Tesla May Be Just Getting A Year-Long Head Start For Self-Driven Cars, Or Not Getting Approval In Europe
Long Only, Tech
Seeking Alpha Analyst Since 2016
Summary
- The Tesla Model 3 is to contain a "Full Self-Driving Hardware". Is this really true? This is based only on cameras, radar and ultrasonic sensors.
- LiDAR systems, which are generally considered necessary, are missing.
- Factory "Giga Berlin": In Europe, it is not Tesla that decides whether a car may be operated "Full Self-Driving", but the motor transport authorities.
- A "full self-driving" mode, which primarily relies only on cameras, does not appear to be approved in Europe. Tesla could fail without a LiDAR module like Boeing.
- However, Tesla has the unique opportunity to exclusively acquire a new LiDAR module (high-resolution, small, cheap) that will probably be necessary for approval in Europe.
Tesla (TSLA) wants to deliver a “Full Self Driving” mode for Tesla cars via software update this year. Is Tesla allowed to do this in Europe? This question seems not to have been discussed yet. Except by Tesla itself, a "Full Self Driving" mode, which is primarily based only on cameras, is generally considered unsafe and unacceptable. At least for road traffic in Europe, it therefore seems almost impossible for Tesla systems to be approved for use on public roads without a LiDAR module. Because Testa does not decide on car type approval, i.e. the permission to activate the systems in road traffic, as obviously even many Tesla shareholders believe. It is decided by the national motor transport authorities. Here, Tesla seems to want to disregard the authorities, in that Tesla wants to determine for itself what is required for a safe system. Similar to Boeing, which was also not so particular about security. The author cannot judge whether a certification without LiDAR module is possible in the USA. However, since Tesla wants to start operating the new factory "Giga Berlin" in Europe as early as 2021 and probably sell up to 500,000 cars per year in Europe, Tesla can no longer ignore the requirements in Europe without having unsaleable cars if the competition sells by the motor transport authorities licensable models with LiDAR systems. This article now shows that Tesla was absolutely right that there were no usable LiDAR systems available until now, because they had a much too low resolution, were much too big and also much too expensive. But the development does not stop there. A new company in the LiDAR market will soon be launching not only the world's highest resolution, smallest and cheapest LiDAR module for cars. The company and the LiDAR product line are also currently for sale. It will be shown that the car manufacturer or supplier who will own this technology in the future could have a competitive advantage for many years and could use patents to prevent similar systems from being used by competitors. It is now up to Tesla whether it will be Tesla or, for example, Toyota or Volkswagen. Tesla shareholders also have the opportunity to acquire a direct stake in the company, which is currently traded on the stock exchange at far below its value.
Tesla - used systems for autonomous driving
Tesla uses three different technologies in Model 3 to enable autonomous driving from Tesla's point of view:
- Cameras
- Radar (front only)
- Ultrasonic sensors
Here are the corresponding specifications from Tesla:
Source: Tesla
Tesla distinguishes between "autopilot" and autonomous driving:
Source: Tesla
Regarding autonomous driving, Tesla claims that the Model 3 already contains the necessary "full self-driving hardware":
Source: Tesla
This functionality is to be added by a software update alone:
Source: Tesla
Who decides which functions of a car may be used on public roads?
Do the motor transport regulatory authorities in Europe see it the same way? A car type approval for it in the European Union (EU) cannot yet be granted, as the functionality for autonomous driving does not yet exist. It depends on the legal requirements in Europe, for example:
- Europe
- France
- Switzerland
- Germany
- Great Britain, which often has laxer guidelines, is no longer of any importance, as Great Britain is no longer a member of the European Union.
Effects on insurance premiums
Not only the motor transport authorities, but also the car insurance companies will pay very close attention to whether the hardware is also suitable to safely guarantee the "Full Self Driving" mode. After all, "Full Self Driving" or autonomous driving in level 5 (compare later) also means that the driver is allowed to sleep while driving.
Possible claims for damages against Tesla in Europe
If Tesla is unable to provide this functionality via software updates, e.g. because the motor transport regulatory authorities consider this software solution to be unapprovable due to insufficient hardware support, Tesla may be subject to claims for damages, as Tesla is already advertising the functionality. This could potentially lead to a procedure comparable to the " Volkswagen Emission Scandal". Buyers will then have significantly impaired cars because the functionality will not be available. At VW, the promised level of emission control was missing, in this case the "Full Self Driving" functionality.
Now there are hardly any Tesla cars in Europe. The registration numbers are very low. In the first half of 2020, only about 33,000 new Tesla cars were registered. However, this should change with the completion of the new factory near Berlin.
