Students developing systems for data gathering
THE SANTA CLARA
January 29, 2014
A student group in the Santa Clara Robotic Systems Laboratory has been developing robots that work together to collect data from an environment, such as depth and temperature.
The technology, called adaptive sampling, would allow scientists to bundle together information taken from the sensors of a group of robots, obtained from locations such as the ocean floor. The information could then be used to build a map of the surrounding area.
“Adaptive sampling is not knowing the location to which you are going, but knowing the condition that you seek, like trying to go out and find the location of maximum temperature,” said Christopher Kitts, associate professor and director of the Robotic Systems Laboratory, who is overseeing the project.
Kitts and his students have used robots to map the depth of Lake Tahoe for the past five years. In 2013, the group experimented with robotic kayaks in Lake Tahoe equipped with adaptive sampling technology.
“We’ve developed new approaches and have demonstrated in the field the ability to ‘climb’ or ‘descend’ a gradient in order to go to the maximum or minimum point in a field,” Kitts said. “To our knowledge, we’re the first group to ever do this in the field with a real, naturally occurring environment.”
Ben Margolis, a graduate engineering student involved with the project, said this technology could be used for search and rescue missions, transportation technologies and surgical procedures.
“Adaptive sampling allows for mapping of the environment, like significant patterns or sources of heat, pollutants or other chemical or biological markers,” Margolis said. “This type of mapping would allow us to continue to learn about and care for the world we live in.”
The first step to using adaptive sampling technology is being able to control the formation of a group of robots, Kitts said. For the past five years, his group has experimented with wheeled robots, boats, underwater robots, aerial drones and blimps. The group has also used the technology to follow an environment’s contours in experimental settings.
Kitts said they have also moved the technology along crests and trenches during simulations.
“Collectively, these capabilities are the basic strategies required to do large-scale exploration, so the fact that we are a leader in this area is pretty exciting,” Kitts said.
Alumnus Thomas Adamek, who has worked in the Robotic Systems Laboratory for the past four years, said the technology could help scientists accomplish things that would be previously unattainable without significant time and resources.
According to Kitts, adaptive sampling could also be useful during environmental disasters, such as chemical spills, and the exploration of areas such as hydrothermal vents at the bottom of the ocean floor.
Contact Sophie Mattson at firstname.lastname@example.org or call (408) 554-4852.