CARLA-GS: Decoupling Representation, Reasoning, and Physics Simulation for Autonomous Driving Corner-Case Synthesis

• URL: http://arxiv.org/abs/2607.07601v1
• Date: Wed, 08 Jul 2026 16:20:19 GMT
• Title: CARLA-GS: Decoupling Representation, Reasoning, and Physics Simulation for Autonomous Driving Corner-Case Synthesis
• Abstract summary: Corner-case generation is inherently a multi-case problem spanning visual representation, scene reasoning, and vehicle trajectory generation and control.<n>To unify these aspects within a single framework, we propose CARLA-GS, a modular corner-case synthesis pipeline.<n>This design enables high-level semantic reasoning, low-level physically executable motion, and corner-case generation within a unified pipeline.
• Score: 3.0621944624163695
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Safety evaluation for autonomous driving is dominated by rare, safety-critical interactions, motivating simulators that can deliberately synthesize corner cases with photorealistic observations. Corner-case generation is inherently a multi-source problem spanning visual representation, scene reasoning, and vehicle trajectory generation and control. Prior knowledge- and model-based approaches typically focus on scene or trajectory components in isolation, while diffusion-based methods attempt end-to-end generation but still struggle to ensure spatiotemporal consistency and physical realism. To unify these aspects within a single framework, we propose CARLA-GS, a modular corner-case synthesis pipeline that decouples visual representation, semantic reasoning, and physics-based execution while maintaining tight cross-module coupling. Starting from real driving data, we reconstruct an editable gaussian scene with additional geometry-consistent constraints. A multi-agent LLM then performs scene-level reasoning to identify risky interactions and generate intent-level waypoint trajectories, while the low-level motion control is delegated to CARLA, where a PID controller ensures kinematic and dynamic feasibility. The simulated vehicle states are finally re-projected into the gaussian scene for ego-centric rendering. This design enables high-level semantic reasoning, low-level physically executable motion, and photorealistic corner-case generation within a unified pipeline. Experiments on the Waymo Open Dataset show, both quantitatively and qualitatively, that our framework enables controllable corner-case generation and produces photorealistic, spatiotemporally consistent videos aligned with semantic intent and physically feasible motion.

