**Lighthouse: The Next Generation of Positional Tracking**

Now that's just a cup that's the optimal point for the two base stations, five meter range, and the software skip how its scaled so far right, correct? Yes, so that's a good shippable solution. It gives you a good experience in full 360 tracking basically you can't it's very difficult to include it you can experience it yourself right. It's you know there's a matter of you you have the two controllers overlapping each other they they're smart enough and it's not very computationally intensive because you said it's just basic math triangulation.

The room in that system like how do how do you know where the walls are in the room? Okay so the walls in the room are they no, because the base stations give you an absolute coordinate system you can work out where the walls are. So at the moment we have a calibration procedure where you use the headset or the controller to actually measure things in the world. You measure the ground you can measure go okay I don't want to go any further than this click and that gives you the geometry of the room. Now there are other techniques so we want to talk about in the future about how you would work out saying wear something that's moving in the room might be but for the moment fix geometry is pretty simple to scan in just by touching off points and I imagine the patient's need to be locked in place you don't want those things to move at all.

Now can they be self-power like with a USB cable and a power brick and yeah that that's exactly how they design. They don't talk to the computer at all they just simply produce the signal that the tracking tracked objects can interpret so we've got ones that will run off a power brick like like a battery in for about a day but there'll be wired they'll take power and that's pretty much it.

These things are going to be in great number out in the world very soon but essentially by Christmas so a lot of people are going to want to use them for non VR applications robotics and micro flight applications like what copters. Obviously a very obvious way they didn't want to use absolute tracking system we want to facilitate that by I mean anyone can licensed lighthouse really. The only entrance into lighthouses to not violate the standards so that all systems are can interoperate with each other will be offering technical information about how to utilize the system will be offering as much as possible their components is off-the-shelf device things that you just buy digi-key and put into your project.

The limitations of lighthouse for for the space that you can use it and I could you have a set up in a room like this? Yes you can totally set up in a room like this there obviously take a fair number of base stations with a current design to operate out to that. There are many ways that you can scale lighthouse including you can increase the range by making it give you slower updates for example one of the like the limitations is the sink we can power the lasers already can go out to 20 meters easily so isolate the 20 meters radially it's huge right it's a 40 meter circle that's that's like a stadium yeah whereas the sink blinkers are a little bit weaker than that. You can have sync repeaters as a bunch of different ways you can make it scale or you can use RF now that's things that we should probably talk to the community about how they're actually going to use it and they can give us feedback on what's probably the best way to take the technology.

Who else have you talked to about possibly licensing lighthouse for their obligations is it consumer electronics is it research. We've been a lot of people who spoke to us about it a lot of researchers a lot of consumer developers as well and hopefully their experiences will feed back to how we can use it for to improve the VR experience as well absolutely so all all users of lighthouse will feed back into the the general you know protocol and set up at the system will probably come up with some kind of standards committee that will have to be the steward at the standard going forward.

**The Future of Lighthouse**

We've been a lot of people who spoke to us about it a lot of researchers a lot of consumer developers as well and hopefully their experiences will feed back to how we can use it for to improve the VR experience as well. Absolutely so all all users of lighthouse will feed back into the the general you know protocol and set up at the system will probably come up with some kind of standards committee that will have to be the steward at the standard going forward.

**Conclusion**

Lighthouse is a next generation positional tracking technology that is designed to provide accurate and reliable tracking for a wide range of applications. It has been developed by Oculus VR, and its technical details were recently explained in an interview with Alan. The system uses a combination of lasers and software to track the position of objects in 3D space, and it can be used in a variety of settings, including virtual reality environments.

One of the key benefits of Lighthouse is its ability to provide accurate tracking over long distances, making it ideal for applications such as robotics and micro flight. It also has the potential to improve the VR experience by providing more realistic and immersive interactions between users and virtual objects. The technology is still in development, but it shows great promise for the future of positional tracking.

