As the name suggests, neuromorphic computing uses a model that’s inspired by the workings of the brain, and is a practise that uses significantly less energy than a conventional computer chip. Most hardware today is based on the von Neumann architecture, which separates out memory and computing.
But because chips of this type have to shuttle information back and forth between the memory and CPU, they waste time, so to compensate, more chips are added to the circuit to make them faster. Momayez’s colleague, Jekan Thanga, an associate professor of aerospace and mechanical engineering, is the lead on HEART and he is adapting the neuromorphic learning architecture technique to fit on a new generation of miniature sensors that will one day drive the autonomous robots.
“It’s really exciting to be at the forefront of a new field,” said Moe Momayez, interim head of the Department of Mining and Geological Engineering at Arizona. To test the design, the team have created low-cost, rapidly designed, 3D-printed rover prototypes that when perfected, could assist astronauts on the moon with laborious mining tasks.
The HEART system not only will train robots to work together on mining, excavation and even building tasks, but it will also allow the robots to improve their collaboration skills over time. Neuromorphic computing on the other hand models the way the brain works through “spiking” neural networks that can convey information in both the same temporal and spatial way as the brain can and so produces more than one of two outputs.
The team plan to build and train the robots here on Earth, so they can practice before heading to the Moon. Ultimately, the researchers envision a fully autonomous swarm of robots that doesn’t need to receive instructions from Earth to mine materials and construct simple structures. “The idea is to have the robots build, set things up and do all the dirty, boring, dangerous stuff, so the astronauts can do the more interesting stuff,” Thanga said.
Instead of the robot drilling through the rock to extract ore, Momayez has developed an electrochemical process to drill through rock five times faster than any other method. “By going through this process, we help perfect these artificial creatures whose job it is to do the mining tasks,” Thanga said. While Thanga concentrates on the autonomy side, excavation of materials is being handled by Momayez.
News Highlights Space
- Headline: NASA funds a project to mine the Moon with swarms of autonomous robots
- Check all news and articles from the Space news information updates.