Am going to assume that's what meant by simulation is 2-d. And am going to assume that the challenge was to move from 2-d to 3-d. If so, I would have appreciated a little insight into how this was done. There are geometries that are used for robot movement, such at cat(0) complexes, that you can analytically perform robotic movement in 2-d space but you can't in 3-d (even with sampling methods). So, just a few math equations to explain what's actually happening would have been appreciated.

😂 we've come full circle

I'm an inventor that's started to work on industrial robotics. IMO, the problem with the bin packing robot is that you haven't deconstructed the problem to take advantage of a robot's strength. For instance, one advantage could be a very long arm. So, rather than place the items into the bin, as a human would, the items are placed from the top of the bin. That is, they're placed in order from bottom to top. This can only be possible if a robot has a large arm. There are other advantages such as the use of gravity to filter and push the items together, that only a robot could do that a human can't. In any case, thanks for the presentation.

Interesting topic definitely worth watching thanks for sharing.

well, this is not earth, but a world of posers, should be called poserion

Fascinating. This could be the future of machines.

Where's the MIT robot? This is all sh*t. The Big Lie of Academia is that all they do is write and accept each others' papers while industry has software that talks to people and answers any question that you ask and writes software to do whatever you ask, and robots that walk, run, jump and play sports like ping pong and soccer. MIT is supposed to be the center of AI and Computer Science but they don't have any of that. and right now academia is still having AI conferences and presenting papers instead of results. Ha Ha HA what utter nonsense and BS!!! They're also BS in set theory and mathematical proofs. They says that ZF set theory can prove anything that can be proven, but they have no examples of even common mathematical proofs. When you ask them, they say their proof is so long they can't give it and won't even summarize it or give the beginning and end. They say they have automatic theorem provers but there are no KZfaq videos with any examples and the manuals don't give any. They are all fraud and spending United States government research money on fraud. They should be in jail for cheating the government out of money.

Hey MIT robotics can you please put some courses on robotics which are taught on MIT. This is a request from robotic students in UK

I imagine a robot without motors, it needs to be a microbot that can hover just because of air currents like the dandelion seed . But it needs to be intelligence , I mean that it can move some arms in order to go forward backward left or right . Could be cool idea

I dont understand anything prof

"All you need" at the end of your paper's title is all you need.

순조로운 발표 좋앗다

That's the kind of complexity required to overcome having only 9 fingers.

amazing!! love it. Thank you so much MIT Robotics and Professor Russ for such a great talk.

I guess most research in robotics is on the algorithms . right ? Its like thinking millions applications of robotics arms and then do a paper . The first country that can have the best humanoid robot will win the race ...

thanks for sharing

❤ Great

Loved it! Great talk.

I have searched everywhere for "switching system factor graph" and could not find any reference 11:38

Very visual, very fun! ❤

Enfin des gens qui travaillent... je me demandais si cela allait arriver... Merci.

👍

Well delivered, Dr. Koltun.

Yoooooo Kev, i wondered all these years where went off to brada!

Captivating presentation. MuJoCo MPC is a fantastic project. 56:34 MuJoCo 3 is GPU parallelizable!

Loved the talk❤

Amazing! Keep going guys!

Cool!

what math is this? I want to learn it so I can understand what's happening here

we went to the same college. happy to see her make it

Wow!

I really like Yuval's work and his presentation which inspired me a lot, especially the metaphor of surfing on the cost function.

Good

sim-real-sim-real.repeat

I dropped out to work on motion cueing and specifically because of RL via. simulation, it's been approachable. What used to be state of the art (which in hindsight, just 5 years ago seems so 👣i.e. no sensors or feedback in the control-loop), to be influenced so easily is another testament to the quiet revolution he's talking about. The reason is the resources it took: (1 person) + Internet + LLMs eventually. Excuses are, not to use AI 🤷‍♂

Great work guys.

Some perspectives I never considered during my research...Great talk!

amazing work! thanks for sharing :)

5.8mil 78 likes. This is amazing an freaking. Military maby are looking for this with attention. Imagine bigger flying objects moving like that autonomously every where.

Great talk. Particularly given the last question/answer, and the work being done in the commercial space building robot simulator technology (NVIDIA, AWS, others).

too bad the videos in the presentation doesn't work.

astounding talk.

Lmoed at the neural network question. Shows how under-appreciated classical control is in a world of "just throw neural net at it"

PEMROGRAMAN C DAN C++

HTML PHP SQL

ROBOT BEAM

3 DERAJAT KEBEBASAN LENGAN ROBOT FOOD & LIFE ASDL 0^0 10 1

You guys arent the only ones working on this kzfaq.info/get/bejne/grePZrurtrmnYY0.html

@Yuval Tassa Hello Dr. Yuval from DeepMind Thank you for the presentation. A small criticism: Python in MuJoCO ==== USB Type C in APPLE I am interested to know more: 1- more comparison between “MuJoCo Versus Isaac” that one of the students asked Dr. Yuval at the end of the presentation. 2- Please tell me more about the limitations of MJPC that you mentioned such as: a. Learned policies b. Learned Value function c. Hierarchical control 3- Regarding my Ph.D project, I am very interested in knowing about MujoCo’s abilities in the projects of robot interaction with the environment and measuring data via virtual sensors that are provided by MuJoCo. Dear viewer, Please, If you have any information or a source for even one of them, please leave a message for me. Thanks in advance

Very nice presentation