Everyone agrees that LiDAR is needed - except Tesla
The subject of whether or not LiDAR is necessary cannot be discussed here in full depth. However, the press coverage is already clear and the motor transport approval authorities will probably make similar decisions. They will not allow the participants to be endangered just because Tesla lacks the necessary hardware.
An example from the articl e „At Tesla’s“Autonomy Day” event for investors on Monday, Elon Musk was full of trash talk for his competitors and their technology”:
LIDAR, the light beam sensor that practically everyone views as an essential ingredient for self-driving cars, is “a fool’s errand,” according to Musk.
There will be further details:
Practically every other company trying to bring self-driving cars to the road — including Ford, Uber, Waymo, and GM Cruise — relies on a suite of sensors comprised of LIDAR, cameras, and radar. These companies argue that LIDAR can do things that cameras and radar cannot, while also providing overlapping capabilities to the things those sensors can do. These capabilities, known as redundancies, are extremely important for fully driverless vehicles as they provide an important backstop in the event of a failure.
The regulatory authorities will take a similar view. In addition:
Musk’s argument about LIDAR being useless with cameras because it replicates the visible light spectrum is “just wrong,” says Sam Abuelsamid, a senior analyst at Navigant, a technology consultancy. The human eye will respond to wavelengths from about 380 to 740 nanometers, while LIDAR can respond to the higher range of 905 to 1,550 nanometers. And new types of LIDAR, like the continuous beam FM sensor from Blackmore, can instantaneously measure velocity while also reducing computational latency, Abuelsamid notes. Layering sensors with different capabilities, rather than just relying on a purely vision-based system, is “ultimately a safer and more robust solution,” he said.
The article is worth reading and should be used to understand the arguments of both sides.
AI is no solution for image recognition
If you still believe Tesla's promises afterwards, you should read this article "Pixel patterns irritate the AI of autonomous vehicles". After that no person will hardly be able to keep its opinion. Especially Tesla shareholders should press for safer cars.
Tesla vehicle accidents with fatalities that might not have happened with LiDAR
There have already been several accidents involving Tesla vehicles that could probably have been prevented by using - sufficiently good - LiDAR technology. However, it is possible that at that time there was no sufficiently good LiDAR technology available on the market. But that is currently changing. For this very reason, it is necessary that Tesla does more for safety and additionally uses LiDAR.
Electric carmaker Tesla says a vehicle involved in a fatal crash in California was in Autopilot mode, raising further questions about the safety of self-driving technology.
Another example:
A preliminary report into a fatal accident involving a Tesla Model 3 in the US has found Autopilot had been engaged 10 seconds before the crash.
The autopilot is not a "self-driving" mode. But even it would be able to detect obstacles better if it had LiDAR.
See also: " List of self-driving car fatalities" and " Tesla Deaths" (not all are related to autopilot).
Why does Tesla really not like LiDAR?
The author does not believe that Tesla considers LiDAR to be useless - this is probably just pretending. The real reason might be the following statement of Mr. Musk:
What’s more, he went on, “anyone relying on LIDAR is doomed. Doomed. Expensive sensors that are unnecessary. It’s like having a whole bunch of expensive appendices... you’ll see.”
That is why the report also states:
LIDAR can be incredibly expensive, and it’s a costly bet for most companies. They can make the job of selling self-driving cars to customers practically impossible thanks to the added costs.
Thus, LiDAR for Tesla is probably simply (still) too expensive at present.
Especially since a car needs not only one but several LiDAR sensors (compare later). But currently, LiDAR sensors cost $1,000 to $25,000 when purchased, and would then be included in the end customer price at least four times the price per unit. Thus, a LiDAR system would cost between $10,000 and $100,000 more per car. Tesla certainly wants to avoid this. However, it would probably endanger the safety of all road users.
Other reasons against LiDAR
LiDAR modules are currently not only far too expensive. They are also much too large and not powerful enough. This means that LiDAR modules provide 3D models of the environment in much too low resolution. Therefore, Tesla is quite right that they are currently hardly useful because they cannot capture smaller objects. There is no point in having a LiDAR system detect a wall, but not a bicycle with thin frames that is standing on the road. Or the branch that juts out of a tree onto the roadway and threatens to smash through the windscreen in a collision. In addition, current LiDAR systems cannot be installed without affecting the aesthetics of the car and other vehicle characteristics such as the drag coefficient value. Some even have to be mounted on the roof of the car.
LiDAR enables a greater range of cars
However, a major advantage of LiDAR is that it provides 3D points and thus 3D objects. This means a three-dimensional environment without having to calculate it first. Cameras only provide two-dimensional images without depth information, from which a three-dimensional environment has to be calculated in real time with a lot of effort. The very powerful computers required for this require a lot of power and thus reduce the range of electric cars in particular, such as those offered by Tesla.
