Thank you for shortening down to 5 minutes almost the entire course of MS Optoelectronic i took this year. Amazing how concepts such as a MZ modulator and integrated photonics can be explained in such an easy way.
Ted Ed is phenomenal. I’m constantly impressed and thankful for your work in making knowledge so accessible and digestible. I’ve learned so much from watching this channel, and your videos have prompted me to delve deeper into different topics I don’t think I ever would’ve gotten the opportunity to. Thanks for all your hard work❤️
Great job, you have done a great deal of emphasis on the sensor part of a self-driving car. However, you did not mention at all the decision-making mechanism, which is in my opinion, the real challenge of self-driving, I think that video on deep learning would be highly appreciated.
Ted-Ed describes things with so much detail in a way that no other science youtube channel does. Thank you for that. You are an amazing teacher and educator
Yeah except...visual resolution of lidar (like color) is abysmal-especially for now. Like Elon said, the roads are made for eyes: rely on vision first, supplement with lidar and ultrasonics. I do have to say, mega respect to Ted-Ed, and it’s cool that they talking about the actual electronics of it.
@@jerry3790 LIDAR is inherently better. With camera you're making the machine operate as a human would. That's like designing a car with legs. Sure, it would be cool. I love AT-ATs. But wheels are just so much more practical on a machine. Let a machine be a machine and not emulate nature when there is a better way. Eventually I expect to use a combination of LIDAR with camera backup for extra safety. An IR camera that sees through fog and in pitch darkness that is.
LIDAR is better for depth resolution than stereoscopic photogrammetry. Sensor fusion wouldn't be and issue since the sensors and camera are static relative to each other
As someone who doesn't drive for health reasons and wants a fully autonomous vehicle in the future, this was really informative on how self-driving cars actually work! Thanks Ted-Ed :)
@@jasperizak123 That's not true at all. They implemented it into the Dragon Cargo to dock with the ISS. He just said it's useless in a model driven by cameras and neural networks.
어릴 때는 공상과학 영화에서만 나오는 내용이라고 생각했지만, 지금 실제로 시험 운행 중에 있는 자율주행 자동차들을 보면 정말 기술이 많이 발전했다는 것을 느끼고 있습니다. 저는 자율주행 자동차들이 내장되어 있는 카메라로 주변 환경을 본다고 생각했는데, 이 영상을 보니 오히려 박쥐가 초음파를 이용해 길을 찾는 방법과 더 유사한 것 같습니다. 유익한 영상 감사합니다!
I studied physics then dropped out of my full ride with my kid years ago. I did develop an ego because of it because I thought I knew everything. Finding years’ old videos of Ted that I hadn’t realized I missed completely, I realize I know nothing.
Old country roads, take me home To the place, where I belong In West Virgunia Or should I say: Old self driving car, take me home To the place where I belong In west virginia
1:53 I would suggest thickening the beam, and using a color other than green to represent light bouncing back. The green beam is harder to see because it is against a blue background, and it practically disappears if viewed on the lowest resolution, 144p.
LiDAR is not the only solution to self-driving. It has a lot of advantages, but also just as many disadvantages. The sensors are very expensive. Resolving the detailed features of objects in front of it is not very precise from off-the-shelf low-res hardware which a lot of these cars are equipped with. Plus, the current transportation infrastructure is created for visual perception, because it's driven by people who use visual perception to navigate vehicles; there is something to be said for a solution that uses visual data to navigate over LiDAR.
“Imagination is more important than knowledge. For knowledge is limited to all we now know and understand, while imagination embraces the entire world, and all there ever will be to know and understand.” - Einstein.
Nice presentation, but the real question is what happens when multiple lidars interfere with one another. The chances are low because of the time scales, but if there's a sufficient number of self-driving cars, there's got to be lidar pulse sync at some point between 2 or more cars.
My English is bad , so... hope you guys can understand my meaning , I have a question , if one super short laser into the another LIDAR system will have problem?
Isn't that mostly a cost question though? If you'd build self driving car with all the best tools, no matter the cost, LiDAR would probably be used. Sense Tesla also want their cars to be affordable, they're using technologies that are cheaper to implement today, at least that's what I've heard. Once LiDAR becomes more affordable and possible at size scales described in this video, Tesla will probably switch to use LiDAR, but will probably still use their software that they're currently pushing to develop and refine.
