I can't make my P4P fly away - a thread of a different color...
- Thread starter Harleydude
- Start date
Fascinating and useful video. However, I fly a regular, very simple orbit mission in Autopilot. On one occasion (and one occasion only, a couple of weeks ago the P4P just started to wander off course and away from the orbit. I paused the mission, took over to regroup then restated mid-flight. After this the mission completed successfully. Conclusion? Autonomous flights aren't foolproof (at least not in Autopilot) and the operator needs to concentrate and standby. It may relate to the different operating methods of the software. Autopilot relies on the device constantly updating the AC. I believe that MP and possibly GS Pro in waypoint or mapping modes, upload the complete mission which the AC then flies, no more RC input until you take over to land. Not sure on Litchi methodology.
Same for Litchi... except for 'gimbal interpolation', there has to be a connection between the AC and RC. When there's no connection, the gimbal angle won't change between 2 waypoints.
I have a hundred acres and with my p3p using litchi flying it straight down to the back and making a survey of the property lines I have lost signal and litchi continued the mission and returned it back to the home point and hovered as it was supposed to do. This is a large area that is east to west as a rectangle it has trees no higher than 70 ft and it is basically flat. We followed the Drone in the pickup so that we were able to maintain line of sight the whole time just for those who would wonder.
I can answer that for you. I'm positive Map Pilot will continue its mission past 30% (my setting for level 1 warning). Map Pilot sends the drone back home based on estimated flight times based on a full battery. So if you send the bird out with a 75% battery, Map Pilot will continue flying past 30%. I've not pushed a mission past 28%, and have never landed lower than 20%, so I can't tell you what happens @ 10%. I'm sure it would land, and not RTH.
Regarding GS Pro, I'm about 80% sure it behaves the same way, but I would have to test.
D
In theory, yes. I turn off OA when flying manually. I don't trust it and it find it causes erroneous behavior when I fly near stuff (or even when I fly AWAY from stuff). What I have found is that SOMETIMES Map Pilot and (I think) GS pro turns all that stuff back on. I can't seem to find a pattern.
For automated flights in a rural areas, I activate "RTH Obstacle Check" and enable "Vision Positioning." I find that Vision Positioning CAN be an asset when landing. After a mission, I generally will either add additional input by either manually descending the bird during RTH, and/or speeding it forward. OR...I'll just throw the switch into Sport Mode and fly home manually. All that said...
As of this writing, I haven't utilized Obstacle Avoidance for any autonomous missions. I'm pretty cognizant of the bird's location and altitude vs. my surroundings during all missions, even if the bird is far beyond LOS. I put a lot of trust in the telemetry data, which hasn't lied to me yet <knock on wood>. Plus, I always utilize the camera feed to insure I'm not flying into something like a mountain or building. My right C2 button is set to move the camera from vertical (mapping position) to horizontal (RTH position). With all the real-time tools at our disposal, a conscious, cognizant pilot shouldn't have a need for Object Avoidance. That said...
I recently did a job that had a 65' building close enough to me that it obstructed view of the drone. In addition, for the first flight, video feed was horrible. Turns out, we were in a massive WiFi zone. For the second mission I switched the bird to 5.8Ghz mode, and all those issues cleared up. But, because video feed was so bad during the first mission, and I had this 65' tall building next to me obstructing view of the bird, I turned on OA. But as soon as I had rock solid video feed, I turned the OA off.
I guess what I would take away from all this is that OA should be thought of as a net, and NOT a primary means of keeping the bird safe. That responsibility rests on the shoulders of the pilot. I think OA tends to make some guys complacent, which leads to crashes and fly-aways...in my humble opinion.
D
Yes and no. As I stated in my video, "Depending on the software used and the settings...."
A quick rundown...
GS Pro uploads missions and flies completely autonomously. GS Pro offers options regarding connection status. But regardless of the option you choose, as long as a connection is maintained, GS Pro WILL finish its mission.
Map Pilot gives the user a choice of "Connected" or "Connectionless" flight. In the case of the former, the software controls the bird via the iPad via the Remote controller. So in this case, not only does the bird need to remain connected to the RC, but the connection has to be a good one. In the case of the latter, all waypoints are uploaded and the bird flies the mission regardless of connection status.
Autopilot, like Map Pilot, has options. Depending on your mission needs, Autopilot may depend on the app to control the bird, in which case connection must be maintained.
I have a fair amount of experience with Autopilot. Every month I fly a waypoint mission for a construction client. Autopilot also controls the camera, so I must fly with a solid connection. I have found that Autopilot is very resource heavy. This tends to heat up the iPad. Because iPads have no active cooling, they cool the CPU by throttling it down. This can cause Autopilot to crash (I have experienced this about 1/2 dozen times). MY experience has been that, when Autopilot crashes, the bird (an Inspire 1, Gen 2) just hovers. This only happens when outdoor temperature is above 75°. During cooler fall and winter months, app-crashing is not an issue. During the summer, I have negated this app-crashing anomaly by storing my iPad in the refrigerator for a few hours pre-mission, and then bringing it to the gig in a cooler with some ice. I then make sure I fly from a shaded area. This works for my Air Pro, but not so much my Air 1. Clearly, the Air Pro's faster CPU doesn't "strain" as quickly as the Air 1 does.
I find it interesting that your bird just drifts away. I've yet to experience that anomaly with any app under any condition with any of my birds; Inspire 1, Mavic Pro, P4P. That said...
I utilize legacy firmware in all my birds, and legacy software most of the time. While I use the latest/greatest GS Pro and Map Pilot, I have locked Autopilot down to the original version that has worked for me for the past year. Once this construction job is finished, I may allow Autopilot to update. I use legacy versions of Go and Go 4.
D
Last edited:
It's a standard speaker stand from a company called "On-Stage." The stands are cheap, light-weight, portable, robust and sturdy.
The holder with a gooseneck is from a company called Arkon. Like the stand, it's very sturdy. Unlike the stand, it's a bit awkward to transport. And while it is relatively sturdy, I wouldn't attach an iPad to the RC while clamped in. The mount "grips" the RC with a spring action. Too much weight in an unbalanced manner, and the RC will slip out. Don't ask me how I know....<:^/
Notice that the RC is level with the antennas @ a 90° angle.
I think the stands are like 30 or 40 bucks each (you can find them at music stores). And the Arkon clamp was 30 bucks via eBay.
D
I agree completely, OA is relied on too much, because it is so heavily over-advertised to the consumer. “It cant be wrecked!” Some say, and then they go on believing that until they wreck it. Solid video. Thanks.
Firmware update on my p4 could have caused me of losing the drone. Latest firmware made it lose its ability to rth. Downgrading to the previous fixed it. Nice video, btw.
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,303
- Reaction score
- 9,247
It's extremely unlikely that your firmware made any difference at all.
It certainly did not cause your drone to lose its ability to rth.
Since you never identified the actual cause of your issue, it seems that there really was no issue in the first place.
Operator misunderstanding is the most likely explanation.
I think you have a comprehension problem. Or just a make -believe expert. Or just plain....slow? Lol. I’m outta this forum.
Similar threads