Disadvantages seen by Tesla disappear just by a new development
This article now shows that the disadvantages seen by Tesla
- exorbitant price
- oversized
- too low resolution
be overcome by a new technology of a new company due to the progressive development.
Enormous opportunity for Tesla
Tesla now has the opportunity to expand its lead in the field of autonomous cars by adding the new LiDAR module and make them ready for registration. This does not only apply to new models. The module is so small and inexpensive that it could also be offered as a retrofit kit for $500-$1,000 (plus installation). The price depends on how many modules are needed and where they would be installed, i.e. whether a component such as the rear-view mirror needs to be modified on the inside.
By exchanging the currently very highly valued Tesla share, Tesla could get hold of the technology without the use of cash.
What are self-driven cars actually?
A self-driving car, also known as an autonomous vehicle (AV), connected and autonomous vehicle (CAV), driverless car, robo-car, or robotic car, is a vehicle that is capable of sensing its environment and moving safely with little or no human input. (Source: Wikipedia)
There are different "levels of driving automation" (compare Wikipedia), ranging from 0 to 5 (excerpts of the definitions of levels 3-5 according to Wikipedia):
- Level 3 ("eyes off"): The driver can safely turn their attention away from the driving tasks, e.g. the driver can text or watch a movie.
- Level 4 ("mind off"): As level 3, but no driver attention is ever required for safety, e.g. the driver may safely go to sleep or leave the driver's seat. Self-driving is supported only in limited spatial areas (geofenced) or under special circumstances.
- Level 5 ("steering wheel optional"): No human intervention is required at all. An example would be a robotic taxi that works on all roads all over the world, all year around, in all weather conditions.
These require both additional hardware and software compared to previous vehicles. Of course, all considerations also apply to trucks.
LiDAR hardware
The software for self-driven cars can only be as good as the data it receives. This means that if the hardware cannot see an obstacle, the software cannot react to it. Under certain circumstances, an accident may occur with possibly injured or even dead people (besides material damage). The software must therefore be enabled to "see". Good enough to see. Humans have eyes for this. Cars can have eyes for it:
- Cameras
- Radar Systems
- LiDAR Systems
- Ultrasonic Systems
Source: Woodside Capital Partners
Cameras are used to capture images, just like the human eye. Unfortunately, they are still far from being as good as the human eye, i.e. even in difficult lighting conditions, fog, etc.
In addition, cameras have the major disadvantage that they only provide two-dimensional images of the environment. So, the depth information is missing, i.e. how far away something is. It is not completely unimportant whether a wall in front of the car is one meter or one kilometer away. Therefore, with cameras consisting of several images, complex calculations from one image/video first have to identify the obstacles and calculate the depth information. Two-dimensional image pixels must therefore be converted into three-dimensional points and three-dimensional objects. This requires an enormous amount of computing power, which in turn reduces the range of the cars due to the high energy consumption of the very powerful computers required for the real-time analysis of the images. This is particularly negative for cars with electric motors. Radar systems are also very useful, but will not be discussed here. That leaves LiDAR systems.
What is a LiDAR system?
A detailed overview of LiDAR technologies and systems is provided by wikipedia. For cars, the following two graphics explain this clearly:
Source: Woodside Capital Partners
Source: Woodside Capital Partners
Why does a car need LiDAR?
Contrary to Tesla's claims, a self-driven car without LiDAR does not appear to be safe or legal. Reasons for using it are:
Lidar balances the weaknesses of today’s sensors. Unlike a camera, it sees just as well in the dark as in the light. It can detect detail better than radar, allowing it to pick out things like pedestrians and cyclists. “You can see this stuff with lidar,” Russell says. A single sensor in a car’s front bumper or windshield might make it a more capable robot than just about anything on dealer lots today. That’s not enough for a fully autonomous system—you’d want 360-degree coverage there—but it could allow for a system that handles highway driving.
Competition, like Waymo
Competitor Waymo offers this LiDAR system:
Source: Waymo LiDAR
Source: Waymo LiDAR
The author assumes that the actual LiDAR system is located in the large box on the roof and is used in addition to the radiator grille, as the latter is probably too bad (see explanations later).
No car manufacturer will offer a car with any kind of system on the roof (be it LiDAR or cameras etc.). There are several reasons for this:
- The appearance of the car is destroyed.
- The car is getting higher.
- Higher energy consumption due to poorer drag coefficient values.
No automaker will also offer cars that have front mounted systems, such as the Waymo LiDAR system in the previous photo. Reasons:
- The aesthetics of the car are destroyed.
- The car becomes longer (not unimportant e.g. in Europe).
- Higher energy consumption due to poorer drag coefficient values.
