Thank you for the lecture. Can you briefly recap questions from the audience to the microphone?
@LeoHoulstonАй бұрын
this is awesome
@sucim3 ай бұрын
Jason Statham teaching robotics was not on my 2024 bingo card
@khkgkgkjgkjh6647
3 ай бұрын
lmao
@user-lh5nz3jf5vАй бұрын
49:35 Can someone teach me why a fully-actuated system can be equivalent to a double integrator (q''=u) ? This means the input u (torque) can directly equal to the second differential of output q (position)? Thanks!
@mishas1653
Ай бұрын
As I understood, If you have a fully-actuated system, that means you can get u from q'' = f1(q,q',t) + f2(q,q',t)u in form u = f2^-1(q,q',t)(q'' - f1(q,q',t)). And now, taking the last equation, you can "change" q'' to any control you want (f.e u*) and passing u to the first equation get q'' = u*, meaning you can fully control accelerations. You can find more info in lecture notes (underactuated.csail.mit.edu/intro.html#section3)
@shankhasubhrasarkar1673 ай бұрын
Sir how can I access the python codes you used in the lecture can you provide the link. can't find it
@forheuristiclifeksh78362 ай бұрын
4:01
@maksym16403 ай бұрын
Where is lecture 2? Can you upload?
@rtedrake
3 ай бұрын
added to the playlist. (we had a covid day and then a snow day, so had to use prerecorded videos for those lecture)
@maksym1640
3 ай бұрын
@@rtedrake ok, thanks! You are awesome!
@junqingqiao80072 ай бұрын
Thanks a lot. It's a very good course but there is room for improvement for the camera.
Пікірлер: 14
Thank you very much! Greetings from Germany.
Thank you for the lecture. Can you briefly recap questions from the audience to the microphone?
this is awesome
Jason Statham teaching robotics was not on my 2024 bingo card
@khkgkgkjgkjh6647
3 ай бұрын
lmao
49:35 Can someone teach me why a fully-actuated system can be equivalent to a double integrator (q''=u) ? This means the input u (torque) can directly equal to the second differential of output q (position)? Thanks!
@mishas1653
Ай бұрын
As I understood, If you have a fully-actuated system, that means you can get u from q'' = f1(q,q',t) + f2(q,q',t)u in form u = f2^-1(q,q',t)(q'' - f1(q,q',t)). And now, taking the last equation, you can "change" q'' to any control you want (f.e u*) and passing u to the first equation get q'' = u*, meaning you can fully control accelerations. You can find more info in lecture notes (underactuated.csail.mit.edu/intro.html#section3)
Sir how can I access the python codes you used in the lecture can you provide the link. can't find it
4:01
Where is lecture 2? Can you upload?
@rtedrake
3 ай бұрын
added to the playlist. (we had a covid day and then a snow day, so had to use prerecorded videos for those lecture)
@maksym1640
3 ай бұрын
@@rtedrake ok, thanks! You are awesome!
Thanks a lot. It's a very good course but there is room for improvement for the camera.
@junqingqiao8007
2 ай бұрын
I can never see the whole black board.