NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation

• URL: http://arxiv.org/abs/2606.03159v1
• Date: Tue, 02 Jun 2026 05:11:05 GMT
• Title: NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation
• Abstract summary: We introduce OmniDreams, a foundation generative world model mid- and post-trained from the Cosmos diffusion model to autoregressively generate action-conditioned videos in real time.<n>By leveraging the rich visual priors of Cosmos and mid- and post-training on 21k hours of driving scenarios, OmniDreams synthesizes complex, unobserved phenomena.<n>Preliminary results indicate that a world-action model post-trained from OmniDreams achieves strong performance on the Physical AI Autonomous Vehicles NuRec dataset.
• Score: 84.88799392695893
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck. In closed-loop simulation, the driving policy model actively interacts with the environment, where its actions dynamically update the simulator state and directly influence the next set of generated sensor observations. While recent reconstruction-based neural simulators offer photorealism, they are fundamentally constrained by their initial captured data and struggle to generalize to highly dynamic or novel scenes. To overcome these limitations, we introduce OmniDreams, a foundation generative world model mid- and post-trained from the Cosmos diffusion model to autoregressively generate action-conditioned videos in real time. By leveraging the rich visual priors of Cosmos and mid- and post-training on 21k hours of driving scenarios, OmniDreams synthesizes complex, unobserved phenomena that are hard for traditional simulators to capture, such as extreme weather and unpredictable dynamic agent behaviors. Crucially, it autoregressively conditions its photorealistic sensor generation on past frames, the current simulator state, and immediate driving actions. Deployed in a closed-loop system with the Alpamayo 1 policy model and AlpaSim orchestrator, OmniDreams acts as a highly responsive, reactive environment, providing a scalable and comprehensive solution for training and evaluating next-generation autonomous driving policies. We additionally show preliminary results indicating that a world-action model (WAM) post-trained from OmniDreams achieves strong performance on the Physical AI Autonomous Vehicles NuRec dataset, surpassing the VLA-based Alpamayo 1.5 research policy model while using only 1/5 the total parameters. These results highlight the potential for a real-time world model like OmniDreams to also serve as a backbone for policy architectures.

Related papers

• Learning the Right Abstraction: Neural Reduced Dynamics for Complex Robot Control [6.192077990532515]
  We advocate the use of less accurate but more expeditious simulations, which might draw on data-driven, e.g., neural dynamics, models.<n>We develop a neural reduced dynamics (NRD) framework that separates the state the model propagates from what can be supplied as an input or recovered analytically.<n>Two case studies instantiate it across three control tasks: terrain-aware HMMWV trajectory tracking on rigid, bumpy and deformable terrain; and goal reaching for a stock tracked vehicle and its front-mounted articulated arm.
  arXiv  Detail & Related papers  (2026-08-19T18:41:27Z)
• DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving [49.11389494068169]
  We present DrivingGen, the first comprehensive benchmark for generative driving world models.<n>DrivingGen combines a diverse evaluation dataset curated from both driving datasets and internet-scale video sources.<n>General models look better but break physics, while driving-specific ones capture motion realistically but lag in visual quality.
  arXiv  Detail & Related papers  (2026-01-04T13:36:21Z)
• SceneDiffuser++: City-Scale Traffic Simulation via a Generative World Model [30.561378506172698]
  We propose SceneDiffuser++, the first end-to-end generative world model trained on a single loss function capable of point A-to-B simulation on a city scale.<n>We demonstrate the city-scale traffic simulation capability of SceneDiffuser++ and study its superior realism under long simulation conditions.
  arXiv  Detail & Related papers  (2025-06-27T07:35:04Z)
• DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation [54.02069690134526]
  We propose DrivingSphere, a realistic and closed-loop simulation framework. Its core idea is to build 4D world representation and generate real-life and controllable driving scenarios. By providing a dynamic and realistic simulation environment, DrivingSphere enables comprehensive testing and validation of autonomous driving algorithms.
  arXiv  Detail & Related papers  (2024-11-18T03:00:33Z)
• OmniRe: Omni Urban Scene Reconstruction [78.99262488964423]
  We introduce OmniRe, a comprehensive system for creating high-fidelity digital twins of dynamic real-world scenes from on-device logs.<n>Our approach builds scene graphs on 3DGS and constructs multiple Gaussian representations in canonical spaces that model various dynamic actors.
  arXiv  Detail & Related papers  (2024-08-29T17:56:33Z)
• DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving [30.024309081789053]
  DriveArena is a high-fidelity closed-loop simulation system designed for driving agents navigating in real scenarios. It features Traffic Manager, a traffic simulator capable of generating realistic traffic flow on any worldwide street map, and World Dreamer, a high-fidelity conditional generative model with infinite autoregression.
  arXiv  Detail & Related papers  (2024-08-01T09:32:01Z)
• GAIA-1: A Generative World Model for Autonomous Driving [9.578453700755318]
  We introduce GAIA-1 ('Generative AI for Autonomy'), a generative world model that generates realistic driving scenarios. Emerging properties from our model include learning high-level structures and scene dynamics, contextual awareness, generalization, and understanding of geometry.
  arXiv  Detail & Related papers  (2023-09-29T09:20:37Z)
• DriveDreamer: Towards Real-world-driven World Models for Autonomous Driving [76.24483706445298]
  We introduce DriveDreamer, a world model entirely derived from real-world driving scenarios. In the initial phase, DriveDreamer acquires a deep understanding of structured traffic constraints, while the subsequent stage equips it with the ability to anticipate future states. DriveDreamer enables the generation of realistic and reasonable driving policies, opening avenues for interaction and practical applications.
  arXiv  Detail & Related papers  (2023-09-18T13:58:42Z)
• VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles [131.2240621036954]
  We present VISTA, an open source, data-driven simulator that integrates multiple types of sensors for autonomous vehicles. Using high fidelity, real-world datasets, VISTA represents and simulates RGB cameras, 3D LiDAR, and event-based cameras. We demonstrate the ability to train and test perception-to-control policies across each of the sensor types and showcase the power of this approach via deployment on a full scale autonomous vehicle.
  arXiv  Detail & Related papers  (2021-11-23T18:58:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.