- At least in Europe, not permitted for road traffic due to the risk of injury to persons such as pedestrians and cyclists from protruding and hard objects.
- Sensitivity to collisions: In case of a frontal collision or even the smallest bumpers, the (very expensive) system will be destroyed because it is located in front of the bumper. The LiDAR system itself becomes the bumper. This inevitably leads to high insurance premiums.
- Rain or dirt can quickly make it "blind". An automatic effective cleaning seems hardly possible with the design Waymo has chosen. In contrast to systems that are integrated in headlights or inside vehicles, e.g. in the rear-view mirror behind the windscreen.
Even the smaller system at the front of the car not only does not look nice, it will have negative properties on the drag coefficient-value, which will become more and more important especially in the future for cars with electric motors. Assuming that, as stated above, it is not suitable for approval in many parts of the world.
Since Waymo LiDAR in its current version is already unapprovable, the Waymo LiDAR system cannot be used in production vehicles and is therefore not for sale.
Technical data for the Waymo system are not known. However, average LiDAR systems only provide a resolution of a few 100,000 to 3 million points per second.
In the opinion of the author, this is not sufficient to ensure safe and reliable detection of smaller obstacles, even if several modules are used. Only larger objects can be detected like other cars. However, a car must also be able to avoid branches that reach into the driveway or come to a halt in front of them. The same applies to children and also animals. It is important that the resolution per second is specified for LiDAR systems. However, since these systems scan the area several times per second, the actual resolution in a few milliseconds is only a fraction of this. However, due to its usual speed, a car cannot wait a second to get a complete image of the surroundings, but may have to react in milliseconds. Even if only 1/10 second is assumed as reaction time, only 1/10 of the resolution would be available, which in turn would only be 10,000 to 300,000 points. For comparison: Full HD has 1920 x 1080 pixels = 2,138,400 pixels. UltraHD/4K even has 3,840 x 2,160 pixels = 8,294,400 pixels - this 60 to 120 times per second. So about 120 to 480 million pixels per second. In contrast to LiDAR systems, today's cameras can record images with the full resolution several times per second. However, the more images, the more processing power is required (compare before).
What other LiDAR providers are there on the market?
It is difficult to get an overview of the LiDAR systems on the market and their specifications. Hundreds of companies are developing LiDAR systems. However, only a handful or less will survive. An overview from the year 2018 can be found in " The Automotive LiDAR Market" by Woodside Capital Partners.
Source: Woodside Capital Partners
Source: Woodside Capital Partners
Source: Woodside Capital Partners
Designs (mechanical, solid state, etc.):
Source: Woodside Capital Partners
Source: Woodside Capital Partners
Source: Woodside Capital Partners
What other LiDAR systems are available?
As of 2018, these LiDAR systems were available with the listed specifications:
Source: Woodside Capital Partners
From the point of view of car manufacturers, the prices charged for the components are probably already alarming. Not even Mercedes, Porsche or Rolls-Royce can build a component for $8,000 - $75,000.
This leaves only the $600 system listed in the previous table. But: Even this one is probably too expensive for almost all car models. Especially when you consider that several modules are necessary for an all-round view as well as a backup in case one of them fails at full speed, as one of the largest automotive suppliers, Bosch, also considers necessary.
Waymo LiDAR system also too expensive
The Waymo LiDAR system is not only a technical failure, as was previously stated, but also an economic flop:
Waymo began manufacturing its own LIDAR sensors in 2011 to reduce cost. At the time, Waymo said it could lower the unit price from $75,000 for an off-the-shelf LIDAR sensor to just $7,500 with its own custom version.
No automaker is going to install a $7,500 LiDAR. Absolutely out of the question.
Likewise, it seems impossible that Waymo or any of the other manufacturers could be able to reduce the cost of their modules to a cost of about $50 - $100 per module acceptable to automakers.
The cheapest LiDAR module currently available is probably one from Luminar for about $500 according to this report from Wired: "Luminar says its new lidar sensor will sell for as little as $500, compared with $75,000 for the industry leader.”
Current data for LiDAR systems can be found here:
- Valeo Ibeo, latest model under development: ibeoNEXT Generic (no resolution specified, except "Very high resolution (e.g. 0.05 deg)" which, with "Large FOV (e.g. 60x30 deg)" specified, is probably only 720,000 points per second.
- Velodyne LiDAR (GRAF): current products with their specifications, including the " Alpha Prime" the "The World's Most Advanced Sensor", but which also has a resolution of only ~2,400,000 (1 channel) points per second (~4,800,000 points for two channels):
Source: Velodyne LiDAR
-
Not listed in the table above are the new products Velarray and Velabit.
