dirty_d
10 kW
I was working on a flight controller a while back. It used an STM32 Arm microcontroller eval board. I was having stability problems the last time I was working on it, so I just ended up getting the KK so I could finally just fly the damn thing 
I admit it is kinda intimidating to fly knowing how heavy it is and that it turning into a 10lb out of control hunk of metal with big sharp spinning propellers is just one loose wire or software bug away, lol. The eval board has a gyro and accelerometer on it already and pin headers, so there was no electrical work, just the software. I think they're only $10 too, and a lot more capable than an arduino. A hardware FPU is really nice to have too.
Here's the code if anyone wanted to mess with it. I'm pretty sure the stability problem is due to not having separate PID parameters for angular rate and leveling. Its just one PID loop that does both with a fixed complementary filter coefficient tying the gyro and accelerometer together. Rate isn't even used, only the absolute angle. I kinda figured it out by using the KK controller, it was very stable with rate control only, but when I turned on auto-level with the default PID parameters, it oscillated out of control like with my controller. I had to turn down the auto-level parameters to get it to stably auto-level.
https://github.com/dirty-d/stm32f3_discovery_quadcopter
This is the part
http://www.digikey.com/product-detail/en/STM32F3DISCOVERY/497-13192-ND/3522185
I admit it is kinda intimidating to fly knowing how heavy it is and that it turning into a 10lb out of control hunk of metal with big sharp spinning propellers is just one loose wire or software bug away, lol. The eval board has a gyro and accelerometer on it already and pin headers, so there was no electrical work, just the software. I think they're only $10 too, and a lot more capable than an arduino. A hardware FPU is really nice to have too.Here's the code if anyone wanted to mess with it. I'm pretty sure the stability problem is due to not having separate PID parameters for angular rate and leveling. Its just one PID loop that does both with a fixed complementary filter coefficient tying the gyro and accelerometer together. Rate isn't even used, only the absolute angle. I kinda figured it out by using the KK controller, it was very stable with rate control only, but when I turned on auto-level with the default PID parameters, it oscillated out of control like with my controller. I had to turn down the auto-level parameters to get it to stably auto-level.
https://github.com/dirty-d/stm32f3_discovery_quadcopter
This is the part
http://www.digikey.com/product-detail/en/STM32F3DISCOVERY/497-13192-ND/3522185