How Lyft Creates Hyper-Accurate Maps from Open-Source Maps and Real-Time Data

#News(Startup) [ via IoTGroup ]

Headings…
How Lyft Creates Hyper-Accurate Maps from Open-Source Maps and Real-Time Da
Why are maps important for Lyft?
What is the role of the mapping team?
Why is having an accurate basemap important?
Driver Localization on the road network: Map-Matching
What are the possible types of map errors in the road network?
Type I Map Error: Map-Matching-based
Type II Map Error: Routing-based
Type I and II map errors for road existence, road direction, turn restricti
How to adapt map-matching to type I errors and find them?

Auto extracted Text……

Driver Localization on the road network: Map-Matching
A traditional map matching algorithm [2] in red, leverages our knowledge of the road network and accurately compute the trajectory of the driver.
What are the possible types of map errors in the road network?
The road network errors that trigger map-matching issues and routing issues, denoted Type I map errors
The road network errors that mostly trigger routing issues, denoted Type II map errors
Because Type I map errors are the only ones that trigger map-matching issues, we can detect them by finding where driver localization is failing on the road network.
A traditional map matching algorithm, in red, fails to detect the missing road, and inaccurately computes the trajectory of the driver on the imperfect road network.
Because Type II map errors are the only ones that triggers routing issues without triggering localization issues, we can detect them by finding where on the road network routing is failing while localization is not failing.
Fig. 4 — In this example, the extra road in dotted line does not cause map-matching errors.
Type I and II map errors for road existence, road direction, turn restrictions
Table 1 — Type I and Type II map errors for road existence.
Table 2 — Type I and Type II map errors for road directions.
At Lyft, the output of the Kalman Filter — the off-road locations — are used to detect Type I map errors, which encompasses missing roads, roads in OSM that are set to the wrong one-way direction, and turn restrictions that should not have been mapped in OSM.
By leveraging weeks of anonymized Lyft’s driver locations when they are connected to the Lyft app, and making a large-scale plot of off-road trajectories, we can highlight areas where we observe Type I map errors.
At Lyft, we found that most of the Type I map errors are due to missing roads (although we have observed a few incorrectly labelled road directions and turn restrictions that trigger Type I map errors)

Read More..
AutoTextExtraction by Working BoT using SmartNews 1.0299999999 Build 26 Aug 2019