Robust Localization for Autonomous Vehicles in Highway Scenes

arXiv cs.RO / 4/27/2026

📰 NewsDeveloper Stack & InfrastructureIndustry & Market MovesModels & Research

共有:

Key Points

Highway localization for autonomous vehicles is highlighted as less studied than urban localization, and directly porting state-of-the-art urban methods to highways can cause performance degradation.
The proposed solution combines a dual-likelihood LiDAR front end (decoupling 3D geometry and 2D road-texture cues), a Control-EKF that fuses steering/acceleration commands to reduce latency and improve closed-loop behavior, and an automated high-cadence offline mapping pipeline.
The work addresses major highway-specific challenges such as environmental changes, heavy occlusion, degraded GNSS signals, and strict downstream accuracy/latency requirements.
A new public dataset spanning urban roads and highways (163 km total) with standardized, product-oriented accuracy metrics and certified ground truth is released to enable fair benchmarking.
Compared with Apollo and Autoware, the system is comparable on urban roads but more robust on difficult highway scenarios, and it has been validated with over one million kilometers of road testing.

Abstract

Localization for autonomous vehicles on highways remains under-explored compared to urban roads, and state-of-the-art methods for urban scenes degrade when directly applied to highways. We identify key challenges including environment changes under information homogeneity, heavy occlusion, degraded GNSS signals, and stringent downstream requirements on accuracy and latency. We propose a robust localization system to address highway challenges, which uses a dual-likelihood LiDAR front end that decouples 3D geometric structures and 2D road-texture cues to handle environment changes; a Control-EKF further leverages steering and acceleration commands to reduce lag and improve closed-loop behavior. An automated offline mapping and ground-truth pipeline keep maps fresh at high cadence for optimal localization performance. To catalyze progress, we release a public dataset covering both urban roads and highways while focusing on representative challenging highway clips, totaling 163 km; benchmarking is standardized using product-oriented accuracy metrics and certified ground truth. Compared to Apollo and Autoware, our system performs similarly on urban roads but shows superior robustness on challenging highway scenarios. The system has been validated by more than one million kilometers of road testing.