Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
NASA’s rovers have revealed exquisite details and close-up images of Mars and the Moon, but exploration of other planets is slow. Most rovers are controlled remotely from the ground, and their precise, energy-efficient design limits the amount of ground they can cover.
now, New search From the University of Basel, ETH Zurich, and the European Space Agency, published Tuesday in the journal Frontiers in Space Technologies, how robots that can “think” for themselves could help us search the surface of the planet much faster.
Because rovers are built and oriented in specific ways, exploration is often limited, making it difficult to collect data across a large area of the planet. It also takes some time to transmit data across the vast distance between Earth and other celestial bodies.
“The rovers are designed for energy efficiency, safety, and moving slowly across hazardous terrain,” the researchers said. statement. “Here, we tested a different approach.”
They experimented with a “semi-autonomous robotic explorer” that could investigate multiple targets and collect data without being told where to go.
“The results show that semi-autonomous robots equipped with built-in tools can significantly speed up the process of resource mining and searching for ‘biosignatures’ (i.e. evidence of life) on planetary surfaces,” the researchers wrote.
So instead of humans controlling robots by remote control over long distances, the researchers found that a robot equipped with the right tools could navigate the surface and collect data in separate locations on its own.
Autonomy test
The researchers used a four-legged robot with a robotic arm. The robot, which was programmed to work without humans, carried a special camera and spectrometer designed to identify different materials.
“[The robot’s]autonomy is limited to executing pre-defined commands, including locomotion, waypoint navigation, device deployment, and data return,” the researchers wrote in their paper.
To test the robot, the team used the Marselabor facility at the University of Basel in Switzerland. This facility simulates planetary surface conditions with a rocky environment and special lightning.
The robot moved through the facility to specific targets, such as certain types of rocks hidden in the landscape. He then used his robotic arm to deploy the scientific instruments and bring back images and data for analysis.
The researchers compared this new strategy to a more traditional approach: scientists direct the robot to explore one target at a time.
The researchers found that multi-objective tasks took the semi-autonomous robot 12 to 23 minutes, while human-guided tasks took 41 minutes to perform similar exploration and analysis.
Moon maneuvers
While most rovers use wheels to move around, using a legged robot could be beneficial, as some lunar resources are located in inaccessible terrain with steep crater walls, which could present a challenge for wheeled robots, the researchers wrote.
NASA has also worked on this Autonomous vehicles It can explore and mine the Moon with minimal human guidance. And while the current Artemis II mission We won’t use a lunar rover, these robots can help Pave the way For a sustainable human presence on the moon.
NASA in December Uses artificial intelligence To successfully plan a training session for the Mars Perseverance rover. The team used Anthropic’s Claude AI models in two demos. The rover crossed obstacles on the surface of Mars and drove about 1,500 feet without any problem.