Announcement

Collapse
No announcement yet.

Weird Issues with Modern Robotics Core Motor Controller

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Hi All,

    I am wondering if anyone used the v2.62 in a tournament. We used it for the first time Yesterday in a championship event, and the result was not good -- the FTC app on the RC phone crashed repetitively. While I cannot say whether 2.62 has anything to do with it or not, that is one of our main suspicions. So, I would like to request the FTC support or those with more internal knowledge of the SDK to go through the logcat files and see if anything suspicious pops out. The log files are available here:
    https://drive.google.com/drive/folde...Ws?usp=sharing

    The logcat files contain some data from our tests in our practice field the night before, but I believe the problem started roughly after 10:30am on 2/5.
    I noticed a Navx-micro related exception in the logcat and notified them. They said navx-micro library is receiving 0 byte data when it is expecting 26 bytes. We do not know if that is the cause of app crash, or if it is a symptom of some other deeper issue.

    Comment


    • #32
      Errors in Logcat

      Have you looked into why your activity could not be stopped? This error from the Android Runtime does not look good and might have played a role in your repeated crashes since the resources might have been locked. Did you try restarting the phone after your repeated crashes?

      Code:
      02-05 10:25:01.653 15321 15321 D AndroidRuntime: Shutting down VM
      02-05 10:25:01.654 15321 15321 E AndroidRuntime: FATAL EXCEPTION: main
      02-05 10:25:01.654 15321 15321 E AndroidRuntime: Process: com.qualcomm.ftcrobotcontroller, PID: 15321
      02-05 10:25:01.654 15321 15321 E AndroidRuntime: java.lang.RuntimeException: Unable to stop activity {com.qualcomm.ftcrobotcontroller/org.firstinspires.ftc.robotcontroller.internal.FtcRobotControllerActivity}: java.lang.NullPointerException: Attempt to invoke virtual method 'void com.qualcomm.ftccommon.FtcRobotControllerService.shutdownRobot()' on a null object reference
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.ActivityThread.handleSleeping(ActivityThread.java:3661)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.ActivityThread.-wrap19(ActivityThread.java)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1532)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.os.Handler.dispatchMessage(Handler.java:102)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.os.Looper.loop(Looper.java:148)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.ActivityThread.main(ActivityThread.java:5459)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at java.lang.reflect.Method.invoke(Native Method)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:728)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:618)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime: Caused by: java.lang.NullPointerException: Attempt to invoke virtual method 'void com.qualcomm.ftccommon.FtcRobotControllerService.shutdownRobot()' on a null object reference
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at org.firstinspires.ftc.robotcontroller.internal.FtcRobotControllerActivity.onStop(FtcRobotControllerActivity.java:315)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.Instrumentation.callActivityOnStop(Instrumentation.java:1301)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.Activity.performStop(Activity.java:6403)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     at android.app.ActivityThread.handleSleeping(ActivityThread.java:3658)
      02-05 10:25:01.654 15321 15321 E AndroidRuntime:     ... 8 more
      02-05 10:25:01.661 15321 15321 I Process : Sending signal. PID: 15321 SIG: 9
      02-05 10:29:52.806 15601 15601 I art     : Late-enabling -Xcheck:jni

      Comment


      • #33
        Hi FTC4855, Thank you for taking a look at the log file. Yes we tried restarting and also tried using a completely different phone without any success.

        Comment


        • #34
          Hi rbFTC,

          Wanted to update with our experience. Before the WI Fox Valley event 2 weeks ago, we did a lot of testing and were getting great scores, but every now and again we'd get a random crash with an error message about a null pointer in some motor method. In 50 tries we saw 4-5 drop outs. 2 days before the event we decided 10% failure was too high and it was worth the risk to switch to 2.62 Alpha. We ran 25 more tests on it before the event and saw no issues. Then, at Fox Valley, we ran 9 matches (all wins!) and saw no issues.

          So we decided to stick with it for WI State last weekend. We did have two problems at WI state, but we have no way of knowing if the issues were the same. Fortunately, any failure is annoying and one may have cost us a match, but we can't know for sure. Late in our first match we hit another robot and we 'disconnected.' The robot ran full speed into the other robot and tried to push it up the ramp. It then blew a battery fuse. (Haven't seen that ever in 7 years) Then in another match a couple hours later, we hit a wall and 'disconnected,' but the drive team was able to recover the robot remotely and still play. At no time did autonomous fail for us due to a random crash, as it had been before 2.62 alpha.

          Those crashes we saw in the event were not rare occurrences in the shop. We think they are a different source than the motor controller null pointer crash. In fact, we are somewhat convinced it is the USB on the driver side phone crashing in such a way that it keeps telling the RC to do the same bad thing. Almost like the old days with Samantha when the robots would continue doing the last joystick action forever.

          Comment


          • #35
            Originally posted by FTC4106 View Post

            Those crashes we saw in the event were not rare occurrences in the shop. We think they are a different source than the motor controller null pointer crash. In fact, we are somewhat convinced it is the USB on the driver side phone crashing in such a way that it keeps telling the RC to do the same bad thing. Almost like the old days with Samantha when the robots would continue doing the last joystick action forever.
            Thank you for sharing your experience with 2.62. I am somewhat intrigued by your observation that the DS phone can cause the RC phone to crash. Is that even possible? If so, we might be having the same situation as we did not see the crash in autonomous mode. We have no experience with Samantha module as we are a 2nd year team.

            Comment

            Working...
            X