- Only the Velobit from Velodyne is said to have an acceptable price of$100, but it is not intended for autonomous cars and will probably not have an acceptable resolution: "Velodyne Velabit is an ideal solution for small platforms like the Robotic Research Pegasus Min, an unmanned autonomous robot that uses lidar for localization, mapping and more. “
Source: Velodyne
- Contd.:
- Small enough for cars, announced four years ago and still not available is Velarray:
Source: Velodyne
-
Contd.:
- Operation and specifications are not known, only general information such as: „With a compact form for seamless integration within a vehicle’s body or behind the windshield, we design the VELARRAY system to produce a robust directional image, day or night to meet our customer’s needs. The VELARRAY system has best-in-class range and resolution, allowing for faster object identification and longer braking distance at highway speeds. Mass production is slated for 2020.“
Source: Velodyne
-
Contd.:
- This application example shows that one LiDAR sensor is not sufficient, but probably six plus more are needed for redundancy.
-
RoboSense:
- RS-LiDAR M1: still much too big, with almost $2,000 in the "simple" version priceless for car manufacturers, only 1,125,000 points per second
- RS ruby, price unknown, probably around $25,000, poor resolution with only 2,304,000 points per second (or 4,608,000 points for two channels)
- RS ruby lite, $12,800, with only 1,440,000 points per second only half the resolution of "ruby" (or 2,880,000 points for two channels)
Source: RoboSense
Source: RoboSense
Source: RoboSense
Source: RoboSense
Source: RoboSense
Source: RoboSense
Source: RoboSense
Source: Ouster
-
Luminar: (Audi uses Luminar systems):
- Will launch the new model " Iris" ("resolution up to 200 points per square degree") in 2022, which will cost about $1,000: „Iris will cost less than $1,000 per unit for production vehicles seeking serious autonomy, and for $500 you can get a more limited version for more limited purposes like driver assistance, or ADAS. Luminar says Iris is “slated to launch commercially on production vehicles beginning in 2022,” but that doesn’t mean necessarily that they’re shipping to customers right now. The company is negotiating more than a billion dollars in contracts at present, a representative told me, and 2022 would be the earliest that vehicles with Iris could be made available.”
- Hydra is available for developers ("resolution up to 200 points per square degree")
-
Innoviz:
- InnovizPro: 178.000 points per second, size: 73x66x165mm, price unknown
- InnovizOne (available 2021): resolution 7.5 million dots per second (single), size: 45x110x95 mm, price unknown
- Aurora: „ Aurora Introduces New LiDAR And Vehicle Platform”: “The specs could be 120 degree field of vision, 450 meters range, 2.4 million points per second for $20,000 per unit.”
-
countless other companies that develop LiDAR systems, all of which, as far as we know, do not offer sufficient resolutions either, such as
- Bosch presented a LiDAR system at the CES 2020, details are unknown. Rumor has it to be MEMS-based. If so, it will have only a fraction of the resolution of Microvision's MEMS LiDAR systems (compare later), because Bosch MEMS projectors also have a much lower resolution.
- Osram and
- Strobe, bought by GM three years ago, no data available until today
- Scantinel: press release, resolution unknown
- many more are listed in " The Automotive LiDAR Market" by Woodside Capital Partners
- and so on
No manufacturer offers (single channel) resolution of more than 3 million dots per second.
All RoboSense products are with $1,898, $12,800 and about $25,000 much too expensive for serial models of car manufacturers. The same is true for Ouster models with prices starting at $3,500, $6,000 and $16,000. Aurora also charges astronomical $20,000, making them uninteresting for Tesla.
From the quantity of competing products, the Veloarray from Velodyne would be the best choice for Tesla. However, this does not seem to be good enough. In particular, Velodyne remains silent on the resolution despite at least four years of development. To get the technology, Tesla would have to take over the company, which was valued at$1.8 billion in July 2020.
However, since companies such as Ford, Hyundai and Baidu are involved there, this does not seem possible.
Funding Type Corporate Round, Money Raised $150M, Pre-Money Valuation $1.9B
Conclusion: The Waymo LiDAR system is not usable for series models of the major car manufacturers. The same applies to almost all competitor products. These are usually already far too expensive. In addition, they achieve a maximum resolution of only under 3 million points per second (single channel), which is not sufficient for the safety of the vehicle occupants in the opinion of the author, since not all small objects are reliably detected, especially if they are a little further away from the car.
This is not only the opinion of the author, but also of the professional and financial world. From a report by MergerMarket dated August 04, 2020:
An automotive sector banker noted that while lidar had been very “topical” for a while because it promised to provide autonomous driving systems that could see far ahead with high accuracy, “I don’t think anyone has been able to deliver a solution that delivers that kind of functionality” yet. These technologies could deliver highly effective autonomous driving systems in the future but still need more development, he explained.
