 Okay, welcome everyone to today's training on ST Open Market approach to laser beam scanning LBS for near-to-eye displays. Today we will be hearing from Marco, and he will be starting any minute now. So Marco, go ahead and get started. Hello. Let me start here. Okay. Hello everyone. Good morning. Good afternoon. My name is Marco Angelici, and I'm presenting. I'm responsible for the MAMS Macro2ator Business Unit at ST Microelectronics, and I'm going to present today the ST approach to laser beam scanning for near-to-eye displays for the Open Market. First of all, thank you all for participating to this online developer conference event. I know this is one of the first presentation of the event. So I want to give you a welcome and thank you for joining us. Let me start with a few data. Where we start from is basically from market analysis. There are so many already mobile augmented reality users worldwide. We are talking about more than 1.7 billion people connected to the phone and using phones for augmented reality. So this is mobile augmented reality. What I'm going to talk here is how to move from mobile to wearable. So we are talking about enabling the head mounted displays for augmented reality. And talking about that, what are the key requirements we have in mind in particular? Okay. It seems that we have some technical difficulties. Eduardo, are you ready? Okay. Yes. So can you hear me? Eduardo Galicius speaking here. Okay. Perfect. So let me try to take over. And this is the beauty of having a live presentation. So I'm the backup for Marco. And okay. So Marco is describing basically key requirements for AR wearable devices. So basically here in this live, we try to summarize which are the key requirements that we have seen working with, you know, key partners and our understanding. So I will scroll down quickly, but basically we definitely see that brightness is a key requirement. We clearly understood that for factors or small factor is also extremely important in order to enable these technologies to be wearable all day long, possibly, okay, on the head. Also power consumption is also extremely important because, you know, definitely battery life would be a key requirement. And then we need to enable that. Life weight is also extremely important. So that's why having a projection module that is small compact and very light are key parameters. And then also latency and field of view are extremely important. So on these, we might have some differences depending on if you're talking about mixed reality or only augmented reality. But overall range and field of view are extremely critical parameter as well as resolution because definitely we want to enable at the end of the day an all day pair of glasses that people will not only like to go around with, but will provide also bright, crispy and good looking images with the additional information, the augmented information that will be available to the final user. And at the end, an important, very important parameter is also the IBOX side. Because again, if these information are too small or we need to be basically constrained, but by a too small IBOX, we might not be able to really reach the wider population. Overall, if we look at all these requirements, what we did in the past, now I'm here, you will see a slide that describes that is how we can enable all of that. And so if we move to the laser beam scanning technology, this is, we believe, the key enabling technology that can really enable all those requirements that we described with the previous slide. So basically, laser beam scanning for near to high displays are really the technology that can enable these multiple factors that can enable compact illumination sources that can enable pretty high brightness. And for this, of course, we have to work with partners for the waveguide and so on. And you will see, we will present the page on that. And also, these laser beam scanning technology enabled by MAMP's mirror as a key component, fundamental component, can also enable low power systems and small, lightweight optical engines. Let me say that as you can see here, there are a couple of real features. So here you can have an idea in perspective to a family, what we are talking about MAMP's mirror and full package on a flex PCB. Then you can see here in relation to a quarter of dollars, the full optical engine that we are able to enable with our components. And on the right, you can see basically the operating principle of a laser beam scanning technology. So we're talking about a complex solution at the end of the day. So there are a lot of pieces involved here. There are semiconductor pieces, there are optical components, and there are also specific manufacturing and precision in assembling all of these components. But overall, and you will see later with the presentation, we can enable all of these with a solution that really we believe is the best today in the market. And I don't know if Marco is back. Can you hear me? Perfect. Yes, Marco. So you're back. So do you want to continue? Okay. Thank you. Sorry for that. So this is having six kids homeschooling at home. Sorry for that. So let me continue for that. And at the base of our MEMS technology for laser beam scanning, the key thing are we are talking about MEMS mirrors. And so when we talk