Posted by VoiceOfReason on September 10, 2008 at 11:45pm
It's hard to understand who here has experience or sklls to do what you all seem to want to do. There is a lot of talk about stuff that might lead to it. But it is unclear if anyone is flying or close to flying real autopilots, or if they are just talking about making their own from scratch (which means you are a long way from it).A drone needs an autopilot. A "real" autopilot means you have to:a) Let your UAV estimate its attitude (roll, pitch, yaw) by itself so it can keep flyingb) Let your UAV be told where to go (mission planning)c) Let your UAV figure out where it is (navigation) and how to get to where it has been told to go (guidance) and then do it (control)d) Integrate, test and fly it without you in the loop. Autonomous.Not trying to rain on your parade, just trying to see what the "base" expertise here really is...
You need to be a member of diydrones to add comments!
I have gone from 0 to 60 on this. I have designed and flown 9 iterations of UAVs I have flown with both MicroPilot and UAV flight systems AP-50 which I was part of developement for (software and beta flight testing). I am currently designing an Unmanned boat autopilot (PSAV)which can be used on both motor and sailing craft. I am a software engineer and an embedded systems engineer. My UAVs (all of them) fly completely autonomously. Though we launch and recover manually as was brought up previously. It is very hard to do that. Well, not hard but very, very expensive to teach an autopilot to do that because mistakes are catastrophic! We tested several methods of lining up for finals and even short finals but without a lake bed, the gps onboard will not give you the accuracy you need to stick your landing on the numbers or even on the field!
I have become very proficient at tuning PID loops as well. As well as programming them. As for the Khalman filter junkies out there... A UAV can fly very well and stable without a bunch of gyros keeping the aircraft level. A roll and pitch gyro to quickly compensate for wind gust is enough! A Yaw axis gyro helps with turn rate.
That is my understanding as well Patrick. And you are right, that was my real point about autonomy. If it isn't a complete soup to nuts operation, it just seems like R/C with bells and whistles.
Just need to get the price well below $30k.
I'm glad everyone got past my "tone", as there was none intended.
From the description I read, I would say he’s talking about fully autonomous flight.
From launch to landing with no pilot intervention (still with PIC capabilities).
I'll ignore the condescending tone and try to answer the question. Some of us here have done it all (including me, both with commercial autopilots and home-builts). Most have not, but the point of this site is to encourage people to learn about aerial robotics from the basics up. If that just means learning embedded electronics, PCB design and GPS programming, along with airframe design and integration, and they never finish an autonomous flight, that's still an education worth having.
But if your question is how many people here understand the definition of autopilot as you've stated it, I think that's practically everyone. Certainly all of our GeoCrawler and Arduino projects qualify.
Replies
Borderhawk UAV
Project: PSAV
I have become very proficient at tuning PID loops as well. As well as programming them. As for the Khalman filter junkies out there... A UAV can fly very well and stable without a bunch of gyros keeping the aircraft level. A roll and pitch gyro to quickly compensate for wind gust is enough! A Yaw axis gyro helps with turn rate.
Just need to get the price well below $30k.
I'm glad everyone got past my "tone", as there was none intended.
From launch to landing with no pilot intervention (still with PIC capabilities).
But if your question is how many people here understand the definition of autopilot as you've stated it, I think that's practically everyone. Certainly all of our GeoCrawler and Arduino projects qualify.