The information, gathered over many different locations, helps scientists understand how and why the world’s ice shelves are melting as well as the dynamics influencing the stability of ice sheets.
Although much of the data can be obtained by satellites, their data isn’t as accurate as good old-fashioned, on-site measurements. The drawback is that static, ground-based weather stations limit the number of locations that scientists would like to cover. Worse, many of the locations are on volatile ice sheets with the possibility of cracking, shifting, and filling with water not safe for scientists or high-cost equipment.
SnoMotes, developed at the Georgia Institute of Technology in cooperation with Pennsylvania State University, operate as a team, autonomously collaborating to gather assigned measurements from many different locations. They use cameras and sensors to navigate. Ayanna Howard, associate professor in the School of Electrical and Computer Engineering at Georgia Tech and project lead, previously worked with rovers at NASA’s Jet Propulsion Laboratory.
It took a few tries when the research team first set out to build a rover to record environment data. The first rover was delicate and ineffective. The team decided they needed something that could withstand constant abuse. So the second attempt was a sturdy, remote-controlled toy snowmobile kit already primed for snow and designed for a child’s rugged use. Howard’s group installed a camera and all computing and sensors inside the 2-ft-long, 1-ftwide snowmobile. The result was a tough but inexpensive rover.
By using existing kits and adding extras like sensors, circuits, AI, and a camera, the team created an expendable rover that wouldn’t break the research budget if it were lost on a mission. Similar rovers under development are more expensive. The cost per rover might also limit the number of rovers a team could deploy at any one time. And losing an expensive rover might blow out research budgets. The loss of a SnoMote during a mission won’t break the bank.
The first phase of the project focused primarily on mobility and communication. Later versions will include a better sensor package developed by Derrick Lampkin, assistant professor in the Geography Dept. at Penn State, and Magnus Egerstedt, associate professor at Georgia Tech. Lampkin studies ice sheets and how changes in climate contribute to changes in these large ice masses, so he is well versed in the types of sensors SnoMote needs.
The Georgia Tech team has created three working SnoMote models to date, but as many Sno- Motes as needed can work together on a mission. Two key innovations in the rovers are a new method of location and work-allocation communication between robots and maneuvering in icy conditions.
Once on site, the rovers position themselves at strategic locations to ensure they cover all assigned ground. They have two different methods to decide amongst themselves which positions each takes to get all desired measurements. The first method is an “auction” in which robots “bid” on a desired location based on proximity and how well their instruments are working or whether they have the necessary instrument. For example, one may have a damaged wind sensor while another may have low batteries.
The second method fixes the robots to certain positions in a net pattern that stretches to fit the targeted location.
Another key innovation on the robotic vehicle is its ability to find its way through the snow. While most rovers use rocks and other landmarks as guides, a continuous white sheet with geographic features buried under several inches to several feet of snow make navigation difficult, at best. Howard’s students discovered that lines formed by snow banks could serve as markers to help SnoMote track distance traveled, speed, and direction. SnoMotes also navigate via GPS if there are no snowbank visuals.
A heartier SnoMote will be needed to handle the well-belowzero climates of Antarctica and other arctic areas. The new rovers would include heaters to keep circuitry warm enough to function and sturdy plastic exteriors that wouldn’t become brittle in extreme cold.