This is just fucking amazing. Decades since people have been looking for this and you guys did it. Simple presentation but still draws your attention.Definitely a shareable video.
1. IMU + laser scanner->sensor fusion 2. states estimation->extended kalman filter or unscented kalman filter 3. trajectory for vehicle to follow-> some kind of predictive control 4. to over come the computational difficulty they use approximate dynamics model-> they are not using first principle model to do control. So the control is most probably some kind of reinforcement learning or approximate dynamic programming.
I'd love to see this tech in replacement of classic rollercoasters. Imagine sitting in a computer controlled plane that flies crazy fast through the forest, avoiding trees by tenth of an inch! What a thrill would that be!
Fermat's last Theorem is the transformer bumblebee Robot. Original equation: z^n=x^n+y^n. Mean: z^(n-3)*z^3=x^(n-3)*x^3+y^(n-3)*y^3. Using the formula z^3=[z(z+1)/2]^2 - [z(z-1)/2]^2 to convert z^3 become the exponent 2. Then using the formula [z(z+1)/2]^2=1^3+2^3+........+z^3 to convert the exponent 2 become the exponent 3. Repeated several times with the same method. The transformer bumblebee Robot was created according your own structure. Simplest format about Transformer Bumblebee Robot. Using two formulas: z^3=[z(z+1)/2]^2 - [z(z-1)/2]^2 And define x
5. There are two trajectories in the video. The blue is planned trajectory (feed-forward part). The red is the real trajectory. Along the time, the vehicle is not deviating from the desired trajectory too far, so there must be some kind of nonlinear feedback (I prefer).
you bet. bootstrap like crazy. what he said in the video was an extremely high level description of something much more finicky, mostly the state estimation. marrying the LIDAR with accelerometers and gyroscopes sounds like a huge pain, however a system like this would be a huge boon to something like APM, the arduino pilot open source project
i've been meaning to look at ROS, didn't realize they had features that varied already. that's only half the story though, the other half is the state estimation using that simplified aerodynamics model. although that might be possible using a rudimentary physics engine for games
There isn't a human R/C pilot alive that could make that parking garage run at those speeds. Too many human factors involved for it to ever happen. Pretty sick stuff MIT!
can the problem for dynamic navigation be fix by adding some kind of motor that gives the back tail more pivoting motion, or would it make the plan loose balance.
I'm very interested in this type of autonomous flight for my quadcopter. Which sensor used? Do I need sensors costing hundreds dollar? Is it opened-source? Thanks
This is only possible with todays fast processors, light weight lasers, and advanced batteries, Only more proof that self driving cars will be in the future.
Umm... You couldn't have made it look like a carrier pigeon? Now how can we use this in a hack? Greetings to all MacGregorites and Class of '75 members!
At MIT students do engineering to make new inventions and discoveries...But in India students do engineering because- 1. their parents want them to get an engineering degree 2. they didn't get admission anywhere else Pity on them...
This is just fucking amazing. Decades since people have been looking for this and you guys did it. Simple presentation but still draws your attention.Definitely a shareable video.
1. IMU + laser scanner->sensor fusion
2. states estimation->extended kalman filter or unscented kalman filter
3. trajectory for vehicle to follow-> some kind of predictive control
4. to over come the computational difficulty they use approximate dynamics model-> they are not using first principle model to do control. So the control is most probably some kind of reinforcement learning or approximate dynamic programming.
This is one of the coolest things I've seen from MITnews, I'm going to look into building one of these, but with some modifications to it!
"Go that way, really fast. If something gets in your way, turn."
I'd love to see this tech in replacement of classic rollercoasters. Imagine sitting in a computer controlled plane that flies crazy fast through the forest, avoiding trees by tenth of an inch!
What a thrill would that be!
Fermat's last Theorem is the transformer bumblebee Robot.
Original equation:
z^n=x^n+y^n.
Mean:
z^(n-3)*z^3=x^(n-3)*x^3+y^(n-3)*y^3.
Using the formula z^3=[z(z+1)/2]^2 - [z(z-1)/2]^2 to convert z^3 become the exponent 2. Then using the formula [z(z+1)/2]^2=1^3+2^3+........+z^3 to convert the exponent 2 become the exponent 3.
Repeated several times with the same method.
The transformer bumblebee Robot was created according your own structure.
Simplest format about Transformer Bumblebee Robot.
Using two formulas:
z^3=[z(z+1)/2]^2 - [z(z-1)/2]^2
And define x
awesome - this is mind boggling ...... this is the future of autopilot on commercial aircraft
5. There are two trajectories in the video. The blue is planned trajectory (feed-forward part). The red is the real trajectory. Along the time, the vehicle is not deviating from the desired trajectory too far, so there must be some kind of nonlinear feedback (I prefer).
Massive props. Wonderful research model, and awesome execution. And the video very professionally presents the results. Nice package. Sharing this.
The parking garage flights are mindblowing... i'd love to see the source code on that.
you bet. bootstrap like crazy.
what he said in the video was an extremely high level description of something much more finicky, mostly the state estimation. marrying the LIDAR with accelerometers and gyroscopes sounds like a huge pain, however a system like this would be a huge boon to something like APM, the arduino pilot open source project
I think the camera is just for observation, the range finder and the inertial measurement unit are gathering all the data
This is incredible.