**About Lighthouse**

Lighthouse is a next generation positional tracking technology developed by Oculus VR. It uses a combination of lasers and software to track the position of objects in 3D space, and it can be used in a variety of settings, including virtual reality environments. The system is designed to provide accurate and reliable tracking for applications such as robotics, micro flight, and more.

**The Future of Positional Tracking**

Lighthouse represents an exciting new development in the field of positional tracking technology. Its ability to provide accurate and reliable tracking over long distances makes it ideal for a wide range of applications, from virtual reality environments to robotics and micro flight. As the technology continues to develop, we can expect to see even more innovative uses for Lighthouse in the future.

**Innovative Applications**

Lighthouse has the potential to revolutionize a wide range of industries and applications. From virtual reality environments to robotics and micro flight, this technology is poised to make a major impact on the world. As researchers and developers continue to explore the possibilities of Lighthouse, we can expect to see even more innovative uses for this technology in the future.

**Conclusion**

Lighthouse is an exciting new development in the field of positional tracking technology. Its ability to provide accurate and reliable tracking over long distances makes it ideal for a wide range of applications, from virtual reality environments to robotics and micro flight. As the technology continues to develop, we can expect to see even more innovative uses for Lighthouse in the future.

"WEBVTTKind: captionsLanguage: enhey it's norm from Testim I'm here at Maker Faire 2015 and we ran into valve software this is Alan gates you're working on valve stem VR project specifically the lighthouse right correct yes I was basically the architect of the lighthouse tracking system and lighthouse is one of the things that may seem VR when we try to GDC this year with the HTC vive it's such a great experience first a quick primer on how lighthouse works that's a lighthouse Space Station key tell me about the anatomy of this okay so here you have the sync linker this is an infrared beacon that essentially floods the entire space it's a track that emits timing synchronization pulse that then all the receivers in the room can use to time the sweeps of laser light as they cross and determine their angle to the base station so here you have rotating we column rotors or fly wheels depending on whether you're a mechanical machine or not but basically you've got laser beam coming down there's a mirror inside that diverts the light out through this long lens this generates a fan of light and the fan of light moves across the world horizontally and vertically in sequence from that the sensor out in the world can determine its angle to the base station and if you have more than one sensor or more than one base station in one sensor you can calculate a position in three-dimensional space and in terms of the sensor or what type of sensor is it for example this is a prototype of what's going on top of the CPR controller and you have a lot of these sensors yes this object has 19 sensors on it each one will individually pick up whatever base station signals that it can see and from that you get a bunch of angles that say okay this base stations at you know from the base station to my sensor I'm at and azmuth and an elevation relative to that by knowing the configuration of sensors on this object we can do you know a triangulation problem solved to work out not only its position but also its orientation in space so we think of this as a for example a smartphone picking up satellite signals from GPS satellites they're orbiting the Earth and triangulating its position so a much higher degree of accuracy now these sensors have are they expensive to make a lot of logic that goes with them now they're relatively simple each sensor has a photodiode now we've also invested quite heavily in making just an ASIC solution for baking down all the electronics so it will be very small and very simple thing to implement basically someone will just grab one of these photo diodes one of these pieces of silicon they can stick on a board and they're done and that whole system here to import maybe a profile how these sensors are arranged the structural form of this is I mean that it has to be in a rigid form it can't change throughout the use yeah and so for the basic tracking that we're talking about here yes it has to be rigid you can track angles on on to single points if you have more than one base station in view but that they don't have the spatial redundancy as much less you have multiple base stations so finger tracking would be problematic anyway because it's very easy to include your hands but for evens like the back of your hand you can put enough sensors on that to track it reasonably well now as far as requiring five sensors to get a pose yeah you do initially require five sensors to get a pose once you have a pose we can do lock in tracking a bit better than that in terms of the virtual world then but these sensors can be on anything they can be on fixed objects chairs keyboards mice so you know when you're wearing a HMD wear those row objects are relative to you in the space right absolutely so we're putting a lot of work now into optimizing the sensor placement and sensor covering so that you can disguise them in an object and put around the edge of your cell phone or in your keyboard there's obviously lensing effects associated with optical