ROS Resources: Documentation | Support | Discussion Forum | Service Status | Q&A answers.ros.org

Autoware-v1.11 about CNN segment error

when I click lidar_cnn_baidu_detect, i got this:
ERROR: cannot launch node of type [lidar_apollo_cnn_seg_detect/lidar_apollo_cnn_seg_detect]: can’t locate node [lidar_apollo_cnn_seg_detect] in package [lidar_apollo_cnn_seg_detect]

here is version info:
ubuntu16.04,kinetic,autoware-v1.11

my laptop:
8th-i7, 8GB RAM, 250GB SSD storage ,1060 6GB

i have installed caffe and do $ make distribute correctly.
1 ./run
2 play—pause—Setup,TF,Vehicle model
3 computing tab,-----lidar detetor—lidar_cnn_baidu_detect—app—select home/gz/apollo-cnn/cnnseg/velodyne64/deploy.prototxt and /home/gz/apollo-cnn/cnnseg/velodyne64/deploy.caffemodel—OK
4 click it √
5 then i got error:
1202306242
here is testual info:

(runtime_manager_dialog.py:7550): Gtk-WARNING **: Unable to show ‘app’: Operation not supported
[‘roslaunch’, ‘lidar_apollo_cnn_seg_detect’, ‘lidar_apollo_cnn_seg_detect.launch’, ‘points_src:=/points_raw’, ‘score_threshold:=0.7’, ‘network_definition_file:=/home/gz/apollo-cnn/cnnseg/velodyne64/deploy.prototxt’, ‘pretrained_model_file:=/home/gz/apollo-cnn/cnnseg/velodyne64/deploy.caffemodel’, ‘use_gpu:=True’, ‘gpu_device_id:=0’]
pid=9281
sched policy=OTHER prio=0
ERROR: cannot launch node of type [lidar_apollo_cnn_seg_detect/lidar_apollo_cnn_seg_detect]: can’t locate node [lidar_apollo_cnn_seg_detect] in package [lidar_apollo_cnn_seg_detect]
[ INFO] [1556243148.813975844]: [visualize_detected_objects] objects_src_topic: /detection/lidar_detector/objects
[ INFO] [1556243148.814427305]: [visualize_detected_objects] object_speed_threshold: 0.10
[ INFO] [1556243148.814801297]: [visualize_detected_objects] arrow_speed_threshold: 0.25
[ INFO] [1556243148.815190093]: [visualize_detected_objects] marker_display_duration: 0.20
[ INFO] [1556243148.815612949]: [visualize_detected_objects] label_color: {R:255.0, G:255.0, B:255.0, A:1.0}
[ INFO] [1556243148.815996558]: [visualize_detected_objects] arrow_color: {R:0.0, G:255.0, B:0.0, A:0.8}
[ INFO] [1556243148.816415116]: [visualize_detected_objects] hull_color: {R:51.0, G:204.0, B:51.0, A:0.8}
[ INFO] [1556243148.816833559]: [visualize_detected_objects] box_color: {R:51.0, G:128.0, B:204.0, A:0.8}
[ INFO] [1556243148.817160784]: [visualize_detected_objects] model_color: {R:190.0, G:190.0, B:190.0, A:0.5}
[ INFO] [1556243148.817546084]: [visualize_detected_objects] centroid_color: {R:77.0, G:121.0, B:255.0, A:0.8}
[ INFO] [1556243148.819916727]: [visualize_detected_objects] object_src_topic: /detection/lidar_detector/objects
[ INFO] [1556243148.820394643]: [visualize_detected_objects] markers_out_topic: /detection/lidar_detector/objects_markers
[ WARN] [1556243447.247893983, 1427157650.332930903]: Shutdown request received.
[ WARN] [1556243447.247922631, 1427157650.332930903]: Reason given for shutdown: [new node registered with same name]

And if i run this cnn segment node from terminal:

gz@cc:~$ rosrun lidar_apollo_cnn_seg_detect lidar_apollo_cnn_seg_detect _network_definition_file:=/home/gz/apollo-cnn/cnnseg/velodyne64/deploy.prototxt _pretrained_model_file:=/home/gz/apollo-cnn/cnnseg/velodyne64/deploy.caffemodel _points_src:=/points_raw
[rospack] Error: package ‘lidar_apollo_cnn_seg_detect’ not found

here i have several question :

  1. why i got the error above:

ERROR: cannot launch node of type [lidar_apollo_cnn_seg_detect/lidar_apollo_cnn_seg_detect]: can’t locate node [lidar_apollo_cnn_seg_detect] in package [lidar_apollo_cnn_seg_detect]

when run apollo_cnn_seotog node from runtime manager,and how fix it?

