Right now, it's kind of a Wild West out there, says Scott Simmons, executive director of the Standards Program of the Open Geospatial Consortium (OGC). The sea change in mapping is the huge proliferation of sensors of variable quality now available”there are lots of good ones, and some that are only slightly better than what I would call ˜hobbyist' level. Finding ways to compare, assess and integrate data from different sources is an area crying out for better standards, and we're just getting started.
Simmons reached out to me after reading the January/February 2016 UAS Innovations column comparing the rapid emergence of UAS technology to that of GPS and laser scanning.
At first I was going to take issue with the comparison to GPS, but as I thought about it, I realized it was a good one, he says. We should pay attention to the way GPS went from specialized technology to suddenly being everywhere. It really is seeming possible now that small drones equipped with cameras and other sensors will be widely deployed, by professionals and amateurs, and that the data will be used to create maps. And just as there were successes and failures with GPS data, there are going to be similar challenges with data from sensors deployed on UAS platforms.
Simmons and I engaged in a wide-ranging conversation in which he made several fascinating points, including the following:
- OGC is starting to talk with UAS users and firms about the provenance of data and how to make them more usable. At last year's UAV expo in Las Vegas, we basically went from booth to booth and asked, ˜What do you need?' notes Simmons. And the most common answer was ˜a way to describe data quality in consistent fashion so apple-to-apple comparisons can be made.' In other words, end users need to be able to compare data from a relatively low-res photo-capture sensor and a high-end laser scanner in ways that allow both to be used intelligently on the same project.
- End users would like a data format that enables rapid exchange and analysis of flight-planning data, adds Simmons. That's tricky, because much of the innovation here is proprietary, and manufacturers can be reluctant to share. But knowing whether the programmed flight was in a spiral or some other algorithmic pattern is incredibly helpful, especially if a project requires several visits. There's a need to understand how an area was covered. For example, monitoring projects where quantities or movements are being tracked, and data from many flights will be compared over time.
- UAS is probably different enough to require specialized standards. There's obviously overlap with, say, aerial photogrammetry, and much of the work done on filing protocols can be brought in easily. But for actual flights and data gathering, the rules are different, compared to planes or satellites, and right now we have no good way to describe UAS missions.
- Photo capture is really booming because the algorithms are getting really good, he says. That's a change”the early stuff was not really usable. Now it's getting accurate, and point clouds from photo capture are very useful. A lot of people are beginning to feel that laser scanning will become something of a boutique process.
- It's true that smartphones produce terrible location data in terms of accuracy, but here's the thing: We've discovered all these uses for less-precise data. I remember the first time I used a handheld GPS receiver and how thrilled I was to come within three feet of a mine-shaft entrance I would not have otherwise spotted. In the same way, widespread, low-quality data from hobbyist drones likely will find useful applications. Great work is being done in the EU to implement ˜citizen science' data from browsers and phones, and it's possible now to tag data for quality. So even though the data is gathered somewhat casually, it's being used intelligently.
- In the long term, (UAS data) should lead to a mapping revolution, but we're not going to get there if we can't easily distinguish between good and bad data. There is definitely a role for OGC and organizations like ISO. But one way or another, it has to get done, and I'm cautiously optimistic that it's going to happen.
Author's Note: My thanks to Scott Simmons for sharing his views with me. Any other readers interested in sharing can reach me at [email protected].
Angus W. Stocking is a licensed land surveyor who has been writing about infrastructure and technology since 2002; e-mail: [email protected].
