So you want a directional aerial, on a turntable?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Two suggestions..

1) LoRa has a sub group of users who work at accurately spatially tracking an emitter without using GPS. There are OTS bits and pieces for doing it using time stamping and inertial monitoring.

2)Non-GPS tracker

Keith Cress
kcress - http://www.flaminsystems.com

Is this a follow-on to a previous post? Seems similar.
https://www.eng-tips.com/viewthread.cfm?qid=369406

Are you asking a question or promoting an idea? I can't decipher.

Antennas on a turntable is a good choice of words, though another description includes a servo motor drive that rotates the small form factor antenna (s) assembly to create equal received signal strength on each side of the boom.

Outdoors, the solution is much easier, indoors, GPS has limitations. Indoors or outdoors,near field, propagation radiation has issues. Triangulated UWB is one solution for both indoors and outdoors, but is not user friendly, and latency can be an issue.

No, this is only a question, not a promotion.

DF Steer utilizes a cathode display, though for this application, the cathode is not needed, only the black box and schematic. DF Steer is not thought to be the Holy Grail, though by adding additional technologies the total concept may save many man hours.

There are at least several threads about indoor radio location.

From 2012: https://www.eng-tips.com/viewthread.cfm?qid=320765
From 2013: https://www.eng-tips.com/viewthread.cfm?qid=340586
There are probably others...

Rather than a directional antenna on a turntable, you'd be further ahead with a synthetic Doppler system.

There are endless possibilities for indoor navigation, with very little commercialization to date. One exception is Google Indoor Navigation where the indoor localization relies upon some sort of crowd sourced Wi-Fi mapping. It works well in most shopping malls.

FYI, if you copy the thread number at top of the thread into your post, the site will automatically pull the thread title: thread236-438942: Tracking the emitter

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

Ah. E.g. "thread236-438942: Tracking the emitter"

Thank you.

---

Further details to assist others: Upon Preview or Submit Post, the plain text string "threadNNN-NNNNNN" magically becomes the expanded link including the thread title.

Doppler, as I know it, is great, especially for long range uncluttered applications. However, it may get confused as the distance shortens down to 1 meter,(propagation radiation) compounded by bounce. By inserting a ping on the carrier and adjusting the ping Hz. as the distance decreases, reliability increases somewhat.

www.Deca-wave.com has had success with their stacked frequency UWB, but it requires multiple anchors with limitations. The concept, for example, is used in the Pixio camera (www.Move-n-see.com) Though Deca-Wave has a solution for propagation radiation, latency is questionable.

www.soloshot.com is another servo drive tracking system that uses differential GPS outdoors, and possibly bluetooth indoors...not yet released. GNSS3 may be capable of performing indoors/outdoors, but little information is available. Subject to being available for commercial applications, GNSS3 may possibly be piped indoors.

Zebra/Motorola serves most of the NFL stadiums, their system uses something like 250 anchors and 3 engineers to operate. The player wears a tag on each shoulder pad, the anchors track the tags and the PC commands the camera you see from time time. (Zebra phoned recently to see how far we had gotten. They know we already have profound UWB.)

Though I've talked cameras, that is not the application. Accurate user friendly technology with low latency and juice load is the goal.

How do the camera pointing algorithms used on drones work? They will keep the camera on target no matter how the drone maneuvers.

One solution, used on Arducopter, is to use a gymbal mounted camera. I'd have thought the drones that are body orientation agnostic wouldn't need the gymbals.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Time-Of-Flight is often the best approach. With the wide bandwidths and fast processors, it's brought the distance increments to a fraction of a meter.

Just by way of another example, the Wi-Fi Analysis apps available for smartphones can typically show the distance to each hotspot. They seem to be quite accurate (our house has multiple hotspots), to within about the nearest meter. So it's a small step to combining several distances to work out the relative location.

--

Pointing a camera at a fixed location on the ground can be done based off of the vehicle's complete navigation solution (position and orientation). The modern chip sensors have built-in gyros and accelerometers, combined with GPS it's straightforward to point to a given location on the Earth. A remaining detail is having accurate local terrain height, as the system needs that info to compute the direction to intercept the desired point. A small drone might use a single value for local height, easy to capture at start-up. Larger vehicles covering huge distances get far more complicated.

Pointing at a moving target is probably a vision tracking scheme. These sometimes offer the option of drawing a little box around the moving target on a video display.

"One solution, used on Arducopter, is to use a gymbal mounted camera. I'd have thought the drones that are body orientation agnostic wouldn't need the gymbals."

There are two issues with a fixed installation:
> Pointing angles may require flight orientations that are not consistent with good flight performance, e.g., if you are directly overhead and want to shoot down with a forward looking camera, you'd need to fly at close to 90-deg roll.

> Rotorcraft have vibration (jitter) that limit the ultimate resolution and sharpness of imaging sensors, but a gimbal can perform stabilization that can improve the jitter performance

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

I appreciate you guys thinking through this. I've worked with several of the big names puzzling this together. The deeper I get, the more it becomes the "Holy Grail". DJI took a hard look at solving for the emitter going one way, while the receiver goes the opposite direction. DJI could benefit from having such performance. Their answer is; their performance yield is 60-80%.

I have a mutual NDA with Leddar-tech.com. They have an interesting technology that should compliment future applications, though, so far, doesn't fit this project.

Google is thought to have collaborated with Soloshot, but as said earlier, Soloshot is pending technology release.

So far, DF Steer, plus other embedded technology, is interesting. FYI, FAA, flight service stations (FSS were the primary users of DF Steer.

There is a lot to be said for facial recognition, acoustics, MEMS (Bosch is thought to be working on their own solution)and MIT has their new PC technology....