Progress Blog

Session: June 19
    • Tried socket communication programming
    • Refactored Toter mode code
    • Followed state-transition coding. Introduced states for each case
    • Problem: Mission sometimes skips some waypoints
    • Initial fixes: 
      • Backup the remaining commands into a new list instead of just pointing to the original one
      • Resetting vehicle.commands.next to 0 doesn’t work. It keeps its previous value and takes some time to update to the correct one causing the mission to go wrong
      • Finally, placing cmds.next = 0 before vehicle.flush() seemed to work
    • Found a way to capture mavlink message when a waypoint is reached (adding a listener)
      • https://github.com/dronekit/dronekit-python/issues/231
      • https://github.com/dronekit/dronekit-python/blob/master/docs/guide/mavlink_messages.rst
    • Accidentally (in the similar reference above) found that instead of creating a new auto mission, it can automatically pause and resume
      • Need to set ms_restart parameter to 0 to resume mission after loiter, 1 if restart
      • refactored code
    • MERGED TOTER CODE WITH SOCKET COMMUNICATION
      • Used pickle module to send receive list 
      • Client: s.send()/receive()
      • Server: conn.send()/receive()
    • Pushed everything to git repo

    ToTer simulation with socket communication




    Session: June 17
      • Tried socket communication programming
      • Test sample client-server code from: https://www.youtube.com/watch?v=PYBZtV2-sLQ&t=622s
        • Laptop is both client and server
      • Test code in raspberry pi (as server)
      • Modified setup: Erle brain - client code, RPI - server code
        • Erle brain will send signal for rpi to start sensing
        • RPi will reply the sensed data to erle brain
        • Erle brain will send data to server via gsm modem


      Session: June 10-11
        • Played around with dronekit python. Tried modifying mission_basic.py in attempt to merge auto and loiter
        • Tried changing mode from auto to loiter: success but the mission won’t continue
        • Saved waypoints to temporary var and reinitiated mission after 3 secs of loiter
        • Success!



        Session: May 28
          • Setup sitl for windows (to try changing of mode via dronekit)
          • Followed tutorials:
            • http://ardupilot.org/dev/docs/sitl-native-on-windows.html
            • http://ardupilot.org/dev/docs/building-setup-windows-cygwin.html#building-setup-windows-cygwin
          • git clone https://github.com/ardupilot/ardupilot.git
          • cd ardupilot
          • git submodule update --init --recursive
          • Build ardurover:
            • https://github.com/ArduPilot/ardupilot/blob/master/BUILD.md
            • ./waf list_boards
            • ./waf configure --board sitl
            • ./waf list
            • ./waf rover
            • Output in /build/sitl/bin/ardurover.exe
          • Run sitl:
            • cd Tools/autotest/
            • python sim_vehicle.py -v APMrover2
          • MavProxy started
            • Connected to tcp:127.0.0.1:5760
          • Wrong home loc
            • python sim_vehicle.py -v APMrover2 -l 14.1622,121.242,584,353
            • MavProxy started
              • Connected to tcp:127.0.0.1:5760
          • Open Mission Planner: connected
          • Test dronekit code
            • On CMD: python simple_goto.py --connect udp:127.0.0.1:14551
              • Contains GUIDED mode to a single point
            • Arming fails
              • Set ARMING_REQUIRE to 0
          • SUCCESS!
          • Up next: retrieve GPS data during mission



          Session: May 7
          • Setup network configuration for Odette and jedidiah
          • Through Erle Brain’s Odette-wifi:
            • ssh thru erle
            • Check /etc/hostapd/hostapd.conf
            • In /etc/dnsmasq.conf increase dhcp-range to allow more than one connection:
              • Change dhcp-range
          • In jedidiah:
            • vncviewer <rpi-ip>:1
            • Edit etc/network/interfaces
            • Comment original iface block
            • Insert:
              • iface wlan0 inet static
                •    wpa-ssid Odette-Wifi
                •    wpa-psk <password>
            • Reboot
          • To test:
            • vncviewer <erle-ip>:1
          • Success! RPi connects to Odette-wifi on startup


          Session: March 20
            • Tune speed and throttle and steering: http://ardupilot.org/rover/docs/rover-tuning-throttle-and-speed.html
            • Preliminaries:
              • Change Switch 8 to Acro > Manual > Loiter
            • Tuning Steps:
              • Throttle/Speed Tuning
                • Manual > Run throttle > Switch 5 > Check CRUISE_THROTTLE and CRUISE_SPEED
                • Config > Full Param Tree > GCS_PID_MASK = 2 > Write Param
                • Flight Data > Tuning checkbox > Double click on graph > Select pidachieved and piddesired
                • Acro Mode > Drive up and down > See red vs blue line
                • Config > Basic Tuning > Increase P and I together
              • Turn Rate Tuning
                • Config > Full Param Tree > GCS_PID_MASK = 1> Write Param
                • Flight Data > Tuning checkbox > Double click on graph > Select pidachieved and piddesired
                • Check max turn rate: Manual > Drive around to tight circles > Check blue on graph
                • Config > Full Param Tree > ACRO_TURN_RATE = result
                • Acro > Drive up and down doing tight circles > See red vs blue line
              • Tuning is successful though turn rate is poor due to non optimal assembly
            • Loiter:
              • Loiter SUCCESS though not tested fully due to errors in steering

