FlowCalib: LiDAR-to-Vehicle Miscalibration Detection using Scene Flows

2026-01-30

AI Summary

This study addresses the safety-critical calibration errors caused by rotational misalignment between LiDAR sensors and vehicle bodies—a problem often overlooked by existing methods that primarily focus on inter-sensor calibration while neglecting mounting deviations of LiDAR relative to the vehicle frame. To tackle this, we propose the first scene-flow-based framework for LiDAR-to-vehicle misalignment detection, which exploits systematic motion cues induced by rotational misalignment in static objects across consecutive point clouds, enabling misalignment diagnosis without additional sensors. Our approach integrates neural scene flow priors, handcrafted geometric descriptors, and global flow features within a dual-branch network to simultaneously predict binary misalignment status both globally and along individual rotational axes. Experiments on the nuScenes dataset demonstrate the robustness of our method and establish a new benchmark for sensor-to-vehicle misalignment detection.

Technology Category

• Computer Vision: Scene Analysis & Understanding
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Security and Privacy: Large-scale security measurements

Abstract

Accurate sensor-to-vehicle calibration is essential for safe autonomous driving. Angular misalignments of LiDAR sensors can lead to safety-critical issues during autonomous operation. However, current methods primarily focus on correcting sensor-to-sensor errors without considering the miscalibration of individual sensors that cause these errors in the first place. We introduce FlowCalib, the first framework that detects LiDAR-to-vehicle miscalibration using motion cues from the scene flow of static objects. Our approach leverages the systematic bias induced by rotational misalignment in the flow field generated from sequential 3D point clouds, eliminating the need for additional sensors. The architecture integrates a neural scene flow prior for flow estimation and incorporates a dual-branch detection network that fuses learned global flow features with handcrafted geometric descriptors. These combined representations allow the system to perform two complementary binary classification tasks: a global binary decision indicating whether misalignment is present and separate, axis-specific binary decisions indicating whether each rotational axis is misaligned. Experiments on the nuScenes dataset demonstrate FlowCalib's ability to robustly detect miscalibration, establishing a benchmark for sensor-to-vehicle miscalibration detection.

Problem

• LiDAR-to-vehicle miscalibration
• rotational misalignment
• sensor calibration
• autonomous driving
• scene flow

Innovation

• LiDAR-to-vehicle calibration
• scene flow
• miscalibration detection
• neural prior
• dual-branch network