Yes I got the copter up on my command and lost control as soon as it got off the ground. So the original throttle input you hear was my control. I tried to shut the motors off at least 4 times to so luck.
I still don't believe everything fails but that doesn't mean any flyer should be thought of to "not fail" or that you should be able to purchase anything and have it "not fail". I was just talking about a blank statement that was made. Look at the majority of the commercial airplanes out there. 99.9% don't have a failure that results in a crash but that doesn't meant that planes or drones don't crash or that I was making the assumption that anything that doesn't fly shouldn't crash. Obviously with a failure 1 out of every 60 flights isn't a very good percentage and I'm just more frustrated than anything. I have been looking at higher end systems and many of them have dual or triple redundancy with very robust recorders. The frustration of these latest failures is compounded by the inability to have precise data logging with it. I worry about what we are going to be have to do once regulations come out. Will the FAA have established auto pilot systems that you need to use, or have specific type of data logging technology? So while I am trying to find an alternative to the MK boards for "better" reliability, could it mean that we all need to start over because of dictated regulations? Not saying that's a bad thing especially if it will make things safer. Just a thought that goes through my mind as I go through this process now. Could we as flyers have too much emphasis on the filming equipment and not enough on the flying equipment and I say that with the most respect. I know that many of us would purchase a 3-6K system it proved to be better. We (including myself) focus so heavily on what cameras to use and the equipment to fly the smoothest but should we be investing more into the higher end auto pilots? To clarify when I say auto pilot solution I am referring to the term auto pilot as the term used for just about every flight control system that has some sort of automation for PH or AH. I do not use way points but I do like to have position hold and altitude hold for some of what I do. I think some of us (including myself) shrug off the thought of spending 10K or higher on an auto pilot system but many of won't think twice about spending big $$$ into a gimbal. Could our logic be slightly flawed? IS the ratio of flight control to camera performance investment skewed? As a growing proffessional these are the things that I am thinking about as an owner, pilot, and business man.
I completely understand where you're coming from, Dave. We're in a tricky place at the moment because most of us would love to ramp up our "game" commercially, but in the US we have major hurdles in our way. And this is also hampering the hardware/software end of things. This is because if I'm a manufacturer of high-end flight electronics for UAVs, I'm hesitant to a) invest heavily in R&D for a market (at least in the US) that's underperforming due to lack of regulation and/or standards and b) show my cards to my competition before there's a solid commercial opportunity. Maybe that's why we haven't seen the Sasquatch of flight control systems (the vaporous Synapse). Commercial and military aircraft stay in the air despite numerous failures because of well-engineered redundancy. Flight logging is an ancillary issue. What we really need is to design a commercial UAV that can withstand the failure of almost any component and land safely (or even continue operating). Right now we have things like LiPo batteries, I2C buses and power distribution boards that can easily fail and cause a machine to instantaneously drop out of the sky. This is unacceptable moving forward. For example, why has nobody built a multirotor with true redundant power systems? Two batteries, two power distro boards, two sets of ESCs, dual flight controllers, dual radios, etc.? Because none of us would buy it, that's why. And if you raise your hand and say, "I'll build one", good luck...this is no easy task. I believe it's going to take a serious R&D effort. There's VC money out there trickling into companies doing this kind of thing, and the big boys (Boeing, et al) are probably already there. But again, we scoff at the $50K-$100K birds because the US commercial market is DOA at the moment. But the problem is this: to design and build a small UAV that's got engineering comparable to commercial and military aircraft would require a significant investment, and the resulting product(s) would cost dramatically more than any of us is currently spending. Are we ready for a $25,000 airframe? If you want a (more) bulletproof experience, that's what it's going to take. But right now, that just doesn't exist.
We're still at the biplane, doped canvas, steel wire with turnbuckles, stage of sUAS development. As you say, Steve, we will witness the maturation of this industry over the next few years in the same way we witnessed the maturation of the personal computer business. Hobbyist prices are good, but the Fifth Law of Thermodynamics holds true: you gets what you pays for. True redundancy only happens when we have triple systems that vote on what the aircraft should do and we can tolerate the failure of a redundant system. I'm going to start on a Cinestar 8 HL redundant test platform system over the holidays -- I want to get insights into the failure modes. I will be using MK avionics including the MK Double Quadro Cool Redundance. As you say, a dual LiPo system with isolation capabilities is a must -- if one LiPo goes to the Happy Hunting Ground, you cannot tolerate it dragging the "good" one down (in every sense of the word down). Until then, we are with a different level of of reliability. Of course, us pilots (if you fly the way I do) are probably the weakest link... Andy.
Talking about redundancy issues i spied this on the quadrocopter llc website this morning. https://www.facebook.com/quadrocopt...9191474120836/838649782841665/?type=1&theater. continue......
I saw this the other day. Zero UAV has a dual redundant system. If it works as advertised this is the type of system I have been waiting for. Has a real time comparitor that is supposed to measure the performance of both systems and if the primary looks like it is having issues it auto reverts to the secondary one. If any issues arise at start up it will not takeoff. http://www.fpvmodel.com/zero-uav-ge...dundancy-autopilot-for-multi-rotors_g480.html
Interesting find, Shaun. Reading the sales pitch: "Meanwhile it is quite a reliable autopilot ..." : Hmmm. I was with you until "quite." "Automatic parachute opening in emergency." Nice. Wonder if the pilot can also activate it? "Automatic dodging landing gear." Yeah! Automatic dodging sounds great! (I wonder what it dodges?) "Aerial rocker arm function with 256 waypoints available." Errr. Rocker arm? Anyone got one and tried it? How's the customer support? Andy.
I know Jason Toth of rvrd cinema is now using them and he used to use wkm and a2 systems. I researched it a bit on rcgroups and from what i gathered it still seems too new to the market and not tried and true yet. Someone needs to make a multi rotor frame that is able to glide if its motors fail, basically an osprey but with 4 motors and the ability to glide like an airplane.
Tristan, I'm no engineer, but the amount of wing area required for a 10kg (or so) multirotor to glide would be enormous. Of course, if I could forcibly eject my 3-4kg of LiPos, it'd work better, but that'd be frowned upon, right?
Using the airplane analogy there is a difference between a home built and a Boeing. The FAA won't let the home built be used for commercial flight operations. Failure rate is much higher. I live next to a small airport. One crash every couple of years with a home built involved. Difference in engineering and price is self evident. Andy is correct. We are in the Orville and Wilbur era. Low cost is the antithesis of 99.99% reliability.
Most of our ships would be considered normal if they had one '9' and fantastically successful if they had two. 4-5 should be the goal. It's going to take some serious work to get there.
I read on another site that the dodging function starts yawing the copter to keep a fixed landing gear out of the shot. That sounds like a gimic of course. This may not be the holy grail of autopilots, but I do like where it is finally going.
This kind of ship may be where things are headed. I like the versatility. Not sure if this is the ideal aerial photo or video platform, though. And this video doesn't show it landing. I hope that is just an editing choice and not out of necessity.
I suspect a ship like that will be quite hard to hover in PH with wind -- the "wind sail" effect will be pretty large. I'm more of the school of thought that VTOL and fixed-wing will be separate aircraft types for the foreseeable future and VTOL will make for better video platforms because of the "hang time." But then what do I know? Andy.