Shortly:
Spot was conspicuous by its absence from Marc Raibert’s keynote presentation at Automate 2022. But that didn’t stop the captive audience of robotics and automation enthusiasts from lulling the tricks the robot dog could pull off in a video clip.
For the most part, attendees would have been familiar with Spot’s YouTube videos posted over the years, where their dance antics and gaffes fuel viral PR campaigns. In a video celebrating Hyundai Motor Group’s acquisition of Boston Dynamics in 2021, Spot dances with K-BTS, the pop sensation. In another, the Rolling Stones join a pack of robot dogs that mimic Mick Jagger’s dance moves.
At first glance, it’s antithetical that the founder/president of Boston Dynamics is up to the task of sanctioning a dancing robot dog. Don’t make mistakes; Raibert, a roboticist with a PhD in robotic techniques that model biological behavior, is keenly aware of the mass-market appeal of choreographed videos, where the use of legs and arms is used as a form of expression and alters perceptions of what a robot can do. do.
Humans tend to anthropomorphize robots, whether they come from Boston Dynamics or anywhere else, and public misperceptions can be attributed in part to a lack of daily fun with robots. “The fact is, we’re not too far from cellular robotics acting on exact operations, but we’re making progress,” said Raibert, who spent decades creating robots with complex mobility, dexterity and perceptual intelligence before designing Spot to navigate challenging conditions, climb stairs or expand dangerous human hard work where classical automation has failed. .
What the dancing robot demonstrates, Raibert explained, is that cell manipulation is one of robots’ greatest opportunities. He said static handling is effective in some types of manufacturing, but remains limited.
“Spot doesn’t just roast your bread; he roasts your bread, delivers your margaritas, does a lot of things,” Raibert said.
READ MORE: Universal Robots’ cobot is redesigned for palletizing
Spot is a flexible four-legged cellular platform that weighs 60 pounds. It’s compact, rugged and portable, and designed to go where humans go. The robot dog is an omnidirectional and customizable platform that can be adapted to a variety of other responsibilities. ; it can be arranged with a variety of payloads for knowledge gathering, such as thermal imaging, LiDAR or 3D scanning for responsibilities such as creating virtual twins.
Spot has incorporated sensors on all 4 sides that navigate through undeniable functions. In commercial applications, the platform can be incorporated with a thermal camera, an arm for gluttonous items (some see it as a head), a Spot CAM (which comes with an optional PTZ camera), Spot CORE I/O (dedicated processing for subsequent calculations on the robot) and Spot GXP (force regulation and Ethernet port).
As it moves, responsibilities are no longer carried out in a constant location, allowing companies to think of service as robotics. you have a base that can be moved,” Raibert said.
Part of Boston Dynamics’ go-to-market technique relies on early adopters to help tame use cases across industries, from mining, power and utilities, manufacturing, oil and fuel to NASA’s Jet Propulsion Laboratory. According to Raibert, there are approximately 1000 points; one hundred universities doing experiments; and in a variety of commercial Spot services for responsibilities such as preventive maintenance inspections, remote sensing, thermal, radiation and fuel leak detection, and scanning services to expand virtual twins.
IMAGE GALLERY: A collection of robots
A human pilots the robot dog so it can gather information and combine a road map. Spot can be assigned the task of walking the trail independently. This procedure is applicable when Spot is pre-programmed for “tours and reads” in environments where repeatable and regime inspections are required or when you want to scan a large number of appliance parts. In facilities where existing appliances want to be inspected and adding constant sensors would be economical, Spot can read readings from analog indicators.
Raibert characterized Spot’s frame manipulations for the maximum component as athletic intelligence (how the robot perceives its environment and purposes through movement, balance, and maneuvering around obstacles) as opposed to cognitive intelligence, which comes to event planning. Boston Dynamics has an interest in leveraging any intelligence to bolster a fully functional hardware system.
Prior to the launch of Spot’s announcement in 2020, Boston Dynamics turned to Rightpoint, a Chicago-based IoT consultancy, to fill a gap in its user experience. “They employed a pretty bulky computer that I had to carry on my shoulder and walk with, and it just wasn’t a wonderful delight to drive or Spot,” said Ben Johnson, senior vice president of virtual products at Rightpoint.
His team collaborated with Boston Dynamics to design, design and expand site-line manipulation using an Android pill or joystick directly to where Spot was headed, as well as the use of remote, or non-line-of-sight manipulation. Use cameras to see what Spot sees in parallel with the expansion of Boston Dynamics’ (Scout) cloud-based software.
READ MORE: The energy-efficient leg works like a bird
From where Johnson is located, Spot is essentially a symbol processor and robot that moves from point A to point B. “A big component of the control formula is treatment,” he said. Among the insects his team needed to correct were how to turn sensor knowledge into an easy-to-use control experience, as well as Spot’s line of sight through reimagining Spot’s camera positioning and the symbol the user sees. “intersected,” Johnson explained.
Mobile gaming provided inspiration. ” When thinking about joystick controls, we also looked to think about a driving experience that included the ability to see the camera,” Johnson said. In addition to a direct view that would allow the user to drive the robot to see where the robot was headed, his team would figure out how to use symbol processing in the pill to turn the robot’s two symbol streams into a symbol that a human operator would expect to see. The existing version, a composition of several symbols, “a big hurdle to overcome,” Johnson said.
“But robotics is difficult; when engineers think of a connected product, a big component of the challenge is creating the physical manifestation of servo motors, motors and positioning systems,” Johnson said. a thin layer. Most of the time, you want to know the fundamental mechanics, and this can be difficult.
The robot dog has gone through many iterations since then, and Rightpoint continues to help build a safer, more physically powerful platform. “As a company, we have gravitated where we expect high-fidelity experiences,” Johnson said.
Spot made an appearance at IMTS 2022, where he presented an exhibition to the delight of the participants. During the presentation, Eric Foellmer, vice president of marketing for Boston Dynamics, shared a slide illustrating the 1900 New York City Easter Parade. The procession consisted of horse-drawn carriages, for one automobile. His next slide, a symbol taken only thirteen years later, showed the opposite: the parade consisted of a carriage cavalcade and a horse and a single horse carriage. “The auto industry has reshaped the entire parade in just over a decade,” Foellmer explained, hoping that the audience will perceive that the arrival of robots in our daily lives has also triggered a turning point.
In addition to Spot, Boston Dynamics is at the forefront with two other apps. Stretch is an advertising robot designed for boxes and boxes in warehouse, shipping and logistics applications. Atlas is an R robot
“Adding cognitive intelligence to the robot is actually going to be important, and I’m starting to work on that problem,” said IMTS’ Raibert. “This means that robots want to perceive their environment. Robots want to be aware of their own behavior. So if they spill boxes, they’ll have to come up with a plan to pick them up. I think one day they will be able to see another robot or user do a task and figure out how to do it. themselves. And they will also have to use reasoning and cognitive sense.
Raibert’s IMTS presentation in June concluded with a cliffhanger: “Stay tuned. . . »
Those who paid attention would have guessed imminent plans. In August, Hyundai and Boston Dynamics announced an initial investment of $400 million to create the Boston Dynamics AI Institute. Located in the Kendall Square network of studios in Cambridge, Massachusetts, the institute will be led through Raibert with the purpose of investing resources in 4 disciplines, namely cognitive AI, sports AI and biological curtain design, and ethics and politics.