There will be consolidation in automotive lidar as companies work refining these products, he said. There aren’t many deep-pocketed lidar companies, he added. He named Innoviz as one of the main companies that has a contract with tier 1 auto companies, as well as Continental AG (OTCPK:CTTAY), Magna International (MGA), and Valeo (OTCPK:VLEEY).
Tesla: Unique chance for a big advantage in autonomous driving over the competition
These days the sale of the company Microvision (MVIS) is probably entering its final phase. The potential has remained undiscovered so far.
Microvision already offers a solid state LiDAR system for smart home devices, which has a standard 15.5 million dots per second resolution at a range of 10 meters, with an optional 20 million dots per second resolution, and is smaller than a smartphone:
Source: Microvision
On the basis of this, Microvision is now developing a version for automobiles that has a range of over 200 meters and offers a resolution of 20 million points per second even in full sunlight. This is about 7 times (!) the resolution of all competing products (compare market overview before). Microvision confirmed this on August 04, 2020 in the already mentioned report of MergerMarket:
MicroVision is planning a demonstration of its lidar technology for automotive applications, which was originally set for November 2020 but has been pushed back to 1Q21 because of the pandemic, he said. It is not a finished product but a demonstration that the company’s technology can capture 20m data points per second; current lidar systems can capture about 1m to 3m, he said. That increased data sensing capability will improve automotive “active safety” features such as automatic emergency braking, active cruise control, and sensing the difference between a car on the road swerving or changing lanes, and advance other capabilities, he explained.
This module is based on the Microvision MEMS technology, which has been proven over many years, and probably allows even higher resolutions if the Gen 4 engine is used instead of the Gen 3 engine.
It is so small that it can be mounted in the rear-view mirror behind the windscreen. Just like in headlights etc. This maintains the aesthetics and the drag coefficient-value of a car.
Due to the COVID 19 pandemic, Microvision had not received a $100 million order for interactive projectors for smart speakers earlier this year (interview by MergerMarket on August 4, 2020 with Microvision's CEO):
In February, MicroVision had been in the “final closing” for a supply agreement with a customer in its interactive display segment for a smart speaker related product, Sharma said. But as the COVID-19 outbreak spread through China, uncertainty about future consumer demand led the customer to put the deal on hold, he said.
Previously, Microvision had already got into financial difficulties due to management errors. For example, the company sold a development to Microsoft (MSFT) at approximately $50 million below development costs..
As a result, employees had to be dismissed. As a result, the supervisory board decided to sell the company.
Subsequently, the board decided to engage an advisor to explore alternatives.
The entire company or the individual product lines are sold:
- AR
- smart glasses
- projectors
- interactive projectors
- Consumer LiDAR
- Automotive LiDAR
Of course, only the product line for automotive LiDAR is of interest to Tesla. See the two articles " Microvision Serves 5 Future Markets" and " Microvision: In The Right Markets At The Right Time".
Microvision introduces the LiDAR system in this video:
Details: Since the system is so small, it can be installed in the rear-view mirror (!) behind the windscreen, for example. The size is comparable to that of smartphones. Size of the consumer version: 64 x 26 x 23 mm (compare excerpt from the previous data sheet). The installation situation is marked here:
Source: Microvision on youtube
It has a range of over 200 meters in full sunlight and a resolution of 20 million points per second, which is approximately seven times higher than that of the competitor products listed above:
Source: Microvision on youtube
For this purpose, it supports a "Dynamic Field of View":
Source: Microvision on youtube
At a glance:
Source: Microvision on youtube
On the basis of
- Resolution
- Size
- Price (about $50-$10; similar to the consumer LiDAR and projectors)
Microvision has the only LiDAR system on the market that is suitable for series models of car manufacturers.
Only the one from Microvision offers the required resolution, size and price, so car manufacturers would install it. Without this system, probably no software for self-driven cars can be sold.
In addition, Microvision has nearly 500 patents for MEMS applications, so it is very likely that competitors will not be allowed to replicate such a system.
Added to this is Microvision's great time advantage, which allows for a shortened time-to-market. Although development has been delayed for a short time, it will probably be possible to make up the shortfall quickly by increasing the development department after a company acquisition. Even with a backlog, Microvision still has a very large lead over all its competitors.
So, whoever owns this LiDAR system will have a decisive competitive advantage. It may even make it possible to sell software for self-driven cars in the first place, as the software is otherwise "blind" unless you want to rely on cameras and possibly radar systems alone. A self-driven car without a LiDAR system does not appear to be eligible for registration in many parts of the world, especially in Europe. Note: It is not Tesla that decides what may be sold, but the national motor transport authorities.
