2019 Dynamic locomotion synchronization of bipedal robot and human operator via bilateral feedb... 2
Вставка
- Опубліковано 17 вер 2020
- Dynamic locomotion synchronization of bipedal robot and human operator via bilateral feedback teleoperation
Video from MIT
Abstract
Despite remarkable progress in artificial intelligence, autonomous humanoid robots are still far from matching human-level manipulation and locomotion proficiency in real applications. Proficient robots would be ideal first responders to dangerous scenarios such as natural or man-made disasters. When handling these situations, robots must be capable of navigating highly unstructured terrain and dexterously interacting with objects designed for human workers. To create humanoid machines with human-level motor skills, in this work, we use whole-body teleoperation to leverage human control intelligence to command the locomotion of a bipedal robot. The challenge of this strategy lies in properly mapping human body motion to the machine while simultaneously informing the operator how closely the robot is reproducing the movement. Therefore, we propose a solution for this bilateral feedback policy to control a bipedal robot to take steps, jump, and walk in synchrony with a human operator. Such dynamic synchronization was achieved by (i) scaling the core components of human locomotion data to robot proportions in real time and (ii) applying feedback forces to the operator that are proportional to the relative velocity between human and robot. Human motion was sped up to match a faster robot, or drag was generated to synchronize the operator with a slower robot. Here, we focused on the frontal plane dynamics and stabilized the robot in the sagittal plane using an external gantry. These results represent a fundamental solution to seamlessly combine human innate motor control proficiency with the physical endurance and strength of humanoid robots.
robotics.sciencemag.org/conte...
copyright belongs to authors and publisher - Наука та технологія
All of yall in the comments need to chill. Just appreciate this work for what it is. There's no way you're going to be able to buy CAD, code, or hardware from MIT smh
most university work is open source, they all use the same or similar algorithms for IK, and you can buy the hardware and actuators for a few bucks, model the limbs and 3D print them, and put the code together in an IDE or system like ROS and anyone can make robots. It has nothing to do with buying anything or money its about knowledge and advancement. Anyone can put this together just like a university student can, you just have to know the same shit. So ya people are going to be curious about how he put this together
1) this is NC State not MIT 2) maybe you can’t but lots of people can
coz you loser XD
Self tuning fuzzy pid, ik, some neural training to pre bias, a solver network w/ consequentialist goap... yeah idk, this can be replicated. It's very awesome though and people clearly want to get into this :3
very robust, makes me want to learn what else you achieved.
Brilliant work! I subscribed after watching this video!
That's incredible. Having suffered serious back injuries years ago there are things that robot are doing I cannot....I am very impressed by its abilities.
You know.. far far too often we take things for granted.. every day, mundane things like, standing on two legs....walking...traversing steps... that is we take them for granted until we can no longer preform those functions with issues. Appreciate your abilities.... for non are guaranteed... nor will they all last indefinitely !!!
Really great work!
Definitely LIKE!
Looks awesome!
fantastic work.
Fantastic! Great!
so impressive!
Write the log test was impressive nice work
Brilliant.
fantastic .. i hope we can do the same Achievement in using hands .. legs took a lot of Time
Fantastic 👏
ah this is perfect.
Amazing
which control schemes did you use? It reminds me of the inverted pendulum.
Skateboarding robots will be a thing...
Will you build the top torso with arms?
Is there a gyroscope on the torso or a tensioned cable (didn't see one though)? Stepping sideways is fine, but how does it maintain foward/barwards stability without stepping forwards of backwards?
Teleoperated from a tube in the back
@@DantalionNl Ah! Hiding in plain sight :D Thank you for pointing it out.
@@alexandrepv ; I read in the description that it was sagitally stabilised by a gantry, but like you, I was damned if I could see it!
nice!!!!!
wow, it very nice, Can you share me what to do for make this project, thanks you
which actuator ?
Will it open source to people?
It almost seems like, with the camera angle, that you're trying to hide the metal rod that connects the robot to the desk behind it.
There is a cable, that’s it, but whatever makes you happy
@@jackhaley6332 You can see where there is a metal rod attached to the leg of the table in the background that extends to the robot. It's most visible at the 0:34 second mark.
@@Lardzor thats my bad, didnt see it initially, it is still a neat concept, just doesnt seem to be equipped with forward back stabilization in the video, thanks for pointing it out
mark my word's.. one day a A.I: took a body.. and she will rew-play all those videos... or torturing small robots.. she will end us XD
Throw a couple wheels on each foot and you've got Metal Gear Mk ii from Metal Gear Solid 4
WO AZ IGEN !!
Cool, how do I build/program one of these at a small scale for a fr hundred dollars, donde source code and models I can 3d print, other parts list.
he is able to stay in balance
When will the code for this stuff be available for purchase or otherwise.
This work is from MIT.
@@biomechatronicsandintellig3133 so essentially never
@@daddyplankton5855 You'll get to see it IRL exactly once, if you survive the initial bombardment.
@@daddyplankton5855 : It might be even longer than that.
Skynet is coming 😁
0:01 it can twerk
what is the grace if it has a tube in the back that holds it so that it does not fall forwards or backwards
yes min 0:35 and 0:41
I am not sure what the wording of your comment means, but the reason it is restrained this way is covered in the thorough description.
Please provide information about this robot
put this guy on a skateboard
Hanzhen harmonic drive gear , strain wave reducer. robot joint,
over 30 years experience
Do a kickflip!