[FYP Log] Terrain Analysis System for a Picosatellite / CubeSat

[note, I do NOT have any schematic drawings or documents here. This is just an informative log!]

This is meant to be an experimental log of my final year project, with the hopes of sharing and gaining ideas and knowledge with CEans here 😀 Though, I have no idea what are the implications of putting FYP info public before it is actually finished. Oh well, it's just a log 😉

So, what's my project about? Well, I was aiming to continue my imaging work from CanSat more seriously, so I thought of making a more sophisticated imaging system. This time, liasing with other students, it's for a CubeSat. Mostly categorized as a Picosatellite, they have a mass of under 1kg.

More info: Cubesat

Objectives:

Develop a small low cost, high resolution imaging system
Combine images with onboard telemetry using "packets" and transmit via a digital downlink with a range between 500 m - 100 km
Perform image processing techniques to stitch images and analyze terrain data

Preliminary Scope:

Low altitude target. Since this is our university's first CubeSat, we still have a long way to go before putting the CubeSat into proper orbit. Our first testing height will be 500 m.
COTS components will be used most of the time, though am evaluating the cost effective ones.
Since SAR or similar forms wont be used, terrain analysis would not consist of 3D data. This might be compensated by stereo imaging though, if I feel like pursuing that 😛

Concept of Operation

Hopefully with the nadir of the CubeSat perpendicular to the ground, the CMOS camera will autonomously take pictures, periodically. Once a picture has been successfully taken, it will be saved on the on board SD card.
The image will then combined with telemetry data such as GPS and accelerometer readings. All this will be transferred to the data modem via a 115kbps UART connection. These processes are controlled via a microcontroller.
The data modem will downlink the data to the ground station at maximum of 288kbps. This varies with distance and atmospheric conditions. Depending on how high the CubeSat is, a tracker might be needed. But, that is beyond the scope of part 1 of my FYP. Perhaps next semester. The ground station will perform the mosaicking of the images to make a larger image map. Edge detection (and/or another process) will be performed to analyze the soil condition for terrain analysis.

My FYP will be solely limited to the imaging system. For the actual CubeSat project, that is under the purview of other students. However, I am still closely tied with the communications and electrical aspects of it.

Replies

raj87verma88

Interesting. Quite interesting.
Ashraf HZ

raj87verma88
Interesting. Quite interesting.
😀

I originally wanted to do something with MEMS, but I have to admit this is waaaay more fun to do. We plan to utilize an RC Helicopter to do some initial camera shots *grin*
Ashraf HZ

Will aim to purchase imaging components by next week 😀

List:
- CMOS image sensor with JPEG compression engine
- 16mm Lens
- PIC microcontroller + supporting components
xheavenlyx

Post for Auto subscription. Need to keep a lookout on this thread 😀
Ashraf HZ

Added Concept of Operation entry
gohm

righteous dude, thanks & please keep us updated!
Ashraf HZ

[Update]

RC helicopter acquisition successful. Having a lot of fun flying it 😛 JPEG Camera and AeroComm Data Modem Kit are en route to Malaysia from US.

Funding is still a little volatile. The department has given a budget limit of $280 for my imaging equipment alone, but I'd have to order through them. That'll take more than a month, so I'll only use it to purchase stuff I need next sem. For now am using my own money and will claim through other channels later.
Ashraf HZ

Pic updates!

#-Link-Snipped-#

The MD 500 helicopter. Between it and the transmitter is a test CMOS camera to be implemented on it next time.

#-Link-Snipped-#

This is part of our working area, before the actual "storm". You might be able to see some electromagnet coils being made, which were attempts to test the magnetorquer concepts.
silverscorpion

wow, your work area is quite good. Keep up the good work
Kaustubh Katdare

Awesome! The can't take my eyes off the heli! Totally jealous of you, Ash. Where did you get it from? I wish to order one for my drawing room pleasures.
Ashraf HZ

Haha, thanks.

