Skip to main content
  1. Home
  2. Emerging Tech
  3. News

Bouncing robot reaches new heights, artfully dodges moving obstacles

Add as a preferred source on Google
SALTO - Teaching an old robot new tricks

Meet Salto, a tiny and adorable bouncing robot developed at UC Berkeley. The bot weighs just 100 grams (3.5 ounces) and is about a foot long, and it can move through an environment by bouncing and jumping.

Recommended Videos

Salto was first developed in 2016, and back then could only jump off the ground once then bounce off wall. The latest version of the robot has learned lots of new tricks, and now it can do hundreds of jumps in a 10-minute period, can jump up to 4 feet high, and can travel forward at eight to 10 miles per hour. It is more intelligent too, and can dodge obstacles, even moving targets.

Ph.D. candidate Justin Yim spent four years working on the bot, and it’s now sophisticated enough that he can take it for short walks around the university campus.

“Small robots are really great for a lot of things, like running around in places where larger robots or humans can’t fit,” Yim explained in a statement. “For example, in a disaster scenario, where people might be trapped under rubble, robots might be really useful at finding the people in a way that is not dangerous to rescuers and might even be faster than rescuers could have done unaided. We wanted Salto to not only be small, but also able to jump really high and really quickly so that it could navigate these difficult places.”

Robotics graduate student Justin Yim led the effort to program Salto with sophisticated control software that lets the robot master complex maneuvers. UC Berkeley photo by Stephen McNally

The bot was developed using motion capture technology to test its capabilities. The information about how the robot should move is calculated on a laptop and sent wirelessly via radio to a control board inside the robot. This means that Salto can hit a particular spot on a surface very accurately, which enables it to perform complex jump maneuvers.

“It’s worked better than any other of our other robots we’ve had so far,” Professor Ron Fearing of the UC Berkeley Department of Electrical Engineering and Computer Science said in a statement. “These fast accelerations let us move on surfaces where a conventional robot would just fall right off.”

The next challenge for the team is to get Salto to work on uneven or complex surfaces like grass or gravel. They are also considering adding an arm to the bot to let it interact more with the environment, such as grabbing onto branches and swinging upward.

Georgina Torbet
Georgina has been the space writer at Digital Trends space writer for six years, covering human space exploration, planetary…
Robot dogs were supposed to help humans. Now humans are being asked to babysit them
One startup offered $20 an hour to walk its robots, with some heavy lifting required when the dogs overheat
GO2 robot dog

San Francisco startup Dimensional advertised for people to walk robot dogs through city streets. The New York Times reported that the job could include carrying a roughly 33-pound robot back to the office if it overheated. A lift home is an awkward requirement for a labor-saving machine.

Keeping the hardware running is only one challenge. Anthropic's robotics tests found that language models mostly failed when controlling joints directly, but did better when directing trained controllers. Giving an AI agent a body doesn't teach it to handle ordinary household objects.

Read more
Forget smartwatches; this smart bed could soon help keep you safe from being a fall risk
Who needs another wearable when your mattress could help spot fall risks?
Bed

Soon, a mattress protector may be more than a bed cover. Australian medtech company Sleeptite and researchers at RMIT University have received a US patent for FlexiSense, a smart bedding technology that tracks body position and posture to help caregivers identify when older residents may be at risk of falling.

The technology is being developed for Sleeptite’s in-bed monitoring system, REMi, and was recently on display at The MedTech Conference in Boston. A peer-reviewed study in Flexible and Printed Electronics says the sensors can also be made at scale without sacrificing performance or durability.

Read more
Anthropic’s sensitive internal data was exposed, and AI helped make it happen
Even the company behind Claude isn't immune to AI-powered security threats, as researchers discovered some worrying vulnerabilities.
Claude Anthropic Featured

AI companies spend enormous amounts of time convincing us their technology is safe and secure. But what happens when those same companies become targets of increasingly sophisticated AI-powered attacks? Anthropic, the company behind Claude, is now facing that question. According to a Washington Post report, cybersecurity startup Cathedral discovered vulnerabilities that let its researchers access sensitive information in Anthropic's internal systems.

The researchers used AI agents to identify potential entry points in less than 12 hours. Interestingly, they didn't need to crack Claude itself. Instead, they exploited weaknesses in third-party business applications that Anthropic uses for everyday operations. Anthropic's security testing program disclosed the findings, which allows outside researchers to identify and report vulnerabilities.

Read more