Parkour Atlas
Boston Dynamics1 HN recommender · recommended in 1 month, 2018–2021
demo · general · 2018–2021 on HN
A short Boston Dynamics clip of the Atlas humanoid robot doing parkour, jumping over a log and leaping up steps using whole-body control and computer vision to locate markers.
HN cites it as a striking demonstration of how far humanoid robot agility had come.
Up next
3 mo on HNLow Latency Routing in Space
HotNets Anonymous
4 HN recommender · 2018–2021
4 mo on HNFeynman: What things really are like FUN TO IMAGINE 1
Christopher Sykes
4 HN recommender · 2011–2013
11 mo on HNWhat Happened to Horses Is Happening to Us
CGP Grey
11 HN recommender · 2014–2026
6 mo on HNZipline - How Rwanda Built A Drone Delivery Service
Real Engineering
9 HN recommender · 2019–2023
5 mo on HNAmazing Invention- This Drone Will Change Everything
Mark Rober
5 HN recommender · 2023–2024
4 mo on HN7 Minutes of Terror: The Challenges of Getting to Mars
NASA Jet Propulsion Laboratory
5 HN recommender · 2016–2021
4 mo on HNSlaughterbots
Stop Autonomous Weapons
4 HN recommender · 2018–2026
4 mo on HNDrone Delivery Was Supposed to be the Future. What Went Wrong?
Wendover Productions
4 HN recommender · 2022–2023
3 mo on HNGiant Robot Project, Part 1
Jaimie Stuff
3 HN recommender · 2011–2020
4 mo on HNSci-Fi Short Film “Slaughterbots” | DUST
DUST
5 HN recommender · 2021–2026
4 mo on HNThe Fastest Maze-Solving Competition On Earth
Veritasium
5 HN recommender · 2023–2024
3 mo on HN7 Minutes of Terror: Curiosity Rover's Risky Mars Landing | Video
VideoFromSpace
3 HN recommender · 2017–2021
4 comments from Hacker News
I still can't understand why Google sold Boston Dynamics, it baffles me. These guys have been pushing the boundaries of robotic motion for years. I see it as another example of Google's inability to focus on anything that is not core services. Their latest video of Atlas doing parkour is just insane [0]. You'd think that with Google's focus on AI they'd want to own and keep under a tight lid tech as impressive as BD's. [0] ▶ this video
It's interesting to see how this has progressed from their video two years ago [ ▶ this video ] and one year ago [ link ]
Bipedal walking and running on uneven terrain has been demonstrated years ago - Boston Dynamics has a lot of nice videos; e.g. here's a four year old video on walking link , here's two year old videos on running link and ▶ this video - perhaps it's debatable if it's "full human level" but it seems sufficient to me for most purposes. IMHO any improvements to that would be mostly for sensors (seeing what terrain is there even in poor conditions e.g. fog/rain/snow) and modelling environment (understanding which objects might break or slide if you step on them), not walking/running as such. The main problem for bipedal robots that makes them still impractical is the hardware expense (wheels are simpler and cheaper) and the power supply required, so for most use cases it's more efficient to use something other than a bipedal robot and there's limited business application and future revenue in scaling up research demos of bipedal walking to practicality, so most people who are working on walking algorithms are doing so in simulated virtual environments (where we have algorithms that can learn walking and running "from scratch" through experimentation) and not building very expensive hardware.
While full control is obviously better for some things, I wonder if overall this isn't better. It would reduce the training from having to adapt your brain to control a foreign body, to just issuing simpler commands, and might be safer too. Imagine using Atlas[1] as the exoskeleton, you just think about going forward and it navigates the terrain for you. [1] ▶ this video