We are using an ex computer lab as our "HQ", so we've been stocking it up with the necessary tools, food, and even some sleeping mattresses for the "all night" work. From the experience of the CanSat hell, we know what we need for CubeSat 😛

Unfortunately, it'll be tough for us to bring in spectrometers and oscilloscopes, because they have to be used and kept at their respective labs due to their very expensive nature. I plan to get a personal one that connects to computer, costs about $180 I believe.
Ashraf HZ

Alrighty, here are some new pics, first one is the image sensor we plan to use, and the other, my precious... one of the transceivers that got stuck at customs. Those stuff are awesome.

#-Link-Snipped-#

#-Link-Snipped-#

#-Link-Snipped-#
gohm

sweet! thanks for the pics and update Ash.
xheavenlyx

Man, I wish I had access to such a beautiful and sexy lab! 😀 Eat, breathe, sleep and dream electronic heaven!!

Keep us updated Ash.
Ashraf HZ

lol 😛 Wait till we get the CE Labs at the CE HQ! We'll stock it with the best equipment and tools ever!

[UPDATE]
Awesome components from Texas Instruments has arrived. One is a 19 channel ADC PDIP chip! The output is SPI 😀 Others include a Lithium Ion/Poly charger and controller, as well as an I[sup]2[/sup]C bus extender.

The telemetry components (GPS, back up transmitters, etc) and solar panels are on their way. Should arrive by next week.

Me and my friend tested the MD500 RC heli again, at the university's small helipad area. Wind was minimal, so the flights were great. I've finished with the test CMOS camera/transmitter module, so I'll put it on the heli next time. We expect to get live video from the air if things are successful.

Other things to complete includes the solar tracker (to move the CubeSat's panels towards the sun), and a solar lipo battery recharging circuit. The latter looks to be super hard, so that one will take up some time. Also, I've yet to interface the JPEG CMOS camera with the data transceiver. I seem to have some signalling issues.

My first FYP report is due on March 24th, so I'll need to really finish as much as I can. Not too concerned about finishing the hardware bit 100%, as I have the next semester (FYP II) to do that.
Ashraf HZ

Time for some new pics!

Fig 1- Test CMOS Camera module. Used for transmitting live video wirelessly

Fig 2- Our 2nd RC Helicopter, more faster and powerful than our first. One of the blades is broken after it collided with it's own tail shaft. Not sure how that happened 😛 Behind it is our beloved MD500.

Fig 3- Close up of the main shaft. The silver thing with the blue top is the electronic gyroscope.

Fig 4- Some of the tools we use. The component under inspection is the camera we used in our previous CanSat project. Its busted at the moment, after an SMT inductor came off.

Fig 5- The electronic engineer's best buddies: The Multimeter and the Soldering Gun. The soldering gun we are using this time is temperature adjustable, since we are now working with more delicate electronic components.
silverscorpion

wow, cant take my eyes off the pics...
Btw, in the third pic, Wat's the use of the gyroscope? Is is used for stabilization of the
heli??
Ashraf HZ

Yep, its used for stabilization purposes 😀 Gyros are usually used in 6 channel heli's, as they are able to perform more complex maneuvers (like flying upside down, etc).
silverscorpion

cool.. I'd sure like to see it all in action soon.. Keep it coming!!
Ashraf HZ

Guess what? More pics! Its crunch time, as I have to finished some stuff before submitting my FYP1 report. Was working on the solar tracker today, took longer than I expected.

Fig 1- Testing the AC4424 Transceivers and the CMOS Camera with MATLAB

Fig 2- Part of the prototype solar tracker completed. On the left is my beloved rotary tool 😀

Fig 3- Some of the Amorphous Silicon solar cells. In direct sunlight (at sunset), each can produce 10mA @ 3V, enough to drive an LED.

Fig 4- Me and my friend finally managed organize the hundreds of components we had laying around 😛
raj87verma88

Cooooool...
durga ch

using my all time favorite word for approciating most amazing stuff- AWESOME!
Ashraf HZ

Thanks, ya all 😀
Ashraf HZ

Final updates! Taking a short break from the project as I've got industrial training. Enjoy the pics!

