3
$\begingroup$

We are using a Koro robot for our PC based automation solution. But sometimes the robot is getting the command but refuses to respond. Then I get a serial communication timeout error. The error is happening for a random type of commands and it is also not happening all the time making the troubleshooting difficult.

I doubt the driver problem. How do you approach this problem.?

Thanks

$\endgroup$
4
  • $\begingroup$ Can you clarify the type of serial comms used? RS232, USB etc $\endgroup$
    – Andrew
    Nov 9 '12 at 6:35
  • $\begingroup$ It is RS232 connection $\endgroup$
    – Jaggi
    Nov 12 '12 at 0:29
  • $\begingroup$ When you say "the robot is getting the command" does that mean that you simply sent the command, or that you have some confirmation that the robot received it? Is the timeout error caused by failing to receive an expected response to your command? $\endgroup$
    – Ian
    Nov 12 '12 at 14:56
  • $\begingroup$ I dont have a confirmation that the command is received. It happens only when I run my main program. But when I try to send the commands through a separate software provided by the console $\endgroup$
    – Jaggi
    Nov 16 '12 at 7:07
4
$\begingroup$

Welcome to Robotics... Hopefully, this will give you some pointers, although your situation is not easy to debug remotely :)

Things to consider (and I appreciate its a non-repeatable problem):

  1. Does it only happen when battery levels are low, or also with full charge (voltage low)
  2. Does it only happen when the robot is active, or also from stationary (processor overload)
  3. What else is the control PC doing? (PC overload)
  4. Have you checked that the comms link is correct (voltage levels, numbers of bits, protocol etc) - Check with an oscilloscope.
  5. Noise interference (EMC) if the motors are active
  6. Can you log what is received at the robot?
  7. Is the receive buffer on the robot big enough for the biggest command sequence? (buffer overflow)
  8. Have you checked the cable, both for continuity and that it isn't loose.

As things stand, it is difficult to be able to determine...

$\endgroup$
3
  • $\begingroup$ Thanks for the reply. The robot was working well perfectly before last week. I am debugging now and get back to you with more information. $\endgroup$
    – Jaggi
    Nov 12 '12 at 0:35
  • 1
    $\begingroup$ I found out the problem was with the serial communication card..I connected the robot through the direct PC com port 2. It was working fine.. $\endgroup$
    – Jaggi
    Nov 20 '12 at 0:58
  • $\begingroup$ Doh! Thanks @Jaggi - such an obvious first check, I'd overlooked it! $\endgroup$
    – Andrew
    Nov 20 '12 at 6:22
3
$\begingroup$

I found out the problem was with the serial communication card..I connected the robot through the direct PC com port 2. It was working fine

$\endgroup$
2
  • 1
    $\begingroup$ Grrr, I almost suggested that you try different com ports, it's amazing that in 2012 this is still a problem! $\endgroup$
    – Mark Booth
    Nov 20 '12 at 2:03
  • $\begingroup$ Thanks for the update... and further to (and agreeing) Mark's comment, this is a reminder to check the basics first, no matter how obvious it may seem. And this is likely to become a bigger issue as built-in RS232 becomes even more rarer :( $\endgroup$
    – Andrew
    Nov 20 '12 at 6:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.