about MEMS at SD, we are talking about the long history as he has been developing manufacturing and selling MEMS for 20 years. You can see from this graph, an example with purple color, starting from actuators, MEMS actuators in 2000 for microfluidics, and then moving with the old wave of inertial sensor, environmental sensor, microphones, etc. To go back again to big investment in the actuators technology, in particular for MEMS mirror, MEMS speaker, autofocus cameras and ultrasound piezo actuators. We are committed to volumes, 24 billion devices have been shipped in those 20 years in terms of MEMS and is making SD a leader, an absolute leader in this market. Talking about again MEMS laser beam scanning, we say the key pillar is MEMS mirror and SD is the only company selling, producing, developing and selling all kind of actuated MEMS mirror from electrostatic to electromagnetic and piezoelectric. So SD can cover the whole band of requirements of our customers from low power to high performances, capability to move big masses, very high field of view. So we can really cover the full flavor in terms of high field of view and in terms of power consumption. Sorry to stay here for a while. SD is the undisputed leader with 12 million mirrors shipped to date on the market. Where do you find our mirrors? I just put the flavor of some components where you can, some products where you can find our mirrors, our laser beam scanners, starting in 2012. So this is what is publicly available, what can be found publicly. So we acquired Bitendo in 2012 with all the IPs on laser beam scanning technologies. And then since then we have been developing and taking all the applications from Pico projection to 3D scanning and 3D scanning for inductor, consumer up to automotive. We also have worked and we are selling our laser beam scanning in the augmented reality world, of course, for example in the medical augmented reality. But also you can find SD manufactured MEMS mirror in different augmented reality customers. The first customers in the world announcing MEMS mirror laser beam scanning technologies like North recently acquired by Google is using SD MEMS mirrors. And recently tech inside showed that inside also the HoloLens tool you can find two MEMS mirrors that are manufactured by SD. We continue this activity and the reason why we are the leaders here is because SD has a full portfolio. It can't be considered one stop shop for laser beam scanning for augmented reality and not only for augmented reality from our customers standpoint. In particular, if you look on the left side here you see a block diagram. Basically, a laser beam scanning system is made of an electronic subsystem that is basically controlling video processing and controlling lasers and controlling the mirror to deflect the lasers in the right place and to get the feedbacks and the control loops. And on the right side you see the optical engine where mirrors, lasers and the optics are basically made. And SD has invested in all those domains, both in the electronic and the optical engine. And SD is providing the full components for both. But the laser diodes that is the only part I would say is missing in this picture because SD is not in the laser diodes business. But as I listed on the right, out of MEMS mirrors, so it's not just MEMS mirror, out of MEMS mirrors SD is developing and selling the MEMS drivers for each of the technology I was mentioning before, electromagnetic, piezoelectric, static. And in particular we focus on energy recovery and performances. So thanks to our piezo mirrors, we have developed energy recovery adiabatic drivers able to cut power consumption to a very low number. It is a fundamental for this application. Talking about laser diodes, we discover also laser diode drivers, we discover that we were missing the right performances in the market and we decided to invest also there. We have a family of products that are the best in class. We have a product with 300 picosecond, rice time and full time with an ultra low power mode of functionality, few nanoseconds to go in and out of sleep mode so that we can really switch off the device when it's needed. Where few pixels are black and there is no projection on those pixels, the laser diode are really switched off. And basically this is enabling the full system performances per walk is required. The pixel clock is 300 megahertz, so again it's the best in class family for laser diodes for this application. And we have developed two devices. One is three channels for RGB, red, green, blue, and one is four channels to include infrared that can be used for several applications. On top of that there is also the, as I mentioned, control loop and the video control. So as the in-house full capability and again coming back to the acquisition of V10 in 2012, we started from those IPs and we developed more and more IPs here in the hardware and software control loop for mirrors, for lasers, eye safety, synchronization between binocular displays projection and all the video processing capability. So this is all the service we provide to our customers. Last but not least, we have recently invested also in the relay optics. Relay optics is the piece of optics or the optical components that shape the beam out of a laser beam scanner in order to