often suffer from homogeneity, limiting model generalization, and unsolvability, where hallucinated scene graphs create physical mismatches with task goals. To address these issues, we propose Scene-graph Construction Optimized by Plan Error-correction (SCOPE), an automated framework designed to generate diverse and executable TAMP datasets. SCOPE incorporates multi-level diversity metrics to maximize the information entropy of scenarios. Crucially, it introduces a Plan Feedback Error-correction pipeline that utilizes planner execution failures as feedback signals to retroactively rectify logical inconsistencies in scene graphs, ensuring physical feasibility. Experimental results demonstrate that SCOPE exhibits greater robustness to erroneous scene graphs and generates higher-quality data. When utilized for LLM fine-tuning, SCOPE-generated data yield a downstream TAMP model that achieves an absolute 12.33% gain in success rate over its counterpart fine-tuned with data generated by DELTA, a state-of-the-art PDDL-based planner.