empty planning trajecoty problem, when debuging control..

[Addi2020]

I checked many different recorded file, all of them gives the same error.
can somebody explain this issue?

System information

Linux Ubuntu 16.04
Apollo version (3.5.0)

[daohu527]

how did you record the file ?
When you start to record the bag you should check in the dreamview, Be Sure there is an planning path, then start to record the bag.

[Addi2020]

we use the actual car setup for recording the data. and YES, before recording all the necessary topic that are required for control topic are turned on. as you can see in the cyber_monitor (picture).
now when i am debugging the control module offline using the recorded data file it give the Control topic error (msg: estop for empty planning trajectory ) and planning topic error (Planing error msg: found collision with obsticle)

[daohu527]

There is a planning message in cyber monitor not mean the planning message is right.
You should first see the planning path in the dreamview which is the green line. The planning message is already right. Then you can start to record the message .

[Addi2020]

I understand , when i did the recoding it shows the green line. i just get the new recorded file and now play it offline. still it gives the same issue of planning.

{control log file... and planning lod file given below}
Log file created at: 2020/04/09 18:07:56
Running on machine: in_dev_docker
Log line format: [IWEF]mmdd hh:mm:ss.uuuuuu threadid file:line] msg
I0409 18:07:56.426316 18519 class_loader_utility.h:78] registerclass:ControlComponent,apollo::cyber::ComponentBase,/apollo/bazel-bin/modules/control/libcontrol_component.so
I0409 18:07:56.447973 18519 control_component.cc:42] Control init, starting ...
I0409 18:07:56.486577 18519 control_component.cc:48] Conf file: /apollo/modules/control/conf/control_conf.pb.txt is loaded.
I0409 18:07:56.486606 18519 control_component.cc:50] Conf file: is loaded.
I0409 18:07:56.486624 18519 controller_agent.cc:36] Only support MPC controller or Lat + Lon controllers as of now
I0409 18:07:56.487129 18519 lat_controller.cc:86] Using LQR-based Lateral Controller
I0409 18:07:56.489782 18519 lat_controller.cc:260] Lateral control gain scheduler loaded
I0409 18:07:56.493561 18519 lat_controller.cc:159] LQR-based Lateral Controller begin.
I0409 18:07:56.493647 18519 lat_controller.cc:160] [LatController parameters] mass_: 2080, iz_: 4208.3, lf_: 1.4224, lr_: 1.4224
I0409 18:07:56.494626 18519 controller_agent.cc:92] Controller init done!
I0409 18:07:56.494824 18519 lon_controller.cc:140] Control calibration table loaded
I0409 18:07:56.494844 18519 lon_controller.cc:141] Control calibration table size is 1297
I0409 18:07:56.496496 18519 controller_agent.cc:92] Controller <LON_CONTROLLER> init done!
I0409 18:07:56.558313 18519 control_component.cc:98] Control resetting vehicle state, sleeping for 1000 ms ...
I0409 18:07:57.558420 18519 control_component.cc:104] Control default driving action is STOP
I0409 18:07:57.574544 18533 control_component.cc:216] Reset Controllers in Manual Mode
I0409 18:07:58.896467 18528 control_component.cc:216] Reset Controllers in Manual Mode

I0409 18:10:43.247560 18524 control_component.cc:216] Reset Controllers in Manual Mode
W0409 18:10:44.136178 18523 control_component.cc:365] planning has no trajectory point.
E0409 18:10:44.136226 18523 control_component.cc:168] Control input data failed: planning has no trajectory point. planning_seq_num:2274
W0409 18:10:44.136309 18523 control_component.cc:207] planning has no trajectory point.
W0409 18:10:44.136463 18523 control_component.cc:256] Estop triggered! No control core method executed!
E0409 18:10:44.136474 18523 control_component.cc:315] Failed to produce control command:planning has no trajectory point. planning_seq_num:2274
E0409 18:10:44.137316 18526 control_component.cc:315] Failed to produce control command:planning has no trajectory point. planning_seq_num:2274
E0409 18:10:44.138537 18523 control_component.cc:315] Failed to produce control command:planning has no trajectory point. planning_seq_num:2274

W0409 18:10:44.260157 18537 control_component.cc:256] Estop triggered! No control core method executed!
W0409 18:10:44.394456 18532 control_component.cc:256] Estop triggered! No control core method executed!

{planning log file...}

Log file created at: 2020/04/09 18:07:52
Running on machine: in_dev_docker
Log line format: [IWEF]mmdd hh:mm:ss.uuuuuu threadid file:line] msg
I0409 18:07:52.401577 18434 class_loader_utility.h:78] registerclass:PlanningComponent,apollo::cyber::ComponentBase,/apollo/bazel-bin/modules/planning/libplanning_component.so
E0409 18:07:52.421152 18434 planning_component.cc:41] NaviPlanning create
I0409 18:07:52.429826 18434 navi_planner.cc:80] In NaviPlanner::Init()
I0409 18:07:52.430138 18434 navi_planner.cc:87] Created task:NaviPathDecider
I0409 18:07:52.430174 18434 navi_planner.cc:87] Created task:NaviSpeedDecider
I0409 18:07:52.430205 18434 navi_path_decider.cc:61] Maximum speeds and move to dest lane config:
I0409 18:07:52.430222 18434 navi_path_decider.cc:65] [max_speed : 34 ,max move dest lane shift y : 0.45]
E0409 18:07:52.514729 18434 planning_component.cc:100] planning conponet init ok
W0409 18:08:00.191423 18438 navi_planner.cc:284] last frame is empty
I0409 18:08:00.193207 18438 navi_speed_decider.cc:489] add speed limit for centric acceleration with kappa: 0 v_max: 1.79769e+308 v_preffered: 1.79769e+308 s: 0
W0409 18:08:00.196379 18438 channel_buffer.h:70] channel[/apollo/prediction] read buffer overflow, drop_message[12] pre_index[2] current_index[14]
I0409 18:08:00.204833 18438 navi_speed_decider.cc:489] add speed limit for centric acceleration

[Addi2020]

but when I play the recorded file, this green line (planning) turn OFF and ON respectively. what does it mean?

Is this the problem of hardware?

[daohu527]

When playing the bag, do not play the message of perception, only play the message needed by the control module. And also you should just open the control module.

1. Chassis
2. ADCTrajectory
3. LocalizationEstimate

The problem may be caused by repeated messages.

1. The running module generated a message
2. The bag playing a message too

Then there may be two duplicate messages, if the planning module receive message from perception message in the bag, it will produce a message too, so pls check the message and the running module when running simulation.

[Addi2020]

Thank @daohu527..
this issue is solved. I check other recording file data,

1. the problem was in recoded file.
2. i use -k to kill the control topic then open only control topic in dreamview （all other topics in dreamview is closed）. (cyber_recorder play -f bag0409/20200409165051.record.00015 -k /apollo/control )
   Now I can debug the control module.

[Avps1]

Thank @daohu527..
this issue is solved. I check other recording file data,

1. the problem was in recoded file.
2. i use -k to kill the control topic then open only control topic in dreamview （all other topics in dreamview is closed）. (cyber_recorder play -f bag0409/20200409165051.record.00015 -k /apollo/control )
   Now I can debug the control module.

Hi @Addi2020,
Can you tell me how you are debugging the control module from the recorded file ?
Are you using VSCode?
I am trying to debug the perception module using VScode, but I cant set breakpoints
#10968

Thanks in advance.