Amateur Active Rocket Stabilization System

[Coldreactor]

Hello, I am Jeremy Hall. I am a 17yo High School Senior and have been working on this project for literal years and am coming close to completion. I am active stabilizing a rocket by use of gimbaling parts of the fins of the rocket to guide it in a straight line up. This is an ambitious project for anyone to undertake but I have spent many nights researching and studying before I even laid down a trace or ordered a single part. I would love to share it with you guys and let me know if you have any questions and such and I will be sure to answer them.

 

Project Links: Hackaday, Github

[Hackentosher]

Pics? 

[Coldreactor]

It's not finished, I only have the body tube so far. I am working on getting the fins ordered. I have the control PCB already ordered and on the way.

[Hackentosher]

1 hour ago, Coldreactor said:

It's not finished, I only have the body tube so far. I am working on getting the fins ordered. I have the control PCB already ordered and on the way.

wait this has been in the works for years and you're just getting around to building the first prototype? What work have you done?

[Coldreactor]

1 hour ago, Hackentosher said:

wait this has been in the works for years and you're just getting around to building the first prototype? What work have you done?

First and foremost, this is not an easy task by any stretch. It is hard when you haven't even taken calculus to try and understand differences between filtering techniques(Kalman, Complementary, Mahony, Madgwick) Just studying those takes a huge chunk of time. Plus I have to learn microcontroller development, which is another tough task that is not in the faint of heart. Right now I am in the middle of porting the code to bare AVR C, and then after I do that I will have to port it again to a 32 bit arm architecture. This is not to mention all the nights of debugging and being frustrated out of my mind. Another thing I had to learn is PID control. When I first started this I though, I need a set point based on what I am given out so I should use a PID controller. Turns out this is not the case, I have to use direct linear relations for the servos(actually not direct linear relation, I am working on code to do it based on velocity). Another good chunk was allocated to the fact that I had to learn Electrical Engineering. A huge portion of time was allocated to this. Electrical Engineering for a 15, 16 17 year old is not an easy task. I had to learn how to use components, resistors capacitors and how to design and layout a PCB. I have nothing in the terms of funding, all I have to work on is in the form of Donations and freelance embedded programming jobs I sometimes get.

I have done a whole lot of studying and a whole lot of work to get up to this point. I have spent these years wisely working on this.
(FYI, I also had life get in a way a bunch. Moved schools, and cities 3 times in the course of this)

[CaleB_]

Fucking rektd

[Hackentosher]

15 minutes ago, Coldreactor said:

snip

Hmm, I would think a PID controller would be a good choice for dealing with errors and correcting them. As for components, I would think you could get away with using an Arduino if the processor is fast enough as it has plenty of IO and capability for servos and sensors.

[Coldreactor]

1 minute ago, Hackentosher said:

Hmm, I would think a PID controller would be a good choice for dealing with errors and correcting them. As for components, I would think you could get away with using an Arduino if the processor is fast enough as it has plenty of IO and capability for servos and sensors.

Ya, I am using a Uno with a shield that breaks out the pins easily for the Servos. I will be switching to a smaller board that I made which is basically a shield for the Arduino Pro Mini. Then I will be switching to an ATSAMD21G18 ARM Cortex M0 board that I will be designing soon that will intergrate LiPoly charging, have Servo Connections and such nicely done and have some mounting holes(Oh ya, and MicroSD reading/writing). The reason I need the 32bit microcontroller is for the fact that I am running out of space and speed on the 8bit Atmega328p. This is rare for projects, but mine has a lot of processing of data and needs that extra horsepower. It will work fine as a barebones system but I really want to get Datalogging in the future so I am going to switch.

[Hackentosher]

9 minutes ago, Coldreactor said:

 that will intergrate LiPoly charging, 

Why do this when these exist https://www.adafruit.com/products/2465? Or just remove the battery when it needs to be charged?

[Coldreactor]

Just now, Hackentosher said:

Why do this when these exist https://www.adafruit.com/products/2465? Or just remove the battery when it needs to be charged?

What I have been starting to think is taking an Adafruit Feather M0 Datalogger and building a wing for that because that has LI-Po charging and SD card built in with the M0. I'l see when I get designs started.

[Hackentosher]

Just now, Coldreactor said:

What I have been starting to think is taking an Adafruit Feather M0 Datalogger and building a wing for that because that has LI-Po charging and SD card built in with the M0. I'l see when I get designs started.

Again, why don't you just remove the battery when it needs to be recharged?

[Coldreactor]

It's an option, but I rather just have the convenience of just plugging USB in and having it all right there. It doesn't take up that much space/complexity/money so why not? I'l leave it as an option because it was part of my original plan, I have a charger module that can do it. I will do what is better/easier/cheaper for me. On top of that, the servo's are gonna have their own separate power supply so it doesn't cause any issues with the processor, so I might as well charge them standalone.

[CaleB_]

Fucking rektd again

[straight_stewie]

17 hours ago, Coldreactor said:

Hello, I am Jeremy Hall. I am a 17yo High School Senior and have been working on this project for literal years and am coming close to completion. I am active stabilizing a rocket by use of gimbaling parts of the fins of the rocket to guide it in a straight line up. This is an ambitious project for anyone to undertake but I have spent many nights researching and studying before I even laid down a trace or ordered a single part. I would love to share it with you guys and let me know if you have any questions and such and I will be sure to answer them.

