Plenty of volunteer mapping is based on satellite data. People justifiably get irritated when they actually do a physical survey and then it gets overwritten remotely.

As far as the imagery, DigitalGlobe provides relatively up to date imagery for most of the planet. The errors are probably more from overinterpreting the satellite data than from it being obsolete.

Not defending grab, but osm has lots of false negatives. Satellite image does not have to be outdated - you can get less than 6 months old satellite image accurate enough to extract roads and buildings from Digital Globe or Airbus.

Shameless plug: At tensorflight.com we have deep learning models that can do just that.

6 months old is outdated, depending on how fast things change on the ground.

Even 2 months old can be outdated (e.g. can be missing houses built since, or showing houses that no longer exist). In some areas, this can be a very significant effect.

For most use cases it's more cost effective to use existing, slightly old imagery. On the other hand, for some use cases, it's necessary to get a very fresh image and commision an airplane flight (or drone if the area is very small) - e.g. hurricane damage mapping.

It's more cost effective if your choices are to use the two vintages of imagery, sure. I understand why imagery can be stale.

My point was that OSM data can be much more accurate than the imagery. Or not, depending on location.

OpenStreetMap has a lot of things left to map. Sometimes, some correct data is better than no data.

I guess it depends on the sat image source which leads to costs.

I'm sure you can use the newest/best data for your company because you can pass on the costs to your clients.

We indeed pass the cost of imagery to customers, but it's flexible. I.e. given the location and customer requirements we have a decision tree that picks the best image source from multiple providers based on freshness, spatial resolution, cost, and some other factors.

How does that tree approximately look? Which providers are there?

> Which providers are there?

In the "high spatial resolution" - i.e. 50 cm per pixel or less the well-known providers everyone is using are Digital Globe, Airbus or Nearmap. Above 50 cm it starts to get hard to confidently see enough detail.

Google maps terms of service actually prohibits using satellite imagery for feature extraction ( https://cloud.google.com/maps-platform/terms/ ).

Bing is the best on the cheap end. Actually, Bing donated imagery for OSM mapping purposes ( https://wiki.openstreetmap.org/wiki/Bing_Maps#Bing_Aerial_Im... ).

I think that they wouldn't use only satellite imagery, but gps data from the fleet and user reports. See also https://qz.com/1481849/grab-southeast-asias-biggest-ride-hai...

When OSM got started, the idea was to use GPS track data almost exclusively, or at least for the finer resolution stuff. The satellite imagery available freely at the time wasn't that good anyway.

But cleaning a bunch of GPS tracks can be a real mess. Many GPS tracks don't include the error/accuracy measurements and just pretend to be infinitely-small points scattered in the general vicinity of what might be a road. I think people tend to overestimate how accurate their satnav or phone's GPS is, because in most uses it's "locked" to a road or other map feature.

Anyway, not an intractable problem by any means, but preprocessing and cleaning GPS data to derive the actual terrain can be quite difficult. Paying somebody to click repeatedly on a satellite image might be temptingly easier.

That was by necessity, IIRC: almost everything else would be under restrictive licenses. The open data landscape has changed a lot since.

Aren’t you able to purchase fresh sat images on request? I’m pretty sure I’ve seen these services online.

Fresh sat images are much more expensive than old ones

How much?

In many places, OSM can be improved with satellite imagery

Yes, agreed. In built-up areas it's not like roads move very often.

I wonder if someone couldn't do "deltas" of satellite imagery to produce some sort of heatmap of how quickly various areas are changing. That would give you an idea of what areas are 'safe' to map based on satellite imagery, and what areas are changing too quickly and need to be mapped some other way.

It would also give you a way of automatically flagging stale data on the vector map; if something hasn't been updated in 12 months and you know from imagery that the area is under heavy development, those map features are probably not trustworthy anymore.

Development Seed has a project Urchn that was trying to do something like that: https://medium.com/devseed/make-sense-of-urban-change-fabb9f...

The idea was to take satellite imagery of the same location from different timestamps and I believe use deep learning to determine if anything in the imagery changed.

I don't think they've released any code for that project yet though.

But for determining whether or not the imagery could be used in a certain region, it might make more sense to just compare the timestamps of the vector data and the imagery. For instance, if the road doesn't appear in the imagery, but the vector data timestamp is after the imagery timestamp, then keep the road. And don't add a road if it was covered by a map segment that was deleted some time between the imagery timestamp and the current time.