Site Network: Home |

Great introductory video to Hololens and Mixed Reality. Of particular note is the description of the waveguide optics at 3:43 minute mark.


Posted on June 12, 2018 by Brad Sams in Hardware with 28 Comments

It was a little over a year ago, that I broke the news that Microsoft had stopped the development of HoloLens V2 and was pushing forward to V3. Since then, I have been asked about a million times for more information about the upcoming device but if you know where to look, the company is developing the hardware in broad daylight.
Based on documents I was able to view, Microsoft is still targeting a Q1 release of the next gen headset. Additionally, the company is referring to the project internally as Sydney.
The device, according to the documents, will be lighter, more comfortable to wear, and have significantly improved holographic displays. But most importantly, it will cost significantly less than the current version of the HoloLens.
All of this is the natural evolution for the second generation of a piece of hardware but the good news is that it’s still on track for a release early next year.
What I don’t know is if the Q1 date is for general availability or for a developer preview. I could see Microsoft seeding the newer higher-end devices to developers prior to the general release of the hardware but that is only speculation at this time.
Based on the documents I was able to view, Microsoft is approaching the MR/VR market as a must-win market. This is likely because they are on the sidelines of the smartphone segment and missing out on this generation of devices would inflict serious long-term ramifications for the company about being anything more than a cloud company.
For Microsoft, they currently have an advantage in this space as they are shipping hardware but Apple, Google, Magic Leap, and many others are investing in this space heavily which means they won’t be the only competitor in this segment for long.

New patent promises to double Field of View of HoloLens v2


Besides the cost, the biggest issue with the Microsoft HoloLens is the Field of View (FoV) which at 35 degrees has been described as looking at the world through a mail slot.
Microsoft has not made it a secret that they are working on HoloLens v2, and has been explicit about the improved Holographic Processing Unit and improved Kinect-based Depth Sensing Unit.
What Microsoft has not talked about much however has been the optics of the device, but now a new patent suggests Microsoft may have achieved the breakthrough they have been after.
Titled “MEMS LASER SCANNER HAVING ENLARGED FOV”, the December 2016 patent applications explains the method below:
A MEMS laser scanner is disclosed for use in a near-eye display including an increased field of view (FOV). In embodiments, one or more polarization gratings may be applied to the mirror of the MEMS laser scanner, which polarization gratings may be configured according to the Bragg regime. Using light of different polarizations, the MEMS laser scanner is able to expand the FOV without increasing the range over which the mirror of the scanner oscillates.
The process is illustrated as below:


MEMS LASER SCANNER HAVING ENLARGED FOV

[excerpted]

Abstract: A MEMS laser scanner is disclosed for use in a near-eye display including an increased field of view (FOV). In embodiments, one or more polarization gratings may be applied to the mirror of the MEMS laser scanner, which polarization gratings may be configured according to the Bragg regime. Using light of different polarizations, the MEMS laser scanner is able to expand the FOV without increasing the range over which the mirror of the scanner oscillates.

Assignee:
Microsoft Technology Licensing, LLC (Redmond, WA, US) 
Publication Date:
06/21/2018
Filing Date:
12/16/2016
View Patent Images:

The image light is generated by a display engine 140 which emits image light in a step 300 that is modulated on a pixel-by-pixel basis by the controller 124. In embodiments, the display engine 140 may be a commercially available assembly, such as for example the PicoP™ display engine from Microvision, Inc. of Redmond, Wash.

