I’m pretty analytical, so there’s a fair amount of detail to convey here, needless to say I have dissected this in my head as much as I can, I’ve reviewed the flight capture etc… I think I have some goodies to share…
First, a bit about myself. – Chris, Inspire 1 owner since February. I have accumulated 13 hours flying time, across 130 flights, and flown my Inspire 60km now. So I have a little bit of experience, I’m not an entire noob – I’m also not an expert and I’m always trying to learn more, which is what has encouraged me to detail this and share it…
Some background points, (I’ve made these in the video too):
- I’m flying on the latest firmware, currently version v1.2.1.06
- I’ve had 7 or 8 flights on this firmware already without any issues.
- All of my batteries, bird and RC were all updated, no issues.
- It remains on my SD card.
- I had two flights today, one following another, the first flight was perfect, no issues at all.
- I compass calibrated before flight 1 but not before flight 2.
- Between the flights, for no reason what-so-ever, I checked the IMU calibration (I did not recalibrate it, I just checked it), it confirmed that it did not need calibrating.
- Between the flights only the battery was swapped, the bird was placed in the exact same take-off position as the previous flight, which is on the edge of my raised deck.
- At the time I took off, I did something unusual, I applied forward stick at the same time as taking off.
- It’s worth noting that the bird was pointing forward, (in a northerly orientation) when I took off, and I kept it facing in that orientation until I applied the very first of my rotational inputs – which is annotated in the video at 02:33. Up until that point the compass direction is showing my bird as if it was facing east, WHICH IT WAS NOT…
- I took the bird out of GPS mode and put it into ATTI mode to try to see if it was moving by itself due to erroneous GPS inputs. This may have helped as it settled down a bit; you can imagine the confusion in the Flight Controller if it was trying to hold a position and it saw itself as drifting to the NORTH (while it thinks it’s pointing to the EAST), so it thinks it needs to generate a RIGHT direction movement. When in fact it would need to generate a BACKWARD direction movement. – Anything like this, I figured GPS mode probably wasn’t helping so I ditched it to try to recover this in ATTI mode.
- My inputs were not having the expected effect. You can see this if you slow the video down or watch it bit by bit…
- I’ve been practicing balancing the unit in ATTI mode against the wind, facing in various different directions, for hours.
- I gathered some safe height to ensure I didn’t hit anything while I worked this all out…
- Inputs in ATTI mode were having some effect, but my inputs weren’t the only thing directing its flight, I suspect the flight controller was still misinterpreting some odd inputs / measurements. In ATTI mode the FC will still try to hold the unit level, I would have thought that this is done more based on the gryo’s, and accelerometer inputs, not utilising the compass much as it’s not obliged to hold it’s heading in ATTI mode (you can start a rotational spin in ATTI and it will continue with a degree of inertia).
- On play back of the flight log I have noted the compass bearing issue, I held a northerly orientation until 1:10 (flight time) / 2:33 (video time). At this point I put in my first rotational input, it appears that the compass jumped back into the correct orientation at that point, as it does end up correctly indicating the orientation which it finally landed in. Beyond this rotation, it took me another 10 seconds to commit to landing it, it was down on the ground by 1:25 mins.
Lessons learnt?:
- I’ve heard this one before and I’ve lived by it: LEARN TO FLY IN ATTI mode, my hours of practicing hovering in one place in ATTI mode, countering the wind, may well have been what enabled me to get this on the ground.
- Another pre-flight check to add to the list. – See if the compass direction is showing the appropriate orientation.
Based on the following assumptions:
- That the compass heading within the replay of the flight log is accurate. If it’s not then all of the compass based considerations are a misnomer… I have considered that the Pilot app may have a bug with regards to the compass heading (when replaying a flight).. I’ll have to look into this separately.
After landing the Inspire, I went out for a bike ride and cleared my head (Stella saidI should do something with the adrenaline)… When we got back I recalibrated the IMU, recalibrated the compass and took it for another fly, during which it behaved perfectly, it hovered absolutely still in GPS mode, compass pointed in the correct direction, it didn’t miss a beat.
I’ll be posting this to the Inspire FB group and hoping for some detailed, insightful advice from the internet experts. 🙂
Here’s the video showing the flight log… (It contains annotations).
With regards to readying oneself – and practicing in ATTI mode… a few weekends ago, I went through 4 batteries back to back, chose a period with a reasonably steady breeze and got my bird hovering safely. Flicked it into ATTI mode in this reasonable wind and did the following.
– Practiced steadying it and trying to hold it in one spot. Once I got the hang of that, turned it, faced me and repeated. – Then turned it to the right, then the left, then repeated in all directions… That lasted about two batteries.
For battery three I got my wife to randomly shout out directions to face, when she shouted, I rotated it, and faced a different way, I had to respond to the changing environment, and battle the wind in all sorts of orientations.
For battery four, I rotated the quad constantly, quite slowly, countering the wind as I rotated it. All the time holding it steady in place. I changed directions of rotation quite frequently.
By the end of my hour of practice, I was able to do this reasonably well… I’m going to repeat this exercise again at some point soon…