Related papers

• IOI: Decoupling Kinematics and Physics for Interactive World Models [46.3330122411516]
  We propose IOI, a hybrid interactive world model integrating kinematics with learned physical dynamics.<n> IOI introduces explicit kinematic guidance, computing kinematics sequences for accurate motion trajectories.<n> experiments on the RoboTwin benchmark validate IOI across kinematic fidelity, out-of-distribution, and policy evaluation.
  arXiv  Detail & Related papers  (2026-06-22T13:09:34Z)
• MAPLE: Latent Multi-Agent Play for End-to-End Autonomous Driving [62.43744546817599]
  Vision-language--action (VLA) models are effective as end-to-end motion planners, but can be brittle when evaluated in closed-loop settings.<n>We propose MAPLE, a novel framework for reactive, multi-agent rollout of a dynamic driving scenario in the latent space of the VLA model.<n> MAPLE achieves state-of-the-art driving performance on Bench2Drive and demonstrates scalable, closed-loop multi-agent play for robust E2E autonomous driving systems.
  arXiv  Detail & Related papers  (2026-05-13T23:35:14Z)
• Real2Sim: A Physics-driven and Editable Gaussian Splatting Framework for Autonomous Driving Scenes [4.759823735082845]
  Real2Sim is an unified framework that combines 4D Gaussian Splatting (4DGS) with a differentiable Material Point Method (MPM) solver.<n>Real2Sim explicitly reconstructs dynamic driving scenes as temporally continuous Gaussian primitives, supports instance-level editing, and simulates realistic object-object and object-environment interactions.<n> Experiments on the Open dataset validate Real2Sim's capabilities in rendering, reconstruction, editing, and physics simulation, demonstrating its potential as a scalable tool for data generation in downstream tasks such as perception, tracking, trajectory prediction, and end-to-end policy learning.
  arXiv  Detail & Related papers  (2026-05-13T14:26:25Z)
• Part-Level 3D Gaussian Vehicle Generation with Joint and Hinge Axis Estimation [19.45064081855758]
  Simulation is essential for autonomous driving, yet current frameworks often model vehicles as rigid assets and fail to capture part-level articulation.<n>We propose a generative framework that synthesizes an animatable 3D Gaussian vehicle from a single image or sparse multi-view input.<n>Our method addresses two challenges: (i) large 3D asset generators are optimized for static quality but not articulation, leading to distortions at part boundaries when animated; and (ii) segmentation alone cannot provide the kinematic parameters required for motion.
  arXiv  Detail & Related papers  (2026-04-06T18:16:12Z)
• Optimization-Guided Diffusion for Interactive Scene Generation [52.23368750264419]
  We present OMEGA, an optimization-guided, training-free framework that enforces structural consistency and interaction awareness during diffusion-based sampling.<n>We show that OMEGA improves generation realism, consistency, and controllability, increasing the ratio of physically and behaviorally valid scenes.<n>Our approach can also generate $5times$ more near-collision frames with a time-to-collision under three seconds.
  arXiv  Detail & Related papers  (2025-12-08T15:56:18Z)
• AD-GS: Object-Aware B-Spline Gaussian Splatting for Self-Supervised Autonomous Driving [29.420887070252274]
  We introduce AD-GS, a novel self-supervised framework for high-quality free-viewpoint rendering of driving scenes from a single log.<n>At its core is a novel learnable motion model that integrates locality-aware B-spline curves with global-aware trigonometric functions.<n>Our model incorporates visibility reasoning and physically rigid regularization to enhance robustness.
  arXiv  Detail & Related papers  (2025-07-16T11:10:57Z)
• LLM-based Realistic Safety-Critical Driving Video Generation [4.537331974356809]
  We propose a framework that automatically synthesizes driving scenarios within the CARLA simulator.<n>The framework has flexibility in scenario scripting, efficient code-based control of traffic participants, and enforcement of realistic physical dynamics.<n>Our approach enables controllable scenario generation and facilitates the creation of rare but critical edge cases.
  arXiv  Detail & Related papers  (2025-07-02T00:45:19Z)
• GeoDrive: 3D Geometry-Informed Driving World Model with Precise Action Control [50.67481583744243]
  We introduce GeoDrive, which explicitly integrates robust 3D geometry conditions into driving world models.<n>We propose a dynamic editing module during training to enhance the renderings by editing the positions of the vehicles.<n>Our method significantly outperforms existing models in both action accuracy and 3D spatial awareness.
  arXiv  Detail & Related papers  (2025-05-28T14:46:51Z)
• RealEngine: Simulating Autonomous Driving in Realistic Context [60.55873455475112]
  RealEngine is a novel driving simulation framework that holistically integrates 3D scene reconstruction and novel view synthesis techniques.<n>By leveraging real-world multi-modal sensor data, RealEngine reconstructs background scenes and foreground traffic participants separately, allowing for highly diverse and realistic traffic scenarios.<n>RealEngine supports three essential driving simulation categories: non-reactive simulation, safety testing, and multi-agent interaction.
  arXiv  Detail & Related papers  (2025-05-22T17:01:00Z)
• Bench2Drive-R: Turning Real World Data into Reactive Closed-Loop Autonomous Driving Benchmark by Generative Model [63.336123527432136]
  We introduce Bench2Drive-R, a generative framework that enables reactive closed-loop evaluation.<n>Unlike existing video generative models for autonomous driving, the proposed designs are tailored for interactive simulation.<n>We compare the generation quality of Bench2Drive-R with existing generative models and achieve state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-12-11T06:35:18Z)
• Multi-Modal Fusion Transformer for End-to-End Autonomous Driving [59.60483620730437]
  We propose TransFuser, a novel Multi-Modal Fusion Transformer, to integrate image and LiDAR representations using attention. Our approach achieves state-of-the-art driving performance while reducing collisions by 76% compared to geometry-based fusion.
  arXiv  Detail & Related papers  (2021-04-19T11:48:13Z)

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.