ModalAI Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    Outdoor navigation problem

    Qualcomm Flight RB5 5G Drone
    2
    22
    785
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • Eric KatzfeyE
      Eric Katzfey Dev Team @Amin Bassiri
      last edited by

      @Amin-Bassiri Looks like lots of GPS accuracy checks failing. Can you set EKF2_GPS_CHECK to 1 and see how that works?

      Amin BassiriA 1 Reply Last reply Reply Quote 0
      • Amin BassiriA
        Amin Bassiri @Eric Katzfey
        last edited by

        @Eric-Katzfey I changed, but it seems the problem is the GPS device, because when we change the GPS drone can go to position mode, but we still have a problem with MAG, we have to turn off the MAG, The drone can not go to position mode, and also we have too much drift.

        when we set EKF2_GPS_CHECK to 1:
        rb5:~$ journalctl -b --no-pager -u voxl-px4
        -- Logs begin at Thu 2023-03-02 12:58:05 UTC, end at Mon 2024-05-27 13:54:15 UTC. --
        May 27 13:54:15 m0052 voxl-px4[1896]: ta rate: 100000 B/s on udp port 14556 remote port 14557
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic telemetry_status
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [mavlink] partner IP: 127.0.0.1
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic obstacle_distance
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic vehicle_visual_odometry
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic timesync_status
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic actuator_controls_status_0
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_ev_hgt
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_ev_pos
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_ev_vel
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_ev_yaw
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [mavlink] mode: Normal, data rate: 100000 B/s on udp port 14558 remote port 14559
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Marking DeviceNode(telemetry_status) as advertised in process_remote_topic
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [px4] Startup script returned successfully
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [gps] Sending initial CFG_VALSET. Attempt: 2
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Marking DeviceNode(vehicle_command_ack) as advertised in process_remote_topic
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [gps] Sending initial CFG_VALSET. Attempt: 3
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic vehicle_rates_setpoint
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [gps] u-blox firmware version: SPG 3.01
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [gps] u-blox protocol version: 18.00
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [gps] u-blox module: NEO-M8N-0
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [mavlink] partner IP: 127.0.0.1
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Marking DeviceNode(sensor_gps) as advertised in process_remote_topic
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic home_position
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_gps_status
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_gnss_hgt
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_gnss_pos
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_aid_src_gnss_vel
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic mission_result
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic ping
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic vehicle_global_position
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_baro_bias
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: horizontal velocity unstable
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: horizontal velocity unstable
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic estimator_ev_pos_bias
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: height estimate not stable
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: Found 0 compass (required: 1)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Preflight Fail: height estimate not stable
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Marking DeviceNode(mavlink_log) as advertised in process_remote_topic
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Armed by RC
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [logger] Start file log (type: full)
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [uORB] orb_advertise_multi: failed to set queue size
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [logger] [logger] /data/px4/log/2024-05-27/13_52_37.ulg
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [logger] Opened full log file: /data/px4/log/2024-05-27/13_52_37.ulg
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Event dropped (80, 65526)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Advertising remote topic logger_status
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Dropped 65446 events (seq=65526)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Takeoff detected
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Event dropped (65527, 80)
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Dropped 71 events (seq=80)
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Event dropped (65527, 80)
        May 27 13:54:15 m0052 voxl-px4[1896]: WARN [mavlink] Dropped 71 events (seq=80)
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic yaw_estimator_status
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Advertising remote topic hover_thrust_estimate
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Landing detected
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Disarmed by RC
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [uORB] Marking DeviceNode(parameter_server_set_value_request) as advertised in process_remote_topic
        May 27 13:54:15 m0052 voxl-px4[1896]: INFO [muorb] SLPI: Marking DeviceNode(parameter_server_set_value_response) as advertised in process_remote_topic

        Eric KatzfeyE 1 Reply Last reply Reply Quote 0
        • Eric KatzfeyE
          Eric Katzfey Dev Team @Amin Bassiri
          last edited by

          @Amin-Bassiri For some reason the journalctl output doesn't have all of the output from voxl-px4 startup. It should show the attempts to find the magnetometer that seem to be failing. The command in the startup script is qshell ist8310 start -R 10 -X -b 1. If you run that command from the px4 console you can see the output. Or from Linux shell px4-qshell ist8310 start -R 10 -X -b 1.

