NASA’s Curiosity rover has been employing an upgrade to its robotic “brain.”
Space.com reports that the Mars rover is now capable of choosing its own targets of study, thanks to improvements in the robot’s artificial intelligence program.
Ground controllers beamed a software update to Curiosity in October 2015, called the Autonomous Exploration for Gathering Increased Science (AEGIS). The implementation of AEGIS marked the first time that artificial intelligence had been tested on a remote probe.
The AEGIS software makes Curiosity better able to identify rocks that scientists want to study. Ground control cannot maintain constant contact with the rover, and even when in contact, a round-trip signal takes about eight minutes to travel. AEGIS saves significant time by allowing Curiosity to make autonomous decisions rather than waiting for directions from Earth.
Without the AEGIS program, Curiosity would target an item of interest about 24 percent of the time. With AEGIS is use, the rover identifies desired targets about 93 percent of the time. The increase in targeting accuracy and the amount of time saved by AEGIS make Curiosity’s performance much more efficient.
AEGIS uses Curiosity’s navigation cameras and Chemistry and Camera instrument (ChemCam) in tandem to identify objects of interest. When the rover locates an object that falls within mission parameters defining something that bears study, the ChemCam uses a laser spectrometer to analyze the composition of the object.
The success of AEGIS demonstrates how similar AI programs could be used for further remote space missions.
“The farther you go in the solar system, the longer the light time delay, the more decisions need to be made on the spot,” AEGIS lead engineer Raymond Francis said.
Study results were published in the journal Science Robotics.