Ground Penetrating Radar (GPR) localization has gained increasing attention in autonomous driving due to its resilience against surface appearance and weather conditions. However, existing LGPR methods based on single-scan matching often suffer from limited sensing range and frequent mismatches, particularly in the absence of complementary sensors. To address these challenges, we propose the first dedicated sequence-based matching framework for LGPR localization, which systematically exploits temporal continuity for robust place recognition. The proposed framework comprises four key components: A learning-based pretraining module to suppress weather-induced variations in GPR signatures; An information-theoretic correlation method for estimating lateral deviations; A velocity-constrained search strategy to enforce spatial consistency during coarse alignment; and A re-ranking mechanism to refine the final matching outcome. Unlike existing methods that focus solely on descriptor design, our framework is modular, extensible, and emphasizes the sequential nature of localization. It supports both handcrafted and learning-based GPR features, enabling fair and reproducible comparisons across different extraction pipelines. Experiments on both public and in-house datasets show that our method significantly improves localization accuracy, validating its robustness, adaptability, and the critical role of sequence modeling in LGPR-based localization.

As a promising technology for autonomous driving, localizing ground-penetrating radar (LGPR) is a vehicle localization method that relies on prior maps and couples deeply with subsurface features. However, the unique characteristics of GPR data often lead to a significant number of mismatched candidates during localization. Previous learning-based GPR place recognition methods have primarily relied on 2-D convolutional neural networks (CNNs), which struggle to effectively capture critical temporal information, limiting further performance improvements. To address this limitation, we propose a spatial-temporal U-shaped network (STU-Net) that leverages 3-D CNNs to simultaneously extract spatial and temporal features from GPR image sequences. In addition, residual dense blocks (RDBs) are integrated into the network to enable multiscale feature extraction. Extensive experiments conducted on publicly available datasets demonstrate that our STU-Net achieves state-of-the-art performance, outperforming existing methods with significant improvements.

Localizing ground-penetrating radar (LGPR) benefits from deep subsurface coupling, ensuring robustness against surface variations and adverse weather. While LGPR is widely recognized as the complement of existing vehicle localization methods, its reliance on prior maps introduces significant challenges. Channel misalignment during traversal positioning and time-dimension distortion caused by non-uniform platform motion degrade matching accuracy. Furthermore, rain and snow conditions induce subsurface water-content variations that distort ground-penetrating radar (GPR) echoes, further complicating the localization process. To address these issues, we propose a weather-resilient adaptive spatio-temporal mask alignment algorithm for LGPR. The method employs adaptive alignment and dynamic time warping (DTW) strategies to sequentially resolve channel and time-dimension misalignments in GPR sequences, followed by calibration of GPR query sequences. Moreover, a multi-level discrete wavelet transform (MDWT) module enhances low-frequency GPR features while adaptive alignment along the channel dimension refines the signals and significantly improves localization accuracy under rain or snow. Additionally, a local matching DTW algorithm is introduced to perform robust temporal image-sequence alignment. Extensive experiments were conducted on both public LGPR datasets: GROUNDED and self-collected data covering five challenging scenarios. The results demonstrate superior localization accuracy and robustness compared to existing methods.

Ground-penetrating radar (GPR) , as a sensor for mapping and localizing subsurface features, has gained significant attention in robotic localization for complex environments. However, varying weather conditions change the subsurface dielectric constant, which weakens the registration correlation between real-time images and maps, thereby compromising localization stability. We design an image registration framework for localizing ground penetrating radar (LGPR) in multi-weather. Specifically, we first propose an adaptive hyperbolic receptive field that aims to significantly enhance the robust features in the GPR images, while improving both discrimination capability in map and reliability for image registration. Then, an image alignment module is introduced to eliminate the time-delay blurring problem caused by the variation of dielectric constant under multi-weather conditions. The proposed method was evaluated on three distinct datasets (simulated, publicly available, and self-constructed), demonstrating significant improvements in registration and localization performance. The average correlation coefficient on simulated data achieved a enhancement from 0.3796 to 0.7406 compared with baseline, validating the effectiveness of feature enhancement. Furthermore, measured datasets exhibited 15% higher average registration and localization accuracy than baseline. These results demonstrate that the method provides a reliable guarantee for the stable operation of LGPR systems in complex environments. The datasets will be released at: https://github.com/Eazin-bz/dataset_AHFT.git.