search

Yadunund / rmf_test_scenarios

Launch files to fleet adapters
0 stars 0 forks source link

office_doors_conflict #5

Open Yadunund opened 4 years ago

Yadunund commented 4 years ago

Robots stuck negotiating passage through a door. Fleet adapter crashed with error

[ERROR] [magni_fleet_adapter-10]: process has died [pid 18413, exit code -9, cmd '/home/yadu/ws_rmf_demos/install/rmf_fleet_adapter/lib/rmf_fleet_adapter/full_control --ros-args -r __node:=magni_fleet_adapter --params-file /tmp/launch_params_n6ffauoj --params-file /tmp/launch_params__cycdc7w --params-file /tmp/launch_params_aa7rpq92 --params-file /tmp/launch_params_kd8tpm4_ --params-file /tmp/launch_params_fc1tf_xi --params-file /tmp/launch_params_3rc0osgp --params-file /tmp/launch_params_inwa0mp9 --params-file /tmp/launch_params_5ojx0h4a --params-file /tmp/launch_params_tttqx1mv --params-file /tmp/launch_params_504qd_3u --params-file /tmp/launch_params_li2t5qjf --params-file /tmp/launch_params_tmkdy9jv --params-file /tmp/launch_params_su8ovh0c --params-file /tmp/launch_params_a3w5m3zh'].

Terminal printout attached. office_doors_conflict.txt

image

Workspace: rmf_core: develop rmf_demos: develop rmf_schedule_visualizer: master traffic_editor: master

Workspace built with command CXX=g++-8 colcon build --cmake-args -DCMAKE_BUILD_TYPE=Release

Launch steps:

ros2 launch demos office.launch.xml 

ros2 launch ~/rmf_test_scenarios/scenarios/launch/office_doors_conflict.launch.xml
mxgrey commented 4 years ago

I suspect this will be fixed with the latest changes.

Yadunund commented 4 years ago

Following the new changes, the scenario was running well for quite a while. I saw negotiations for the case above to resolve successfully. But after a while, the fleet adapter crashed again.

image

yadu@yadu:~$ ps aux | grep rmf_traffc_schedule
yadu     31142  0.0  0.0  21532  1116 pts/5    S+   12:50   0:00 grep --color=auto rmf_traffc_schedule

Terminal output:

terminal_output.txt

mxgrey commented 4 years ago

Can anyone provide some indications of how long the simulation are running before the crash happens? And what Real-Time Factor are the tests running at? I've had the simulation going for 30 minutes at RTF 1, and I haven't seen a crash.

Yadunund commented 4 years ago

In this case it ran for ~20mins with RTF 1.0.