BACKGROUND

Various types of computing, entertainment, and/or mobile devices can be implemented with a near-eye transparent or semi-transparent display. Near-eye displays may include a transparent or semi-transparent display through which a user may view the surrounding environment, and also see images of virtual objects (e.g., text, graphics, video, etc.) that are generated on the display to appear as a part of, and/or overlaid upon, the surrounding environment.
Near-eye displays have conventionally been implemented using spatial light modulation (SLM) systems including for example liquid crystal on silicon (LCoS) display engines and digital light processing (DLP) display engines for generating an image. LCoS and DLP display systems project all pixels in an image simultaneously, modulating the amplitude, phase, intensity or polarization of light across the image. Another emerging technology is microelectromechanical (MEMS) laser scanners. MEMS laser scanners conventionally include a laser light source including for example red, green and blue laser diodes directing RGB laser light to a MEMS mirror capable of deflection about two orthogonal axes.
In contrast to LCoS and DLP displays, MEMS laser scanners typically generate a two-dimensional raster scan image pixel by pixel for each image frame. The laser light source is synchronized with the bi-axial MEMS mirror drivers so that bi-axial deflection of the MEMS mirror directs laser light from the light source to the respective pixels in the raster scan, as the RGB laser light for each pixel is modulated to thus generate the desired light content of each pixel in the image.
The bi-axial range of motion of the MEMS mirror in a near-eye display laser scanner establishes the size of the field of view (FOV) that the laser scanner can generate. However, various factors impede the pivoting range of motion of a MEMS mirror during the scanning of an image frame. These factors include for example the mass of the MEMS mirror, as well as the opposing forces of air (or other gas) against the mirror surface as it pivots. Currently, MEMS mirrors in near-eye display laser scanners commonly achieve a range of motion of about 30 degrees, and an FOV of about 35 degrees.

SUMMARY

Certain embodiments of the present technology relate to a MEMS laser scanner for use in a near-eye display including an increased field of view (FOV). In embodiments, one or more polarization gratings may be applied to the mirror of the MEMS laser scanner, which polarization gratings may be configured according to the Bragg regime.
The one or more polarization gratings diffract polarized light from a laser image source in two different directions, depending on the polarization of the light, according to a time-division multiplexed scheme. The MEMS scanner pivots back and forth through its range of motion about an axis to complete one full stroke. During the time that the MEMS scanner pivots about an axis through a first half of its stroke, the laser image light may be polarized for example as LHC polarized light. The one or more polarization gratings may be tuned to allow a zero order of the LHC polarized light to pass straight through the grating un-diffracted and reflect off the MEMS mirror at an angle equal to the angle of incidence. As the MEMS scanner pivots through its first half stroke, the un-diffracted zero order light traces out a first portion of the FOV.
During the time that the MEMS scanner pivots about the axis through a second half of its stroke, the laser light may be polarized for example as RHC polarized light. The one or more polarization gratings may be tuned to diffract a first order of the RHC polarized light in reflection off the MEMS scanner at some angle greater than the mirror angle. As the MEMS scanner pivots through its range of motion, the diffracted first order light traces out a second portion of the FOV. The first and second portions of the FOV may overlap and combine to provide an enlarged overall FOV.



This block diagram of the "Interactive PicoP" was posted to the MVIS Reddit a few days ago. The items marked in GREEN appear to be MVIS proprietary components.

Vertical Licensing Strategy & Components Synergy

The company has established a licensing strategy based on the use case or application of their technology, rather than the components themselves. This strategy allows the company to leverage the exact same component into differentiated vertical solutions, with each vertical having potentially one or more licensee. 

If we look at the recently announced display-only license agreement, we can see that the company receives a $10M upfront payment for the exclusive license to manufacture and sell MVIS modules for display-only applications. The key phrase in the announcement is this one (emphasis mine):

“We believe our licensee's manufacturing prowess and use of our Laser Beam Scanning technology will assist MicroVision in lowering product costs and securing new customers in the other verticals that we serve.”


The proprietary components in the Interactive Projection application include:
  • MVIS MEMS Control & Laser Feedback ASIC
  • MVIS Video ASIC
  • MVIS Time of Flight ASIC
  • MEMS Mirror
The remaining aspects of the solution such as the laser diodes, photodiodes, power management chips, etc. we can consider to be off-the-shelf components, available from various sources. 