systems that we need to need to be able to manage and we're getting pretty good at that we already have as far as designing an object we have software if you give us like an STL file you have straight for your 3d printer we can work out what the optimum arrangement of sensors is and show you how well it will track we have software to do that and we'll probably release that fairly shortly now were you there Viet Valle when you experiment the QR code room yes absolutely so our early trading system was the QR codes but obviously that's not a very shippable solution very few people make me some enthusiasts would love to paper their walls with those things but most people that's not kind of why how did you get from the leap from place or a wall with QR codes for marking where it's an optical system where the headset is looking at the room to basically pulling in signals so it's sort of a mathematical duel right you can turn the whole system around inside out and it's still a sort of an inside-out tracking system but it's using a beacon instead of a QR code on the wall it's far more shippable solution because instead of having to pay for things everywhere you can just put a couple of beacons up and you've got tracking wherever you can see the beacons now the beacon call your car lighthouse because it's a laser emitter how does it map the room or how does it spread out across the room so basically it has a spinning mirror that makes a fan beam of laser light that sweeps across the room so that works in one dimension in two dimensions you need two beams and you can have an object with sensors on it that can determine its orientation by essentially triangulation it's pretty simple mathematically but it's not completely straightforward when you start talking about three dimensions and multiple poses and you've developed it so you only need two beacons and a certain number of the trackers on whatever you want to track in order to get real accurate low latency positional accuracy what does that trade up how many markers you need so you actually only need one base station and up to five central basically five senses to get a good pose to start with once you've got as tracking pose is like a lock you can then actually use less sensors course you can coast on the IMU for a little bit we have in experiments with the track controller we've got a right down to one sensor and the IMU as long as you keep moving it around it tracks reasonably well three is basically a minimum for a good pose if you're taking information not just from the sensors receiving the light from the beacons combining other information of the IMU to compensate whenever there is obstruction because you need line of sight yes yeah it does need line of saw like all optical systems it can be included so the idea of having more than one base station is primarily redundancy not actually accuracy in the case of when you've got controllers in front of you if you turn around away from the base station they can't see the base station anymore but if you have another view another angle that the lights coming in from then it's much more difficult to occlude them and do the best issues communicate with each other in terms overlapping so there or is it just unique signals they can so there's a couple of different models actually a lot of different modes that lighthouse can operate in the modes that you've seen at GDC they were synchronized with each other by or a wire that it's not the only mechanism they can operate in they can operate asynchronously from each other where they run at different speeds or they can use differ area frequencies where there's sort of synchronous or that can be completely synchronous as well in all these different modes are just different ways to either increase accuracy or scale so that the kids can work each other so scalability is a big thing with the lighthouse system because you talked about needing you you only really need the one in the five sensors but you can have more lighthouses how to envision it being scaled out so it was designed from day one to be very scalable because basically the beacons currently have a range of about five meters that's primarily limited by the optical power of the the synchronization signal the in order to scale you can basically just add more that the software for that is still somewhat work in progress but the the general design of the system completely allows you know any number of base stations to be used so the current talking point of you know a 15 foot by 15 foot room right now that's just a cup that's the optimal point for the two base station that five meter range and the software skip how its scaled so far right correct yes so that's a good shippable solution it gives you a good experience in full 360 tracking basically you can't it's very difficult to include it you can experience it yourself right it's you can almost put the thing under your arm and it still works yeah yeah you know there's a matter of you you have the two controllers overlapping each other they they're it's smart enough and it's not very computationally intensive because you said it's just basic math triangulation now how does the room then in that system like how do how do you know where the walls are in the room okay so the walls in the room are they no because the base stations give you an absolute coordinate system you can work out where the walls are so at the moment we have a calibration procedure where you use the headset or the controller to actually measure things in the world so you measure the ground you can measure go okay I don't want to go any further than this click and that