Robotics · 2026-10-08
A Review Of Robotic World Models For Dynamic Environments Based On Factor And Scene Graphs
Marco Giberna, Miguel Fernandez-Cortizas, Jose Luis Sanchez Lopez, Holger Voos
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
The authors review graph-based robotic world models for dynamic environments, focusing on factor graphs (uncertainty-aware geometric estimation), scene graphs (semantic relational abstraction), and hybrid combinations. They structure the survey around (I) representations, (II) pipelines to construct/update them, and (III) how they are exploited for downstream tasks, highlighting dynamics modeling, uncertainty propagation, observability under minimal sensing, lifelong maintenance, and evaluation gaps.
Why it matters
Robotic autonomy needs internal world representations that remain accurate when environments change over time. Factor and scene graphs were developed largely for static settings; this review addresses how they can represent dynamic motion and change, how hybrid designs couple geometry and semantics, and what open challenges remain—especially uncertainty propagation from perception layers through the representation.
Method
- Organizes the review around representations (factor graphs, scene graphs, hybrid models), construction/update pipelines, and exploitation for downstream tasks.
- Defines robotics dynamics broadly as “any directly or indirectly perceivable motions,” distinguishing direct vs indirect perceivability.
- Highlights hybrid architectural coupling patterns: layered factor-backbone stacks, single unified optimizable graphs, and loosely coupled parallel architectures.
Limitation
The authors identify a “lack of datasets and evaluation protocols that ground world-model quality in downstream task performance under dynamics.”
Abstract (from arXiv)
Models based on graphs have emerged in robotics as a powerful foundation for internal world representations, where factor and scene graphs are among the most prominent model types found in the related literature and in successful robotic solutions. Initially, many of these models were assuming static environments as a simplification. Herein, factor graphs mainly provide uncertainty-aware geometric estimations while scene graphs enable a structured semantic abstraction. However, real-world robotic environments are often dynamic, posing severe challenges for purely static world representations. Therefore, this review presents a comprehensive view on how dynamic aspects of real-world environments can be addressed in such graph-based world models. We organize our assessments around three main aspects: (I) suitable representations, (II) pipelines to construct and update the representations, and (III) their exploitation for downstream tasks. We review approaches that are either based on factor or scene graphs, but put special emphasis on novel approaches that combine both types to form hybrid models. We mainly analyze how different types of dynamics can be modeled herein, and categorize common architectural patterns. Finally, emerging trends and open challenges are identified, including uncertainty propagation from learned perception through the representation layers, the observability of dynamic-entity motion and scale under minimal sensing, scalable lifelong maintenance, and the lack of datasets and evaluation protocols that ground world-model quality in downstream task performance under dynamics.
Related papers
- UniFlowWAM: A Unified World-Action Model with Visual Semantics and 4D Geometry
- T2-z: Test-Time Optimization of Environment Latents for Robotic World Models
- GreenWAM: A World Foundation Model for Robotics
- Robotic Video World Models: A Survey of Applications, Research Challenges, Future Directions
- RoboJEPA: Scaling Robotic Latent World Models
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-07-27 · Indexed by Pub-AI: 2026-10-08 UTC
A Review Of Robotic World Models For Dynamic Environments Based On Factor And Scene Graphs
Preprint posted: 2026-07-27 · Original record ↗
- arxiv
- 2610.08800
Explicit source versions (1)
- v1 · 2026-07-27
- Preprint available
Source date: 2026-07-27 · First detected: 2026-10-25 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
Models based on graphs have emerged in robotics as a powerful foundation for internal world representations, where factor and scene graphs are among the most prominent model types found in the related literature and in successful robotic solutions. Initially, many of these models were assuming static environments as a simplification. Herein, factor graphs mainly provide uncertainty-aware geometric estimations while scene graphs enable a structured semantic abstraction. However, real-world robotic environments are often dynamic, posing severe challenges for purely static world representations. Therefore, this review presents a comprehensive view on how dynamic aspects of real-world environments can be addressed in such graph-based world models. We organize our assessments around three main aspects: (I) suitable representations, (II) pipelines to construct and update the representations, and (III) their exploitation for downstream tasks. We review approaches that are either based on factor or scene graphs, but put special emphasis on novel approaches that combine both types to form hybrid models. We mainly analyze how different types of dynamics can be modeled herein, and categorize common architectural patterns. Finally, emerging trends and open challenges are identified, including uncertainty propagation from learned perception through the representation layers, the observability of dynamic-entity motion and scale under minimal sensing, scalable lifelong maintenance, and the lack of datasets and evaluation protocols that ground world-model quality in downstream task performance under dynamics.