Daniel McKinnon PHD was the product development force behind 3DR’s Site Scan and is now the Director of Product Development at geospatial analytics company Orbital Insight.
You’ve always had a great drive to talk to real users in the field and figure out just what they’re looking for and why. What do you think are some very underserved niches that could benefit from specialized drone technology?
1) Agricultural spraying. I knew these numbers like the back of my hand during my Agribotix days. It is very expensive and dangerous to apply chemicals to by airplane when the conditions prevent terrestrial spraying. Heavy-lift drones can apply concentrated chemicals far more cheaply, more precisely, and more immediately than airplanes with greater safety and lower environmental impact. I’m looking forward to watching DJI evolve their Agras platform.
2) Dangerous delivery. I believe that drone delivery as a whole is largely overhyped for the immediate future, but there are certain scenarios where society spends an enormous amount of money and risks an enormous number of human lives to move a parcel from one point to another. These scenarios include mountain rescue, resupply behind enemy lines, delivery of life-saving medicines to areas stricken by disaster, and many more. We should identify situations where a human pilot is flying a multi-ton flying machine to deliver a few hundred pounds and replace that human and multi-ton flying machine with a drone.
You’ve worked on two very different sides of the aerial data industry, both in drones at Agribotix and 3DR, and in satellite imagery with Orbital Insights… What are the key differentiators between the two, why would users prefer one to the other?
From an analytics perspective, the businesses and technologies are largely identical–imagery comes in and analytics come out. Satellites obviously offer a larger catalog that is captured effortlessly at the expense of resolution and timeliness. That said, this tradeoff implies enormous differences in the business model.
Despite enormous advances in drone hardware and software that have enabled nearly autonomous operation, flying a drone still represents a significant hurdle for most companies. A company interested in integrating drone imagery into its decision-making process will either have to make a significant investment in operations or pay a premium to a drone operator. Satellite imagery arrives in a processing queue without effort and for minimal cost.
Large, macro-scale problems that can be solved with satellite imagery should be. If the resolution or the timeliness limitations imposed by orbiting spacecraft require the use of drones, then companies should be prepared to foot the bill associated with their operation.
Do you think there will always be a place for both drone and satellite data, or does the drone industry need to think ahead to when satellite imagery or other technology sources will be able to provide competitive data?
There will always be a place for both drone and satellite data. Commercial satellites are limited by international treaty to 30 cm GSD, which obviously precludes their imagery from solving quite a few business problems. In addition, a typical high-resolution satellite costs hundreds of millions of dollars to design and launch, so we will not see a flock of Worldviews orbiting the earth anytime soon. On the flip side, even fully autonomous large fixed-wing drones will never cover the earth the way satellites do. At Orbital Insight, we image more than a quarter million parking lots in the United States for our retail product. Unless the skies become blackened with drones, I see the job of the satellite as quite secure for these applications.
However, I do believe that the idea of a drone analytics company or a satellite analytics company will become obsolete. The pipeline to extract rooflines from satellite images to service the insurance market is nearly identical to that from drones. The analytics derived from the former would likely be purchased by an underwriter and those from the latter by a claims agent, but there is no reason a company couldn’t or shouldn’t offer both. Rather than differentiating by data source, the analytics companies will begin to differentiate by customer vertical served.
You spearheaded development of one of the best niche solutions available for aerial technology with Site Scan… Do you have any recommendations for new companies developing solutions for a specific vertical?
Talk to your customers. Seriously. Just talk to your customers, listen to what they have to say, and understand their problems. Go so far as to even literally step inside their shoes. Part-way through Site Scan development, I became a surveyor-in-training so I could get comfortable stepping on to any job site and empathize with the problems our customers were facing.
While Site Scan began as a broad drone mapping platform, it evolved into a specific tool for helping engineering service providers survey their earthworks projects through relentless customer development. Nearly every business activity at 3DR from engineering backlog to customer success process was driven by 3DR users. This focus allowed 3DR to outmaneuver entrenched competitors.
You’ve hacked together some projects yourself you’ve featured on your blog and forums like DIY drones. Whats one of your favorite DIY/hacked projects you’ve done?
I love building fun software projects. I’ve done some neat things with drones including hacking together the first Solo/QX-1 prototype that later became Site Scan R10C and building a tool to check on the games at the local basketball courts. More recently, I scanned satellite imagery of the entire planet to find secret signs of civilization in remote places. That said, my current favorite is unpublished. I’ve been writing algorithms to profit from inefficiencies on and between various cryptocurrency exchanges. I’ll post this one after all of our alpha disappears 🙂
What technology would you be excited about hearing is integrated into drones in the future?
The end of the era of innovation in sensor-carrying drones is in sight. Changes have been largely incremental over the past couple years. Currently, commercial drones carrying almost every type of sensing payload are already on the market hunting for vertical-specific solutions.
I am excited for the next generation. I believe that this will center around communications. From the tactical repeater that could be deployed to connect firefighters in adjacent valleys by radio while combating a blaze to the large-scale atmospheric satellites that the large tech companies believe will connect the distant corners of the earth to the internet, I am excited to see our flying robots evolve into a tool to further connect humanity.
What are some interesting trends you’re excited about in satellite imagery over the next few years?
Synthetic aperture radar. These satellites use radar to image through clouds and are not limited to day-time observation. While the resolution does not match the high bar set by E/O birds operated by companies like Digital Globe and Airbus, I am certain that the ability to image the entire earth in all conditions will unlock some new insights into the state of the planet.
What are some of the most exciting companies or technologies in GIS that you’ve come across in last year?
Orbital Insight is doing some pretty amazing work processing imagery from most of the world’s earth observation satellites into economic insights that are driving decisions across top businesses across the globe. Present company excluded, I have been most excited by the types of insights that have been extracted from voluntarily-collected phone telemetry data. I am sure that learning how the world exercises in the now-infamous Strava map and how the world travels using Uber Movement will lead to incredible changes in society itself.
If you’re new to the satellite world, whether as a new team member at a company, or a client looking into satellite data for the first time, where would you recommend learning about the industry?
The real commercial satellite industry, what many in the tech world call new space, is just lifting off (sorry, I couldn’t help myself). While Digital Globe has been selling imagery for a couple decades, their customers were largely governments and government-like organizations. Only recently have satellite imagery and analytics been available and beneficial to every business on the planet. The nascency of the industry means that information is quite fragmented. The Orbital Insight Confluence page is probably the best place on the planet to learn about satellite analytics, but if you haven’t joined our team yet, there are good webinars online hosted some of the major space hardware companies.