arxiv.org·2026/10/19
自动驾驶3D建模须在有限算力下持续估计道路几何、障碍占用、交通参与者行为与观测不确定性；现有重建可合成逼真新视角，但视觉保真不足以刻画驾驶所需的几何可靠性，生成式世界模型能扩展场景却可能引入无测量支撑的内容。DriveWorld3D联合时序视频学习、多传感器测量约束与离线生成仿真：在线组件统一维护道路表面、体素占用、动态实体和道路关系的世界状态，并分离物理占用、可观测性与证据来源，动作条件未来预测和风险敏感规划将状态连到驾驶决策；离线组件用NeRF或3D Gaussian Splatting重建记录场景，再用生成模型扩展长尾条件，但禁止把生成内容作为几何真值。评估指定数据划分、传感器消融、故障注入、配对闭环测试和真实到仿真一致性评估。中心假设是显式证据管理可在匹配计算预算内减少危险假空闲空间与遮挡风险，同时保持可接受驾驶效率；所有性能声明尚待实验验证。
为什么重要 如果其「显式记录并约束证据来源」的假设成立，就能减少危险的自由空间误判；但摘要中没有一项量化结果，因此落地价值待验证适合 做自动驾驶占用感知、世界模型与闭环仿真的算法工程师
自动驾驶的三维建模必须在有限计算资源下持续估计道路几何、障碍物占据、交通参与者行为以及观测不确定性。现有重建方法可以合成逼真的新视图，但视觉保真度不足以刻画驾驶所需的几何可靠性。生成式世界模型可以扩展场景并预测未来观测，但可能引入未被测量支持的内容。我们提出 DriveWorld3D，一个结合时序视频学习、来自多传感器的测量约束和离线生成式仿真的框架。
其在线组件维护统一的世界状态，包括道路表面、体积占据、动态实体和道路关系，同时分别表示物理占据、可观测性和证据来源。动作条件化未来预测和风险敏感规划将这一表示与驾驶决策连接起来。其离线组件使用神经辐射场或 3D Gaussian Splatting 重建记录的场景，然后使用生成模型扩展长尾条件，同时限制将生成内容用作几何真值。
我们规定了数据划分、传感器消融、故障注入、成对闭环测试以及真实到仿真一致性评估。核心假设是，在匹配的计算预算内，显式证据管理可以减少危险的虚假自由空间预测和遮挡风险，同时保持可接受的驾驶效率。所有性能声明仍需经过实验验证。
关键词：autonomous driving; 3D occupancy; sensor fusion; uncertainty; generative world models; NeRF; 3D Gaussian Splatting; closed loop simulation
Three dimensional modeling for autonomous driving must continuously estimate road geometry, obstacle occupancy, traffic participant behavior, and observational uncertainty under limited computational resources.
Existing reconstruction methods can synthesize realistic novel views, but visual fidelity does not sufficiently characterize the geometric reliability required for driving.
Generative world models can expand scenarios and predict future observations, yet may introduce content unsupported by measurements.
We propose DriveWorld3D, a framework combining temporal video learning, measurement constraints from multiple sensors, and offline generative simulation.
Its online component maintains a unified world state comprising road surfaces, volumetric occupancy, dynamic entities, and road relations, while separately representing physical occupancy, observability, and evidence provenance.
Action conditioned future prediction and risk sensitive planning connect this representation to driving decisions.
Its offline component reconstructs recorded scenes using neural radiance fields or 3D Gaussian Splatting, then expands long tail conditions using generative models while restricting the use of generated content as geometric ground truth.
We specify data partitioning, sensor ablations, fault injection, paired closed loop testing, and real to simulation consistency assessment.
The central hypothesis is that explicit evidence management can reduce dangerous false free space predictions and occlusion risk within a matched computational budget while retaining acceptable driving efficiency.
All performance claims remain subject to experimental validation.
Keywords: autonomous driving; 3D occupancy; sensor fusion; uncertainty; generative world models; NeRF; 3D Gaussian Splatting; closed loop simulation
EVIDENCE / METHODS
要点
自动驾驶 3D 建模须在有限算力下持续估计道路几何、障碍物占据、交通参与者行为和观测不确定性
原版排版与图表 · 手机无法内嵌时可在浏览器打开。PDF 插件内选文不能传给网页助手。