It would take practice to do that with an onboard camera and a remote human controller
i've been meaning to look at ROS, didn't realize they had features that varied already. that's only half the story though, the other half is the state estimation using that simplified aerodynamics model. although that might be possible using a rudimentary physics engine for games
That's amazing. I wonder if DARPA employed your team.
They went on to create the worlds best Self Flying drone. Checkout Skydio.
Awesome work!
Is it a custom onboard camera you used, it looks quite good namely + it has to be lightweigt? Right?
Amazing! A breakthrough to the control system of autonomous aerial vehicle.
There isn't a human R/C pilot alive that could make that parking garage run at those speeds. Too many human factors involved for it to ever happen. Pretty sick stuff MIT!
I really wish you would put credits on these films.
Awesumm....how do you combine camera info with the laser unit to know about its localization wrt to the environment????
Designing, building and supervising those machines.
This is great. Are there any factors for wind or otherwise in the algorithm?
I'll get really excited about this when they show that it is able to correct its path to avoid moving obstacles.
Very cool - *_Good work guys!_*
Amazing work 👍. Can this platform and setup choose it's own route within varied course that will accommodate it's passage?
yeah just set the waypoints
You mean you want to cut and paste the code and see if you can be on par with MIT engineers.. The guy pretty much told you how to do in the video.
gratz to the skydio guys for this piece of history :)
what the heck are you doing here
@@NicholasRehm congratulating my students on their accomplishments at MIT
Great. MIT is just great.
Especially a extra thrill if it is done with Microsoft Windows !
that sir is brilliant
Great work! amazing.
innovation at it's finest
You should be able to sell this system.
Very Impressive Flight Platform, I wonder if takeoff and is landing fully autonomous as well?
i have mastered traveling forward in time at a nearly steady rate.
there was no traveling forward in time in terminator
what else to expect from MIT :)
well done - that is impressive
nice innovation to prevent terrorists from attacking other tall buildings with airplanes
can the problem for dynamic navigation be fix by adding some kind of motor that gives the back tail more pivoting motion, or would it make the plan loose balance.
it would be aerodynamically wrong
These planes MUST be on the next mission to Mars.
Great work!
This is totally a tunnel mission from Ace Combat.
well, same can be said for rollercoasters too.
How can someone dislike this ?!
I'm very interested in this type of autonomous flight for my quadcopter. Which sensor used? Do I need sensors costing hundreds dollar? Is it opened-source? Thanks
google it.
My Investment Trading System Works very similar to this.... but i designed my trading system a few years back..
PLEASE RELEASE THE CODE OPEN SOURCE! I WANT TO EXPAND ON THIS SO BADLY
Yeah I'd love to have my hands on the code's
to a grad student who has dabbled in robotics, this is truly impressive.
Can you make it dodge a bullet fired at it from a distance of 100 yards?
this would be cool to buy
I just hope MIT won't strap a bomb and let it fly away
Possible to share the source code?
lol they would never
Bloody marvelous
Given how thin the Martian atmosphere is, that might be tricky.
quick, before Skynet arises...
Forgotten but actual
This is pretty amazing actually.
What the closest plane like this can buy on the internet or what plane is this
The B1 had much of this technology, but in the 70s
Why thank you :)
This is only possible with todays fast processors, light weight lasers, and advanced batteries, Only more proof that self driving cars will be in the future.
Oehcs Inc. there’s self driving cars
I for one welcome our new robot overlords
So this is how the use the gym at MIT...
That is pretty amazing.
well done!
Impressive
is John Connor already born ?
This seems like the beginning of the little terminator airplane killing machines!
So very cool!
This will render courier services dead.
So, how it land on the floor?
Marrying LIDAR, accelerometers and gyroscopes is already implemented on ROS. Look for laser_scan_matcher
What will be left for humans to do?
awsome guys!
Very good!
Open Steer?
Awesome!
Still space flight
not rc anymore. its a super fancy free flight/paper airplane!
well done.
too bad for humanity, John Connor isn't born yet.
Skynet is watching you 1:30
Real nice!!!
humans need the most supervision. when left on their own, humans have put nuclear bombs on devices such as these
So cool
thumbs up if you guys want them to post the flight video
Flies better than I do! Then again, I'm not even saying how many planes I've crashed. :-P
thats been around for decades.....flying at supersonic speeds!
Umm...
You couldn't have made it look like a carrier pigeon?
Now how can we use this in a hack?
Greetings to all MacGregorites and Class of '75 members!
... Bruce Wayne is an MIT grad.
figure out time travel and we'll be able to watch Terminator in real 3-D.
Use flight control what ia this?
Awesome
There are already robots that feed the dogs.
At MIT students do engineering to make new inventions and discoveries...But in India students do engineering because-
1. their parents want them to get an engineering degree
2. they didn't get admission anywhere else
Pity on them...
...and then SkyNet ! hehe... great !
滿酷的!
Now the NSA drones can fly into buildings.
I’d love to see the crash videos.😉😁