Fragility Robotics

ExTwitter post from Agility Robotics includes a video of a bipedal robot performing a task of moving boxes in a simulated factory setup in what looks like a trade show. The robot crashes on itself at the end in what appears to be a failure of its legs.

So many things happening in this ExTwitter post, let’s unpack.

It’s quite interesting for a robotics company to put up a social media message of its own product failures, especially when that failure happens in a product demo setting at a trade fair. Knowing the price per square footage of these trade shows, a setup with a conveyor belt and box shelf is no small marketing budget. Failures in the lab are ok, and there’s a history of robodog video bloopers, but failures when you’re trying to convince a large crowd to buy your tech, maybe much less so.

So, marketing probably thought, Well, that’s a million dollar fuckup, so let’s change strategy and use this to our advantage. We’ll make it a viral social media event. And while we are at it, let’s make our own metrics: “99% success rate over 20 hours“.

That robot was not going to get back to work without serious repairs, so forget having it moving boxes for the rest of the day. Since we’re talking about human-looking robots replacing humans doing machine jobs, might as well expect robothings to do the work 24/7, as metric of success, not 20/6 or 20/4, in that particular context. So let’s rewrite that as “82.5% success rate over 24 hours” if robothing gets repaired in a few hours. “20.7% success rate over 4 days”, if you forgot to bring a set of “quick change limbs” to the show.

Lastly, I can’t stop looking at the crowd standing on the other side of the conveyor belt, witnessing the scene. The lack of response or interest in what just unfolded is palpable. No one seems surprised, amused or alarmed. Even what looks like members of the sales team from [Fr-]Agility Robotics barely turned around to see what was happening behind their back and then just ignored their flagship robotic product having a melt-down.

Staging robots in manufacturing settings is boring. Breaking a leg is no way to impress.

More than meets the vision sensor

Waymo, the robotaxi company from Alphabet/Google, broke the first law of Asimov.

Way more interesting is to read how the robocompany describes the incident:

“The cyclist was occluded by the truck and quickly followed behind it, crossing into the Waymo vehicle’s path. When they became unoccluded, our vehicle applied heavy braking but was not able to avoid the collision,” Waymo said.

https://boingboing.net/2024/02/07/waymo-autonomous-car-hit-bicyclist.html

Let me emphasize that: “the cyclist crossed into the Waymo vehicle’s path“. That’s such an engineering thing to say. It’s your 2 tons metal box on wheels that does not have a small moving vehicle hidden by a larger one in its computation vision model. Your software calculates a trajectory to pass behind that truck. Oops, there was a cyclist there. But it’s the cyclist who crosses your path? How convenient.