Collecting cloud of several million precisely positioned 3D points is kind of easy. Figuring out the actual object from that cloud, now THAT is a challenge!
I wish it's applicable to blind people for navigation especially on detecting the height of obstacles... So they know what next step is... Well, there's a limit to everything
I'm actually a team leader for a self driving racing car competition in the UK 'FS-AI'. This is a top quality video, however, LiDAR hasn't quite been explained perfectly becuase even the best LiDAR radars only see a very limited slice of the world. Whilst we have LiDAR it's kinda useless, the limited slice that the LiDAR can see makes it very difficult for object recognition as there simply isn't enough data coming from the LiDAR to make any reasonable assumption of the object. We use the Robosense V16 Lidar, which, despite it's high price tag is simply not useable. To get better resolutuion you have to pay vast amounts of money even though they fundemantly have the same issue. What we use instead is stereo vision (Zedd Camera) as you're able to make better object detection useing each image from the camera and using machine vision and the principles of the Fundemental Matrix, we're able to rebuild a higher quaility map of the world.
But zedd cam that doesn't imtegrated by some kind of 3d lidar, doesn't give accurate results especially under extreme weather conditions. How did you manage to do that?
@@selene8734 Firstly, neither does LiDAR in those situations. We solved it using multiple cameras, and a model train on sensor fusion for the various weather conditions
Brilliant video. But i believe detection is not the main problem currently, but the main problem is to identify and track objects reliably. And also the ethical dilemma (which is already explained in another ted-ed video).
When the laser bounces off of things, doesn't it travel away from the laser? How does the car receive the laser? Does it really travel straight back, even when the surface isn't perpendicular?
"The robot cars cannot accurately predict human behavior, and the real problem comes in the interaction between humans and the robot vehicles,”. This was in response to the woman that was killed by a self driving car in AZ.
How it would detect fast moving obstacles like an animal running across the road ,we humans will switch to right if it's running from right to left ,but what will the automated car do ? Will it automatically stop until it crosses?
Being the best in pioneering self driving currently, Elon Musk doesnt support Lidar in cars. He says its worthless and that cameras are far better. Elon also said that he isn’t against Lidar, as SpaceX use them for space station docking, but it just doesnt make sense in cars.
The problem with Lidar that Musk describes is that they are (currently) expensive and large. However, in the near future devices such as those shown in the video will become available, probably at a fraction of the current cost of lidar systems. At that point, any autonomous vehicles would probably choose lidar's superior accuracy and depth perception over the advantages offered by cameras.
@@JWFitz just "seeing" the object is HALF the battle, then you need to identify what it is... think a plastic bag blowing across a highway. Telsa is using cameras (with color) and big data to solve the second half of the problem. They have billions of miles worth of data identifying animals, people, bikes, etc... where the companies using Lidar only have a depth map.
This narration and animation can make anything interesting
Next “the science of taking a poop” or “how to make an average sandwich”
IKR. Like I wish we could like watch videos in school like this! If we did I would be able to learn a lot better. I would be making 100 A's.
I love the narrator's voice too
Ted-Ed always coming with the answers to question I ask in my head but never say out loud
Yoi banget
MATH Genius that was beautifully said
The funny thing is that Tesla don't use Lidar due to its high cost and low reliability
You really pretty Dani
Thank you for shortening down to 5 minutes almost the entire course of MS Optoelectronic i took this year. Amazing how concepts such as a MZ modulator and integrated photonics can be explained in such an easy way.
Ted Ed is phenomenal. I’m constantly impressed and thankful for your work in making knowledge so accessible and digestible. I’ve learned so much from watching this channel, and your videos have prompted me to delve deeper into different topics I don’t think I ever would’ve gotten the opportunity to. Thanks for all your hard work❤️
I love the tone of this Narrator the most, I could listen to him all day
*A better question*
How does a Tow Truck drive backwards at 80 km/h and becomes a secret agent
The best movie.
I understood that reference
Great job, you have done a great deal of emphasis on the sensor part of a self-driving car. However, you did not mention at all the decision-making mechanism, which is in my opinion, the real challenge of self-driving, I think that video on deep learning would be highly appreciated.
Self-driving car: "sees" a pedestrian with a club
Self-driving car: "Why do i hear boss music?"
self-driving car- *switches to manual
gamer in car-*it's showtime
The club: Alright! Here we go again
@Aaron Yu **Jazz music stops**
purification in progress
Ted-Ed describes things with so much detail in a way that no other science youtube channel does. Thank you for that. You are an amazing teacher and educator
Yeah except...visual resolution of lidar (like color) is abysmal-especially for now. Like Elon said, the roads are made for eyes: rely on vision first, supplement with lidar and ultrasonics. I do have to say, mega respect to Ted-Ed, and it’s cool that they talking about the actual electronics of it.
Then maybe pursue more scientific qualifications and publish a video yourself?
Lidar doesn't have anything to do with color. It just generates a point cloud that is converted to a 3d image
Real N He does have a point though, while Lidar as a technology is cool, camera technology is getting to a point where it’s no longer necessary.
@@jerry3790 LIDAR is inherently better. With camera you're making the machine operate as a human would. That's like designing a car with legs.
Sure, it would be cool. I love AT-ATs. But wheels are just so much more practical on a machine. Let a machine be a machine and not emulate nature when there is a better way. Eventually I expect to use a combination of LIDAR with camera backup for extra safety. An IR camera that sees through fog and in pitch darkness that is.
LIDAR is better for depth resolution than stereoscopic photogrammetry. Sensor fusion wouldn't be and issue since the sensors and camera are static relative to each other
Havent thought of this before! The scientists and engineers who invented LIDAR are really innovative. Can LIDARs see color too?
wonderful animation as always!
As someone who doesn't drive for health reasons and wants a fully autonomous vehicle in the future, this was really informative on how self-driving cars actually work! Thanks Ted-Ed :)
Ted-Ed always answers my everyday questions. What a splendid RUclips channel!
I have seen CARS movie enough times to believe cars have eyes in their windows and the ventilation as their mouth
Makarov -Father of fairies top ten smartest people in the world
👌🏻👏 congrats 🎉
Yes and I can feel whether the car is angry or happy.
But Tesla doesn’t use Lidar, so we need another video
Yup. Elon does not like LiDAR
He is using golems
They need another title like "HOW LIDAR WORKS". Self driving is not exclusive to cars using LIDAR.
@@jasperizak123 That's not true at all. They implemented it into the Dragon Cargo to dock with the ISS. He just said it's useless in a model driven by cameras and neural networks.
how do you know 🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓🤓
무인자동차가보는 방법을 구체적으로 배워보는 좋은 시간이 되었습니다. 무인자동차의 과학적원리가 참 대단해 보였습니다. 저의 궁금증이 하나 더 해결되었습니다. 좋은 영상 감사합니다.
Some of the best narration on RUclips
어릴 때는 공상과학 영화에서만 나오는 내용이라고 생각했지만, 지금 실제로 시험 운행 중에 있는 자율주행 자동차들을 보면 정말 기술이 많이 발전했다는 것을 느끼고 있습니다. 저는 자율주행 자동차들이 내장되어 있는 카메라로 주변 환경을 본다고 생각했는데, 이 영상을 보니 오히려 박쥐가 초음파를 이용해 길을 찾는 방법과 더 유사한 것 같습니다. 유익한 영상 감사합니다!
This channel deserves more likes and views.
4:58
Person who manufactured the car: "Am I a joke to you?"
Machines do most of that work as well to be honest.
@@ghostderazgriz But their work cannot be counted as labour.
@@stormysamreen7062 exactly?
A vey satisfying video.
Another gem from TED after long time 💎❤️
TED-Ed awsering question, that i may never use in real life.
This why they're the best!
What a great and simple explanation of solid state lidar, I am impressed!
Forever answering questions I never asked but am fascinated by the answer
I think we can all agree that, this person is our favorite ted-ed narrator.
I highly recommend Lex Friedmans series on autonomous vehicles. Great introduction into the self driving space.
2:27 "Use something else"
Literally
Oh I just noticed that it's labeled that way xD that's hilarious.
I studied physics then dropped out of my full ride with my kid years ago. I did develop an ego because of it because I thought I knew everything. Finding years’ old videos of Ted that I hadn’t realized I missed completely, I realize I know nothing.
Keep posting these type of animations to help us learn practically......
That is actually a question I still wonder until now, though, thank you.
As always sooting voice of narrator and intelligible animation
We take technology as granted we don't realize how complex it is until someone make video like this.
Maybe not all.
This makes you appreciate your eyes. Your eyes and brain does this naturally.
Ted Ed needs to open a school these vids are so great
Crazy! These people are incredible
👌
I love you people who create this video.
Good learning....ted make rockz...brave voice. Clear and sharp to listen..
The best answers and information
*They are watching us - ALWAYS*
Old country roads, take me home
To the place, where I belong
In West Virgunia
Or should I say:
Old self driving car, take me home
To the place where I belong
In west virginia
You are best at explaining
1:53 I would suggest thickening the beam, and using a color other than green to represent light bouncing back. The green beam is harder to see because it is against a blue background, and it practically disappears if viewed on the lowest resolution, 144p.
Damn ! You're uploading this video right after i created a self driving car program !
My IQ has risen! Thank you for answering the questions I ask myself in my head!!!!
TedEd astounds yet again!
#TedEd-at-its-best!
it is just awesome
2020 vision?
I see TED-Ed is cultured organization as well.
I can’t see into the future because I don’t have 2020 vision.
keep up the good work, thanks for explaining complex things simply,
thanks ted-ed i was planing to make a project on it and you just made it more intresting
Love your animation 😇 and the narrator.
LiDAR is not the only solution to self-driving. It has a lot of advantages, but also just as many disadvantages. The sensors are very expensive. Resolving the detailed features of objects in front of it is not very precise from off-the-shelf low-res hardware which a lot of these cars are equipped with. Plus, the current transportation infrastructure is created for visual perception, because it's driven by people who use visual perception to navigate vehicles; there is something to be said for a solution that uses visual data to navigate over LiDAR.
“Imagination is more important than knowledge. For knowledge is limited to all we now know and understand, while imagination embraces the entire world, and all there ever will be to know and understand.” - Einstein.
thanks for the lesson ted
Nice presentation, but the real question is what happens when multiple lidars interfere with one another. The chances are low because of the time scales, but if there's a sufficient number of self-driving cars, there's got to be lidar pulse sync at some point between 2 or more cars.
*As our mobile phones know where we are, where we should go and where is the nearest coffee shop.*
We are being constantly tracked.
I love those quotes before videos 📕👏
Another amazing video with great animation ted-ed...
Iam fan of your channel.. It gives me more knowledge
Thank you ted-ed...🙏👍
My English is bad , so... hope you guys can understand my meaning ,
I have a question , if one super short laser into the another LIDAR system will have problem?
this is some hardcore electrical engineering, humans did come a long way
TED-Ed again with super simple animation and great explanation, Keep Going >>
thank you TED-Ed for this wholesome break from the James/Tati drama
Except Tesla doesn't use Lidar and is the best self-driving car on the road today
Tesla cars aren't self driving, only Level 2
@@jojodroid31 Elon Musk is offended by this video. He shat on Lidar at the latest tesla presentation xD
Lidar is costly, heavy and bulky, and not completely reliable. Elon is right
@@shivamtyagi677 From your answer I recon you watched Real Engineering's recent video.
Isn't that mostly a cost question though? If you'd build self driving car with all the best tools, no matter the cost, LiDAR would probably be used. Sense Tesla also want their cars to be affordable, they're using technologies that are cheaper to implement today, at least that's what I've heard.
Once LiDAR becomes more affordable and possible at size scales described in this video, Tesla will probably switch to use LiDAR, but will probably still use their software that they're currently pushing to develop and refine.
I've literally been asking this question in my mind a lot haha!
Mesmerising !
I'm gonna take my *self driving car* to the old town roooooad
Collecting cloud of several million precisely positioned 3D points is kind of easy. Figuring out the actual object from that cloud, now THAT is a challenge!
‘Anyone relying on lidar is doomed,’ -Elon Musk
Daamn, J'allais dire la même chose!