  1. i use baidu’s cnnseg velodyne-64(deploy.prototxt & deploy.caffemodel),and rosbag is autoware demo ‘sample_moriyama_150324.bag’ which include velodyne-32 pointcloud, is this reason lead to error?

  2. when i run autoware node or package from terminal, always got error :

[rospack] Error: package ‘lidar_apollo_cnn_seg_detect’ not found

not only this lidar_apollo_cnn_seg ,the same as other autoware computing node.how what shall i do if i want to rosrun this autoware node like common ros node?

anyone can help me? many thanks :slight_smile:

@defendergz

This means the node wasn’t built. Very likely because you hhave an incorrect setup of either Caffe and/or CUDA.

You can have a look at your build logs. The reason of why this node wasn’t built is located in the Autoware/ros/log/latest/lidar_apollo_cnn_seg_detect directory.

@amc-nu hello! thank u for your answer!
i check

Autoware/ros/log/latest/lidar_apollo_cnn_seg_detect

and find that i have built lidar_apollo_cnn_seg_detect node:

e[35me[1mScanning dependencies of target lidar_apollo_cnn_seg_detecte[0m
[ 20%] e[32mBuilding CXX object CMakeFiles/lidar_apollo_cnn_seg_detect.dir/nodes/cluster2d.cpp.oe[0m
[ 40%] e[32mBuilding CXX object CMakeFiles/lidar_apollo_cnn_seg_detect.dir/nodes/lidar_apollo_cnn_seg_detect_node.cpp.oe[0m
[ 60%] e[32mBuilding CXX object CMakeFiles/lidar_apollo_cnn_seg_detect.dir/nodes/cnn_segmentation.cpp.oe[0m
[ 80%] e[32mBuilding CXX object CMakeFiles/lidar_apollo_cnn_seg_detect.dir/nodes/feature_generator.cpp.oe[0m
[100%] e[32me[1mLinking CXX executable devel/lib/lidar_apollo_cnn_seg_detect/lidar_apollo_cnn_seg_detecte[0m
[100%] Built target lidar_apollo_cnn_seg_detect
e[35me[1mScanning dependencies of target testse[0m
Built target tests
[ 0%] Built target jsk_footstep_msgs_generate_messages_lisp
[ 0%] Built target jsk_footstep_msgs_generate_messages_py
[ 0%] Built target jsk_footstep_msgs_generate_messages_nodejs
[ 0%] Built target actionlib_msgs_generate_messages_lisp
[ 0%] Built target autoware_msgs_generate_messages_cpp
[ 0%] Built target actionlib_generate_messages_nodejs
[ 0%] Built target sensor_msgs_generate_messages_eus
[ 0%] Built target actionlib_msgs_generate_messages_cpp
[ 0%] Built target actionlib_generate_messages_lisp
[ 0%] Built target actionlib_msgs_generate_messages_eus
[ 0%] Built target actionlib_msgs_generate_messages_nodejs
[ 0%] Built target tf2_msgs_generate_messages_py
[ 0%] Built target topic_tools_generate_messages_py
[ 0%] Built target std_msgs_generate_messages_nodejs
[ 0%] Built target std_msgs_generate_messages_cpp
[ 0%] Built target sensor_msgs_generate_messages_nodejs
[ 0%] Built target geometry_msgs_generate_messages_cpp
[ 0%] Built target tf_generate_messages_py
[ 0%] Built target jsk_recognition_msgs_generate_messages_eus
[ 0%] Built target roscpp_generate_messages_py
[ 0%] Built target tf_generate_messages_lisp
[ 0%] Built target actionlib_generate_messages_cpp
[ 0%] Built target rosgraph_msgs_generate_messages_py
[ 0%] Built target jsk_recognition_msgs_generate_messages_cpp
[ 0%] Built target actionlib_msgs_generate_messages_py
[ 0%] Built target geometry_msgs_generate_messages_nodejs
[ 0%] Built target geometry_msgs_generate_messages_eus
[ 0%] Built target roscpp_generate_messages_cpp
[ 0%] Built target nodelet_generate_messages_py
[ 0%] Built target roscpp_generate_messages_eus
[ 0%] Built target bond_generate_messages_eus
[ 0%] Built target geometry_msgs_generate_messages_py
[ 0%] Built target geometry_msgs_generate_messages_lisp
[ 0%] Built target std_srvs_generate_messages_lisp
[ 0%] Built target std_msgs_generate_messages_lisp
[ 0%] Built target tf_generate_messages_cpp
[ 0%] Built target std_srvs_generate_messages_eus
[ 0%] Built target autoware_msgs_generate_messages_py
[ 0%] Built target tf2_msgs_generate_messages_cpp
[ 0%] Built target rosgraph_msgs_generate_messages_cpp