gives you the geometry of the room now there are other techniques so we want to talk about in the future about how you would work out saying wear something that's moving in the room might be but for the moment fix geometry is pretty simple to scan in just by touching off points and I imagine the patient's need to be locked in place you don't want those things to move at all now can they be self power like with a USB cable and a power brick and yeah that that's exactly how they design they don't talk to the computer at all they just simply produced the signal that the tracking tracked objects can interpret so we've got ones that will run off a power brick like like a battery in for about a day but there'll be wired they'll take power and that's pretty much it now you guys are Maker Faire cuz you're obviously looking at using lighthouse in applications outside of virtual reality how do you envision people Baker's using lighthouse as a positional tracking system very good questions so obviously these things are going to be in great number out in the world very soon but essentially by Christmas so a lot of people are going to want to use them for non VR applications robotics and micro flight applications like what copters are obviously a very obvious way they didn't want to use absolute tracking system we want to facilitate that by I mean anyone can licensed lighthouse really the only entrance into lighthouses to not violate the standards so that all systems are can interoperate with each other will be offering technical information about how to utilize the system will be offering as much as possible their components is off-the-shelf device things that you just buy digi-key and put into your project you mentioned quadcopter so is that is that something like outdoor use what are the limitations of lighthouse for for the space that you can use it and I could you have a set up in a room like this yes you can totally set up in a room like this there obviously take a fair number of base stations with a current design to operate out to that there are many ways that you can scale lighthouse including you can increase the range by making it give you slower updates for example one of the like the limitations is the sink we can power the lasers already can go out to 20 meters easily so we isolate the 20 meters radially it's huge right it's a 40 meter circle that's that's like a stadium yeah whereas the sink blinkers are a little bit weaker than that you can have sync repeaters as a bunch of different ways you can make it scale or you can use RF now that's things that we should probably talk to the community about how they're actually going to use it and they can give us feedback on what's probably the best way to take the technology who else have you talked to about possibly licensing lighthouse for their obligations is it consumer electronics is it research we've been a lot of people who spoke to us about it a lot of researchers a lot of consumer developers as well and hopefully their experiences will feed back to how we can use it for to improve the VR experience as well absolutely so all all users of lighthouse will feed back into the the general you know protocol and set up at the system will probably come up with some kind of standards committee that will have to be the steward at the standard going forward so awesome thank you so much Alan for explaining a little bit more about the lighthouse how it works and your vision for how it might be used in VR and other places thank you so that's it for Maker Faire 2015 with more stuff from this show on test comm please subscribe to our YouTube channel on norm Justin only see guys next time byehey it's norm from Testim I'm here at Maker Faire 2015 and we ran into valve software this is Alan gates you're working on valve stem VR project specifically the lighthouse right correct yes I was basically the architect of the lighthouse tracking system and lighthouse is one of the things that may seem VR when we try to GDC this year with the HTC vive it's such a great experience first a quick primer on how lighthouse works that's a lighthouse Space Station key tell me about the anatomy of this okay so here you have the sync linker this is an infrared beacon that essentially floods the entire space it's a track that emits timing synchronization pulse that then all the receivers in the room can use to time the sweeps of laser light as they cross and determine their angle to the base station so here you have rotating we column rotors or fly wheels depending on whether you're a mechanical machine or not but basically you've got laser beam coming down there's a mirror inside that diverts the light out through this long lens this generates a fan of light and the fan of light moves across the world horizontally and vertically in sequence from that the sensor out in the world can determine its angle to the base station and if you have more than one sensor or more than one base station in one sensor you can calculate a position in three-dimensional space and in terms of the sensor or what type of sensor is it for example this is a prototype of what's going on top of the CPR controller and you have a lot of these sensors yes this object has 19 sensors on it each one will individually pick up whatever base station signals that it can see and from that you get a bunch of angles that say okay this base stations at you know from the base station to my sensor I'm at and azmuth and an elevation relative to that by knowing the configuration of sensors on this object we can do you know a triangulation problem solved to work out not only its position but also its