A good project for someone: automatic rocket-tracking motorized camera mount

[mikec]

they claim they are "tracking" the rocket but I say that is bs.

"Tracking" is a camera term meaning following the subject, they don't claim to be actively sensing the rocket's position.

 

[Derek]

"Tracking" is a camera term meaning following the subject, they don't claim to be actively sensing the rocket's position.

but that system isn't following the subject, it is following a predefined plot (or maybe curve is the right term) based on the weight of the rocket and the motor thrust curve. if you read their faq, you can figure out how they figure out when to move the camera. hopefully the rocket just happens to be in the frame at that time.

don't get me wrong, it is a clever system but there isn't any real tracking going on. 99+% it probably works great. but it's usually that other small percentage when something goes wrong that you'll probably miss and wish you hadn't.

their business model is interesting, too. they really are just offering a video recording service of your launch, not the equipment. wonder why?

 

[cerving]

Probably more than most hobbyists would shell out... if they showed up at LDRS they'd probably make a bundle on the service, assuming that they can get a BLM permit in time.

 

[ksaves2]

Well................ I bet if there were a lot of takers, I'm sure Doug would work something out with multiple launches planned. Might be worth it for one's landmark cert flight. Kurt

 

[Eyesinthesky]

but that system isn't following the subject, it is following a predefined plot (or maybe curve is the right term) based on the weight of the rocket and the motor thrust curve. if you read their faq, you can figure out how they figure out when to move the camera. hopefully the rocket just happens to be in the frame at that time.

don't get me wrong, it is a clever system but there isn't any real tracking going on. 99+% it probably works great. but it's usually that other small percentage when something goes wrong that you'll probably miss and wish you hadn't.

their business model is interesting, too. they really are just offering a video recording service of your launch, not the equipment. wonder why?

Perhaps it is using the term 'tracking' loosely but you are correct that it is not actively tracking the rocket but it is tracking the path of a rocket that is taking off at some acceleration. If you are looking for a completely active tracking system using radar or one that optically follows a rocket on any path, you had better have a much bigger budget than I have. I use commercially available products and put them together for a few hundred dollars for each pitch mount (not including the cameras). But you do bring up a good point, if the rocket does not accelerate at what is set, it will not film the rocket very well. The biggest issue that I have seen with the set up is trying to determine what the acceleration should be set.

I use to just ask the flier what is the initial acceleration of their rocket and that rarely worked. At best they might provide the maximum acceleration but mostly they would never look at the initial acceleration. I keep a table of many commercial motor thrust curves and their initial thrust and weigh the rockets just prior to flight. Then just using the thrust to weight ratio (minus 1) works very well for the first half of a second. After that, the rocket is so high errors in the angle of the camera is trivial.

Well................ I bet if there were a lot of takers, I'm sure Doug would work something out with multiple launches planned. Might be worth it for one's landmark cert flight. Kurt

I don't consider myself to have a business considering I would never make any money doing this. I think of myself providing a service that I like to do for other fliers. I tried very hard to go to LDRS this year. But the rules from the BLM and ROC made it impractical. Besides the vendor fee, there is also a filming permit required if the images are to be resold and the permits starts at $500. I have not seen enough interest to over come this so I made other arrangements to cover the IREC launch. But I would attend a launch if there was enough interest.

 

[bobkrech]

none of these toys are going to be fast enough mechanically to track a rocket on ascent. even if you had a mount that could, the lag from the gps is too great to track at ascent speeds.

tracking on descent is possible with gps, but the optics on phones/gopros/etc. won't be good enough to actually see the rocket.

however it can be done optically but the hardware required is very expensive.

Sorry to disagree, but I've done the math and the optics. The servos on the product I referenced are plenty fast enough to track a rocket. The description of the electronics is very poor and it may well be that the tracking system is using the transmitter in the rocket as a beacon for tracking. If it does not, it's not that difficult to make a null circuit detector for tracking purposes. As for the cameras, a decent quality video camera with a decent 10X magnification lens is about what you need, as well as a baseline of ~1000' from the launch pad. One you get to the point where elevation is not changing rapidly, you could have a remote method to increase the magnification of the zoom lens.

You might be interested to know that the downrange NASA launch videos are taken by man cranked optical tracking system.

Bob

 

[bobkrech]

for an example, see this:

[YOUTUBE]jtSUwP0znbo[/YOUTUBE]

Your video proves my point. The video last 8 seconds, and the launch occurs at 1 second, and the track is less than 180 degrees. The average tlle slew rate is less than 180/7 ~ 25 degrees/second. The "toy" tracker has a tile rate of 240 degrees per second, almost an order of magnitude more than observed tile rate in the video recorded by the million dollar IR missile tracker...... and the "toy" tracker has a 112 degree per second tilt rate which again is an order of magnitude faster than observed in the video.......