[ 0%] Built target pcl_msgs_generate_messages_eus
[ 0%] Built target jsk_footstep_msgs_generate_messages_cpp
[ 0%] Built target autoware_msgs_generate_messages_lisp
[ 0%] Built target rosgraph_msgs_generate_messages_nodejs
[ 0%] Built target tf_generate_messages_eus
[ 0%] Built target sensor_msgs_generate_messages_py
[ 0%] Built target roscpp_generate_messages_nodejs
[ 0%] Built target pcl_msgs_generate_messages_nodejs
[ 0%] Built target std_srvs_generate_messages_nodejs
[ 0%] Built target rosgraph_msgs_generate_messages_lisp
[ 0%] Built target rosgraph_msgs_generate_messages_eus
[ 0%] Built target std_msgs_generate_messages_py
[ 0%] Built target roscpp_generate_messages_lisp
[ 0%] Built target tf_generate_messages_nodejs
[ 0%] Built target std_srvs_generate_messages_py
[ 0%] Built target std_msgs_generate_messages_eus
[ 0%] Built target actionlib_generate_messages_eus
[ 0%] Built target pcl_ros_gencfg
[ 0%] Built target tf2_msgs_generate_messages_eus
[ 0%] Built target nodelet_generate_messages_cpp
[ 0%] Built target tf2_msgs_generate_messages_lisp
[ 0%] Built target tf2_msgs_generate_messages_nodejs
[ 0%] Built target dynamic_reconfigure_generate_messages_cpp
[ 0%] Built target dynamic_reconfigure_generate_messages_lisp
[ 0%] Built target dynamic_reconfigure_generate_messages_nodejs
[ 0%] Built target actionlib_generate_messages_py
[ 0%] Built target dynamic_reconfigure_generate_messages_py
[ 0%] Built target dynamic_reconfigure_gencfg
[ 0%] Built target nodelet_generate_messages_lisp
[ 0%] Built target nodelet_generate_messages_nodejs
[ 0%] Built target bond_generate_messages_cpp
[ 0%] Built target bond_generate_messages_lisp
[ 0%] Built target nodelet_generate_messages_eus
[ 0%] Built target topic_tools_generate_messages_cpp
[ 0%] Built target topic_tools_generate_messages_eus
[ 0%] Built target nodelet_topic_tools_gencfg
[ 0%] Built target pcl_msgs_generate_messages_lisp
[ 0%] Built target bond_generate_messages_nodejs
[ 0%] Built target dynamic_reconfigure_generate_messages_eus
[ 0%] Built target pcl_msgs_generate_messages_py
[ 0%] Built target topic_tools_generate_messages_lisp
[ 0%] Built target sensor_msgs_generate_messages_lisp
[ 0%] Built target topic_tools_generate_messages_nodejs
[ 0%] Built target bond_generate_messages_py
[ 0%] Built target autoware_msgs_generate_messages_eus
[ 0%] Built target pcl_msgs_generate_messages_cpp
[ 0%] Built target std_srvs_generate_messages_cpp
[ 0%] Built target autoware_msgs_generate_messages_nodejs
[ 0%] Built target jsk_footstep_msgs_generate_messages_eus
[ 0%] Built target jsk_recognition_msgs_generate_messages_lisp
[ 0%] Built target jsk_recognition_msgs_generate_messages_nodejs
[ 0%] Built target jsk_recognition_msgs_generate_messages_py
[ 0%] Built target sensor_msgs_generate_messages_cpp
[100%] Built target lidar_apollo_cnn_seg_detect
e[36mInstall the project…e[0m
– Install configuration: “Release”
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/lib/pkgconfig/lidar_apollo_cnn_seg_detect.pc
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/share/lidar_apollo_cnn_seg_detect/cmake/lidar_apollo_cnn_seg_detectConfig.cmake
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/share/lidar_apollo_cnn_seg_detect/cmake/lidar_apollo_cnn_seg_detectConfig-version.cmake
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/share/lidar_apollo_cnn_seg_detect/package.xml
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/bin/lidar_apollo_cnn_seg_detect
– Set runtime path of “/home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/bin/lidar_apollo_cnn_seg_detect” to “”
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/share/lidar_apollo_cnn_seg_detect/launch
– Installing: /home/gz/Autoware/ros/install/lidar_apollo_cnn_seg_detect/share/lidar_apollo_cnn_seg_detect/launch/lidar_apollo_cnn_seg_detect.launch

