Gradient-NBV: Gradient-based Local Next-best-view Planning for Improved Perception of Targeted Plant Nodes

Akshay K. Burusa · Eldert J. van Henten · Gert Kootstra

Paper

https://github.com/akshaykburusa/gradientnbv/assets/127020264/dfa1f2a9-f07c-4af0-84ef-7ea20a7cb61b

Installation

Prerequisites

Ubuntu 20.04
ROS Noetic
At least 8GB of GPU VRAM

Clone the repository

git clone https://github.com/akshaykburusa/gradientnbv.git

ROS dependencies

Quick install:

cd gradient_nbv
sudo ./install_ros_pkgs.sh

Manual install:

sudo apt install ros-noetic-moveit
sudo apt install ros-noetic-ros-controllers
sudo apt install ros-noetic-trac-ik

Python packages

Quick install:

cd gradient_nbv
conda env create -f install_conda_pkgs.yaml

Manual install:

conda create -n grad_nbv python==3.8.10
conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
pip install pyyaml==6.0.1
pip install rospkg==1.5.1
pip install opencv-python==4.9.0.80
pip install scipy==1.10.1
pip install pytransform3d==3.5.0
pip install open3d==0.18.0

Compile

cd gradient_nbv
catkin_make -DCMAKE_BUILD_TYPE=Release

Execute

Bring up the simulation environment with the robot (ABB robotic arm + Realsense L515 camera):

roslaunch abb_l515_bringup abb_l515_bringup.launch

Start the gradient-based local next-best-view planner in a new terminal:

conda activate grad_nbv
roslaunch viewpoint_planning viewpoint_planning.launch

Citation

@inproceedings{burusa2024gradient,
  title={Gradient-based local next-best-view planning for improved perception of targeted plant nodes},
  author={Burusa, Akshay K and van Henten, Eldert J and Kootstra, Gert},
  booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)},
  pages={15854--15860},
  year={2024},
  organization={IEEE}
}