MVIS' proprietary components in the Display-Only applications include:
  • MVIS Video ASIC
  • MEMS Mirror
  • MVIS MEMS Control & Laser Feedback ASIC
So, 3/4 of the MVIS proprietary components are the same between the Interactive and Display-Only PicoP, with only the MVIS Time of Flight ASIC being the difference.

(It stands to reason that the Display-Only and AR solutions include largely the same set of MVIS components but it's possible they may add some new proprietary components specific for AR).

Now I'm not sure what's going to be in the LIDAR solution, but if I were to guess I'd say it doesn't need the MVIS Video ASIC as the LIDAR module's primary function is to generate "beautiful point clouds" rather than interpret an external video source.

So that would mean that MVIS' proprietary components in the LIDAR applications include:
  • MEMS Mirror
  • MVIS MEMS Control & Laser Feedback ASIC
  • MVIS Time of Flight ASIC
So what this boils down to is that 75% of the MVIS components are the same between Display-Only and Interactive Projection. And this means that volume sales of those MVIS components to the Display-Only licensee will create significant cost reductions for Interactive Projector licensee(s) and for the OEMs that go-to-market with both flavors of MVIS solutions.

Likewise, the MEMS Mirror and MEMS Control & Laser Feedback ASIC that is inherent in all of the solutions will be 100% synergetic as all of the vertical markets will use those components. So the LIDAR solutions will benefit from the adoption of Display-Only, Interactive Projection and AR solutions by collectively driving cost out of the MEMS and MEMS Control elements of the solution.

Deal Template

So, this is all good, and these benefits will accrue to all of the go-to-market partners. We can also consider the previously mentioned Display-Only exclusive license as a potential template for deals in the other verticals.

The established template looks like this:

Application: Display-Only ($$$)
Term: 5 years worldwide exclusive
Upfront License Fee: $10M
Minimum Volume Requirement: $15-20M/year in MVIS product revenue (as described on the Q1 conference call). That $15-20M per year of sales will apply to the MVIS Proprietary components in the Display-Only solution:
  • MVIS Video ASIC
  • MEMS Mirror
  • MVIS MEMS Control & Laser Feedback ASIC

Thought Exercise

We have seen that the Interactive Projection solution is the big kahuna for the 2019 "probable" revenue, rated at "$$$$" vs "$$$" for Display-Only. The note on the slide says the $ signs are 'directional and not scaled', so here is a thought exercise using the established template from the Display-Only license deal, and the quantity of dollar signs on the slide as direction for the other applications they are ready to market in 2019:

Application: Interactive Projection ($$$$)
Term: 2 years worldwide exclusive
Upfront License Fee: $20M
Minimum Volume Requirement: $20-26.66M/year in MVIS product revenue. That $20-30M per year of sales will apply to the MVIS Proprietary components in the Interactive Projection solution:
  • MVIS Video ASIC
  • MEMS Mirror
  • MVIS MEMS Control & Laser Feedback ASIC
  • MVIS Time of Flight ASIC
Application: Consumer LIDAR ($$)
Term: 5 years worldwide exclusive
Upfront License Fee: $6.66M
Minimum Volume Requirement: $10-15M/year in MVIS product revenue. That $10-15M per year of sales will apply to the MVIS Proprietary components in the LIDAR solution:
  • MEMS Mirror
  • MVIS MEMS Control & Laser Feedback ASIC
  • MVIS Time of Flight ASIC
This deal trifecta, if it were to happen according to the established template, with the relative directional sizing, would result in total upfront license fees of $36.66M, and minimum product revenue of $45-65M/year, driving volumes across the four proprietary MVIS components, well into the millions.

But, it's just as important to have the right licensees as it is to have a smart strategy, so coming out of the gate here, a lot depends upon who this unnamed-for-now Display-Only partner may be, how motivated they are to make their $10M investment pay off, and how strategic the MVIS solution is to their business. 


With a new CEO taking over, I've listened with great interest to how Perry Mulligan would position the company going forward. Microvision's new positioning has been established as "The IO of AI" which I think is an intriguing concept for sure.

