Robot Videos: Baseball Bots, UBTECH Walker, and More

Video Friday this is your weekly selection of amazing robotics videos collected by your friends on IEEE spectrum robotics. We're also posting a weekly calendar of upcoming robotics events for the next few months. Please send us your events to turn on.

SOSV Robotics match: December 1–5, 2025, ONLINE

IKRA 2026: 1–5 June 2026, VIENNA

Enjoy today's videos!

Researchers from the RAI Institute have created a low-impedance platform for studying dynamic robot manipulation. In this demonstration, the robots play a game of catch and engage in batting drills, both with each other and with experienced humans. The robots are capable of reaching speeds of 70 miles per hour. [112 kph]approaching the speed of a strong high school pitcher. The robots can catch the ball and shoot at short distances (23 feet). [7 m]), requiring quick reactions to catch balls thrown at speeds up to 41 mph. [66kph] and hit balls flying at speeds up to 30 mph [48kph].

It's a nice touch with the custom “RAI”. baseball gloves, but what I really want to know is how long a pair of robots can keep themselves entertained.

[ RAI Institute ]

This week's top acronym is GIRAF: Greatly Increased Reach. AnyMAL Function. And if this hand looks like magic, that's because it is, although by carefully pausing the video you can see it in action.

[ Stanford BDML ]

DARPA completed the second year of study DARPA Triage Challenge on October 4, awarding top scores to DART and MSAI in the systems and data competitions, respectively. The three-year competition aims to revolutionize medical triage in mass casualty situations where medical resources are limited.

[ DARPA ]

We propose a robot-independent reward function that balances achieving the desired end pose with minimizing impact and protecting critical parts of the robot during reinforcement learning. To make the policy robust to a wide range of initial fall conditions and allow the specification of an arbitrary and unseen final pose during inference, we introduce a simulation-based strategy for sampling initial and final poses. Through simulated and real-life experiments, our work shows that even bipedal robots can perform controlled soft falls.

[ Moritz Baecher ]

Oh look, still humanoid acrobatics.

My prediction: Once the humanoid companies run out of mock-up dance moves, we'll start seeing some fancy stuff that takes advantage of the degrees of freedom that robots have that humans don't. Guys, you heard it here first.

[ MagicLab ]

I challenge the next company that shoots a video with the lights off to just cut it to a completely black screen with a little “Success” meter in the corner that keeps going up and up.

[ Brightpick ]

Thank you Gilmarie!

The terrain is cool and all, but can we talk about the trailer instead?

[ LimX Dynamics ]

Probably very picky German birds receive special birdhouses made by robots with extremely high precision.

[ TUM ]

All these UBTECH It turns out that the Walker S2 robots were not fake.

[ UBTECH ]

It's more automation than what we actually call robotics now, but I could still watch it all day.

[ Motoman ]

Brad Porter (Cobot) and Alfred Lin (Sequoia Capital) discuss the future of robotics, artificial intelligence and automation on Human[X] The conference was moderated by CNBC's Kate Rooney. They explore why collaborative robots How AI is transforming physical systems, the role of humanoids, changes in the job market, and investment trends are now accelerating to shape the next decade of robotics.

[ Cobot ]

Humanoid robots have long captured our imagination. Interest has skyrocketed with the realization that robots are getting closer to performing a wide range of labor-intensive tasks. In this discussion, we reflect on what we have learned from observing the factory floor and why we have come to believe that the pursuit of generalization in manipulation—both equipment and behavior—is not just interesting, but necessary. We'll discuss the AI research topics we're studying at Boston Dynamics to move this mission forward and identify opportunities in which our field must collectively invest more to transform the humanoid vision and reimagining of manufacturing into a practical, economically viable product.