Arduino-based Helicopter UAV

butterS_ · Nov 05, 2009, 01:19 AM

Hi all, long time lurker, first time poster. I'm an Electrical Engineering student and a hobby RC Heli flyer. I've had my helicopter for awhile now, due to budgetary constraints (ie. I'm a cheap bastard :P) I got it from Hong Kong on ebay. It's a Walkera Dragonfly 4#1, 4-channel non-CCP cheapie. When it comes to flying it, I've been marginally successful at best to get it stably flying around (the slightest wind pretty much puts it out of commission). Ever since I started doing embedded electronics at uni I've been keen on trying to turn my helicopter into a UAV. I'm also hoping to get involved in the UAV Outback Challenge which my uni (QUT) competes in.

Now, I was thinking the best way of starting off with such a project would be to get the helicopter first working semi-autonomously so that using an IMU, the craft would be able to stabilize itself at a basic level in order to make it easier to fly.. (getting pitch/roll/yaw hold), then take weighted control inputs from the radio controller. Once that's achieved, then complicate things further by starting to build in a GPS, a wireless datalink to monitor sensor output and add waypoints on-the-fly, etc. etc. But I'm getting ahead of myself.

Now I'm hoping to work with the Arduino platform, starting off with a Duemilanove for pre-flight testing, then transition to a Mini for lighter payload, and using a 6DOF 3-axis accelerometer, 3-axis gyro for my IMU. I know it's been attempted before, but I haven't seen any finished products, at least in English - I saw a couple of videos from a Spaniard who seemed like he was giving a good go of it. Anyway I guess I'll cut to the chase.. I've been playing around with interfacing the Arduino to the helicopter servos, and the IMU I have but I've kinda stalled at finding a rationale for calculating controller gains.

I was thinking before I go for the first test flight, that I would model the helicopter first using MATLAB to simulate the impulse response of different gains with the aim to find a good starting point for tuning my gains. Now I know I won't find the perfect practical value this way, but at least it'll give me a good start, and also a good theoretical concept if I decide to use this project for my thesis next year. However, being an Electrical student, I don't really have any experience modelling dynamics equations. I guess the point of this post was to see if anyone else here had attempted a similar project, or at least had some pointers for modelling the dynamics of a helicopter (I have found a few resources but for the most part they go over my head).

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Build Log Hades - PC [SBC] based UAV	cupsster	UAV - Unmanned Aerial Vehicles	25	Sep 15, 2009 12:42 PM
Discussion Arduino based quadrocopter?	kaldak	Multi Rotor Helis	3	Jul 02, 2008 12:57 PM
Discussion UAV COTS based Flight control system hardware design	shahmaan	UAV - Unmanned Aerial Vehicles	0	Jul 11, 2007 05:16 PM
Discussion Yacht Based UAV - How To	sharkyjones	UAV - Unmanned Aerial Vehicles	14	May 16, 2007 07:39 AM
Question Electric Glider based UAV flight times?	vector_vortex	UAV - Unmanned Aerial Vehicles	8	Dec 13, 2006 08:51 AM

Nov 05, 2009, 01:19 AM	#1
butterS_ Registered User Join Date: Nov 2009 Posts: 1	Arduino-based Helicopter UAV Hi all, long time lurker, first time poster. I'm an Electrical Engineering student and a hobby RC Heli flyer. I've had my helicopter for awhile now, due to budgetary constraints (ie. I'm a cheap bastard :P) I got it from Hong Kong on ebay. It's a Walkera Dragonfly 4#1, 4-channel non-CCP cheapie. When it comes to flying it, I've been marginally successful at best to get it stably flying around (the slightest wind pretty much puts it out of commission). Ever since I started doing embedded electronics at uni I've been keen on trying to turn my helicopter into a UAV. I'm also hoping to get involved in the UAV Outback Challenge which my uni (QUT) competes in. Now, I was thinking the best way of starting off with such a project would be to get the helicopter first working semi-autonomously so that using an IMU, the craft would be able to stabilize itself at a basic level in order to make it easier to fly.. (getting pitch/roll/yaw hold), then take weighted control inputs from the radio controller. Once that's achieved, then complicate things further by starting to build in a GPS, a wireless datalink to monitor sensor output and add waypoints on-the-fly, etc. etc. But I'm getting ahead of myself. Now I'm hoping to work with the Arduino platform, starting off with a Duemilanove for pre-flight testing, then transition to a Mini for lighter payload, and using a 6DOF 3-axis accelerometer, 3-axis gyro for my IMU. I know it's been attempted before, but I haven't seen any finished products, at least in English - I saw a couple of videos from a Spaniard who seemed like he was giving a good go of it. Anyway I guess I'll cut to the chase.. I've been playing around with interfacing the Arduino to the helicopter servos, and the IMU I have but I've kinda stalled at finding a rationale for calculating controller gains. I was thinking before I go for the first test flight, that I would model the helicopter first using MATLAB to simulate the impulse response of different gains with the aim to find a good starting point for tuning my gains. Now I know I won't find the perfect practical value this way, but at least it'll give me a good start, and also a good theoretical concept if I decide to use this project for my thesis next year. However, being an Electrical student, I don't really have any experience modelling dynamics equations. I guess the point of this post was to see if anyone else here had attempted a similar project, or at least had some pointers for modelling the dynamics of a helicopter (I have found a few resources but for the most part they go over my head).