Impact of Localization Errors on Label Quality for Online HD Map Construction
Authors: Alexander Blumberg, Jonas Merkert, Richard Fehler, Fabian Immel, Frank Bieder, Jan-Hendrik Pauls, Christoph Stiller
Organizations: Institute of Measurement and Control Systems, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany · FZI Research Center for Information Technology, Karlsruhe, Germany
High-definition (HD) maps are crucial for autonomous vehicles, but their creation and maintenance is very costly. This motivates the idea of online HD map construction. To provide a continuous large-scale stream of training data, existing HD maps can be used as labels for onboard sensor data from consumer vehicle fleets. However, compared to current, well curated HD map perception datasets, this fleet data suffers from localization errors, resulting in distorted map labels. We introduce three kinds of localization errors, Ramp, Gaussian, and Perlin noise, to examine their influence on generated map labels. We train a variant of MapTRv2, a state-of-the-art online HD map construction model, on the Argoverse 2 dataset with various levels of localization errors and assess the degradation of model performance. Since localization errors affect distant labels more severely, but are also less significant to driving performance, we introduce a distance-based map construction metric. Our experiments reveal that localization noise affects the model performance significantly. We demonstrate that errors in heading angle exert a more substantial influence than position errors, as angle errors result in a greater distortion of labels as distance to the vehicle increases. Furthermore, we can demonstrate that the model benefits from non-distorted ground truth (GT) data and that the performance decreases more than linearly with the increase in noisy data. Our study additionally provides a qualitative evaluation of the extent to which localization errors influence the construction of HD maps.
Figures & tables
Fig. 1 : Ring evaluation input (colored) over the entire scene (grey). Displayed are the map elements road boundaries (red), centerlines (green), pedestrian crossings (blue), and dividers (pink). Rings plotted as dashed with the currently observed ring (red) and other rings (grey).
Fig. 2 : Flowchart with an overview from map to evaluation. The noise application is described in Section III-A , details about the ground truth (GT) and training can be found in Section IV-A and we introduce our distance-aware evaluation metric in Section III-B .
Fig. 3 : Qualitative example with ground truth (blue) and noisy localization (red) with Ramp noise (left), Gaussian noise (middle) and Perlin noise (right).
Experiment
Noise parameters
APdsh
APsol
APbou
APcen
APped
mAP
ϵL
σL
ϵR
σR
NR
HC
Baseline B
-
-
-
-
-
-
36.7
54.4
49.6
48.3
40.0
45.8
Ramp R1
2 m
-
1°
-
50 %
✓
34.8
52.1
46.0
46.1
36.5
43.1
Ramp R2
2 m
-
0°
-
100 %
✗
32.0
49.0
46.5
44.3
33.6
41.1
Ramp R3
2 m
-
0°
-
100 %
✓
30.6
46.4
43.7
42.7
33.1
39.3
Ramp R4
2 m
-
1°
-
100 %
✓
29.8
45.6
44
43.1
31.8
38.9
TABLE I : Comparison between prediction results for the noise types Ramp, Gaussian, and Perlin evaluated using Average Precision (AP). As noise parameters, we defined the maximum translation error ϵL , maximum angular error ϵR , for the Gaussian noise case standard deviation in translation σL and angular σR direction, noise ratio NR as the percentage of altered ground truth, and heading correction HC as correction to the now altered driving direction.
Fig. 4 : Qualitative prediction results for the Baseline B (trained without noisy GT labels), Ramp R1−R3 , Gaussian G1−G3 , and Perlin P1−P3 for the camera input (top left, Baseline results are projected into the image) and the ground truth map GT .
Fig. 5 : Gaussian G1 (GT distortion and prediction) and Baseline B1
Fig. 6 : Ramp R2 (GT distortion and prediction) and Baseline B1
Fig. 7 : Perlin P2 (GT distortion and prediction) and Baseline B1
Fig. 8 : Perlin P3 (GT distortion and prediction) and Baseline B1
Accurate, up-to-date representations of road structures are critical for the safe operation of autonomous vehicles. Existing systems rely either on costly, maintenance-heavy high-definition (HD) maps which compromise safety when outdated, or purely sensor-based online mapping which struggles with long-range reliability and occlusion. Systems incorporating map prior information into online mapping seek to overcome drawbacks of both approaches by combining them in some way. We propose 'Driveline To HD Map' (D2HDMap), an online mapping system that injects a lightweight, non-visible driveline prior to guide the estimation of visible road structures such as lane dividers, road boundaries and crosswalks. This prior incurs less effort to create and update compared to full HD map priors used in other approaches. We also show that training with such a prior can improve generalization at inference time when no prior is available. Ablation studies conducted on the nuScenes and Argoverse 2 dataset demonstrate that models trained using a driveline prior largely retain performance even when priors are not available. On a geographically disjoint split, D2HDMap achieves 44.8 mAP, surpassing recent state-of-the-art. Additionally, noise-aware training substantially increases robustness to realistic localization error.
Seojun Shon, Chikao Tsuchiya, Dhaval Bhanderi +3
Nissan North America · Nissan Advanced Technology Center - Silicon Valley, Nissan North America, Santa Clara, CA 95014, USA
High-definition (HD) maps are essential for autonomous driving systems. In constructing such maps, onboard multi-view camera images, standard-definition maps and satellite images provide crucial information. However, due to the modality and perspective differences among these data sources, existing methods often struggle to effectively align and fuse them, making online HD map construction still challenging. To address these issues, we propose Driver2Map, an online HD map construction model inspired by human drivers. Unlike existing HD map construction models that utilize only two modalities, our Driver2Map can simultaneously exploit three modalities. Specifically, we propose a "two-stage alignment" strategy to reduce spatial misalignment across different modalities. Additionally, we introduce "Pose-Guided BEV Fusion", a BEV (bird's-eye-view) generation module that leverages camera pose information to adaptively weight multi-view features, thereby effectively suppressing cross-view feature overlap during BEV generation. Also, we design a "Pretrained Prior for Map Refinement" module to refine the initial prediction by learning map structure priors, thus improving the HD map prediction under dynamic occlusions. Extensive experiments demonstrate that Driver2Map outperforms existing methods on both IoU and AP metrics.
Autonomous vehicles rely on map information to understand the world around them. However, the creation and maintenance of offline high-definition (HD) maps remains costly. A more scalable alternative lies in online HD map construction, which only requires map annotations at training time. To further reduce the need for annotating vast training labels, self-supervised training provides an alternative. This work focuses on improving the latent birds-eye-view (BEV) feature grid representation within a vectorized online HD map construction model by enforcing geospatial consistency between overlapping BEV feature grids as part of a contrastive loss function. To ensure geospatial overlap for contrastive pairs, we introduce an approach to analyze the overlap between traversals within a given dataset and generate subsidiary dataset splits following adjustable multi-traversal requirements. We train the same model supervised using a reduced set of single-traversal labeled data and self-supervised on a broader unlabeled set of data following our multi-traversal requirements, effectively implementing a semi-supervised approach. Our approach outperforms the supervised baseline across the board, both quantitatively in terms of the downstream tasks vectorized map perception performance and qualitatively in terms of segmentation in the principal component analysis (PCA) visualization of the BEV feature space.
Jonas Merkert, Alexander Blumberg, Jan-Hendrik Pauls +1
Institute of Measurement and Control Systems, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany