Tesla VS Waymo - Who Will Win the Race to Full Self Driving? + LiDAR VS Computer Vision

In the current

race

toward our

self

-

driving

future, who

will

be the first company to offer

full

self

-

driving

features? I'm Johnathan Stewart and welcome to Cleaner One. I recently read an article about autonomous driving and this article had a graph from Navigant Research. their leaderboard and it showed which companies, in their opinion, were leaders in autonomous driving and those that were lagging, interestingly they had Tesla lagging behind, they had GM and Ford much more in the lead, so I took a few minutes and dove in in the data. for each of these companies and I found out very quickly that actually the only way it had some serious data, some serious autonomous miles and really showed promise towards autonomous driving, of course knowing what I know about Tesla and their autopilot and how it all works.

I know they are leaders too, so in this video I want to take a moment and talk about the two main leaders in the

race

to autonomous driving, Mo and Tesla, and I want to focus on some key metric technology differences and determine them based on. Based on this data, who

will

probably be the first to drive

full

y autonomously, so I think the first category that's important to talk about is the sensor suite. What sensors does each of these companies use to achieve fully autonomous driving? it can function on its own in its environment. One of the main differences of the sensors in the system using MO versus Tesla is Tesla's

computer

vision

versus a LIDAR.

A religious system like MO LIDAR, just if you don't know, is simply light detection and ranging. system that bounces lasers off objects at millions of pulses per second and then the car measures the changes in distance and the laser pulses bounce back to give these readings. Instead, a camera-based system uses

computer

learning to train the neural network, allowing the car to recognize items. on the environment and react accordingly here's a clip of Elon Musk talking about

lidar

versus cameras at autonomy day earlier this year everyone is going to get rid of

lidar

it was my prediction walk my words I should point out that I don't actually hate a lot of light or as much as it may seem, but at SpaceX their basic dragon uses lidar to navigate to the space station or to the dock, not only that we, SpaceX, developed their own lidar from scratch to do that and I spearheaded that effort personally because in that scenario lidar makes sense, it's fucking stupid, it's expensive and unnecessary and, as Laurie said once you saw the

vision

, it's worthless, so you have expensive hardware that's worthless in the car, we have radar front which is low cost and is useful especially for occlusion situations, so if there is fog or dust or snow, the radar can see through that, if you are going to use active photon generation, do not use the length of visible wave because once you use passive optics, you have taken care of the entire visible wavelength. and things you wonder if you want to use a wavelength that is occlusive and penetrating like radar, so routine coating is just active generation of photons in the visual spectrum.

If you are going to do active photon generation, do it outside the visual spectrum on radars. the radar spectrum, like twenty point eight millimeters versus 400 to 700 nanometers, it's going to be much better occlusion penetration and that's why we have a Ford radar and then we also have alpha 12 ultrasound for near field information in addition to the eight cameras and The young Ford radar needs the radar in all four directions because that's the only direction you go very fast, so I mean we've got through these moving times as always, we're sure we have the right set of sensors in case that we add something more.

The advantages of a camera-based system are, of course, that it is very economical to deploy cameras in a car; it can be trained through computer learning to recognize objects, so this system can be trained to recognize objects. Drop the car somewhere or have never been before and it could still orient itself a camera based system allows the car to read the signs our current road system is set up for human vision of course and therefore a system that it is based on human vision of course a camera that is computer vision allows it to read signature and react accordingly like a human would and of course the cameras allow for a very stylish built in design that does not take away from the aesthetics of the car .

The disadvantages of the camera-based system are only that it requires a lot more computer learning. The work and the data entry, and this time and computing power are very intensive, so now that we know the difference between the lidar-based system and the camera-based system, I want to go over the sensors in that way that is currently not They use them in their cars to navigate. around their environments, so of course, as we mentioned before, they use lidar sensors on the front, top and back of the car. They also have a vision system so you have a camera to help in the surroundings and driving the car.

They have a radar system in both the front and the rear and then they have supplemental sensors that they describe as their audio detection system that can listen to police and emergency vehicle sirens and they also add a GPS so that Know where you are based on a map. Tesla's Autopilot sensor suite. includes eight cameras, twelve ultrasonic sensors and a front radar system, the ultrasonic sensors are simply used to measure distance using ultrasonic waves, it lets you know how close you are to an object, of course they also use the cameras and radar to a view of the environment around it, the second category I want to talk about is the approach of how each of these companies approaches the problem of solving fully autonomous driving, so currently we know that wham-o is using a system based on lidar and they have to currently map the entire section that way, Moe can drive, they have to go through and make 3D maps for an area and then they have what they call their geofencing, which means you can just operate wham-o on the functions full self-driving cars on this geo-fenced map and here's an image from way Mo's website and it shows what their current self-driving territory is in the Phoenix metro area and what they hope to expand to in the future, so the focus of Mo is to use lidar and Geofencing and mapping specific areas of the city and I will allow their cars to be used in those areas.

Tesla, on the other hand, is using computer learning and computer vision, which means that once its feature is complete with fully autonomous Tesla vehicles, it will be able to drive autonomously almost anywhere in the world, no geofence. What about the cost of these vehicles? Of course, cost will be important as they implement these features in cars, whether that means a network of robot taxis or whether they decide to sell these vehicles. For the public too, the cost will be high. Currently the retail price of the MO Chrysler Pacifica minivan that they use to convert it into their Way Mo taxi retails for around $40,000, it's a hybrid and if you look at that.

In the Pacific, a 2019 hybrid costs about thirty-nine thousand nine hundred dollars, then of course you realize that there are multiple lidar sensors and you start adding about $7,500 for each of them, plus the computer system and the rest of the technology. put on the car and a very conservative estimate would be that it would cost about a hundred thousand dollars for this car, but I guess it's actually a lot more, but at a minimum, a hundred thousand dollars for each one of these taxis that they put on the road Tesla. On the other hand, it has a model three sr+ which has a retail price of thirty-nine thousand nine hundred and ninety dollars so the same base price of the chrysler pacifica hybrid minivan in that way use but the Tesla car already includes all the chips all the sensors and everything you need for fully autonomous driving also adds to the fact that the current leases for the Model 3 do not allow you to purchase the car because Tesla has mentioned that they will make these cars their network of Robbery taxis.

Basically they are getting these cars at a deep discount where the customer has already taken on much of the depreciation of that vehicle so the actual cost to Tesla will be much less and this doesn't even take into account the fact that Tesla makes between 15 and a 20 percent margin on its cars, in addition, another important factor to compare between the two companies is the size of the current fleet. How many cars does each of these companies have in real world situations? According to my research, it currently has 600 autonomous vehicles. On the road they have an order for 62,000 more Chrysler Pacifica Z' and they have also placed an order for 20,000 Jaguar I Paces.

Tesla, on the other hand, has over 600,000 cars on the road right now with hardware version 2 and higher. that are capable of driving themselves once Tesla releases those features via a software update on Lex Friedman's website. He put up a nice graph that he created and you can check it out. I'll put a link in the description, but I estimated that there are currently six hundred and twenty-five thousand, five hundred and seventy cars with autopilot for hardware plus and that was as of the third quarter of 2019, so by the end of the fourth quarter of 2019, of course, there will be about a hundred thousand more cars on top of that, so surpassing the 700 thousand mark by the end of 2019, now the next important thing to consider is the data that this fleet collects, so of course, as we mentioned, wham-o has six hundred vehicles, Tesla has over six hundred thousand and is very close to 700 thousand vehicles on the road collecting data and giving them information, so Moe recently posted that they have driven over ten million miles on real world roads autonomous since 2009 and while that seems impressive, especially when you compare that to Freeman also estimates that the rest of the Tesla competition will reach two billion miles driven on Autopilot with full self-driving features included by January 2020, so Which when you look at this for a minute, 10 million sounds impressive, but then you realize that Tesla by the end of this year will have data for over two billion miles driven with their hardware systems.

This is an incredible amount of data and is a huge benefit for Tesla engineers who are working on the computer learning and computer vision that Tesla recently launched. features such as navigating on autopilot, which allows the Tesla vehicle on autopilot on the road to change lanes and also be able to take exits on its own. They also launched features like smart summoning that allows the vehicle to leave a parking lot and come meet you. on the sidewalk or come pick you up wherever you are and it's a pretty incredible feature that is a step towards full autonomous driving so in the race towards autonomous driving data is king as you know the real world is very unpredictable, people humans do very irrationally. things and don't always follow the traffic laws and the laws that appear on the signs, so the more real-world data you can collect, the better it will allow you to anticipate and solve the equations that are needed for all corner cases, for example. which you can solve them and make sure the cars react accordingly as they should.

Tesla has all the real-world data for its global fleet. If you think about that for a minute, they have a global fleet, they have data not only. for the United States, Canada, Europe, around the world, they have this data that they are currently building, billions of miles of data, as well as these 10 million miles of data. Wow, that's a lot of data, it's only for specific areas that have been mapped, they have data. For the Phoenix metro area primarily, Tesla is also working to solve computer vision and has hundreds of thousands of owners helping them with this data.

You have to pay your engineers to drive all these miles, map and get these areas geographically. It's fenced in and that's very expensive and time consuming, so based on all these data points that we've discussed, who's going to win the race, I think it becomes very obvious when you think about the data, when you think about the features and you think about the number of cars on the road with the hardware necessary to be fully autonomous. I think it's becoming pretty obvious that Tesla is poised to get there faster than any other company. Currently, Wham-o is operating at SAE Level 4.

They have a list of what they consider autonomous driving, they have levels from 0 to 5, currently they say Tesla is officially operating at level 2 and wham-o is operating atlevel 4, so at first glance you would say, well, of course, Mo is a leader. doing level 4 Tesla is only at level 2, but this level 4 is only in geofenced areas that have been mapped very intensively and they are using really expensive sensors to put in these cars to achieve this once everything is done . of a fleet and the cost of preparing maps for these areas is very substantial. Tesla on the other hand is at level 2, but once they figure out the computer vision they are good to go, all they have to do is just flip the switch and the entire fleet. is ready to be autonomous, its biggest problem right now to reach fully autonomous driving is the millions of edge cases that exist in the real world, as we mentioned before, the real world is unpredictable and Tesla currently has to train its neural network to deal with With all these corner cases and in my opinion Tesla will probably solve full autonomous driving and have a full feature system which may not be approved for use yet but they will have a full feature system for purposes of 2020 and then.

We will have over a million cars ready instantly, so as I mentioned, I think Tesla is ready to get there much faster than anyone else. I think they will be complete by the end of 2020. I think regulation and things like that will take quite a while. It's still a while away from officially allowing it, but maybe by the end of 2021 2022 the regulators will approve it and it could be ready for you to fall asleep and let the car drive you around the city. Companies like way mo and others that use lidar systems. I think we are still five years to a decade away from achieving this.

Thank you very much for watching this video. I hope you learned a few things and I hope you enjoyed reviewing the data comparing these two. businesses, if you like the video, consider subscribing so you can see future content posted in the future. Also, if you liked it, go ahead and click the Like button which helps more people watch the video. Thank you so much.