Rover for Testing Systems the Ground

3689583151?profile=originalHello everyone from the Aerial Vista Challenge Team. We have been trucking along on out competition UAV.  We are real happy with the progress and we are getting all knids of stuff in from sponsors.  This week we received our Arsov AUAV3 and in a couple of weeks we should be getting our AUAV X-1 Autopilots in.  We will really be talking about those a lot in the near future.  I would also like to add that M.A.R.S. Parachutes is sending us a mini parachute and we should be getting it in anyday.

Today I want to talk about something a little bit different.  Rovers!!!  We have a challenge here in the Texas Panhandle. I know that a lot of people deal with weather and sometimes it is just impossible to fly.  In the panhandle our big challenge is wind.  An average day here sees gusts in excess of 25 to 30 mph.  In RC this has not usually been a big deal.  With 600 size helis or larger 40 mph are navigable.  In RC planes we usually have stable simple systems.  However, in UAV work we are often pushing the boundaries and we have software that is not always stable. So between the winds and unstable programming we have a large possibility of a plane falling out of the air.  Our solution has been to make a ground rover for systems testing.

Obviously there are lot of things with plane AP's that are not able to be tested on the ground.  I want to talk about some of the things that we are hoping to accomplish with this system and what we think others can use them for.  First I will talk about the platform a little.  This is our one day rover.  I assembled this on Sunday.  I was looking for a crawler but I found a REVO that has been modified to be electric with a brushless motor.  This is obviously a very fast machine.  We are dialing down the throttle and of course programming a large amount of exponential into the radio.  We will also tune the steering so that it is mild and we will not flip and damage components.  The nice thing about this setup is that the slipper clutch is locked and so we do not have to worry about excessive wear at low speeds.  These are some things that we used to make it usable.  We needed a cheap flat platform to place electronics.  I decided on a cutting board.  They are inexpensive and made of durable Delrin.  The other components on this setup are a 500 mw 900 mhz av setup and a 433 openlrs rx.  We have the telefly pro for tracking modulation and minimal OSD.  We are starting with these systems.

The Openlrs is a great new addition to long range radio market and a great benefit for those that want more flexibility in programming.  Combined with an OpenTx radio and channel wizards you could have a very robust command and control system.  The disadvantage is that they are not exactly plug and play yet.  So these are for people who are up for debugging and personalizing a system for your own needs.  That being said, we sure do not want to find bugs in the air.  That is why we are starting with such a basic radio setup on the Rover at the moment.  I have flashed a 1W Tx and 100 mw TxRx with openlrsng.  This is the most recent stable publicly available update for the system to date.  We are just range testing and making sure that the channels work like we want them to.  We will actually be hacking a 1WTx into a reciever for the plane.  This will give us a 1W 2way communications link with the plane.  We should be able to receive telemetry and send commands on the same radio frequency at a great distance.  This has been tested with APM by the developer Luxembourg.  We will be releasing his and our work on this hack soon.  Obviously there are a lot of variables at play here so we are going to be doing extensive ground testing of this setup before we put it in the air.

The telefly pro has been added on board because we need to test our tracking system.  We have an MyFlyDream Automatic Tracking antenna and it requires that the telemetry be modulated across the audio and video channels to give us a diversity link with the plane and allows the tracker to follow the planeby coordinates and altitude.  This tracker probably stands in a category all by itself in ability and we are proud to be representing MyFlyDream with this product.  The Telefly Pro can stand alone in the system as we have it on the Rover for now or can be used in conjuction with other AP's and OSD's that do not support modem modulation of the telemetry data.  The MyFlyDream Autopilot is the only system I know that supports both video and audio channel signal with a built in OSD and AP menu without the use of a GUI.  However, it is a simple 8-bit AP like most others to date and leave a lot to be desired to combine with embedded systems and the firmware is proprietary.  While I rank it as one of the best in class for hobby class autopilots, it is not what we desire our selves for development.

We will use the rover to test the long range radio and to get the tracker calibrated and running smoothly.  Both of these will be important to be stable as we integrate them into our airborne platforms.   I also like the idea that we will be able to practice FPV control without worrying about losing orientation in 3D space.  I think this is a system that a lot of people should try before jumping into FPV planes.  It takes a lot of practice and familiarity with the radio and onboard systems control requirements.  Even flying a an RC pilot I have confused myself with multiple switch commands while trying to get used to a new system.  I think testing without worrying about the extra costs of crashing is invaluable.  We will be able to test our multiple vision systems.  We can test the radio environments and the requirements for smoothing a lot of the interference out before actually getting airborne.  We will be able to test embedding a linux based micro-controller for our sensory systems.  All this can be done without worrying about sacrificing an airplane.  We can perform critical tests regardless of the weather or wind.  If you are new to FPV, autopilots or are a developer looking for a stable test platform, a rover is an in inexpensive platform that can save a lot of headache, time and trouble.  We built this for a few hundred dollars, not counting testing systems.  It is durable and easy to deploy.  