The Microvision CEO on 05 August 2020:
… disruptive nature of our Automotive Lidar that would output more than 20 million points per second at 30 Hz, have a range greater than 200 meters operable in full sunlight, output a velocity field in addition to point cloud, and will include three scanning fields of view in a single Lidar hardware.
Effect on Tesla
According to a report by " MergerMarket", there is a strong interest from "top tier" companies in Microvision's LiDAR division. Excerpt from the report dated August 4, 2020:
MicroVision attracts auto and augmented reality bidders for potential sale, CEO says
August 4, 2020
MicroVision, a Redmond, Washington-based laser sensing technology company, is seeing strong interest from bidders in the automotive and augmented reality space, said CEO Sumit Sharma.
The USD 336m market cap company announced in April that it had engaged Craig Hallum to sell all or part of the company, which holds IP and technology for laser sensing and scanning in four segments: automotive, augmented reality, consumer and interactive display.
For Tesla it is hoped that Tesla will bid and secure the LiDAR division of Microvision. This would inevitably increase the value of Tesla due to safer cars and the achievement of an approved system for autonomous driving.
How much should the LiDAR division of Microvision cost?
Here you can find comparative values from comparable companies that are updated almost daily.
The value of the Microvision LiDAR division would probably have to be higher than that of the competition due to the better specification of the module (smaller, cheaper and seven times higher resolution than all others on the market). Although the module is still in development, it can certainly be completed in the short term if Microvision can expand the development team following a takeover.
Velodyne LiDAR was valued at$1.8 billion a few days ago in early July 2020:
Pursuant to the business combination, GRAF will acquire Velodyne through a reverse merger in which Velodyne is ascribed an enterprise value of approximately $1.6 billion and equity value of approximately $1.8 billion.
Luminar (LAZR), which also develops exclusively LiDAR systems, was previously valued at$900 million in November 2019:
To go with this news, Luminar said it has closed its latest round of funding, bringing it to $250 million raised (and a $900 million valuation) to help it boost manufacturing capacity at its Orlando factory.
This rating has just been drastically increased to $2.9 billion: In "Luminar, the Global Leader in Lidar Autonomous Driving Technology, to Be Listed on Nasdaq Through Merger With Gores Metropoulos" the announcement is:
- Luminar delivers the world’s first lidar sensor and software to enable autonomous consumer vehicles and trucking for volume production
- Transaction proceeds will be used to accelerate commercial growth across its 50 partners and the expansion of its product roadmap for a turn-key “full-stack” highway autonomy and proactive safety ADAS solution
- Bolstered by the industry’s first high-volume series production win for self-driving vehicles on highways with Volvo Cars
- Pro forma implied enterprise value of approximately $2.9 billion and market capitalization of approximately $3.4 billion
The evaluation is supported by top companies like Volvo:
- Transaction includes $400 million of cash from Gores Metropoulos (NASDAQ:GMHI) (NASDAQ:GMHIU) (NASDAQ:GMHIW) and an immediate $170 million financing into Luminar, anchored by top tier institutional investors including Alec Gores, Van Tuyl Companies, Peter Thiel, Volvo Cars Tech Fund, Crescent Cove, Moore Strategic Ventures, Nick & Jill Woodman and VectoIQ
The author considers this rating to be very ambitious and too high. Especially if one considers that Luminar can deliver first devices, but these do not seem to be competitive with Microvision's solution, both in terms of technical features and price.
Uber is said to have paid $680 million to Waymo for details of their LiDAR system (which, as previously stated, is completely unusable).
The CEO confirmed in the conference call Q2 2020 that Microvision is also oriented towards such transactions:
As we continue to engage with interested parties and actively explore their desired structures, we are endeavoring to make sure that the value of all they would acquire is demonstrated through our designs, financial models and examples of other relative transactions in a particular market.
Since the Microvision LiDAR system will be technically many times better, smaller and, above all, significantly cheaper to produce than Velodyne's, the price for the Microvision LiDAR division is more likely to be around $1-$2 billion.
However, this is not an amount that should be a problem for Tesla. In fact, it seems extremely reasonable if it would give Tesla a significant competitive advantage for its own software. Not only does the Microvision system make LiDAR useful in cars at all, it is probably also effectively protected by patents.
Due to the current high share price of Tesla, the purchase could be carried out without the use of cash by means of a share exchange and would require only a small number of own shares. An exchange of 1 Tesla share for 100 Microvision shares would probably be a fair deal for both sides.
Opportunity for Tesla shareholders
In comparison to Velodyne LiDAR, Microvision - only the LiDAR division considered - appears massively undervalued on the stock market.
