Can We Trust Our Maps? An Evaluation of Road Changes and a Dataset for Map Validation

Pauls, Jan-Hendrik; Strauß, Tobias; Hasberg, Carsten; Lauer, Martin; Stiller, Christoph

doi:10.1109/itsc.2018.8569249

Cited by 16 publications

(9 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Karlsruhe Aerial Images (KAI): Lane marking detection We use aerial images of a highway section near Karlsruhe, Germany [31], for which the actual lane markings have been annotated manually as polylines. This dataset provides more than 1400 bird's eye view (BEV) images for training and the task is to identify the explicit white lane marking instead of the implicit lane boundaries of TuSimple.…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

YOLinO: Generic Single Shot Polyline Detection in Real Time

Meyer

Skudlik

Pauls

et al. 2021

2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)

Self Cite

View full text Add to dashboard Cite

The detection of polylines in images is usually either bound to branchless polylines or formulated in a recurrent way, prohibiting their use in real-time systems.We propose an approach that transfers the idea of single shot object detection. Reformulating the problem of polyline detection as bottom-up composition of small line segments allows to detect bounded, dashed and continuous polylines with a single head. This has several major advantages over previous methods. Not only is the method at 187 fps more than suited for real-time applications with virtually any restriction on the shapes of the detected polylines. By predicting multiple line segments for each spatial cell, even branching or crossing polylines can be detected.We evaluate our approach on three different applications for road marking, lane border and center line detection. Hereby, we demonstrate the ability to generalize to different domains as well as both implicit and explicit polyline detection tasks.

show abstract

Section: Datasetsmentioning

confidence: 99%

“…In fact, our parametrized system is able to even estimate polylines that head in opposite directions while occupying the exact same pixel space. Facilitating polylines is in general more intuitive as most ground truth data is provided as polylines in modern datasets [7,31].…”

Section: Introductionmentioning

confidence: 99%

YOLinO: Generic Single Shot Polyline Detection in Real Time

Meyer

Skudlik

Pauls

et al. 2021

2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Even though HD maps are supposed to contain data with centimeter level accuracy, that can only be guaranteed the day of the mapping. Indeed, lane markings can be redrawn and road work can temporarily affect the lane boundaries [7], [6]. This may lead to a slight shift of the markings location.…”

Section: Map Reliability Estimationmentioning

confidence: 99%

“…Another issue is that the map has to be accurate. Even though roads do not change much from one day to the next, they can change when markings are repainted or when road work is done [7], [6]. For that reason, autonomous systems require ways to detect changes in the environment that lead part of the map to become unreliable.…”

Section: Introductionmentioning

confidence: 99%

Estimating the reliability of georeferenced lane markings for map-aided localization

Welte¹,

Xu²,

Zinoune

2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Maps can greatly improve vehicle localization using perception sensors that detect features georeferenced in the map. This relies on two assumptions. Firstly, the detected features and the elements of the map have to be correctly associated. Secondly, the features of the map have to be accurately referenced. In this paper, solutions regarding these issues are presented. The case study of localization using a camera detecting road markings is considered. A Kalman smoothing process is used to obtain the best possible estimate of the trajectory that enables to evaluate the reliability of markings stored in the map. A likelihood maximization technique is used to best associate the observed markings to those referenced in the map. By using these two methods, map errors are detected after a first passage in an area and can be mitigated in later passes. Experimental results are reported to evaluate the performance of this approach. It is shown that mapping errors can be correctly handled.

show abstract

“…Building such maps can be expensive but we believe that, in the near future, HD maps of the road network can be deployed on a large scale. Accurate mapping being itself a challenging task, the resulting map often contains errors or is incomplete [6] due to constant evolution of road networks. Such faulty data must be carefully taken into account.…”

Section: Introductionmentioning

confidence: 99%

Improved Data Association Using Buffered Pose Adjustment for Map-Aided Localization

Welte

Zinoune

2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Maps provide an important source of information for autonomous vehicles. They can be used along with cameras and lidars to localize the vehicle. This requires the ability to correctly associate observations to features referenced in the map. The problem is all the more difficult than all observations are not necessarily referenced, and all map features might not be detectable with the embedded sensors. This paper presents an adjustment technique that enables to increase the number of associations that can be made while limiting the chance of obtaining wrong associations. This is achieved by matching observations in batches in a buffer and matching them regularly. Periodically, the observation buffer is used to adjust the trajectory used to match observations. This is done without making any assumption on the association between observations and map features through a likelihood maximization process. The adjusted trajectory then provides the best associations that are used for real-time localization. The method was tested with data recorded on public roads using an experimental vehicle. The results show that thanks to the trajectory adjustment step and the use of an observation buffer, the number of associations that can be made is increased. This also results in greater localization accuracy and consistency with an average error of 0.7 meters at 50 Hz using road markings and traffic signs. This work was carried out within the shared laboratory SIVALab between Renault and Heudiasyc (UTC/CNRS) and co-financed by the Hauts-de-France region (ERDF grant). The equipment is from the Equipex ROBOTEX

show abstract

Can We Trust Our Maps? An Evaluation of Road Changes and a Dataset for Map Validation

Cited by 16 publications

References 19 publications

YOLinO: Generic Single Shot Polyline Detection in Real Time

YOLinO: Generic Single Shot Polyline Detection in Real Time

Estimating the reliability of georeferenced lane markings for map-aided localization

Improved Data Association Using Buffered Pose Adjustment for Map-Aided Localization

Contact Info

Product

Resources

About