ASIMOV: Team Part-Time Scientists are developing Asimov to roam the lunar landscape autonomously, that is, after they navigate the rover the required 500 meters to win the Google Lunar X PRIZE. Image: Courtesy of Part-Time Scientists
More In This Article
Some people try to make the most of their spare time by exercising, volunteering or simply recharging their batteries. Others like to use that time to build robots that can be blasted to the moon and then set free to roam the lunar landscape. A group of engineers and researchers calling themselves Team Part-Time Scientists have chosen the latter, and are building a moon rover named Asimov they hope will win the coveted Google Lunar X PRIZE by early 2014.
As the name implies, at least half of Part-Time Scientists' 100 members are holding down full-time jobs at industrial firms or universities. They are competing against 25 other teams to be the first to land a robotic rover on the moon and have it travel 500 meters over the surface, sending high-definition images and data back to Earth as they go.
Part-Time Scientists' goals are representative of all the teams entering the competition—they are aiming for something more lasting than the $20-million first prize. After guiding Asimov beyond the 500-meter mark, the team plans to switch the rover into autonomous mode for the rest of its lunar exploration, making it the first bot built by anyone to navigate on the moon without human intervention. In addition, the team plans to be the first to use ultrafast graphical-processing units (GPUs)—known for their ability to render complex graphics for video games and scientific simulations—to help their Earth-bound command systems, Asimov's landing vehicle and possibly Asimov itself gather, assess and act on information in as close to real-time as possible.
Autonomous travel is part of Part-Time Scientists' strategy to design and test new technologies that could impact not just future moon travel but other interplanetary endeavors as well. "We don't want to develop something specific for the moon that ultimately is of no use to anyone else," says team founder Robert Böhme, who works as a cyber security advisor for the German government in Berlin. "We want to take useful technology with us and test it on the moon, which is one of the reasons the robot rover will operate autonomously."
Other teams are developing autonomous technology for similar reasons, including Astrobotic Technology, Inc., led by legendary bot-maker William "Red" Whittaker, and the Juxtopia Urban Robotics Brilliant Application National (JURBAN) not-for-profit research organization. Astrobotic's landing system will navigate autonomously by aligning real-time data from cameras and laser with existing satellite imagery of the moon. Meanwhile, JURBAN, like Part-Time Scientists, is designing its rover with autonomous capabilities.
Going for a touchdown
All the Lunar X PRIZE competitors have access to images and data collected by NASA's Lunar Reconnaissance Orbiter (LRO) as well as from previous manned trips to the moon. Still, Asimov will not really know the specific terrain it needs to cross until it touches down on the lunar surface, says Wesley Faler, the team's chief software developer. Based near Detroit, Faler's day job is writing software, including programs for the U.S. Environmental Protection Agency.
"In those last few minutes before landing, we'll be getting a tremendous amount of video data from the moon," Faler says. In addition, the landing vehicle that will set Asimov down on the surface has integrated force-feedback sensors that can analyze soil composition around the landing area. "We're hopeful this combination of visual and tactile data will [yield] an unprecedented level of lunar detail," he says.
Asimov is to land somewhere in the vicinity of the Apollo 17 site, which has fine soil and few large rocks that might impede the rover's progress. "We have an algorithm that calculates rock density that shows we can do a lot of exploring without getting into any trouble, such as getting stuck trying to get around [a rock]," Böhme says. Much of what scientists know about this site comes from the original mission as well as LRO flyovers.
Despite all of the preparation that goes into a moon landing, teams cannot be entirely certain of where their rovers will actually touch down. "We're relying on relatively low-resolution data ahead of the mission and may later find out that there are bumps there that would tip over the lander," Faler says. Böhme adds, "You need to make adjustments as you go, and to make those adjustments you need information."
Move over, rover
The GPUs should give Asimov a distinct advantage over prior rovers, even those NASA has on Mars. The rovers there have performed commendably since setting down on the Red Planet in 2004, but Spirit (when it was operative) and Opportunity (which is still active) must pause frequently to gather, assess and act on information about their surroundings, Böhme says. The rovers require nearly three minutes to process a pair of images—a delay that causes them to move at an average speed of about one centimeter per second.