Thomas Snitch is visiting professor at UMIACS and chairman of the Board of Visitors of the College of Computing, Mathematical, and Natural Sciences at the University of Maryland, CEO and co-founder of GeoQuera, as well as president of Little Falls Associates, an international consulting firm. His research focuses on use mathematic models to use UAS and satellite imagery to combat poaching of endangered species in Africa and Asia. Last year, he and his team put their research to use in South Africa, flying a UAS over a game reserve in South Africa to help protect rhinos.
You’re an economist at heart, and have previously spent time in the State Dept. You’re team even created a program to help the pentagon predict where bombers in Iraq in Afghanistan would use IEDs. What sparked your interest in UAS? How did that experience translate into what you’re doing now?
I did my doctoral work in mathematical economics using models of risk behavior to analyze human behavior. In 1978, my dissertation looked at how terrorists chose their targets. What I did was to look at literally a hundred variables surrounding terrorist attacks on political leaders. Using the Tobin risk model of gambling behavior, I was able to discern that rather than crazy individuals who were simply out to kill, it was clear that terrorists were very rational actors. They carefully assessed the costs, risks and benefits when picking out a target. I have spent the past 35 years looking at other issues using this methodology.
For the last eight years, I worked with GeoEye, a leader in high-resolution satellite imagery. The next logical step is to bring the imagery down to earth and employ UAS to hone in to particular situations.
Flying UAS in foreign countries requires coordination with various agencies and governments. Can you tell us about your experiences in getting approval for your missions? What’s the importance of going through this process?
Just as in the U.S., foreign nations have rules and they must be followed or the consequences can be very ugly. In South Africa, we had a local partner, the Endangered Wildlife Trust, and they were instrumental in helping us with the paperwork to import the UAS into the nation. We had to apply to the Civil Aviation Authority to give us permission to fly in the South African skies and we had to speak with the South African Air Force because they have a base very near to where we were planning to fly.
The key is to have a good local partner who knows the local laws and regulations. You must start the process very early on because the delays can be lengthy. If you fail to follow local rules and regulations, you may end up in prison in a foreign land.
From your experience working with foreign governments, do you have any advice for other looking to do similar projects abroad?
My advice is quite simple – do it right the first time. No shortcuts. No lies. No trying to sneak UAS into a foreign nation. Play by the rules.
Poaching is on the rise. According to the International Fund for Animal Welfare (IFAW), wildlife trafficking is now the 3rd largest organized crime in the world, estimated at $19 billion. But as this market evolves, so too do the methods to combat it. Can you explain, generally, how UAS are being used to help?
A UAS can be very helpful in both finding animals and poachers. We use a mathematical model that is based on a number of variables: animal collar data, ranger patrol patterns, weather, ground intelligence and previous poaching incidents – anything that helps us to statistically recreate the environment that is present during previous poachings. If we can then clearly determine where rhinos are on a given night, we can use a UAS with an infrared camera to circle the animals while looking for poachers that are approaching them. With this information, we can move the ground rangers into positions so that they can intercept the poachers before they can reach the animal.
The model tells us precisely where to fly the UAS to have the most beneficial impact.
Anti-poaching UAS projects involve more than just strapping a camera to a UAS. To make a solution feasible, you have argued that a system will need to utilize “predictive analysis and heuristic modeling to narrow areas to be monitored.” Could you elaborate on what that means/entails?
You must know where to fly and this is where the modeling comes in – Africa is so big that simply launching UAVs into the sky will have no impact. You must learn the patterns of how animals and poachers move and then use the UAS to be your eyes in the sky.
We use very high resolution satellite imagery as a base and then layer on data covering everything from animal collar data, GPS routes of rangers, weather, ground intelligence, previous poaching incidents, moon phases, everything we can get our hands on. This allows us to statistically recreate the environment that was present during previous poaching events.
Heuristic means that we are constantly adding new data to the model – we get data dumps every week – and the model is continually learning. That is, the model looks for changes in the data and teaches itself to monitor new patterns. This is critical because as our work successfully catches poachers, we must believe that their behavior will change. The key is early recognition of these changes so that we can stay one step ahead.
We also know you maintain that any potential solution needs to be “exportable, importable, affordable, maintainable and easy to use in the field.” Could you give an example or two of what this might look like? How do we get here?
An example is the UAV Solutions Talon 120 UAS. Its wing span is 10 feet, so the Government of South Africa licenses the Talon as a hobby aircraft that is not subject to Civilian Aviation Authority. This means you do no need to file a new flight plan for every mission. It is affordable – the entire system of the UAS, cameras, ground package, and computer is less than $25,000. I like to price the UAS under the cost of a car. I can teach the rangers to fly the Talon in a few hours and show them how to repair the UAS in an hour – the system is completely modular so everything pops together with a color-coded system. It is like an IKEA project.
We have to have technology that is appropriate for local environmental conditions. You must consider the capabilities of the locals who will fly and operate the UAS. It must be simple, but get the job done.
The key is to select a level of technology that can get the job done but without unnecessary add-ons that increase cost and make it harder to repair.
As UAS technology continues to develop, how do you foresee it being used by researchers in the future?
I think the opportunities to use UAS for conservation or most anything else is only limited by one’s imagination. This morning, I signed an agreement with the Dian Fossey Gorilla Fund to analyze 13 years of data and then think about using UAS to map where gorillas nest at night – never been done before. The jungle is too thick and only a UAS with infrared capability can track them. Soon, I will be going to Costa Rica for a major conference on illegal fishing. I will be talking about how to use a UAV to target illegal fishing vessels. Our plan is to use the UAV to geo-tag the position of the vessel and film them hauling in the catch at night inside a nation’s economic zone. The UAV will provide the evidence the nation needs to seize the catch and the ship through asset forfeiture.
UAS will become a key tool in literally hundreds, if not thousands, of research applications.