Facet Builds New Street Centerline Database - How Does It Match Up to Current Digital Street Data?
Corporation could be the next NAVTEQ or Tele Atlas, at least for U.S. The company's new offering "includes the public roadways in the U.S. at high accuracy and includes full routing and connectivity." How do the data compare? Joe Francica questioned Andy Munyon, vice president of Business Development to find out.
Joe Francica (JF): You are engaged in capturing street
centerlines for the U.S. similar to what NAVTEQ and Tele Atlas provide.
Did you start with a baseline digital database
like TIGER and enhance the data?
Andy Munyon (AM): No, we did not enhance
existing data. SightMap was created from scratch from the multitude of
miles that our collection vehicles have driven.
JF: Did you drive every street with your fleet of vehicles?
AM: We drive every street within our collection areas. By
the end of 2008, we will have driven every road within the continental
U.S. and Canada.
JF: You mention that SightMap is targeted at the LBS market.
What are you seeing in the market that leads you to believe there is a
need for higher accuracy for routing and navigation?
SightMap has higher spatial accuracy than existing data sources, this
is not what will differentiate our product in the LBS and navigation
markets. We see the LBS and navigation markets calling for more
accurate and rich attribution.
JF: How are you intending to keep the street centerlines
will keep our centerlines updated using a variety of methodologies. Our
collection vehicles are on the road every day capturing new, and
updating existing, data. Furthermore, we will use customer feedback and
both public and private third party data providers to augment our
ability to identify areas where updating is required. Once identified,
we will send a collection vehicle to gather data that will be updated
in the SightMap database.
JF: The key differentiator between your product and the two
leading digital street vendors appears to be accuracy. How much more
accurate is your data and how have you verified this?
navigation and LBS dataset spatial accuracy will set a new benchmark
for digital map data providers. Our accuracy specifications have been
validated by multiple customers. Beyond accuracy, our attribution data
and price will set us apart. SightMap was created using new processes
that are very different from how maps have been made in the past. The
efficiencies in our methodology will result in lower licensing fees for
our customer, in comparison to the current pricing environment.
For the ADAS [advanced driver assistance systems] market, sub-meter
accuracy and elevation (z) data differentiates our dataset. Current
industry-accepted accuracy specifications for ADAS data will not, in
our opinion, support future map-to-machine interfaces in vehicles.
JF: You list several patents for 3D object generation. Can you
explain the applications that you intend to develop?
content is very suitable for 3D object generation, and our patents are
a key component in the automation of 3D model creation. We do not offer
a 3D product. Our customers can license content and patents to create
their own 3D models using our imagery.
JF: You also list a patent for "Method and Apparatus for
Multi-image Analysis." How will this technology be integrated with your
digital street database?
image analysis patents provide key portions of our workflow for content
creation. In addition, they will be fundamental building blocks for
vehicle safety systems and autonomous vehicle features. The attribution
for our ADAS sets will include information about signage, signals,
barriers, crosswalks, etc. The sensor systems on board the vehicles
will capture information from the surroundings and make decisions about
warnings to be relayed to the driver or changes to be made in the
handling of the vehicle. Our patented approaches for acquiring sensor
data on board the vehicle and interpreting the sensor data are
necessary elements for the map-to-machine interface.
JF: Do you intend to license your technology to various LBS
application developers? Can you tell me who has shown the greatest
interest so far? Or will you market directly to carriers and device
intend to license our technology to all map data users. We're under
NDAs [non-disclosure agreements] with all of our customers at this
time. Unfortunately, we can't comment on current or future customers at
JF: You are integrating intersection imagery into SightMap. Is
your image library from only primary roads? Will you be expanding this
image library includes 360-degree imagery from all roads we have
driven, not just primary roads. In addition to intersection images, we
also have imagery of a large number of POIs, addresses/parcels, road
assets, ramps, etc. Anything that you can see from a vehicle, we
capture with our cameras. The database is constantly expanding, 365
days a year. We currently have well over 2 billion images in the
JF: What will be your next steps in
the navigation and LBS markets, our next step will be to expand
coverage into Europe and Canada. For the ADAS market, we will be
focusing on increasing our accuracy to the decimeter level for future
ADAS applications and augmenting existing attribution.