Robo-optics, Organ Clocks, Blood Tech + Reforesting

A looping video of a small, six-legged robot with sensors walking steadily forward along a narrow gravel path in a lush garden setting.

As this robot scuttles down a path, it uses a sensor, chip and tiny AI model inspired by biological eyes and brains to determine its location. The system uses a tenth of the energy of a conventional camera-based system and could work on any robot.

Adam Hines

👁 Low Energy, High Efficiency Remote Sensors

A new seeing robot has entered the chat, and the energy needed to power its vision is only a fraction of what conventional location systems demand. Kathryn Hulick reports for Science News on ​an eyelike sensor incorporated with a chip and tiny AI model​ that can change the remote visual sensing arena.

🤖 Low-Power Superpower

Robots that gather visual information from their surroundings need an eye of some sort. While in some cases a camera and an AI that determines location may serve as that apparatus, that’s not always the most efficient visualizing instrument because a camera is always capturing information even if the environment remains static, needlessly using a huge amount of energy. To save energy, researchers have developed a system called LENS, comprising a sensor, chip and tiny AI model.

LENS uses an existing, commercially available product known as Speck, which is the sensor-chip duo, created by the company SynSense. Akin to a human eye, Speck operates more efficiently than a camera. It pays attention to how the environment changes through a process known as neuromorphic computing, which more closely resembles how our eyes and brain take cues.

Not yet a subscriber?

Access to this content requires an Investors Lab subscription. Sign up for a free trial today to explore exclusive, data-backed insights into scientific breakthroughs set to disrupt industries.