Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Next »

Day recap

  • 5AM flight ICARUS to run autotune

  • Run autotune for 4:30 min, the drone lost all autopilot and dropped about 1-2 meters

  • caused mechanical failures:

    • superstructure mostly undamaged

    • landing gear has a crack, leaving that to monitor

    • weight in the cabin s

  • Drew more than 2amps via BEC that has a 2 amp limiter for the pixhawk is the assumed cause. Repaired by switching out the BEC to one with 8 amp limit (I think)

  • repaired the drone after this

Presentation

  • Went great - we presented last, picked 1st place

  • They liked that we had pitched it to “big city” and followed a presentation as if we were presenting to a mayor of a city instead of technical experts

Path optimization

  • Task 2 code looked good, only a few edits, maximized revenue.

  • Then switched priorities to change priority to distance and simplicity instead of profit

  • NOTE: probably should’ve reassessed this requirement PRIOR to the day we needed to use the algorithm

  • ended up not using the algorithm to prioritize simplicity for the drone

  • Algorithm was still helpful for considering alternates and high revenue options

  • 13 landing pads with 15kg weight limits. Had to fly at least 50% of the 19 total given

  • So fly 10 routes

  • However, was informed that 10kg weight limit was lifted, so we recreated the route to include 10 kg pads as well

  • Made some landing pads easier to get to → makes the flight path look pretty achievable.

  • Order is easiest to hardest

  • Task 1

    • fixed the bug that it flies through the restricted zones

    • Need to test if we can upload a mission while the drone is in flight

    • and the current mission state

    • that’s how the diversions are followed

Flight readiness @ 1400

  • flight termination in fixed wing flight was discussed

  • fixed wing spiral down was spiralling nosedive

  • kill switch → out of sky

  • controller disconnect → spiral down and QLAND, or just QLAND

  • passed the review

  • however 1 kg overweight

    • started taking material off the drone

      • hacksawing 3D prints and drilling things out, removing tape, tack, extra bolts, a redundant phone video system and battery

  • passed the next weight test after 3-4 hours, 14.8 kg

Fixing houston

Flight test at 1830

Flight Test attendees - all

Flight line - Anthony Luo Megan Spee Aidan Bowers (Deactivated) Nathan Green Conall Kingshott

Objectives

  • Validate flight after several failures in autotune over the last few days, including one at 5AM this morning that broke part of the cabin

  • See if the different BEC (5V line power to the pixhawk) fixes the full system shutoff error

Notes

  • checklist was not followed much, we could not take the transmitters out

  • should and can change the checklist to account for this situation

    • plug telem back in when done

    • can walk back n forth with the drone because we are at the airfield

  • Saw carleton fly semi transition with a flexible frame

  • Compass calibration was cool!

  • At the flight line: did control surface checks, motor test, arming check

    • Op flow issue, unplug->plug back in to solve

    • Anthony Luo flew in hover QLOITER, landed, no incident. throttle touchy,

    • Megan Spee also flew hover QLOITER, during the descent to land, rear right motor seemed to drop out and it fell from ~1m

  • As far as Aidan Bowers (Deactivated) can tell, nothing to do with Ardupilot

    • Fell for less than a second

    • was trying to YAW clockwise. Ardupilot wasn’t trying to throttle up or anything though to correct for this

    • Nathan Green says likely ESC or wiring issue

      • Maybe faulty ESC? was it a soggy ESC?

    • Motor didn't STOP spinning it just slowed down and lost throttle

    • Using DSHOT → if it doesn’t receive a valid signal, it will just stop. But it wasn’t stopped or locked (we have locking motors too)

    • Didn’t see this issue on cornflakes, we are using different ESCs, possibly that issue. Overheating? Was only on for a few minutes. The motor that failed was pinned at nearly 100% throttle. around 95%

  • didnt use radios

    • make sure behind flight line ppl arent calling up flight line

    • maybe conall has earpiece and filters behind fline info

  • also didnt take vtx. ended flight before plugin

Action items

Testing ICARUS motor - run it with no props on? See if it cuts out. if so, it’s a signal issue

wire something up to the pixhawk to get ESC stats so we can see what’s up with the rear right motor.

Might abort partway through task 1 if it looks like it’s gonna be bad + if we need to do more testing

Tomorrow - flight window opens @ 10:00

Breakfast 630-730

800 -1000 prep. start doing checklists @ 930. just over an hour to fix issues

  • devices need to charge immediately after breakfast

  • 6 batteries

  • all controllers

  • laptop

  • VTX battery

  • goggles battery

  • No labels