The IR video tracking electronics is probably worth a million dollars, but could be used to drive a $400 "toy" tracker just fine. And you could use a 1 ounce, 1 cubic inch, $8000 640x480 FLIR Systems Quark LWIR video camera to record it........ This camera is an option on our sub 1 pound PSI Tactical Instant-Eye Drone which is the SOF, FBI, and other alphabet soup agencies tactical drone of choice.

Bob

 

[Xrain]

You might be interested to know that the downrange NASA launch videos are taken by man cranked optical tracking system.

Bob

You might be interested to know that Vandenberg at least has upgrade from the man cranked mounts.



This was on Kodiak but the equipment was contracted from Vandenberg. Too bad we cant use these at launches, but the cameras alone were way more than a million. Took amazing footage though. Even these though aren't auto tracking. They use a joystick to manually follow the rocket. They do certainty make models of these that optically track.



If I was going to make an autotracker for our use, I'd use the GPS metric tracking with hotspot centering on the telemetry antenna to zero in on the beacon from the rocket in case GPS lock is lost. Optical tracking would be good, but it is hard to do after the motor burns out since you lose that nice hotspot. Works great on big rockets since they burn for most of the time you can see them.

 

[Derek]

Sorry to disagree, but I've done the math and the optics. The servos on the product I referenced are plenty fast enough to track a rocket. The description of the electronics is very poor and it may well be that the tracking system is using the transmitter in the rocket as a beacon for tracking. If it does not, it's not that difficult to make a null circuit detector for tracking purposes. As for the cameras, a decent quality video camera with a decent 10X magnification lens is about what you need, as well as a baseline of ~1000' from the launch pad. One you get to the point where elevation is not changing rapidly, you could have a remote method to increase the magnification of the zoom lens.

You might be interested to know that the downrange NASA launch videos are taken by man cranked optical tracking system.

Bob

Ok then. go for it!

I know what nasa uses. we provided a lot of the equipment for them. the manual tracking is due to union rules. most of the contractors use automatic systems.

 

[Derek]

Perhaps it is using the term 'tracking' loosely but you are correct that it is not actively tracking the rocket but it is tracking the path of a rocket that is taking off at some acceleration. If you are looking for a completely active tracking system using radar or one that optically follows a rocket on any path, you had better have a much bigger budget than I have. I use commercially available products and put them together for a few hundred dollars for each pitch mount (not including the cameras). But you do bring up a good point, if the rocket does not accelerate at what is set, it will not film the rocket very well. The biggest issue that I have seen with the set up is trying to determine what the acceleration should be set.

I use to just ask the flier what is the initial acceleration of their rocket and that rarely worked. At best they might provide the maximum acceleration but mostly they would never look at the initial acceleration. I keep a table of many commercial motor thrust curves and their initial thrust and weigh the rockets just prior to flight. Then just using the thrust to weight ratio (minus 1) works very well for the first half of a second. After that, the rocket is so high errors in the angle of the camera is trivial.

thanks for verifying what I suspected. Have you experimented with another switch or two further up the rail to sort of double check how things are progressing? it might help with outright failures.

your system is pretty clever and seems well engineered. and you demonstrated that a brushless gimbal is fast enough to track a rocket at launch. great work!

 

[bobkrech]

For those not familiar with aerospace tracking systems, it is important to understand long range tracking cameras are not located near the launch pads for several reasons. If they are located too close to the launch pads, the focal distance changes rapidly and it is difficult to get clear images of the vehicle as it ascends, and you also have to worry about image blurring do to the rapid angular movement of both the rocket and the cameras. Minimum baseline distances for NASA down range tracking cameras are mile, not feet. This 1980's video shows the range of cameras used to track shuttle launches.

[video=youtube;tGtcW0Lt4QE]https://www.youtube.com/watch?v=tGtcW0Lt4QE[/video]

from https://petapixel.com/2013/05/01/ch...otograph-space-shuttle-launches-in-the-1980s/

Pointing accuracy and stability are important but slew rates are not high by camera positioning design, and the telescopic focal ranges are around 5:1 to 6:1. Any automated hobby rocket tracking system should have a minimum baseline of1000' or several times the minimum personnel separation distance, which ever is greater, to minimize changes in image focus and image field of view versus time.

[video=youtube;MOe8oqbWZEQ]https://www.youtube.com/watch?v=MOe8oqbWZEQ[/video]

This is a long range tracking video of STS-133, the last shuttle landing. Note the distances, and the fraction of the frame covered by the shuttle. The best the pros get is a fill factor of 25% to 35%. These images are obtained using ~0.5 meter optics and are not all that clear. You can not get similar images with GoPros.

Slew rate and tracking is not the issue. Optical resolution becomes the limiting factor.

Bob