maximize the in-coupling to the final length, to the final combiner element that is typically a wave guide. So working on laser beam scanning for augmented reality requires an holistic approach. We are talking about STL to simulate with our partners from lasers to the eye. So we simulated and we developed the components in order to maximize the performances, maximize the MTF, the color uniformity, the in-coupling efficiency, and to reduce at the minimum the power consumption. For that reason, it's important to work in an ecosystem. It's not just ST working alone. We have been working for a couple of years with some key partners in the domain of augmented reality and recently we officially announced that we started this laser alliance, laser scanning for augmented reality is the meaning of a laser. And basically we built an ecosystem with at least the first start, starting companies that you can see listed here like Osram and Dispelix, Applied Material, Megawand. They are all leaders in their own segment building the building blocks basically for augmented reality. And as I said before, it's an holistic approach that can enable the mass production at the end of the high quality, high performance components. It's an open alliance. It has just started. More partners are going to join. We have a very good feedback from the market. We have very good feedback from our customers. And again, more and more partners are willing to join with the same purpose of us that is basically to build and create an augmented reality space based on laser-being scanning solutions in order to enable this market that is start ramping up now. Thanks to this alliance and to go a little bit on details on what we have developed so far and we are basically accelerating the what we call the STAR program, the STAR roadmap as the augmented reality roadmap. In January and March of this year, we started with the very first tool of concept and demonstrators to show the possibility to match a laser-being scanner with a waveguide for an open market application. As I said, our customers did already in the past. We wanted to work on something that is smaller and affordable and targeting, as we say, an old-day wearable market. So to do that, we built, as I said, in January and February, some proof of concepts. And now, finally, we have an engineering sample of what we call the STAR-Zero that is available now based on diffractive waveguide, a little static mirrors, and focusing on the miniaturization of the optical engine. So to match real old-day wearable small pair of glasses, lightweight glasses. With that, we work with our partners in the laser diode. They built a module that is very compact. And of course, we work on our laser driver to match these laser diodes and to match also performance-wise and size and consumption-wise what is necessary to achieve those performances, Of course, we have a roadmap. And from STAR-Zero, we move to STAR-1. It is for next year. In the second half of next year, we have already a plan. And more than a plan, we have already some IPs ready and some components validated, where we are moving to our thin-field piece of mirrors in mass production. And thanks to this move from electrostatic mirrors that are in using the STAR-Zero device. Going into STAR-1, we will cut heavily the power consumption. We will reduce the volume compared to something that is already very small. And I'm going to show you in the next slide what does it mean. But we are also, again, improving performances. So we have a roadmap thanks to our thin-field piece of technology that is improving performances in terms of field of view, in terms of power consumption, in terms of resolution, reducing the volume occupation and reducing the power consumption. That's really where we want to be. And that's why we believe laser beam scanning, thanks to the scalability, you can have one mirror opening more. You can increase the opening angle and then you can increase the resolution by working on the frequency of the mirror. You can really increase the scalability without increasing volumes, occupation, and power consumption. Just to show some of the different steps we put in place, here is a presentation, a slide on what we have done in March. It was a technology demonstrator where we took an existing picoprojector from a steam 1.7 cubic centimeter. It was not optimized for recommended reality glasses. And we still built a reference design and worked with this Pelix for a waveguide connection. So we built a pair of glasses that was basically, again, just a technology demonstrator. The waveguide used at that time was a 30-degree field of view waveguide capable of 20 nits per lumens. Resolution was 540 pixels. Our MEMS mirror are electrostatic and the power consumption for these MEMS staff mirror plus driver was in that 100 milliwatts, just to give you an indication. Still using the big discrete laser diode, the 3TO can package laser diodes as into this picoprojector on the bottom left side. And using of the shelf components, basically, we were able to demonstrate, as you can see from a picture in our lab taken from a phone. So it's not a super high quality picture. But what is interesting, at least, we were able to show the capability to project through the laser beam scanning into a diffractive waveguide. Color