orientation in space so we think of this as a for example a smartphone picking up satellite signals from GPS satellites they're orbiting the Earth and triangulating its position so a much higher degree of accuracy now these sensors have are they expensive to make a lot of logic that goes with them now they're relatively simple each sensor has a photodiode now we've also invested quite heavily in making just an ASIC solution for baking down all the electronics so it will be very small and very simple thing to implement basically someone will just grab one of these photo diodes one of these pieces of silicon they can stick on a board and they're done and that whole system here to import maybe a profile how these sensors are arranged the structural form of this is I mean that it has to be in a rigid form it can't change throughout the use yeah and so for the basic tracking that we're talking about here yes it has to be rigid you can track angles on on to single points if you have more than one base station in view but that they don't have the spatial redundancy as much less you have multiple base stations so finger tracking would be problematic anyway because it's very easy to include your hands but for evens like the back of your hand you can put enough sensors on that to track it reasonably well now as far as requiring five sensors to get a pose yeah you do initially require five sensors to get a pose once you have a pose we can do lock in tracking a bit better than that in terms of the virtual world then but these sensors can be on anything they can be on fixed objects chairs keyboards mice so you know when you're wearing a HMD wear those row objects are relative to you in the space right absolutely so we're putting a lot of work now into optimizing the sensor placement and sensor covering so that you can disguise them in an object and put around the edge of your cell phone or in your keyboard there's obviously lensing effects associated with optical systems that we need to need to be able to manage and we're getting pretty good at that we already have as far as designing an object we have software if you give us like an STL file you have straight for your 3d printer we can work out what the optimum arrangement of sensors is and show you how well it will track we have software to do that and we'll probably release that fairly shortly now were you there Viet Valle when you experiment the QR code room yes absolutely so our early trading system was the QR codes but obviously that's not a very shippable solution very few people make me some enthusiasts would love to paper their walls with those things but most people that's not kind of why how did you get from the leap from place or a wall with QR codes for marking where it's an optical system where the headset is looking at the room to basically pulling in signals so it's sort of a mathematical duel right you can turn the whole system around inside out and it's still a sort of an inside-out tracking system but it's using a beacon instead of a QR code on the wall it's far more shippable solution because instead of having to pay for things everywhere you can just put a couple of beacons up and you've got tracking wherever you can see the beacons now the beacon call your car lighthouse because it's a laser emitter how does it map the room or how does it spread out across the room so basically it has a spinning mirror that makes a fan beam of laser light that sweeps across the room so that works in one dimension in two dimensions you need two beams and you can have an object with sensors on it that can determine its orientation by essentially triangulation it's pretty simple mathematically but it's not completely straightforward when you start talking about three dimensions and multiple poses and you've developed it so you only need two beacons and a certain number of the trackers on whatever you want to track in order to get real accurate low latency positional accuracy what does that trade up how many markers you need so you actually only need one base station and up to five central basically five senses to get a good pose to start with once you've got as tracking pose is like a lock you can then actually use less sensors course you can coast on the IMU for a little bit we have in experiments with the track controller we've got a right down to one sensor and the IMU as long as you keep moving it around it tracks reasonably well three is basically a minimum for a good pose if you're taking information not just from the sensors receiving the light from the beacons combining other information of the IMU to compensate whenever there is obstruction because you need line of sight yes yeah it does need line of saw like all optical systems it can be included so the idea of having more than one base station is primarily redundancy not actually accuracy in the case of when you've got controllers in front of you if you turn around away from the base station they can't see the base station anymore but if you have another view another angle that the lights coming in from then it's much more difficult to occlude them and do the best issues communicate with each other in terms overlapping so there or is it just unique signals they can so there's a couple of different models actually a lot of different modes that lighthouse can operate in the modes that you've seen at GDC they were synchronized with each other by or a wire that it's not the only mechanism they can operate in they can operate asynchronously from each other where they run at different speeds or they can use differ area frequencies where there's sort of synchronous or that can be