I have/am worked/working on a similar project for a team at my University. Aerospace (astronautics) and CS are also my majors, although I am only a Freshman. 

 

12 hours ago, Coldreactor said:

It's an option, but I rather just have the convenience of just plugging USB in and having it all right there. It doesn't take up that much space/complexity/money so why not? I'l leave it as an option because it was part of my original plan, I have a charger module that can do it. I will do what is better/easier/cheaper for me. On top of that, the servo's are gonna have their own separate power supply so it doesn't cause any issues with the processor, so I might as well charge them standalone.

It's not about ease, it's about weight. Have you figured out your ΔV? I would highly recommend that you think very seriously about weight in this case. I would also recommend that you have atleast limited engine gimbal capability unless your initial acceleration is very high. Your fins won't be worth a crap until the aircraft is moving at sufficient speed for them to generate enough force to move the aircraft.  Is this your first rocket build?

[Coldreactor]

2 hours ago, straight_stewie said:

I have/am worked/working on a similar project for a team at my University. Aerospace (astronautics) and CS are also my majors, although I am only a Freshman. 

 

It's not about ease, it's about weight. Have you figured out your ΔV? I would highly recommend that you think very seriously about weight in this case. I would also recommend that you have atleast limited engine gimbal capability unless your initial acceleration is very high. Your fins won't be worth a crap until the aircraft is moving at sufficient speed for them to generate enough force to move the aircraft.  Is this your first rocket build?

This is my third rocket build, though my first two were much smaller.
I would have done engine gimbaling but the acceleration is more than great enough.(1.2 second burn time)

About that weight of having the charger on board... The weight of that is insignificant, it does not weigh much if it is included onto the board, I doubt it would change the weight by even half a gram.

Also another point is that I am working on some code to get an approximation of the speed of the rocket using the acceleration and then adjusting for that in the code.

Thanks for the suggestions.

[bob345]

Id be carefull with a system like that. It is very much a legal grey area in most states and outright illegal in others as it can be classified as a 'missile' at that point. Making it for a challenge like TARC or just a personal project?

[straight_stewie]

1 hour ago, bob345 said:

Id be carefull with a system like that. It is very much a legal grey area in most states and outright illegal in others as it can be classified as a 'missile' at that point. Making it for a challenge like TARC or just a personal project?

Basic active guidance systems (of the type we are discussing here) are legal in all states given that the rocket conforms to all other amateur rocketry laws in that state. Where the gray area comes in is "fire control". Can the rocket "see" where it is going? Can it track a fixed or moving point and "chase" it?

[bob345]

8 minutes ago, straight_stewie said:

Basic active guidance systems (of the type we are discussing here) are legal in all states given that the rocket conforms to all other amateur rocketry laws in that state. Where the gray area comes in is "fire control". Can the rocket "see" where it is going? Can it track a fixed or moving point and "chase" it?

You are correct,thanks for clarifying. Its just something you want to be careful with especially with how paranoid some of these lawmakers are nowadays. 

[straight_stewie]

4 minutes ago, bob345 said:

You are correct,thanks for clarifying. Its just something you want to be careful with especially with how paranoid some of these lawmakers are nowadays. 

I'll agree that it's always best to be careful. In my state, we can do whatever we want, which includes UAVs. There are, as usual some stipulations: You must now obey the new FAA laws and rules regarding model aircraft/drones. But we can even have homing drones (like WIFI or Bluetooth homing) or "seeing" drones (like camera guided drones).

Personally, I find these topics quite interesting, as that's why I went to Uni to study aerospace and CS. I want to work on guidance/FCS/Avionics systems specifically (they are very closely tied together)

[Coldreactor]

Yup, I have researched the laws regarding this. It is legal because what I am using is dead reckoning to find the position of the rocket and then controlling it off of that. It has no external guidance features, hence it cannot be used as a missile system. It just goes at the angle defined in the code.

[straight_stewie]

23 hours ago, Coldreactor said:

Yup, I have researched the laws regarding this. It is legal because what I am using is dead reckoning to find the position of the rocket and then controlling it off of that. It has no external guidance features, hence it cannot be used as a missile system. It just goes at the angle defined in the code.

What method are you using to do your initialization? Are you just spinning up the gyros/accelerometers while level in any position, then controlling it relative to a vertical flight path, or do you have any GPS map stuff going on so that you can get accurate flight profiles?

 

[mookie1590]

Bumping so I can check back later and see this. Sounds awesome bro.

[Coldreactor]

On 9/8/2016 at 0:02 AM, straight_stewie said:

What method are you using to do your initialization? Are you just spinning up the gyros/accelerometers while level in any position, then controlling it relative to a vertical flight path, or do you have any GPS map stuff going on so that you can get accurate flight profiles?

 

I am starting the sensors in the level position, and plus they are software calibrated so even if they start in a weird position they are good.

[straight_stewie]

1 minute ago, Coldreactor said:

I am starting the sensors in the level position, and plus they are software calibrated so even if they start in a weird position they are good.

Sounds pretty fun. Right now my team and I are working on breaking the collegiate rocketry world speed record. We will have to break Mach 4.21 to do so. We are hoping to launch around this time next year.

[Coldreactor]

Just now, straight_stewie said:

Sounds pretty fun. Right now my team and I are working on breaking the collegiate rocketry world speed record. We will have to break Mach 4.21 to do so. We are hoping to launch around this time next year.

Woah, very impressive. What is your propulsion system going to look like? Some high performance APCP or a COTS motor?