uniformity here is not perfect, but the images are already good and it's a good starting point. Consider also the waveguide who was a waveguide designed for fixed pixel display like DLP and Delcos. So it was not optimized for laser beam scanning. Since then, we have moved to the next generation, the TSTAR-0, focusing on the dimension of the module. So we developed a new module moving from 1.7 cubic centimeter to something that we call Helen optical engine, much, much smaller. We are talking about 0.75 cubic centimeter. And as you can see from this picture on the right, this is perfectly matching the dimension of a temporal arm for an eyewear solution that is an all-day wearable. What we have built today, we have the engineering samples of this part. I will show you the results later in the next two slides. We are still limited to 30 degrees by a field of view by the waveguide. The optical engine itself with our mirror is already able to achieve a 56 degrees of field of view. So means we are ready also for the next generation waveguides where 40 degrees and maybe 50 degrees will be available. Even if for all the wearable application, we believe that 30 degrees is already a good use case. And we believe that is enough for the big volume of the consumer market. The big difference versus the previous solution on top of the dimension of the module is also on the waveguide side. The efficiency has been improved to 100 nits per lumen. Actually, we are going to receive a 150 nits per lumen at the end of this year. So brightness can be already in the range of 1,000 nits today and will be in the range of 1,500 nits at the end of this year. Just to give you a reference, they are giving HoloLens 2 from Microsoft at 550 nits. So we are targeting all the wearable outdoor application we want to be, as we said at the beginning, in the 1,500 nits brightness. The last point below is on top of the volumes occupations 0.75 cubic centimetre that is making laser beam scanning the smallest display for augmented reality. We also, as I say, we built a new laser diode driver at 300 picosecond riser full time matching this application. And we were able to match and combine this laser driver with a laser diode from our partner Osram in this case that was embedding the three laser edge emitting diode into a single module, a single package with very small dimension. Thanks to that you can see here the integration made by ST. So ST built this reference board. We built all the electronic board. It's 65 millimetre by 10 millimetre dimension. Again, it's a reference design. And we built this black box that is the optical engine. You can see here on the right side, you see some more details. You can see the optical engine is not just lasers and mirrors. The optical engine has all the collimation and combining optics required to collimate the beam out of the lasers and to combine the three colors onto the mirror. On top of that, it also embeds, as you can see, three and four year as a photodiode and the temperature sensor that are helping and necessary to close the loop in order to keep the color control and to keep the temperature performances and the loop for the lasers. And then to keep the performances at the right level. So it's an ultra compact module. It's not just mirror. It's not just lasers. It's embedding the full system to close the loop and control the system. It's just the electronics is missing. That is part of the green board you see in the middle in the center. On the left side, you have a sketch from the laser diodes module from Osram. And of course, the waveguide from these pelix that basically we are today with whom we are working today to integrate and to evaluate the full performances with this small projector. I show you a few pictures so that the start zero exists. We have a few samples now assembled. We have a few units assembled. So we are at the very beginning, but all the tests are very good. You see here some images from our lab in Israel where basically we are able to align and collimate the laser beams and we are able to get out the right performances, right alignment and right balance in terms of color. I show also down an example of a projection. This projection is on the wall. So this is a little projector projecting on the wall. We are able to get a very, very good quality and very good font readability. So the contract is very good. The sharpness is very good. The color balance is very good. And most important, there is no speckle that is one of the key question mark, typically, that may arise when using laser beam scanning solution. What's next? In our roadmap, so start zero is what we have today as samples. We are already showing to key customers. We are going to start delivering parts to our selected customers in January. And as per the alliance that we have built and as per some press releases that we've come soon, you will see a number of partners entering into the game. In particular for manufacturing. So we will have a partner in Taiwan. We already had a joint meeting and the webinar so we can mention the partner is quanta today. Quanta is working to build a full pair of glasses by Q1 next year based on star zero and full pair of glasses means including all the components. And the on this reference platform you're using all is the sensors is the