completely synchronous as well in all these different modes are just different ways to either increase accuracy or scale so that the kids can work each other so scalability is a big thing with the lighthouse system because you talked about needing you you only really need the one in the five sensors but you can have more lighthouses how to envision it being scaled out so it was designed from day one to be very scalable because basically the beacons currently have a range of about five meters that's primarily limited by the optical power of the the synchronization signal the in order to scale you can basically just add more that the software for that is still somewhat work in progress but the the general design of the system completely allows you know any number of base stations to be used so the current talking point of you know a 15 foot by 15 foot room right now that's just a cup that's the optimal point for the two base station that five meter range and the software skip how its scaled so far right correct yes so that's a good shippable solution it gives you a good experience in full 360 tracking basically you can't it's very difficult to include it you can experience it yourself right it's you can almost put the thing under your arm and it still works yeah yeah you know there's a matter of you you have the two controllers overlapping each other they they're it's smart enough and it's not very computationally intensive because you said it's just basic math triangulation now how does the room then in that system like how do how do you know where the walls are in the room okay so the walls in the room are they no because the base stations give you an absolute coordinate system you can work out where the walls are so at the moment we have a calibration procedure where you use the headset or the controller to actually measure things in the world so you measure the ground you can measure go okay I don't want to go any further than this click and that gives you the geometry of the room now there are other techniques so we want to talk about in the future about how you would work out saying wear something that's moving in the room might be but for the moment fix geometry is pretty simple to scan in just by touching off points and I imagine the patient's need to be locked in place you don't want those things to move at all now can they be self power like with a USB cable and a power brick and yeah that that's exactly how they design they don't talk to the computer at all they just simply produced the signal that the tracking tracked objects can interpret so we've got ones that will run off a power brick like like a battery in for about a day but there'll be wired they'll take power and that's pretty much it now you guys are Maker Faire cuz you're obviously looking at using lighthouse in applications outside of virtual reality how do you envision people Baker's using lighthouse as a positional tracking system very good questions so obviously these things are going to be in great number out in the world very soon but essentially by Christmas so a lot of people are going to want to use them for non VR applications robotics and micro flight applications like what copters are obviously a very obvious way they didn't want to use absolute tracking system we want to facilitate that by I mean anyone can licensed lighthouse really the only entrance into lighthouses to not violate the standards so that all systems are can interoperate with each other will be offering technical information about how to utilize the system will be offering as much as possible their components is off-the-shelf device things that you just buy digi-key and put into your project you mentioned quadcopter so is that is that something like outdoor use what are the limitations of lighthouse for for the space that you can use it and I could you have a set up in a room like this yes you can totally set up in a room like this there obviously take a fair number of base stations with a current design to operate out to that there are many ways that you can scale lighthouse including you can increase the range by making it give you slower updates for example one of the like the limitations is the sink we can power the lasers already can go out to 20 meters easily so we isolate the 20 meters radially it's huge right it's a 40 meter circle that's that's like a stadium yeah whereas the sink blinkers are a little bit weaker than that you can have sync repeaters as a bunch of different ways you can make it scale or you can use RF now that's things that we should probably talk to the community about how they're actually going to use it and they can give us feedback on what's probably the best way to take the technology who else have you talked to about possibly licensing lighthouse for their obligations is it consumer electronics is it research we've been a lot of people who spoke to us about it a lot of researchers a lot of consumer developers as well and hopefully their experiences will feed back to how we can use it for to improve the VR experience as well absolutely so all all users of lighthouse will feed back into the the general you know protocol and set up at the system will probably come up with some kind of standards committee that will have to be the steward at the standard going forward so awesome thank you so much Alan for explaining a little bit more about the lighthouse how it works and your vision for how it might be used in VR and other places thank you so that's it for Maker Faire 2015 with more stuff from this show on test comm please subscribe to our YouTube channel on norm Justin only see guys next time bye\n"