The market leading AI platform is Amazon's Alexa service. When you think of an Amazon Echo speaker, its' utility really is to play music from services like Spotify, and also for smart home applications like turning on lights and things like this. Definitely fun and convenient. But for Amazon, which is of course a retailer above all else, this use case doesn't really enable their primary desire, which is to sell more goods to you.

If we could imagine the speaker to have an embedded interactive projection display, we could say something like "Alexa, show me golf shirts", which it could then project on the wall or tabletop, complete with a "touch to buy" button. Now, they could do this with an LCD touchscreen panel but that makes the whole device much larger, all the time. With Amazon's "Echo Show" product, you've basically got a 'Mom's TV in the kitchen' type of thing, which I don't think is what they really want.

I think Amazon's larger vision is to make the devices as inconspicuous as possible, so the AI service is just there, waiting for your command. An embedded interactive laser projector would allow the device to remain small and portable, and the display and interaction to be available on demand, without a big black square sitting there all the time that the display is not in use.

So, we know there is a battle royale in this space, primarily between Amazon, Google and Apple. My view is that Amazon would have the most to gain from adding interactive projection to their device, as it would make transacting on their AI platform much easier, and protect their leadership position in the space by co-opting the disruptive innovation of interactive projection.

Is this a realistic thing to expect?

In this article, Foxconn-Sharp is described as the sole manufacturer Amazon Echo devices

And as it turns out, Foxconn-Sharp also manufactures the Apple HomePod. (This article from 2016 identifies Taiwan's HTC as the Google Home speaker manufacturer.)

One week ago, Foxconn-Sharp describes establishing mass production of new, brighter green laser diodes to 6M units a year, beginning in October, with AI speaker projection displays mentioned as an "assumed application".

In the below presentation, we see the new MVIS CEO pretty boldly laying it on the line with the assertion that interactive display, embodied in the slide by the AI smart speaker application, will be the biggest contributor to 2019's "probable" revenue. "If it is easier to interact, it is easier to transact."

So, we will see. If it is indeed the case that the company's technology, the supply chain's readiness and product demand from a major player converge here, then something pretty unique may be getting ready to happen.











Sumitomo Electric self-identifies as the laser module supplier for Robohon

http://www.sei.co.jp/products/laser-module/

Utilizing the high-density packaging technology that we cultivated in the manufacture of optical communication devices, we developed a full-color laser module that houses three lasers: red, green, and blue, temperature control elements, and monitor PDs in an ultra-compact package. By providing a highly airtight cap seal and ensuring high reliability, it is expected to be applied to various applications such as pico projector, HUD, industrial light source, laser lighting and so on.

Our Products in Sharp's Robohon


Our ultra-compact full-color laser module was installed as a high-definition projector light source (equivalent to HD 1,280 × 720 pixels) in Sharp's mobile robot phone "ROBOHON" released in May 2016. With our high-density packaging technology, it is now possible to mount a laser projector in a small space of ROBOHON head (45 mm).
Robohorn
Robohon
Application example of laser projector
Application example of laser projector






HTF MI recently introduced new title on “Asia-Pacific Green Laser Diode Market Report 2018” from its database. The report provides study with in-depth overview, describing about the Product / Industry Scope and elaborates market outlook and status to 2025. 

Asia-Pacific Green Laser Diode market competition by top manufacturers/players, with Green Laser Diode sales volume, price, revenue (Million USD) and market share for each manufacturer/player; the top players including

OSRAM
Coherent
IPG
Sharp Corporation
Sumitomo
Panasonic
Hamamatsu Photonics K.K
JDS Uniphase Corp
Jenoptik AG
Newport
Rofin-sinar Technologies,Inc
Finisar
Avago Technologies
Nichia
Laser Components

On the basis on the end users/applications, this report focuses on the status and outlook for major applications/end users, sales volume, market share and growth rate for each application, including:
Heads-Up Display
Head-Mounted Display
Projectors
Smartphones
Others


