2015 IEEE Intelligent Vehicles Symposium (IV) 2015
DOI: 10.1109/ivs.2015.7225665
|View full text |Cite
|
Sign up to set email alerts
|

Grid-based online road model estimation for advanced driver assistance systems

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
3
0

Year Published

2016
2016
2020
2020

Publication Types

Select...
5
1

Relationship

1
5

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 11 publications
1
3
0
Order By: Relevance
“…The majority of this error comes from the longitudinal position, where the lateral position error is much less. This demonstrated performance meets the requirements of ADAS and automated driving applications, where there are higher performance requirements on the lateral position than the longitudinal position, for example when matching an object onto a lane, doing lane-change prediction [340,341] or estimating a road model using object data [286]. Furthermore, performance requirements from ADAS are higher for the front area of the host vehicle than for the side and the rear, which also matches the presented results.…”
Section: Discussionsupporting
confidence: 80%
“…The majority of this error comes from the longitudinal position, where the lateral position error is much less. This demonstrated performance meets the requirements of ADAS and automated driving applications, where there are higher performance requirements on the lateral position than the longitudinal position, for example when matching an object onto a lane, doing lane-change prediction [340,341] or estimating a road model using object data [286]. Furthermore, performance requirements from ADAS are higher for the front area of the host vehicle than for the side and the rear, which also matches the presented results.…”
Section: Discussionsupporting
confidence: 80%
“…It is important to note that this road model formulation makes no assumptions about the geometrical shapes lanes can have. Additional information such as the type of lane boundaries can easily be added later by extending the definition of B or p. In contrast to [23] this work is based on the theory of belief functions as brought forward by Dempster and reformulated by Shafer [24]- [26]. Dempster-Shafer theory (DST) is a generalization of Bayesian probability theory and allows to calculate the belief in a specific hypothesis taking all available evidence from different sources into account.…”
Section: Related Workmentioning
confidence: 99%
“…Higher-level functions like prediction and maneuver planning require a continuous and consistent description of the lane geometry and topology graph to work. Different methods to extract these representations from grids were already proposed in [19], [23]. In this paper, lane geometries are extracted by first searching for drivable paths using the belief Bel(L) of m all 1:t with the path planning method with unknown goal pose described in [32].…”
Section: Road Model Extractionmentioning
confidence: 99%
“…There are multiple methods of building detailed representations for the road model, and they can result in different models based on various purposes. For example, in [1,2], the authors presented approaches for grid-based road model estimation for advanced driver assistance systems. Their measurements from sensors are transformed into a grid-based road model and a geometrical description is extracted out of this model by the use of a path-planning based method.…”
Section: Introductionmentioning
confidence: 99%