Which LiDAR scanning pattern is better for roadside perception: Repetitive or Non-repetitive?

2025-10-28

AI Summary

This study investigates the suitability differences between repetitive-scanning (e.g., mechanical/solid-state) and non-repetitive-scanning (e.g., prism-based) LiDARs for roadside perception. To this end, we construct InfraLiDARs—a first-of-its-kind multimodal benchmark dataset tailored for infrastructure deployment—using CARLA, featuring synchronized multi-LiDAR point cloud acquisition. We conduct systematic evaluation using mainstream 3D object detectors including PointPillars and SECOND. Results show that non-repetitive LiDAR achieves detection performance comparable to a 128-line repetitive LiDAR at medium-to-short ranges (≤50 m), while significantly reducing hardware cost. Although its statistical sparsity limits long-range perception, this characteristic aligns well with typical intelligent transportation system (ITS) application requirements. This work establishes the first empirical benchmark and open-source dataset supporting evidence-based roadside LiDAR selection, revealing the practical viability of non-repetitive LiDARs under cost–performance trade-offs.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Large language models for search
• Security and Privacy: Large-scale security measurements

Abstract

LiDAR-based roadside perception is a cornerstone of advanced Intelligent Transportation Systems (ITS). While considerable research has addressed optimal LiDAR placement for infrastructure, the profound impact of differing LiDAR scanning patterns on perceptual performance remains comparatively under-investigated. The inherent nature of various scanning modes - such as traditional repetitive (mechanical/solid-state) versus emerging non-repetitive (e.g. prism-based) systems - leads to distinct point cloud distributions at varying distances, critically dictating the efficacy of object detection and overall environmental understanding. To systematically investigate these differences in infrastructure-based contexts, we introduce the "InfraLiDARs' Benchmark," a novel dataset meticulously collected in the CARLA simulation environment using concurrently operating infrastructure-based LiDARs exhibiting both scanning paradigms. Leveraging this benchmark, we conduct a comprehensive statistical analysis of the respective LiDAR scanning abilities and evaluate the impact of these distinct patterns on the performance of various leading 3D object detection algorithms. Our findings reveal that non-repetitive scanning LiDAR and the 128-line repetitive LiDAR were found to exhibit comparable detection performance across various scenarios. Despite non-repetitive LiDAR's limited perception range, it's a cost-effective option considering its low price. Ultimately, this study provides insights for setting up roadside perception system with optimal LiDAR scanning patterns and compatible algorithms for diverse roadside applications, and publicly releases the "InfraLiDARs' Benchmark" dataset to foster further research.

Problem

• Comparing repetitive and non-repetitive LiDAR scanning patterns for roadside perception
• Evaluating how scanning patterns impact 3D object detection algorithm performance
• Determining cost-effective LiDAR configurations for intelligent transportation systems

Innovation

• Introduced InfraLiDARs' Benchmark dataset for comparison
• Evaluated repetitive and non-repetitive LiDAR scanning patterns
• Analyzed cost-performance trade-offs for roadside perception systems