A Practical Causal Framework for Autonomous Mobile Robots: From Simulation to Real-World Human Environments
A Practical Causal Framework for Autonomous Mobile Robots: From Simulation to Real-World Human Environments
This paper introduces a practical causal framework for autonomous mobile robots that addresses the challenge of transferring simulation-learned causal relations to real-world human environments. The authors demonstrate that by retaining invariant high-level causal structures and updating only model parameters with limited new observations, the system achieves efficient sim-to-real transfer. Empirical results show the framework runs efficiently onboard a mobile robot and significantly outperforms non-causal baselines in navigating complex, human-shared scenarios safely.
The authors present a causal inference framework that transfers high-level causal structures from simulation to real-world human environments by updating only model parameters with limited observations, demonstrating successful navigation on a mobile robot.
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
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
work runs efficiently onboard of a mobile robot and significantly outperforms a non-causal baselines navigating successfully and safely in complex human-shared scenarios.
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org