Point Cloud Recombination: Systematic Real Data Augmentation Using Robotic Targets for LiDAR Perception Validation

TL;DR

This paper introduces Point Cloud Recombination, a novel method for augmenting real LiDAR data with physically accurate, controllable, and repeatable point clouds from laboratory-measured targets, enhancing validation and testing.

Contribution

It presents a systematic approach to augment real-world LiDAR scenes with physically measured point clouds, improving validation, controllability, and realism for perception systems.

Findings

Recombined scenes closely match real sensor outputs.

Enables scalable, repeatable testing with physically accurate data.

Improves system safety and failure analysis capabilities.

Abstract

The validation of LiDAR-based perception of intelligent mobile systems operating in open-world applications remains a challenge due to the variability of real environmental conditions. Virtual simulations allow the generation of arbitrary scenes under controlled conditions but lack physical sensor characteristics, such as intensity responses or material-dependent effects. In contrast, real-world data offers true sensor realism but provides less control over influencing factors, hindering sufficient validation. Existing approaches address this problem with augmentation of real-world point cloud data by transferring objects between scenes. However, these methods do not consider validation and remain limited in controllability because they rely on empirical data. We solve these limitations by proposing Point Cloud Recombination, which systematically augments captured point cloud scenes by integrating point clouds acquired from physical target objects measured in controlled laboratory environments. Thus enabling the creation of vast amounts and varieties of repeatable, physically accurate test scenes with respect to phenomena-aware occlusions with registered 3D meshes. Using the Ouster OS1-128 Rev7 sensor, we demonstrate the augmentation of real-world urban and rural scenes with humanoid targets featuring varied clothing and poses, for repeatable positioning. We show that the recombined scenes closely match real sensor outputs, enabling targeted testing, scalable failure analysis, and improved system safety. By providing controlled yet sensor-realistic data, our method enables trustworthy conclusions about the limitations of specific sensors in compound with their algorithms, e.g., object detection.