          Amin BassiriA 1 Reply Last reply Reply Quote 0
          • Amin BassiriA
            Amin Bassiri @Eric Katzfey
            last edited by

            @Eric-Katzfey i got this error
            rb5:~$ px4-qshell ist8310 start -R 10 -X -b 1
            INFO [qshell] Send cmd: 'ist8310 start -R 10 -X -b 1'
            INFO [qshell] cmd returned with: -1
            INFO [qshell] qshell return value timestamp: 670686689, local time: 670688462
            ERROR [qshell] Command failed
            rb5:~$ px4-qshell ist8310 start

            also qshell ist8310 start -R 10 -X -b 1. same error

            Eric KatzfeyE 2 Replies Last reply Reply Quote 0
            • Eric KatzfeyE
              Eric Katzfey Dev Team @Amin Bassiri
              last edited by

              @Amin-Bassiri Shoot. With px4-qshell ist8310 start we don't see all of the output. That command needs to be run in the px4 shell to see the output (i.e. qshell ist8310 start -R 10 -X -b 1). When I have no magnetometer attached to my VOXL 2 and I run the command I can see that the probe fails. For example:

              INFO  [qshell] Send cmd: 'ist8310 start -R 10 -X -b 1'
INFO  [muorb] SLPI: qshell gotten: ist8310 start -R 10 -X -b 1
INFO  [muorb] SLPI:   arg0 = 'ist8310'

INFO  [muorb] SLPI:   arg1 = 'start'

INFO  [muorb] SLPI:   arg2 = '-R'

INFO  [muorb] SLPI:   arg3 = '10'

INFO  [muorb] SLPI:   arg4 = '-X'

INFO  [muorb] SLPI:   arg5 = '-b'

INFO  [muorb] SLPI:   arg6 = '1'

INFO  [muorb] SLPI: *** I2C Device ID 0x60e09 396809
ERROR [muorb] SLPI: i2c probe failed
INFO  [muorb] SLPI: PX4_qshell: no instance started (no device on bus?)
ERROR [muorb] SLPI: Failed to execute command: ist8310 start -R 10 -X -b 1
INFO  [qshell] cmd returned with: -1
INFO  [qshell] qshell return value timestamp: 2382015392, local time: 2382017339
ERROR [qshell] Command failed
Command 'qshell' failed, returned -1.
              Eric KatzfeyE 2 Replies Last reply Reply Quote 0
              • Eric KatzfeyE
                Eric Katzfey Dev Team @Amin Bassiri
                last edited by

                @Amin-Bassiri Regardless, based on your earlier log file it seems like the magnetometer is being detected and is providing data. But you are still seeing the No valid data from Compass error. The log you provided does show the error_count incrementing which indicates failed i2c transfers from the magnetometer. So I am also suspecting hardware failure of some sort. Let me discuss internally to figure out next steps.

                1 Reply Last reply Reply Quote 0
                • Eric KatzfeyE
                  Eric Katzfey Dev Team @Eric Katzfey
                  last edited by

                  @Eric-Katzfey We can test a couple of GPS / Mag units with RB5 and send known working units to you and see if you can also replicate that success. If so, we can send you 5 more.

                  1 Reply Last reply Reply Quote 0
                  • Eric KatzfeyE
                    Eric Katzfey Dev Team @Eric Katzfey
                    last edited by

                    @Eric-Katzfey Can you send a picture with both the GPS unit and cable? Just want to double check what unit(s) you have. I am currently testing with the mRo GPS / Mag unit on my RB5 running SDK 1.1.2 and it is working fine. I have 2 of them that seem to be fine so we can send those to you to try out on your drone(s).

                    Amin BassiriA 1 Reply Last reply Reply Quote 0
                    • Amin BassiriA
                      Amin Bassiri @Eric Katzfey
                      last edited by

                      @Eric-Katzfey GPS.jpeg

                      Eric KatzfeyE 1 Reply Last reply Reply Quote 0
                      • LaborzaubererTHIL LaborzaubererTHI referenced this topic on
                      • Eric KatzfeyE
                        Eric Katzfey Dev Team @Amin Bassiri
                        last edited by

                        @Amin-Bassiri Really sorry that you guys are having so much trouble with the GPS units! We are now shipping with M10 GPS units so we can send you those and hopefully that will improve the situation.

                        Amin BassiriA 1 Reply Last reply Reply Quote 0
                        • Amin BassiriA
                          Amin Bassiri @Eric Katzfey
                          last edited by

                          @Eric-Katzfey ok, Thanks, we tested GPS M10 and the problem solved.

                          1 Reply Last reply Reply Quote 0
                          • V vjuliani referenced this topic on
                          • V vjuliani referenced this topic on
                          • First post
                            Last post
                          Powered by NodeBB | Contributors