red RF circuits and also the application processor being a now augmented reality glasses for infographic and symbology. It's using an STM 32 application processor. What's next as I mentioned is a star one where basically by middle next year, we will have a very nice evolution of what we believe will be really a killer application and the killer solution for augmented reality. Because we are going to have, as I mentioned, together with new waveguides already available by the end of this year at 150 days per lumens. We will have a true 720p resolution. And thanks to our pin film piezo actuated mirrors, we will have an ultra low power actuation technology. So we will be able to move sense and control the mirrors with 100 milliwatt. That is an unbeatable result. So none nobody in the market can achieve this number already today. With such they mentioned that will be even smaller because those those mirrors will be smaller than the electrostatic mirrors use it into star zero. So we are going to have an estimated 0.65 cubicle centimeter optical engine. The electronics for star one is incorporating most of the passive components and all the external components will be embedded in the driver in the new driver with energy recovery charge. So we will have even a smaller electronic PCB and components. The bill of material, let me say in terms of number of components in terms of cost and also in terms of dimension. And most important, the full solution, not just the mirror, but the full solution, including the electronics, including the lasers, including the mirror, including the control loop will be 50% power consuming versus star zero. So it is already state of the art and best in class power consumption module for laser scanning. So star zero is already below one watt. That was the target we were talking about at the beginning of this presentation. We want to be below one watt. What will happen with star one is that we will be below one watt for a binocular application. So two displays will be able to produce a video projection for near to high displays in less than one watt. And this is really something outstanding and for which we have seen the big traction on laser scanning. And we see several customers coming to us asking for samples and asking for the next generation. I'm not talking here about what next. But it's interesting to see that thanks to our technology, we are already developing higher force capable piezo actuators. That means a better meter per volt displacement force. So we call a degree one of a piezo as the picometer per volt. So it's the capability of displacement as the roadmap in particular in a design center and are in this center. We have just built and for which you will see an announcement on the 28th of October. We have already in place the scalability for next generations. So we are not just closing here. We are basically at the beginning of a journey to complete a bigger, a bit of a picture when we move from star zero to star one, I was talking about going back. I was talking about 30 degrees 35 degrees of field of view. What is interesting here is that our projector. So removing the waveguide, but the projectors as I mentioned before are already capable of more field of view. So star zero is made of two electrostatic mirrors as I was measuring. And as you can see the diagonal field of view of this optical engine, it's already 56 degrees field of view. So it's already the state of the art for augmented reality. When we talk about star one, we are talking about a 65 degrees field of view diagonal with an increased efficient resolution at true 720p resolution and the basically lower voltage and lower power consumption and lower size, lower real estate enabling the next generation module to be at 0.65 optical centimeter. Last, but not least the drivers. I mentioned that is not just a matter of man's mirror, but the technology is combining with very well with energy energy recovery drivers. And we have built already some IPs and we have lots of patents on driving in 50 years of mirrors and controlling them. We have already available for customers that we gave already to some of our customers demonstrator over one over six. So a six time saving power consumption of our rational mirrors that is the more consuming the most consuming mirror. And we are building we are building now a system in package with a new IP that is able to achieve one over 10. And for that reason, thanks to the combination of the driver and the piece of mirror, we are able to achieve the very low power consumption. I was mentioned before for star one. So combination of mirror combination of the driver is enabling the technology roadmap and the product roadmap for next generation augmented reality for which we see a brighter future. So we see really a big interest and we see here now available today a solution that is enabling already small lightweight all day wearable glasses. That's all from my side and let me thank you and again, I will ask you to stay tuned because more things will happen in the next month and we will have more and more updated updates also in terms of partners joining the Alliance Alliance so we expect to have more and more information coming and more and more exciting partners joining us in this journey. Thank you. And thank you for participating.