The ArduPilot development team is proud to announce the release of version 3.5.1 of APM:Plane. This is a minor release with primarily small changes.

The changes in this release are:

  • update uavcan to new protocol
  • always exit loiter in AUTO towards next waypoint
  • support more multicopter types in quadplane
  • added support for reverse thrust landings
  • added LAND_THR_SLEW parameter
  • added LAND_THEN_NEUTRL parameter
  • fixed reporting of armed state with safety switch
  • added optional arming check for minimum voltage
  • support motor test for quadplanes
  • added QLAND flight mode (quadplane land mode)
  • added TECS_LAND_SRC (land sink rate change)
  • use throttle slew in quadplane transition
  • added PID tuning for quadplane
  • improved text message queueing to ground stations
  • added LAND_THR_SLEW parameter
  • re-organisation of HAL_Linux bus API
  • improved NMEA parsing in GPS driver
  • changed TECS_LAND_SPDWGT default to -1
  • improved autoconfig of uBlox GPS driver
  • support a wider range of Lightware serial Lidars
  • improved non-GPS performance of EKF2
  • allow for indoor flight of quadplanes
  • improved compass fusion in EKF2
  • improved support for Pixracer board
  • improved NavIO2 support
  • added BATT_WATT_MAX parameter

The reverse thrust landing is particularly exciting as that adds a whole new range of possibilities for landing in restricted areas. Many thanks to Tom for the great work on getting this done.

The uavcan change to the new protocol has been a long time coming, and I'd like to thank Holger for both his great work on this and his patience given how long it has taken to be in a release. This adds support for automatic canbus node assignment which makes setup much easier, and also supports the latest versions of the Zubax canbus GPS.

My apologies if your favourite feature didn't make it into this release! There are a lot more changes pending but we needed to call a halt for the release eventually. This release has had a lot of flight testing and I'm confident it will be a great release.

Happy flying!

Views: 5503

Comment by zhangpengsir on March 20, 2016 at 4:59pm

Great, many new highlights!

Comment by Darius Jack on March 20, 2016 at 5:00pm

My congratulations.

Any chance for manual describing every feature in details, as one required by FAA under new legislation pending ?


3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 ‘‘(4) each operator of a test site under section 44802; ‘‘(5) the Center of Excellence for Unmanned Aircraft Systems; ‘‘(6) unmanned aircraft systems stakeholders; and ‘‘(7) community-based aviation organizations. ‘‘(d) FAA APPROVAL.—Not later than 1 year after the date of enactment of the Federal Aviation Administration Reauthorization Act of 2016, the Administrator of the Federal Aviation Administration shall establish a process for the approval of small unmanned aircraft systems make and models based upon safety standards developed under subsection (a). The consensus safety standards developed under subsection (a) shall allow the Administrator to approve small unmanned aircraft systems for operation within the national airspace system without requiring the type certification process in parts 21 and 23 of the Code of Federal Regulations. ‘‘(e) ELIGIBILITY.—The standards for approval of small unmanned aircraft systems developed under this section shall set eligibility requirements for an airworthiness approval of a small unmanned aircraft system which shall include the following: 


Comment by Philip on March 20, 2016 at 5:01pm
Comment by Cala on March 20, 2016 at 5:16pm

Cool!!! thank's Tridge, the reverse thrust landing, reverse prop? how it's works? 

Comment by Chris on March 20, 2016 at 6:59pm

Hi Cala, i think they are using ESC's that can be reversed on the fly.  The brushless motors don't mind which way it goes.  The prop would be way less efficient though but should be enough to make it quite draggy to steepen the decent down.

Hi Tridge or Tom, is this the deep stall stuff that was being worked on a while back?  Looked promising but I'm guessing a forgiving airframe is needed?


Comment by Cala on March 20, 2016 at 7:05pm

Thank's Chris, that's great to land over crops with less damage risk :) you need a special esc or common ones are enought?

Comment by Chris on March 20, 2016 at 7:30pm

Hi Cala,

I think their using reversible ESC's for it, so a special ESC as normal ones wont do it.


Comment by JB on March 20, 2016 at 7:46pm

Tridge and team -  Awesome as always!

Reverse thrust yay! ;-) Thanks Tom!

Is it possible to use reverse thrust on a quadplane for the forward motor? I'm thinking steep decent with a quick quadplane flare at the end could reduce hover time for landing and resulting exposure to wind, similar to how a "jump" to forward flight can be used at launch.

Also what is the function of LAND_THR_SLEW parameter (which you've listed twice I think) and LAND_THEN_NEUTRL parameter? Are these for fixed wing or quadplane aircraft? I'm also curious as to what multirotor layouts are now supported (quad, octo obviously) and what QLand does. Is SBUS out still in the works?

BTW It seems that the rear left quadplane ESCs have a higher likelihood of failure than the rest... ;-)  Yesterday one of our dinky little 7g 20A esc melted down (and desoldered itself!) resulting in the quadplane doing a flip at 3m above ground after prolonged hover tests in wind. No damage was done at all, just a new ESC required. We're only running 2206 on 6x4 for 2kg MTOW. Quad setup is around 200g all up!. Flies well with plenty of reserve and the motors run cool, but I think we'll put the 30A ESC version in, that will add an extra 16g though... we're trying to keep the weight down to essentially a flying battery. I suppose every solution comes with it's own set of problems!  ;-) lol.

One last idea for quadplanes: Have you considered using a asymmetric motor v-tail quad layout for better yaw control in wind?




From what I understand reverse thrust is accomplished by using a reversible ESC (SimonK etc) and the prop is forced to spin backwards on decent to reduce approach speed. Very useful in particular for efficient aircraft like wings etc that have good glide ratios, but as a result are hard to land in tight spaces.

Comment by Andrew Tridgell on March 20, 2016 at 8:21pm

@JB, I haven't really thought about reverse thrust landing for quadplanes. I think it should actually work with the current code, as the quad motors will automatically cut in to help at below Q_ASSIST_SPEED. Let me know if you try it!

On LAND_THR_SLEW, there are some docs here:

on LAND_THEN_NEUTRL, docs are here:

both are for fixed wing.

On quadplane frames, it supports all sub-types of quad, hexa, octa and octaquad so far. No support for tri yet. This is just a result of the structure of AP_Motors from the copter code. It was easy to support any from the MotorMatrix class.

On v-tail quadplanes, I haven't looked at that, and probably won't look at it anytime soon, but I'd welcome experimental results from others :-)

Cheers, Tridge

Comment by JB on March 20, 2016 at 9:00pm

Thanks for the info and links Tridge.

I'll be happy to share some more info on the V-tail quad once we have more time after D2. ;-)

Quadplane reverse thrust is going on the test schedule now. I'll need to try it on the old X8 wing too. I'll let you know how it goes!


You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service