            Turn Rate Tuning - March 20, 2019

            Loiter Test - March 20, 2019






















            Session: March 8, 13, 19
              • Tried setting Loiter mode
                • Switched to Mission Planner GCS
              • RC, Accelerometer and Compass Calibration
                • Compass Calibration doesn’t work
                • Tried Live Calibration
              • Tested Loiter outside
                • SUCCESS! Moved the boat around and the propeller and rudder moved on their own in attempt to hold position


              Session: February 13, 14, 20


                Session: February 4
                  • Built ardurover3.5 from source code
                    • Followed this link: http://docs.erlerobotics.com/brains/erle-brain-3/sofware/apm#build-apm-using-waf-cross-compil-



                  Side Project with Interns
                    • Bianca Ruth Bautista and Andric Quinn Baticos
                    • Prototype web app and mobile app developed to receive and visualize data from boat
                    • 18 June-20 July 2018
                      



                    Session: May 29
                       
                       
                       

                      Temperature sensor on-board setup - May 29, 2018

                      Temperature sensor on-board test with motor and props - May 29, 2018
















                      Session: May 25
                        • Another field test was done in the attempt to improve rudder turns
                        • Location: Baker Hall swimming pool
                        • Boat was able to move smoothly but turns are not yet at their sharpest
                        • Up Next: Integrate water temperature sensor

                        Test Sail Part 2 - May 25, 2018

                        Test Sail Part 3 - May 25, 2018























                        Session: May 23
                        • First test for modified boat assembly
                          • The modifications were on the shaft-propeller assembly where the shaft, propeller, and joint from banggood was used
                          • It improved the rotation of the propeller together with the motor
                        • Test was done on the pool
                          • Issue encountered: Boat moves backwards!

                          • Solution: Reverse the polarity of the motor to reverse the rotation
                            • Exchange the red and blue cables
                          • Sail was successful!
                        • Up Next: artificial waves/ auto mode

                        First Successful Sail Part 1 ‎(RC Controlled)‎

                        First Successful Sail Part 2 ‎‎(RC controlled)‎‎
















                        Session: April 13
                          • We have assembled the boat (with the complete Erle Brain components inside)
                          • Tested gps signal in APM
                          • FIRST SAIL ATTEMPT
                            • Location: Baker Hall swimming pool
                            • Tested buoyancy first for a minute
                            • Tested controlling the boat with remote control
                              • Shaft loosens which made it difficult for the propeller to operate properly
                              • Propeller’s hollow area adds resistance
                              • The boat was able to move in a short distance
                              • TO DO: Modify shaft and propeller. Turn rudder to the other side

                          Buoyancy Test (Erle Brain sensors loaded)

                          First Sail - RC controlled



















                          Session: August 30
                            • We have done a buoyancy test on the boat (No electronics first)
                            • Location: Baker Hall swimming pool
                            • On first try, no load was placed on the boat
                            • On the second, a 5lbs load was placed on the boat

                            Buoyancy Test (no load)

                            Buoyancy Test (5lb load)























                            Session: August 19-21
                            • Built a small prototype boat
                            • Material: Polystyrene foam board
                            • UP NEXT: Integrate boat hull with Erle Brain and components
                             
                             
                             
                             

                            Servo Test





                            Session: August 18
                            • Tried building APMrover2 on Erle Brain: SUCCESS!
                            • Installed APM Planner on Windows
                            • Using controller, tried controlling the motor at different throttle speeds: SUCCESS!
                            • After turning off controller, the motor went failsafe
                            • Solution: Initial Setup>FailSafe>Change RTL
                            • UP NEXT: Motor-Propeller assembly. Prototype boat: Catamaran style, 1 propeller, with rudder
                             
                             
                             
                             




                            Session: August 8
                            • Tested temperature sensor DS18B20 on RPi at conference room: Success! 
                            • Set RPi as access point: DONE. Set up VNC: NOT DONE YET
                             
                             
                             





                            Session: August 3
                            • Tested RPi at conference room: Success!
                            • Next step: Buy sensors!
                            • Q: Can I use Erle-brain instead?
                              • Maybe: Read on ArduBoat - https://discuss.ardupilot.org/t/rover-3-0-0-release/8267?source_topic_id=8308
                            Comments