Lane-Level Map Matching Based on HMM

Hansson, Anders; Korsberg, Ellen; Maghsood, Roza; Nordén, Eliza; Selpi,

doi:10.1109/tiv.2020.3035329

Cited by 23 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…But, a road-level view is required for microscopic traffic behavior analysis on intersections. Hansson et al use an HMM for lane-level map matching based on GPS measurements, visual features of the surrounding and change indication information by on-board systems of the vehicle [18]. However, they show that the performance decreases with the absence of the lane-changing indicator signal, which is not available in this work, and that the transition probability matrix needs to be tailored to the digital map.…”

Section: Related Workmentioning

confidence: 88%

“…Due to that, and since this work's overall aim is to extract driving scenarios in the urban domain, a mapmatching-based route estimation approach for microscopic traffic analysis on the road instead of lane level is proposed utilizing a digital map in OpenDRIVE format. For route estimation, a road is represented with the reference line instead of the lane borders as in [18], since each road on an intersection typically represents a driving direction in OpenDRIVE. Thus, the route can be estimated by mapping trajectories on reference lines.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Scenario Mining in the Urban Domain: Exploiting the Topology of a Road Network for Maneuver Annotation and Scenario Extraction

Klitzke¹,

Schicktanz²,

Gimm³

et al. 2022

Preprint

View full text Add to dashboard Cite

Before Automated Vehicles (AVs) can enter the market, academic and industrial partners must tackle system homologation to ensure the functional safety of AVs. For that task, a scenario-driven approach has gained acceptance in recent years since it allows facing the overall system’s complexity. Real-world drivings are a valuable source of scenarios. However, their extraction and analysis are still the subjects of research. This work addresses this issue and proposes a novel methodology to extract routing scenarios on urban intersections. Therefore, trajectories are annotated with routing maneuvers using a semantically enriched road network and map matching for route backtracking. The accuracy of the approach and consistency of the methodology is evaluated with Brunswick’s inner-city ring and by extracting u-turn scenarios on the Application Platform for Intelligent Mobility (AIM) Research Intersection of the German Aerospace Center (DLR).

show abstract

Section: Related Workmentioning

confidence: 88%

Section: Related Workmentioning

confidence: 99%

Scenario Mining in the Urban Domain: Exploiting the Topology of a Road Network for Maneuver Annotation and Scenario Extraction

Klitzke¹,

Schicktanz²,

Gimm³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Hansson et al propôs um modelo oculto de Markov para combinar uma trajetória de medições GNSS ruidosas com as pistas nas quais o veículo registra suas posições [Hansson et al 2020]. O modelo depende de um sinal de mudança de faixa, bem como de informações visuais sobre marcações de faixa próximas.…”

Section: Trabalhos Relacionadosunclassified

“…Com base na análise do estado da arte, identificou-se que a maioria dos trabalhos de map matching não considera a possibilidade de usar localização dos CAVs vizinhos para aumentar a precisão da posição [Li et al 2018] [Kang et al 2020] [Hansson et al 2020. Além disso, os trabalhos de localização estão restritos ao comportamento de fusão de dados em um cenário de comunicação veicular [Lobo et al 2019a [Balico et al 2018].…”

Section: Trabalhos Relacionadosunclassified

Mecanismo Eficiente de Localização Cooperativa para Veículos Autônomos Conectados

Lobato

Souza

Cerqueira³

et al. 2021

Anais Do XXXIX Simpósio Brasileiro De Redes De Computadores E Sistemas Distribuídos (SBRC 2021)

View full text Add to dashboard Cite

A localização robusta e precisa é fundamental para garantir o funcionamento de aplicações de Veículos Autônomos Conectados (CAVs). No entanto, os mecanismos de navegação veicular (GNSS) são afetados por diversos problemas nas áreas urbanas, principalmente devido à refração ou reﬂexão dos sinais de satélite em prédios. Assim, para usar a localização do GNSS e satisfazer os requisitos das aplicações de CAVs, é necessário considerar técnicas de fusão de dados para obter um posicionamento mais eﬁciente por meio da troca de dados entre os CAVs. Neste cenário, este artigo apresenta um mecanismo de fusão de dados para localização veicular cooperativa, denominado de DUELAR. O mecanismo considera o Filtro de Kalman Sem Cheiro em conjunto com um algoritmo de map matching a nível de faixa para corrigir os erros de localização. Os resultados da simulação mostraram que o DUELAR reduz o erro de localização em pelo menos 70% comparado com outras abordagens da literatura.

show abstract

“…Szottka [19] used a particle filter to fuse GPS/INS integration information, lane marking types by a camera, and a digital map. Hansson et al [20] proposed an HMMbased method that infers the driving lane by fusing GPS signals, IMU data, lane marking types by a camera, and a digital map. Using lane marking types is very effective on roads with two or three lanes with different marking types, but it is difficult to determine an accurate driving lane on roads with many lanes with the same marking type as highways.…”

Section: Introductionmentioning

confidence: 99%

Map Matching-Based Driving Lane Recognition for Low-Cost Precise Vehicle Positioning on Highways

2021

View full text Add to dashboard Cite

This paper proposes a map matching-based driving lane recognition system for lowcost precise vehicle positioning on highways. The proposed method finds the position where the road boundaries detected by a LIDAR sensor are best matched with the road boundaries of the precise digital map and then recognizes the driving lane based on the position. To improve the limitations of the existing Chamfer matching method, this paper proposes the following two methods. The first is a method of generating one candidate position for each lane by using the lane information of the map and the left and right lane offsets obtained from the front camera. Second is a simple matching method that converts the LIDAR and map data into arrays and compares only the lateral distances to measure the matching score. In addition, this paper proposes a method for maintaining sufficient road boundary information in every frame by applying a temporal accumulation to the LIDAR data. In the experiment, the proposed method was quantitatively evaluated using a database acquired from sensors mounted on a real vehicle on highways. The experimental results show that the driving lane recognition rate is 99.56%, the lateral position error is 0.49m, and the processing time is 6.4ms. These results prove that the proposed system provides sufficient accuracy and reliability even in the presence of GPS position errors in various highway environments.

show abstract

Lane-Level Map Matching Based on HMM

Cited by 23 publications

References 15 publications

Scenario Mining in the Urban Domain: Exploiting the Topology of a Road Network for Maneuver Annotation and Scenario Extraction

Scenario Mining in the Urban Domain: Exploiting the Topology of a Road Network for Maneuver Annotation and Scenario Extraction

Mecanismo Eficiente de Localização Cooperativa para Veículos Autônomos Conectados

Map Matching-Based Driving Lane Recognition for Low-Cost Precise Vehicle Positioning on Highways

Contact Info

Product

Resources

About