apollo icon indicating copy to clipboard operation
apollo copied to clipboard

Camera module automatically shut down after starting, on the latest master branch

Open GabrielZZZ opened this issue 3 years ago • 2 comments

System information

  • OS Platform and Distribution (e.g., Linux Ubuntu 18.04): Ubuntu 20.04
  • Apollo installed from (source or binary): source
  • Apollo version (3.5, 5.0, 5.5, 6.0): 7.0 master branch
  • Output of apollo.sh config if on master branch: [INFO] Apollo Environment Settings: [INFO] APOLLO_ROOT_DIR: /apollo [INFO] APOLLO_CACHE_DIR: /apollo/.cache [INFO] APOLLO_IN_DOCKER: true [INFO] APOLLO_VERSION: master-2022-02-16-8fc0381 [INFO] DOCKER_IMG: dev-x86_64-18.04-20210517_1712 [INFO] APOLLO_ENV: STAGE=dev USE_ESD_CAN=false [INFO] USE_GPU: USE_GPU_HOST=1 USE_GPU_TARGET=1

Steps to reproduce the issue:

  • Please use bullet points and include as much details as possible:
  • Successfully start, build Apollo without errors
  • Strictly following the instructions from https://github.com/ApolloAuto/apollo/blob/master/docs/howto/how_to_run_perception_module_on_your_local_computer.md
  • When running cyber_launch start /apollo/modules/perception/production/launch/perception_camera.launch, the module first start, then automatically shut down. Here is the terminal output: image Here is the log result from mainboard: image

I want to test the perception-camera module, especially the obstacle-detection capability. I have a recording containing only images generated from Apollo DaoxiangLake Dataset.

The perception, traffic light module appears to be started successfully: image

Supporting materials (screenshots, command lines, code/script snippets):

GabrielZZZ avatar Sep 17 '22 03:09 GabrielZZZ

According to the error message, you started 2 identical nodes, which is the cause of the error.

daohu527 avatar Sep 20 '22 15:09 daohu527

I also encounered this problem. I solved it by launching the motion_service dag separately and remove the this module in this camera perception *.launch file.

I am not sure if a duplicate node is the cause for this. When I encountered this problem, I checked that if the motion_service has been started or not (use command ps aux | fgrep mainboard ). And If a duplicate node is started, there will be very clear messgae telling you that the node is duplicate and it will not be started. What is more, starting a duplicate node will not fail a cyber_launch, it will just tell you that a duplicate node is found and it will not be started again.

ywangwxd avatar Sep 22 '22 06:09 ywangwxd

@ywangwxd Thank you! That solved the problem!

GabrielZZZ avatar Sep 23 '22 07:09 GabrielZZZ