@@karasira2696 trop tard
Exactly what I thought 😂
77
I wish it's applicable to blind people for navigation especially on detecting the height of obstacles... So they know what next step is... Well, there's a limit to everything
I'm actually a team leader for a self driving racing car competition in the UK 'FS-AI'. This is a top quality video, however, LiDAR hasn't quite been explained perfectly becuase even the best LiDAR radars only see a very limited slice of the world. Whilst we have LiDAR it's kinda useless, the limited slice that the LiDAR can see makes it very difficult for object recognition as there simply isn't enough data coming from the LiDAR to make any reasonable assumption of the object. We use the Robosense V16 Lidar, which, despite it's high price tag is simply not useable. To get better resolutuion you have to pay vast amounts of money even though they fundemantly have the same issue. What we use instead is stereo vision (Zedd Camera) as you're able to make better object detection useing each image from the camera and using machine vision and the principles of the Fundemental Matrix, we're able to rebuild a higher quaility map of the world.
But zedd cam that doesn't imtegrated by some kind of 3d lidar, doesn't give accurate results especially under extreme weather conditions. How did you manage to do that?
@@selene8734 Firstly, neither does LiDAR in those situations. We solved it using multiple cameras, and a model train on sensor fusion for the various weather conditions
Brilliant video. But i believe detection is not the main problem currently, but the main problem is to identify and track objects reliably. And also the ethical dilemma (which is already explained in another ted-ed video).
I wonder what quality content I would watch if Ted Ed didn't exist?
It's baffling, we didn't even know how to conduct electricity a few hundred years ago, and now we're at autonomous cars.
I like your videos so much it really is a great educational channel
Quite interesting. Top-notch narration and animation. Loved it❤
I got a self driving car ad before this!
wow thank u so much sir
Oh nice one of the first time to be here so early
What if a selfdriving car meets another selfdriving car? Will they collide??
Thank you
Self-driving car: *sees an insect 10 meters infront.*
Self-driving car: *Drifts like Vin Diesel.*
Nice, I work with lidar to use on self driving car in California
Good video
Whatever a Mach-sender modulator is, it sure sounds like it is cool
To see is one thing, to decide which one to crash to is another...
I'm purchasing this car....
When the laser bounces off of things, doesn't it travel away from the laser?
How does the car receive the laser? Does it really travel straight back, even when the surface isn't perpendicular?
part of gets deflected and some traces back the same path
How can self driving cars see?
Sensor👨🏿🦰
Amazing 👍🏻
I had an ad of safely driving cars before this video
NEXT VIDEO
How we read viewers mind and made our video with such a good animation
"The robot cars cannot accurately predict human behavior, and the real problem comes in the interaction between humans and the robot vehicles,”. This was in response to the woman that was killed by a self driving car in AZ.
Awsome work❤
I love ted ed 💞♥️😍
118 liker and 7 minutes ago uploaded AND NOTIFICATION SQUADD!!!
Ok, but to respond to a traffic light, the car needs to ACTUALLY see, not just measure distance. How does that work?
How it would detect fast moving obstacles like an animal running across the road ,we humans will switch to right if it's running from right to left ,but what will the automated car do ? Will it automatically stop until it crosses?
Thanks
Being the best in pioneering self driving currently, Elon Musk doesnt support Lidar in cars. He says its worthless and that cameras are far better. Elon also said that he isn’t against Lidar, as SpaceX use them for space station docking, but it just doesnt make sense in cars.
was about to comment this. thanks for educating others :) Elon Musk rocks!!!
The problem with Lidar that Musk describes is that they are (currently) expensive and large. However, in the near future devices such as those shown in the video will become available, probably at a fraction of the current cost of lidar systems. At that point, any autonomous vehicles would probably choose lidar's superior accuracy and depth perception over the advantages offered by cameras.
@@JWFitz I am pretty sure they are just going to use both.
@@JWFitz just "seeing" the object is HALF the battle, then you need to identify what it is... think a plastic bag blowing across a highway. Telsa is using cameras (with color) and big data to solve the second half of the problem. They have billions of miles worth of data identifying animals, people, bikes, etc... where the companies using Lidar only have a depth map.
@@stachowi no they don't what age you smoking they just have video footage no one is going to label billion miles not video by hand you Dumbo