I'm Arul's team mate. I can probably provide a wee bit more information.
So...
1. On power up everything is fine and we can control the robot with our PS4 controller.
2. A few minutes after power up one of the front motors engages for no reason and just keeps spinning.
3. The only connection we have to the Pi at this point is via the controller & bluetooth.
4. Arul has configured the Pi to auto-start both ds4drv and your joypad python script on power up.
5. The Pi is running on Robot Operating System (ROS).
6. I have another robot, wired identically ( or so we believe ;) ), that doesn't show this behaviour with the same ROS image.
That's about the extent of my knowledge - if you require any more detail, Arul can provide that. He's driving down from Edinburgh this afternoon and plans to attend the Cambridge Makespace at some point tonight (as do I).
This is not a problem we have seen before so I am not entirely sure what is happening.
The first thing to do is carefully double-check the wiring to the motors and the battery. We have seen some odd behaviours when a motor is wired to both M1 and M2 by mistake for example.
If you have access to a terminal could you try running the following after the problem starts:
I am seeing this same issue. Were you ever able to resolve the issue. I am actually using an xbox controller. But the monsterborg powers up just fine. I put the runMonsterJoy.sh in the autostart and the monsterborg starts up fine. I start to move the monster borg and then all of the sudden it takes off in the reverse direction and won't stop without a re-boot.
Any Ideas?
The image attached shows a screen shot from Putty, not sure what the No Protocol specified means.
Based on your description, it sounds like the controller might be losing connection with the Raspberry Pi. We have seen robots "nudge" forward or backward when a controller disconnects in a way that the Raspberry Pi does not realise before.
The first thing I would try is charging the controller fully (possibly overnight) and see if it makes any difference.
You could also try running the MonsterBorg sat on its motors so that it cannot move and see what happens when the controller goes out of range.
Both the No protocol specified and xcb_connection_has_error returned true messages are from Pygame. They are probably harmless.
You could try the monsterJoy.py script in our Gamepad Library instead, it does not use Pygame.
Having said that I would be surprised if they are related to the problem you are seeing. Pygame related issues tend to either prevent the script working at all or they have no effect.
Laurence Molloy
Fri, 03/29/2019 - 12:36
Permalink
Hi
Hi
I'm Arul's team mate. I can probably provide a wee bit more information.
So...
1. On power up everything is fine and we can control the robot with our PS4 controller.
2. A few minutes after power up one of the front motors engages for no reason and just keeps spinning.
3. The only connection we have to the Pi at this point is via the controller & bluetooth.
4. Arul has configured the Pi to auto-start both ds4drv and your joypad python script on power up.
5. The Pi is running on Robot Operating System (ROS).
6. I have another robot, wired identically ( or so we believe ;) ), that doesn't show this behaviour with the same ROS image.
That's about the extent of my knowledge - if you require any more detail, Arul can provide that. He's driving down from Edinburgh this afternoon and plans to attend the Cambridge Makespace at some point tonight (as do I).
However,
piborg
Fri, 03/29/2019 - 13:19
Permalink
Diagnostic information
This is not a problem we have seen before so I am not entirely sure what is happening.
The first thing to do is carefully double-check the wiring to the motors and the battery. We have seen some odd behaviours when a motor is wired to both M1 and M2 by mistake for example.
If you have access to a terminal could you try running the following after the problem starts:
The output might help explain what is going on.
chuber
Sun, 03/29/2020 - 19:06
Permalink
Same issue
I am seeing this same issue. Were you ever able to resolve the issue. I am actually using an xbox controller. But the monsterborg powers up just fine. I put the runMonsterJoy.sh in the autostart and the monsterborg starts up fine. I start to move the monster borg and then all of the sudden it takes off in the reverse direction and won't stop without a re-boot.
Any Ideas?
The image attached shows a screen shot from Putty, not sure what the No Protocol specified means.
piborg
Sun, 03/29/2020 - 18:53
Permalink
Sounds like the controller might be losing connection
Based on your description, it sounds like the controller might be losing connection with the Raspberry Pi. We have seen robots "nudge" forward or backward when a controller disconnects in a way that the Raspberry Pi does not realise before.
The first thing I would try is charging the controller fully (possibly overnight) and see if it makes any difference.
You could also try running the MonsterBorg sat on its motors so that it cannot move and see what happens when the controller goes out of range.
piborg
Sun, 03/29/2020 - 22:49
Permalink
Pygame error messages
Both the
No protocol specified
andxcb_connection_has_error returned true
messages are from Pygame. They are probably harmless.You could try the
monsterJoy.py
script in our Gamepad Library instead, it does not use Pygame.Having said that I would be surprised if they are related to the problem you are seeing. Pygame related issues tend to either prevent the script working at all or they have no effect.