Skip to content

Latest commit

 

History

History
115 lines (77 loc) · 3.35 KB

install.md

File metadata and controls

115 lines (77 loc) · 3.35 KB

Install

  1. Docker build

Download a cuDNN 8.6.0 installer deb-package from NVIDIA's download page (login required).

Place the cuDNN installer deb-package in this directory and rename it to cudnn-local-repo-$(dpkg --print-architecture).deb.

Download a TensorRT 8.5.3 installer deb-package from NVIDIA's download page (login required).

Place the TensorRT installer deb-package in this directory and rename it to nv-tensorrt-local-repo-$(dpkg --print-architecture).deb.

Build the Docker image with below command.

sh ./docker/setup_env.sh
  1. ROS Noetic
  • vision_msgs
sudo apt-get install ros-noetic-vision-msgs
  1. Support packages
sudo apt install qtmultimedia5-dev

sudo apt-get install ros-noetic-ddynamic-reconfigure

sudo apt-get -y install libflann-dev

sudo apt-get -y install libflann1.9

sudo apt install build-essential libboost-system-dev libboost-thread-dev libboost-program-options-dev libboost-test-dev

sudo apt-get install cmake libblkid-dev e2fslibs-dev libboost-all-dev libaudit-dev

sudo apt-get install libvtk7-dev

sudo apt-get install ros-noetic-joy ros-noetic-teleop-twist-joy \
  ros-noetic-teleop-twist-keyboard ros-noetic-laser-proc \
  ros-noetic-rgbd-launch ros-noetic-rosserial-arduino \
  ros-noetic-rosserial-python ros-noetic-rosserial-client \
  ros-noetic-rosserial-msgs ros-noetic-amcl ros-noetic-map-server \
  ros-noetic-move-base ros-noetic-urdf ros-noetic-xacro \
  ros-noetic-compressed-image-transport ros-noetic-rqt* ros-noetic-rviz \
  ros-noetic-gmapping ros-noetic-navigation ros-noetic-interactive-markers
  1. object_mapping (Skip CUDA Install if use jetpack in Jetson series)

  • CUDA 11.6 [Link]

    • Choose option and follow instructions

  • OpenCV 4.5.2 (With CUDA Build) [Link]

    • Note: Change CUDA_ARCH flag to your NVIDIA Device and OpenCV version
    • Note: If you have conflict with Ros opencv, remove them and install ros depends manual

  • PCL 1.8 [Link]

    • Uncompress and build

  • TensorRT

  • ZED SDK

  • Realsense SDK (2.50.0)

  • Realsense ROS (2.3.1)

git clone -b v7.0 https://github.com/ultralytics/yolov5.git
# create conda envs and install requierments.txt for running gen_wts.py
# stupid scripts

git clone -b yolov5-v7.0 https://github.com/wang-xinyu/tensorrtx.git
cd yolov5/
wget https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s.pt
cp [PATH-TO-TENSORRTX]/yolov5/gen_wts.py .
python gen_wts.py -w yolov5s.pt -o yolov5s.wts
# A file 'yolov5s.wts' will be generated.

cd [PATH-TO-TENSORRTX]/yolov5/
# Update kNumClass in src/config.h if your model is trained on custom dataset
mkdir build
cd build
cp [PATH-TO-ultralytics-yolov5]/yolov5s.wts . 
cmake ..
make

# Generate engine file (engine include 80 class of coco dataset)
./yolov5_det -s yolov5s.wts yolov5s.engine s

# Build and serialize TensorRT engine
./yolov5_seg -s yolov5s-seg.wts yolov5s-seg.engine s

Bug

Pls refer links:

PCL build Errors

PointCloudMapping Errors

Note

  • Make -j4
  • If log too long, use can use this command:
catkin build -j4 &> log.txt