Brains For Bots

 Robotics  

Neurala is using neuromorphic algorithms to control robots with intelligence and autonomy for robotics.  Started by CEO Max Versace, Neurala uses neuroscience to design vision-based systems that may one day be used for rovers on Mars, and robots on Earth.

T he US space agency, NASA recently selected Neurala for a Phase II NASA Small Business Technology Transfer (STTR) Award.  Neurala is developing software for the Mars Rover and other planetary exploration robots to autonomously explore new environments, identify interesting objects, memorize locations of obstacles or objects, learn about objects, build and update an environment map, and return to a safe location.

The video above shows a a neural model controlling an active visual system in a simulated Mars rover and a corresponding robotic platform. The work is a collaboration between Neurala, the Boston University Neuromorphics Lab, and NASA Langley.

Image Source - Neurala

Related articles The Human Brain Project Has Begun Building the Super Brain  Robots, Humans and the Future of Space Exploration  Researchers Develop Neuromorphic Chip With Cognitive Abilities

The Neurala demonstration robot's brain processes visual information in real time, enabling it to do more than simply navigate from one spot to another. This means robots could one day be trusted to make their own decisions when navigating changing terrain on Mars.

"If I'm going to send a rocket that costs billions of dollars, I want to be sure that every millimetre I travel is gonna be safe," says Max Versace, Neurala's CEO and head of Boston's Neuromorphics Lab.

Mark Motter of NASA's Langley Research Center in Hampton, Virginia, told New Scientist that Neurala's approach highlights the difference between automation – in which a detailed, prescribed plan is executed by a machine – and autonomy, where a machine is free to make its own decisions on how to reach its goal. Neurala aims to mimic how human brains recognize objects, accumulate experience and make judgements, Motter says. "This is an interesting approach to autonomy."

The challenge is great.  Planetary exploration requires robots to explore:

• -Without a pre-existing map, because no detailed map has ever been created

• -Without GPS, as there is no positioning system available

• -With low power, since it cannot deplete the spacecraft’s batteries

• -With limited communications, as it can take 5-20 minutes for a signal to reach Mars (depending on orbit)

• -With unknown obstacles that could be risks (rocks — which may have rolled, cliffs, caves, etc.)

Moreover, the use of neuromorphic chips has great implications for the development of artificial intelligence and robotic systems on Earth.  

In the video below, Versace describes Neurala's work in more depth, including how Neurala is working on systems for the home:  (For anyone scared of the future of robotics, note what Versace says at the four minute mark - tongue in cheek, but essentially the truth).



SOURCE  Neurala

By 33rd Square

Subscribe to 33rd Square