Sharp raises the output of the green laser more than 4 times, realizing a bright projector


Assumed application to AI speaker, HUD, HMD

Nozawa Tetsuo = Nikkei x TECH / Nikkei Electronics

 Sharp announced that it will mass-produce in October 2018, developing an output 130 mW semiconductor laser device that emits green light (wavelength 520 nm) without using a fluorescent material. It is the first time for the company to emit green light with an output exceeding 100 mW, "the highest level in the industry with single mode products (beam shape is round)" (the company).
 Sharp announced a green light emitting semiconductor laser device in October 2017, but the wavelength was slightly bluish at 515 nm, and the output was 30 mW. This time, it increased the light output to more than 4 times and at the same time increased the purity of green. The threshold current is 100 mA. The operating current is 300 mA as standard, and the operating voltage is 6.8 V. The efficiency when converting input power to light is assumed to be 6.37%. The beam divergence angle is parallel and 7 degrees standard, vertical and 22.5 degrees same.
Sharp's new and old green light emitting laser comparison demonstration
Right newly developed wavelength of 520 nm, 130 mW output product. The left is an existing element with a wavelength of 515 nm and an output of 30 mW. You can see that the left is slightly bluish.
[Click on image to enlarge]
 The green laser device to be newly mass-produced is "GH 0521DA 2 G" of diameter (Φ) 5.6 mm TO-CAN type and "GH 0521 DA 5 G" of Φ 3.8 TO-CAN type. Both sample prices are tax included and 10,800 yen. Sample shipment in early August 2018, start mass production in early October. The mass production scale is assumed to be 500,000 pieces per month.


This website does not recommend the purchase or sale of any stocks, options, bonds or any investment of any kind. This website does not provide investment advice. Disclaimer and Notices: Disclaimer: This website may contain "forward-looking" information including statements concerning the company's outlook for the future, as well as other statements of beliefs, future plans and strategies or anticipated events, and similar expressions concerning matters that are not historical facts. The forward-looking information and statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in, or implied by, the statements. The information on this website includes forward looking statements, including statements regarding projections of future operations, product applications, development and production, future benefits of contractual arrangements, growth in demand, as well as statements containing words like believe, estimate, expect, anticipate, target, plan, will, could, would, and other similar expressions. These statements are not guarantees of future performance. Actual results could differ materially from the results implied or expressed in the forward looking statement. Additional information concerning factors that could cause actual results to differ materially from those in the forward looking statements are included in MVIS most recent Annual Report on Form 10-K filed with the Securities and Exchange Commission under the heading 'Risk factors related to the company's business,' and our other reports filed with the Comission from time to time. Except as expressly required by Federal securities laws, MVIS Blog undertakes no obligation to publicly update or revise any forward looking statements, whether as a result of new information, future events, changes in circumstances, or other reasons. Legal Notice: Although considerable care has been taken in preparing and maintaining the information and material contained on this website, MVIS Blog makes no representation nor gives any warranty as to the currency, completeness, accuracy or correctness of any of the elements contained herein. Facts and information contained in the website are believed to be accurate at the time of posting. However, information may be superseded by subsequent disclosure, and changes may be made at any time without prior notice. MVIS Blog shall not be responsible for, or liable in respect of, any damage, direct or indirect, or of any nature whatsoever, resulting from the use of the information contained herein. While the information contained herein has been obtained from sources believed to be reliable, its accuracy and completeness cannot be guaranteed. MVIS Blog has not independently verified the facts, assumptions, and estimates contained on this website. Accordingly, no representation or warranty, express or implied, is made as to, and no reliance should be placed on the fairness, accuracy, or completeness of the information and opinions contained on this website. Consequently, MVIS Blog assumes no liability for the accompanying information, which is being provided to you solely for evaluation and general information. This website does not contain inside information, proprietary or confidential information learned or disclosed as part of employment relationships or under nondisclosure agreements or otherwise.