At a share price of $1.71 on August 7, 2020, Microvision has a market capitalization of only about $250 million with less than 150 million shares. This is only 1/7th of Velodyne's valuation. One of the reasons for this is that Microvision is under massive pressure from short sellers.
In fact, Microvision should be much more valuable because of its superior technology, more like $2 to $2.5 billion. With 150 million shares, this would correspond to a share value of about $10 to $15 per share. If the other divisions are added, it could also be $20 per share. Sales can be announced daily. This was stated by the CEO at the announcement of the second quarter 2020 results on August 5, 2020:
I note that Mergermarket recently published an article about our strategic process. The article referred to strong interest from bidders. I would like to clarify that we are engaged in discussions with certain potential interested parties who are at various stage of diligence. We do not plan to make public statement about any bid or a potential transaction unless and until and appropriate agreement is reached.
"Various stages" means that at least one conclusion could be imminent. These are not small companies, but those with sufficient money:
I can share that a focus group of top tier OEMs and technology companies are engaged in exploring and potentially pursuing strategic alternatives, which could include a sale or merger of the Company, acquisition of one or more product verticals, strategic investment and acquisition or licensing of our intellectual property. We believe parties will be particularly interested in our augmented reality and automotive LiDAR vertical.
The technology is very well rated, which should lead to a high price:
We have received feedback from multiple interested parties, recognizing that they think MicroVision technology is the most sophisticated and potentially disruptive that they have reviewed so far.
What Microvision offers is also exactly what Apple has specified as a requirement. About Apple, Reuters wrote:
“They’re not happy with most of what they see,” the first person familiar with the matter said. “They’re looking for a revolutionary design.”
A third person familiar with the matter said Apple is seeking a “design-oriented” sensor that would be sleek and unobtrusive enough to fit into the overall lines of a vehicle.
This reads exactly like a description of the Microvision Automotive LiDAR system, which will even significantly undercut the target price:
The designs Apple is seeking could potentially be made with conventional semiconductor manufacturing techniques, all four people familiar with the talks said.
That has the potential to lower prices from the many thousands to the hundreds of dollars as the sensors are produced in larger numbers, similar to chips in phones and other devices. Apple also wants sensors that can see several hundred meters (yards) down the road.
Tesla could thus consolidate or even extend its lead over competitors by taking over Microvision's Automotive LiDAR.
Probably the Microvision Automotive LiDAR was already introduced at Tesla
Although most Tesla shareholders are unlikely to be familiar with Microvision's products, the author was certainly not the first to come up with the idea that Microvision Automotive LiDAR would be a perfect fit for Tesla. Microvision has approached over 100 companies through a consulting firm. It can be assumed that Tesla was one of them.
The Microvision CEO has now mentioned in a conversation with shareholders that a request for discussion from Microvision has never been rejected by a large company in the last four years, even though Microvision is only a small company. It is known that Microvision is the technological leader in its business areas. This means that it can be assumed that Microvision was already at Tesla. And Tesla could currently be among the bidders for Microvision technology.
Another interesting tidbit, was Sumit talking about how respected and acknowledged MVIS is in the tech world amongst the big boys. (…) But his point was, and he's only been there about four years, is how remarkable and unusual it is for a tiny engineering tech start-up that when they contact the whales and say "We have something we think you will want to see". . . they GET THAT MEETING EVERY TIME. That just doesn't happen for most tiny tech engineering houses. But it does with MVIS. That lead into just a general description about how NONE of these big dogs dispute that MVIS tech, in its core competencies (i.e. LBS), is years ahead of the competition. None of them.
Conclusion
Tesla now has to decide whether or not it wants to have registerable cars for the "Full Self-Driving" mode in accordance with the requirements of the Federal Motor Transport Authorities in Europe and install LiDAR systems. If so, Tesla should not miss the chance to do so.
Microvision, one of the most innovative and exciting companies, is currently on sale due to the consequences of the COVID 19 pandemic.
Although it is unlikely to come cheap at $2 billion or more, it fits very well with Tesla's strategy for autonomous cars and could be acquired cheaply through stock exchange given the current high price of Tesla stock. For example, a stock exchange of one Tesla share for 100 Microvision shares.
Tesla shareholders would benefit from this in the coming years. In the short term, they could also benefit from the acquisition of Microvision shares, which - as mentioned above - already appear to be spot cheap only for the LiDAR division compared to the valuations of other companies such as Velodyne.
A valuation as for Velodyne would result in a share price of $13. A valuation as for Luminar would result in a share price of $19. Including the other divisions, such as smart glasses and AR, it would be approximately $20 per share. In the best case. In worst case much lower, approximately $5 per share.
Analyst's Disclosure: I am/we are long MVIS.
Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.