It’s difficult to tell with our dull human senses, but everything around us is vibrating. Sure it takes more energy to get big objects like bridges and houses humming compared to a telephone pole or mailbox, but make no mistake, they’ve all got a little buzz going on. With their new automated laser, the team behind VibroSight++ believes they can exploit this fact to make city-scale sensing far cheaper and easier than ever before.
The key to the system is a turret mounted Class 3B infrared laser and photodetector that can systematically scan for and identity reflective surfaces within visual range. Now you might think that such a setup wouldn’t get much of a signal from the urban landscape, but as it so happens, the average city block is packed with retroreflectors. From street signs to road studs and license plates, the team estimates dense urban areas have approximately 7,000 reflectors per square kilometer. On top of those existing data points, additional reflectors could easily be added to particularly interesting devices that city planners might want to monitor.
Once VibroSight++ has identified its targets, the next step is to bounce the laser off of them and detect the minute perturbations in the returned signal caused by vibrations in the reflector. In the video below you can see how this basic concept could be put to practical use in the field, from counting how many cars pass over a certain stretch of road to seeing how popular a specific mailbox is. There’s a whole world of information out there just waiting to be collected, all without having to install anything more exotic than the occasional piece of reflective tape.
Retroreflectors are interesting materials, so known for their nature of reflecting light back to its source. Examples include street signs, bicycle reflectors, and cat’s eyes, which so hauntingly pierce the night. They’re also used in the Tilt Five tabletop AR system, for holographic gaming. [Adam McCombs] got his hands on a Tilt Five gameboard, and threw it under the microscope to see how it works.
[Adam] isn’t mucking around, fielding a focused ion beam microscope for the investigation. This scans a beam of galium metal ions across a sample for imaging. With the added kinetic energy of an ion beam versus a more typical electron beam, the sample under the microscope can be ablated as well as imaged. This allows [Adam] to very finally chip away at the surface of the retroreflector to see how it’s made.
The analysis reveals that the retroreflecting spheres are glass, coated in metal. They’re stuck to a surface with an adhesive, which coats the bottom of the spheres, and acts as an etch mask. The metal coating is then removed from the sphere’s surface sticking out above the adhesive layer. This allows light to enter through the transparent part of the sphere, and then bounce off the metal coating back to the source, creating a sheet covered in retroreflectors.
By now we’ve all seen the cheap headsets that essentially stick a smartphone a few inches away from your face to function as a low-cost alternative to devices like Oculus Rift. Available for as little as a few dollars, it’s hard to beat these gadgets for experimenting with VR on a budget. But what about if you’re more interested in working with augmented reality, where rendered images are superimposed onto your real-world view rather than replacing it?
As it turns out, there are now cheap headsets to do that with your phone as well. [kvtoet] picked one of these gadgets up for $30 USD on AliExpress, and used it as a base for a more capable augmented reality experience than the headset alone is capable of. The project is in the early stages, but so far the combination of this simple headset and some hardware liberated from inexpensive Chinese smartphones looks to hold considerable promise for delivering a sub-$100 USD development platform for anyone looking to jump into this fascinating field.
On their own, these cheap augmented reality headsets simply show a reflection of your smartphone’s screen on the inside of the lenses. With specially designed applications, this effect can be used to give the wearer the impression that objects shown on the phone’s screen are actually in their field of vision. It’s a neat effect to be sure, but it doesn’t hold much in the way of practical applications. To turn this into a useful system, the phone needs to be able to see what the wearer is seeing.
To that end, [kvtoet] relocated a VKWorld S8 smartphone’s camera module onto the front of the headset. Beyond its relatively cost, this model of phone was selected because it featured a long camera ribbon cable. With the camera on the outside of the headset, an Android application was created which periodically flashes a bright LED and looks for reflections in the camera’s feed. These reflections are then used to locate objects and markers in the real world.
In the video after the break, [kvtoet] demonstrates how this technique is put to use. The phone is able to track a retroreflector laying on the couch quickly and accurately enough that it can be used to adjust the rendering of a virtual object in real time. As the headset is moved around, it gives the impression that the wearer is actually viewing a real object from different angles and distances. With such a simplistic system the effect isn’t perfect, but it’s exciting to think of the possibilities now that this sort of technology is falling into the tinkerer’s budget.
Some people really put a lot of effort into rigging the system. Why spend years practicing a skill and honing your technique to hit a perfect bullseye in darts when you can spend the time building an incredibly complicated auto-bullseye dartboard that’ll do it for you?
In fairness, what [Mark Rober] started three years ago seemed like a pretty simple task. He wanted to build a rig to move the dartboard’s bullseye to meet the predicted impact of any throw. Seems simple, but it turns out to be rather difficult, especially when you choose to roll your own motion capture system.
That system, built around the Nvidia Jetson TX1, never quite gelled, a fact which unfortunately burned through the first two years of the project. [Mark] eventually turned to the not inexpensive Vicon Vantage motion capture system with six IR cameras. A retroreflector on the non-regulation dart is tracked by the system and the resulting XY data is fed into MATLAB to calculate the parabolic path of the dart. An XY-gantry using six steppers quickly shifts the board so the bullseye is in the right place to catch the incoming dart.
It’s a huge amount of work and a lot of money to spend, but the group down at the local bar seemed to enjoy it. We wonder if it can be simplified, though. Perhaps tracking just the thrower’s motions with an IMU-based motion capture system and extrapolating the impact point would work.