LiDARsim: Realistic LiDAR Simulation by Leveraging the Real World
This addresses the need for realistic sensor simulation in autonomous driving, particularly for rare and safety-critical events, though it is incremental by building on existing simulation approaches with real data integration.
The paper tackles the problem of generating realistic LiDAR point cloud simulations for self-driving vehicles by leveraging real-world data, resulting in a simulator that combines physics-based and learning-based methods to produce accurate simulations for testing perception algorithms and safety-critical scenarios.
We tackle the problem of producing realistic simulations of LiDAR point clouds, the sensor of preference for most self-driving vehicles. We argue that, by leveraging real data, we can simulate the complex world more realistically compared to employing virtual worlds built from CAD/procedural models. Towards this goal, we first build a large catalog of 3D static maps and 3D dynamic objects by driving around several cities with our self-driving fleet. We can then generate scenarios by selecting a scene from our catalog and "virtually" placing the self-driving vehicle (SDV) and a set of dynamic objects from the catalog in plausible locations in the scene. To produce realistic simulations, we develop a novel simulator that captures both the power of physics-based and learning-based simulation. We first utilize ray casting over the 3D scene and then use a deep neural network to produce deviations from the physics-based simulation, producing realistic LiDAR point clouds. We showcase LiDARsim's usefulness for perception algorithms-testing on long-tail events and end-to-end closed-loop evaluation on safety-critical scenarios.