I am sure we will be talking a lot about how much the rover saved our hind ends this year as we continue to progress with our anti-poaching plane.  As always thank you for checking in our progress in the wildlife challenge and I hope that our information can help others in the community with their unmanned systems.  Have a great week.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • I chose to build on this platform  not cheap but i am Very impressed with the quality of construction ,it is nearly immersible straight out of the box  ! The lid is so close fit i have to rap it to seat it and the axle bearings are easy to seal ,this is important because i do live in a Swamp ! 

    It is built like a Tank 1/8" aluminum solid welded chassis 10" wheels  , i added an internal set of axle support bearing  they do not include in the kit  so i can run Tracks made of plastic conveyor chain 4" wide  .I will have to make internal and external growsers  but 40 Lbs  it will be hard to stop it even in heavy brush .

    The large format 26" long  by 22" wide  and heavy weight means i can add a 24" mower deck with swivel casters in the front and mow my yard from inside the house  (seriously)  by remote control , i am getting too old to do it in the summer heat and humidity .

    I am putting an on board quad video processor  so i can get 4 cams on one TX , driving cams will show as 2 small pics over the main cam image .One day it will sport an APM  and do sentry duty with a Thermal imager i already own  , i am starting with a 3  day/night cam setup with the main  pan/tilt  mount on top ,i am still pondering  a double  synchronized  pan/tilt setup though so i can run main color cam and thermal together (laughing evilly  about an OC dispenser  instead of one of the cam`s  for the neighbor hood dogs ( we have a leash law  or an attempted thief  >;) ).

    By pushing to the limit of Part 15 radio  regs in 2.4 and 5.8 Ghz  i hope to get a mile of control distance (distance to Mail box ) out of it with  the tree lines  obstructing  LOS (only one deep one and it is around the house the rest is fairly thin  and open ) That is the fun about Rovers with about $4K in the rover  you really can not Crash it LOL !

    Eventually i will go APM  and " Leave the driving up to it" (rotflmao an old long distance bus line`s  motto )

  • Ok I am convinced I am getting tracks for the rover

  • That sounds like a nice setup you plan Quad.  Mp1 the telefly pro is a stand alone osd.  It is primitive by today's standards but it will handle the gps, current sensor, rssi, and an auxiliary sensor for what ever you wish.  The main function is modulating the gps data to the av channels for tracking.

  • I know they both spit out NMEA messages but i was just wondering if the telefly supports snooping on a source. Does it require both the tx / rx lines ?

  • @Toby Is is possible to hookup the tx line of a ublox gps straight to the telefly pro osd as the gps source ?

  • With regards to taming the blinding speed of the brushless setup, I plan to leave the gearbox in low, install a 6S system like the Turnigy 1/5-1/8 scale brushless system with the biggest pinion gear I can find, then run it on 2S.  Fingers crossed...

  • Thanks Toby.  Based on your description, it is clear that the freewheeling mechanism has been locked also.  I wasn't aware that this was possible.  You have sold me on the Tekno option I think, seems to be the superior choice in terms of cost, time investment, performance and finish.

    The Axial Wraith looks nice, however if you do end up staying with the Revo you can fit it with a sweet set of snow tracks.  These appear to be pretty well put together, but are pretty pricey.


    With regards to stability, swaybars are an option.  Longer a-arms and bigger tyres are also available.

  • Thank you for checking out the blog.  The Rover is not something I believe that we will concentrate too much on but I could be wrong.  We are going to be using it for testing stability of firmware and hardware etc. before we stick the systems in an airborne rig.  The Revo that I purchased was from an individual on facebook so i do not want to act like I am a Traxxas expert.  My setup is a Tekno setup with the stock transmission.  The reverse is disabled in the transmission and the slipper clutch on the transmission plate is locked.  this allows the esc to do the braking and reverse.  In my opinion the Revo is too fast for a long term drone setup and a little narrow in the wheelbase for good stability.  I really would like to get an axial wraith for a long term drone project.  Plus you can get tracks like these.


  • Nice rig Toby!

    I am interested to learn more about this project, as I am currently evaluating driveline options for my own gas to EP Revo conversion in preparation for my first APM rover.

    You mention that the slipper clutch (which is located behind the main spur gear on the gearbox input) is locked.  I am wondering whether you are referring to the centrifugal clutch (which is located in the bell behind the pinion gear)? Are you using a Tekno style mount, or the one from Integy which allows a brushless motor to drive the standard transmission?  If so, does this mean that braking is still done mechanically via the standard disc brake?  Also, how have you dealt with reverse?

    There appears to be a few different options for converting a gas Revo (2.5 or 3.3) to EP brushless.   I really like the idea of swapping out the gearbox for a 3908 which will allow braking and reverse to the dealt with by the ESC.  This option does however require an upgrade to 3.3 chassis (mine is the 2.5, which is 30mm shorter than the 3.3) along with some metalwork in order to get the gearbox to sit right.  I would appreciate any advice you might have on this.  I would love to avoid the custom work if it is not necessary.

    The two images below show my Revo, which has just been torn down.  The top image shows the hardware that is staying, the bottom image shows all the gear that I plan to discard.



This reply was deleted.