Video Friday: Autonomous Drift
Video Friday is your weekly selection of awesome robotics videos, collected by your friends at IEEE Spectrum robotics. We'll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!):
Let us know if you have suggestions for next week, and enjoy today's videos.
In the name of safety, the Toyota Research Institute has been teaching cars to autonomously drift, the idea being that it could be a good way to maintain control of a vehicle in an emergency.
This reminds me of some autonomous-driving research from back in 2009 (!) when Audi teamed up with Stanford to explore autonomy at the limits of vehicle handling, including autonomous drifting. Here's a video from back then:
[ TRI ]
InfraredTags is a system for fabricating objects with embedded codes that are visible only to infrared cameras. These codes can be used for purposes such as metadata or interaction with devices through augmented reality.
This is cool, because it's like a cheat code for robots, potentially allowing computer vision and vision-related tasks by providing additional information on demand.
[ MIT ]
In this video we demonstrate fully autonomous multicontact locomotion for our humanoid robot LOLA. In contrast to our previous multicontact videos, where contact points for the feet and hands had to be specified manually by the user, this time all contacts are autonomously planned by the robot itself based on the perceived environment.
[ TUM ]
The CMU ballbot is controlling its entire body to maintain balance on top of its ball while carrying a heavy payload. The ballbot is actively compensating for the additional payload weight.
Doesn't this look exactly like the stereotype of a robot butler?
[ CMU ]
Of all the research I have seen with humans and drones in close proximity, this is the first where a shield has been used. I think it's a totally reasonable thing. I'm just wondering what happened for the researchers to decide it was necessary, you know?
[ Paper ]
Jimmy Fallon interacts with Boston Dynamics CEO Robert Playter's robot, which shows off a dance in collaboration with BTS.
[ Boston Dynamics ]
I love watching autonomous vehicles drive through urban areas in Asia, because it's such a totally different challenge than we see in the United States.
[ AutoX ]
Multirobot systems such as swarms of aerial robots are naturally suited to offer additional flexibility, resilience, and robustness in several tasks compared with that of a single robot by enabling cooperation among the agents. In this paper, we propose a general-purpose Graph Neural Network (GNN) with the main goal to increase, in multirobot perception tasks, single robots' inference perception accuracy as well as resilience to sensor failures and disturbances.
[ Paper ]
The ExynAeo autonomously explores a waterlogged stope in Alaska as a survey team monitors the robot's real-time progress while they stay warm and dry in their vehicle.
Stope: a combination of the words "subterranean" and "nope," referring to underground areas where you really don't want to go.
[ Exyn ]
The world's largest quadruped robot cluster performance!
I could see the noise of 200 of those things running all at once becoming a little bit scary. Especially if they were chasing you.
[ Unitree ]
SWIFTI works alongside an IRB1300 and a human worker in this demonstration assembling clocks. Designed to support intermittent collaboration between human and cobot, the safety laser scanner on SWIFTI creates safe zones for human workers to approach.
Those are some of the least useful clocks I've ever seen.
[ ABB ]
Soft, flexible fingers are exactly what you want in a meat-handling robot.
[ Soft Robotics ]
Northrop Grumman is getting serious about making sure the military can handle drones. And evil ground robots too, I guess?
[ Northrop Grumman ]
Chen Li from Johns Hopkins gives a talk on "Dynamic multilegged locomotion in complex terrain."
Studies of dynamic multilegged locomotion that began late last century have significantly advanced our understanding of how animals walk and run on relatively simple, flat, rigid surfaces and have led to the burgeoning of multilegged robots that do so dynamically, stably, and efficiently. Here I will review research over the last two decades building on these early insights to further understand dynamic multilegged locomotion in complex terrain.
[ JHU ]
Jessica Burgner-Kahrs from the University of Toronto gives an ICRA 2021 keynote: "I, Continuum Robot."
[ CRL ]