and this is caffe: does it means i have installed caffe successfully?
Screenshot%20from%202019-05-01%2012-18-27

and then i recompile Autoware:

$ cd ~/Autoware/ros/
$ rosdep update
$ rosdep install -y --from-paths src --ignore-src --rosdistro $ROS_DISTRO
$ ./colcon_release

output info in my screen:

Summary: 140 packages finished [11min 9s]
44 packages had stderr output: astar_search autoware_camera_lidar_calibrator autoware_connector autoware_driveworks_gmsl_interface autoware_driveworks_interface autoware_pointgrey_drivers data_preprocessor decision_maker dp_planner kitti_player kvaser libdpm_ttic lidar_euclidean_cluster_detect lidar_localizer lidar_point_pillars lidar_shape_estimation map_file microstrain_driver mqtt_socket ndt_cpu ndt_gpu object_map op_ros_helpers op_simulation_package op_utilities ouster_ros pcl_omp_registration pixel_cloud_fusion points_downsampler points_preprocessor pos_db road_occupancy_processor sick_ldmrs_tools trafficlight_recognizer vector_map_server vision_darknet_detect vision_dpm_ttic_detect vision_segment_enet_detect vision_ssd_detect vlg22c_cam way_planner waypoint_follower waypoint_maker waypoint_planner

then in autoware runtime manager,when i use lidar_cnn_baidu_detect node,
i got the same error again:

ERROR: cannot launch node of type [lidar_apollo_cnn_seg_detect/lidar_apollo_cnn_seg_detect]: can’t locate node [lidar_apollo_cnn_seg_detect] in package [lidar_apollo_cnn_seg_detect]

anyone can help me?? please! thank u!

@amc-nu
my CUDA & CUDnn info is below:
Screenshot%20from%202019-05-01%2012-38-01

gz@cc:~$ cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 5
#define CUDNN_PATCHLEVEL 1
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include "driver_types.h"
gz@cc:~$ 
gz@cc:~$ cat /usr/local/cuda/version.txt 
CUDA Version 9.0.176
gz@cc:~$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2017 NVIDIA Corporation
Built on Fri_Sep__1_21:08:03_CDT_2017
Cuda compilation tools, release 9.0, V9.0.176
gz@cc:~$ 

Thanks for your question. However we ask that you please ask questions on http://answers.ros.org following our support guidelines: http://wiki.ros.org/Support

ROS Discourse is for news and general interest discussions. ROS Answers provides a forum which can be filtered by tags to make sure the relevant people can find and/or answer the question, and not overload everyone with hundreds of posts.

@amc-nu Please help by directing people there instead of just answering here.

@defendergz Thanks for the question with details, please reask your question on answers.ros.org and then edit your post here to link to it so others interested in your solution can find the other thread.

I’ll note that in our support guidelines we ask that you not use screenshots of text. The copy and pasted code is much more useful in terms of searchability and able to be copy and pasted again.