CUDA Toolkit v9.0 or v10.0
CUDNN: Tested with v7.3.1
TensorRT: Tested with 5.0.2 -> How to install
https://github.com/k0suke-murakami/kitti_pretrained_point_pillars/
GPU编译：
$ AUTOWARE_COMPILE_WITH_CUDA=1 colcon build --cmake-args -DCMAKE_BUILD_TYPE=Releaseros启动运行命令方式：
roslaunch lidar_point_pillars lidar_point_pillars.launch
rosbag play test.bag
rviz autoware启动方式：
cd ~/autoware.ai
source install/setup.bash
roslaunch runtime_manager runtime_manager.launch
取消baseline_support，ref添加对应的pfe.onnx和rpn.onnx模型，点击OK
标签：point,onnx,autoware,pillars,manager,相关,runtime From： https://www.cnblogs.com/liuweiweitju/p/18003604