PBD-AG: Persistent Baseline-Delta Active Graphs with Uncertainty-Aware Inspection for Long-Horizon Service Robots

2026-08-11

AI Summary

This work addresses the challenge of constructing a persistent, autonomously updatable, and geometrically verifiable world model for long-term service robots operating in unknown environments—a task hindered by error accumulation, static scene representations, and insufficient 3D geometric evidence in existing approaches. The authors propose a baseline-increment decoupled active graph framework that separates stable static structures from revisable dynamic objects, building a structural baseline through autonomous exploration and performing uncertainty-aware verification via hierarchical object beliefs. A novel reliability-weighted state representation and a geometry-aware visibility gating mechanism are introduced to jointly inform graph-conditioned viewpoint planning, effectively mitigating erroneous deletions under occlusion and enhancing object identity continuity and event recall. Experiments demonstrate significant improvements over baselines in multi-environment simulations, with superior performance in static-object F1 scores, identity continuity, and event recall, and successful integration with onboard systems is validated on a physical robot.

Abstract

Long-horizon service robots require persistent world models that can be built autonomously in unseen environments and revised as task-relevant objects change. Existing methods rely on online mapping, which accumulates localization and observation errors, static scene representations that cannot capture persistent object changes, or holistic vision-language predictions that lack verifiable 3D geometric evidence. We present PBD-AG, a persistent baseline-delta active graph framework that decouples robot-verified stable fixtures from revisable dynamic object events. Under our framework, the robot autonomously bootstraps the structural baseline from onboard exploration and inspects discovered fixtures to ground hierarchical object beliefs. PBD-AG maintains reliability-weighted object states over geometry, semantics, identity, existence, and support relations, utilizing a geometric visibility gate to mitigate false deletions under occlusion. Inspection viewpoints are selected by a graph-conditioned policy that balances target coverage, travel cost, collision risk, and redundant observation. Simulation experiments in multiple environments and under controlled dynamic evaluation show higher aggregate coarse-fixture F1 than capability-matched controls, as well as stronger identity continuity and event recall. A qualitative physical-robot demonstration further illustrates integration with onboard sensing, providing a traceable world model for long-horizon robotic perception.The project page of PBD-AG is available at https://shuobao214.github.io/PBD-AG/

Problem

• long-horizon service robots
• persistent world models
• object change detection
• 3D geometric evidence
• autonomous environment modeling

Innovation

• persistent world modeling
• baseline-delta decomposition
• uncertainty-aware inspection
• active graph policy
• long-horizon robotic perception