Comments

• 

  Bot Thoughts January 27, 2014 at 8:19am

  That sounds really great, TC!  APM:Rover is definitely a platform to beat. I'm really looking forward to AVC this year. I've tried to get my small group to port APM to our board, but... well, it hasn't worked out. :(  I've been working on my code, mostly cleanup, rearchitecting, and most recently giving my gps lag compensation algorithm more scrutiny. It seemed to be working beautifully a couple years ago, top speed ~8m/s, an apparently high degree of accuracy and precision. Hoping to get that tightened up even more and officially measure performance. I'll probably have to switch to my other, much faster, better handling platform if I want to give your Slash a run for its money. :D
• Jon-Paul January 26, 2014 at 11:44pm

  Really interesting update Thomas, looks like you are on to a winning platform for 2014.
  
  What are the competitive improvements you think you have inherited from the Pixhawk?
  
  Interested in using the platform for custom development and need to examine the learning curve required to comfortably develop, however I am concerned about losing the vast community base available with the arduino platform.
• Admin

  Thomas J Coyle III January 26, 2014 at 11:14am

  Hi All,

  If any of the Group members are contemplating running a Pixhawk on their rover in place of an APM, you can rest assured that it will function just fine as I just completed testing a Pixhawk equipped rover.

  I installed a Pixhawk on my Traxxas Slash 2WD brushed rover and ran it over the same test track that I run my APM equipped rovers on. The Pixhawk performed equal to or better than my APM2.5 equipped brushless rover using the v2.44 beta firmware with the low latency GPS configuration. I was able to run the rover at 8m/s and did not have any rollovers in the turns. The chicane performance was flawless even with the NAVL1_PERIOD set to 8.

  When I was at the 2013 AVC I won 2 of the 3 Peloton heats with a maximum cruise speed of 3m/s. This time I feel confident that I can use a cruise speed of between 8 - 10m/s with confidence and successfully navigate the 2014 AVC without a rollover in the turns:-)

  Regards,

  TCIII ArduRover2 Developer   
• Admin

  Thomas J Coyle III January 19, 2014 at 5:21pm

  Hi All,

  We will be starting the New Year with a new release of the ArduRover2 firmware which will be release v2.45.

  The new release will include the ability to add two sonars to the new Pixhawk navigation processor by adding the capability to define PWM pins as GPIO pins and will improve the rover position knowledge by reducing the latency in the GPS data acquisition function.

  I have recently tested the reduced GPS latency code and it has provided a marked improvement in the rover's ability to accurately navigate a tight chicane course while maintaining smooth straight line performance.

  These and several other minor improvements in the rover code will be released in the coming week.

  TCIII

    
• 

  Antonie Kruger January 3, 2014 at 10:51pm

  Hi Guys, I recently acquired a Dagu Wild Thumper 6WD chassis and motor controller. I intend to use it for an autonomous rover project with one of my APM boards, - from Dad to Dad for Xmas :)

  I trust all had a blessed Xmas and have peace and prosperity for the coming year.

  Antonie
• 

  KM6VV December 24, 2013 at 10:09am

  Hi All,

  Best Christmas wishes from me as well!

  Alan KM6VV
• Admin

  Thomas J Coyle III December 24, 2013 at 5:03am

  Hi All,

  I would like to wish all of the members here a safe and happy holiday!

  We have a lot to look forward to in 2014, especially auto tuning and the Sparkfun 2014 AVC.

  Regards,

  TCIII ArduRover2 Developer
• Admin

  Thomas J Coyle III December 12, 2013 at 2:01pm

  Hi All,

  It may seem like there has not been much activity concerning the ArduRover2 firmware development as there has not been any recent updates to the code. However Tridge, Linus and I have been testing the latest ArduRover2 v2.44 beta 1 firmware to see how it performs and to establish baseline values for the operating parameters that have the most affect on navigation performance.

  Presently we are trying to get a good balance between straight line performance (minimal weaving) and tight turns in the chicanes. Basically high NAVL1_PERIOD values give good straight line performance, but result in wide turns in the chicanes. And just the opposite, lower NAVL1_PERIOD values cause weaving during straight line navigation, but result in tighter turns in the chicanes.

  Therefore we are shooting for a balance between the NAVL1_PERIOD, the STEER2SERV_TCONST, the STEER2SRV_D, plus several other less influential parameters that will result in good straight line navigation as well as robust chicane performance.

  More to come as we continue refining those parameters.

  Regards,

  TCIII ArdurRover2 Developer 
• 

  John Moore December 7, 2013 at 8:44am

  Hey everyone, my adrurover has been off line for some time.  Have been dealing with GPS accuracy still which I have concluded its due to the tight confined space I have.  With the drone an gps only 2 to 3 inches away from the brushed motor, I am guessing thats where my issue is.  So I am using that as an excuse to buy a new crawler.  A 1/5 scale Exceed MaxStone with motor on axle configuration.  At 33 inches long, it should give more than enough seperation between the drone hardware an motors.
• Admin

  Thomas J Coyle III December 5, 2013 at 5:41pm

  @mikel-ange,

  I do not know of a twin centered joystick transmitter with 5 to 6 channels, however I suspect that you can add springs to the non-centering joystick in a Mode 1 or 2 transmitter to make it perform like the centering joystick. Though you might have to remove the ratchet from the throttle joystick axis. You might want to make this project of yours a discussion so it will not get lost on the Comment Wall.

  Regards,

  TCIII ArduRover2 Developer