Predictive Spatio-Temporal Scene Graphs for Semi-Static Scenes
Predictive Spatio-Temporal Scene Graphs for Semi-Static Scenes
This paper presents PredictiveGraphs, a framework that combines Perpetua* Bayesian filters with 3D scene graphs to enable robots to reason about spatio-semantic relationships over time in semi-static environments. The method is validated in both simulation and real-world dynamic navigation tasks where the environment undergoes structured changes at bi-hourly intervals over three weeks, demonstrating superior performance in predicting future states compared to baselines.
The authors introduce PredictiveGraphs, a method integrating Perpetua* Bayesian filters into 3D scene graphs to enable tempo-spatio-semantic reasoning for environments with semi-static changes; validation includes simulation and real-world dynamic navigation over three weeks with bi-hourly environmental updates.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
Source：arXiv Robotics — research abstracts · arxiv.org