Fig 1- Test RC Helicopter with mounted CMOS camera and transmitter

Fig 2- The Solar tracker module consisting of the solar sensor board (right) and the 19 port ADC board (left). The solar sensor took a long time to do, since it was an original design. There were nearly 90 solder points 😔

Fig 3- Complete solar tracker module. The ADC chip is Texas Instrument's TLC545.

Fig 4- Lithium Polymer Solar battery charger from Sparkfun

Fig 5- Antenna and camera module mounting on test structure

Fig 6- High efficiency Switching regulator, using TI's PTH08000W

Fig 6- Initial integration of all modules. The awesome structure was made by the two aeronautical girls, using CNC milling. They also did ANSYS analysis on it, though I doubt I could get my hands on it.

Fig 7- Close up of other modules inside the Cubesat. The others are made by my friend including the servo board, OBDH, GPS and magnetometer.

I didn't get to take the pictures of the final integration, unfortunately. My training had already started by the time the others in my group integrated everything properly, including stacking up the modules.

Alright, there goes most of the hardware of our first prototype Cubesat. The hardware is only a part of it, though. A lot of things involve software! That is the tricky stage for us so far. For me, I've yet to complete the coding for image processing and packeting of data, using C programming for the PICs and MATLAB for post analysis in the ground station.

Thanks for following the thread. Will continue the work in July 😀 If theres any questions, feel free to ask!
durga ch

heights of awesomeness!!! 😁
farhanfaisal

Yo, whats the update? no more update?

btw, what radio module are you guys using? what spec?
sumith123

do you have more update..

now its seen very fun...and interesting
Ashraf HZ

Whoops! Forgot about this thread!

farhanfaisal
Yo, whats the update? no more update?

btw, what radio module are you guys using? what spec?
We used the Aerocomm AC4424 modems.
Frequency: 2.4 GHz
Power: 200mW
Data Rate: 192kbps
Range: 4km

We never really tested the range though.

sumith123
do you have more update..

now its seen very fun...and interesting
Well, I finished this project about December last year. Unfortunately we were not able to hoist the camera+AC4424 on the RC Heli for long range testing.. so we had to use data from lab testing, and induced the noise and fading parameters via Matlab. There was also some buffer issues between the camera DSP and AC4424.. after 3 large picture transfers, the AC4424 buffer gets overflowed and wont accept new bytes. I should have used some PIC in between them for buffer control via the CTS and RTS pins of the AC4424 (the camera module only used TX and RX). Anyhow.. at least we managed to process the pics for the mosaicking process!
vishnu priya

Seriously INTERESTING!
sonomer

Does anybody know how that CubeSat chassis, in the pictures displayed above, was made? It looks quite professional (unlike the electronic boards). I like it. I am interested in making or having such a #-Link-Snipped-# made. I possess the #-Link-Snipped-# but do not know how to cut the aluminum plates, bend and/or join them with precision.
P.S.: For me, buying a ready made cubesat structure is out of question because of the price.
Ashraf HZ

sonomer
Does anybody know how that CubeSat chassis, in the pictures displayed above, was made? It looks quite professional (unlike the electronic boards). I like it. I am interested in making or having such a #-Link-Snipped-# made. I possess the #-Link-Snipped-# but do not know how to cut the aluminum plates, bend and/or join them with precision.
P.S.: For me, buying a ready made cubesat structure is out of question because of the price.
I'm not sure exactly how my aeronautical team mates did them, but they fed the CAD design into a computer CNC milling machine. You'll need to find out the particular file format for that. You can try outsourcing the CNC milling to an outside workshop if you don't have them at your college.

You are reading an archived discussion.

[FYP Log] Terrain Analysis System for a Picosatellite / CubeSat

Replies

Related Posts

Running a business online.

Adsense earnings.

Please give me the list of mini projects

Approaching a client.

ARM processor interfacing with USB flash drive