RoSA_SLAM

This repository contains a Python-based Robust SLAM for Arboreal environments, combining 3D point cloud slicing and 2D Hausdorff scan-to-map matching for reliable navigation using only LiDAR data. The system consumes point clouds from a horizontally oriented Velodyne VLP-16 and produces accurate pose estimations, along with a 3D map, without relying on IMU or GNSS measurements.

1. Dependency

The algorithm was tested with:

Operating System: Ubuntu 18.04 LTS
Architecture: x86_64 ROS Distribution: ROS Melodic

System & ROS Dependencies

Component	Version / Notes
ROS	Melodic
roscpp	ROS Melodic default
rospy	ROS Melodic default
tf / tf2	ROS Melodic default
sensor_msgs	ROS Melodic default
geometry_msgs	ROS Melodic default

Python Dependencies

Package	Version
Python	3.8.3
fonttools	4.44.0
ipython	8.12.3
jupyter_client	8.6.0
jupyter_core	5.5.0
matplotlib	3.7.3
matplotlib-inline	0.1.6
numba	0.58.1
numpy	1.24.4
open3d	0.18.0
opencv-python	4.8.1.78
rosbag	1.14.13
rospy	1.14.13
scikit-learn	1.3.2
scipy	1.10.1

2. Build

Prerequisites:

• Ubuntu 18.04 LTS
• ROS Melodic (desktop-full) or melodic-ros-base with pcl_ros and pcl_conversions

Clone the repository and catkin_make:

git clone https://github.com/RAL-UC/RoSA_SLAM.git
cd RoSA_SLAM/
cp -r path_publisher ~/catkin_ws/src
cd ~/catkin_ws
catkin_make
source ~/catkin_ws/devel/setup.bash

3. Project Structure

├── data/                        # Input dataset
├── path_publisher/              # ROS package
│   ├── launch/                  # Launch files
│   ├── rviz/                    # RViz configuration
│   └── src/                     # Source code
├── pictures/                    # imagens of readme
├── pullally_example/            # Python code
│   ├── EKF/                     # EKF and robot model functions
│   ├── hausdorff/               # Trajectory matching
│   ├── utils/                   # Screen and variable utilities
│   └── example_pullally.ipynb   # Main example with output files and figures
├── results_evaluation.ipynb     # Output files and figures of some datasets
└── README.md

4. Example

To run the Python code, use:

example_pullally.ipynb

Download the dataset from Rosbag data_pullally_example.bag of Pullally Dataset and store it in YOUR_DATASET_FOLDER.

roscore
roslaunch path_publisher cloud_pose_mapper.launch
rosbag play data_pullally_example.bag

5. Docker

Prerequisites:

• Docker (v20+ recommended)

Dowload the Docker Image from Docker_RoSA, and load it on the destination computer:

Terminal 1:

docker load < ros-melodic-18_RoSA.tar

After loading the Docker image, start the container using the commands below.

• Replace <PATH_ROSBAG_IN_YOUR_PC> with the absolute path to the directory on your machine that contains the rosbag files. For example: /home/dataset_pullally.

• Execute the following commands:

xhost +local:docker
docker run -it \
  --env DISPLAY=$DISPLAY \
  --env QT_X11_NO_MITSHM=1 \
  --volume /tmp/.X11-unix:/tmp/.X11-unix:rw \
  --net=host \
  -v /<PATH_ROSBAG_IN_YOUR_PC>:/data/rosbags:ro \
  ros-melodic:18.04
roscore

The rosbag directory will be mounted inside the container at /data/rosbags in read-only mode.

Terminal 2:

Identify the running container:

docker ps

• Replace <YOUR_CONTAINER> with the id in your machine. For example: 0f59b1840653.

• Source the ROS environment and launch the application:

docker exec -it <YOUR_CONTAINER> /bin/bash
source /opt/ros/melodic/setup.bash
roslaunch path_publisher cloud_pose_mapper.launch

Terminal 3:

• Attach to the same container and play the rosbag file:

docker exec -it <YOUR_CONTAINER> /bin/bash
rosbag play data_pullally_example.bag

6. Cite

Nazate-Burgos, P., Torres-Torriti, M., Huang, S., & Auat Cheein, F. (2026). **Consistent lidar-only SLAM for legged agricultural robots in arboreal environments via robust dimensionality reduction. Computers and Electronics in Agriculture, vol. 247, 111687, ISSN 0168-1699, https://doi.org/10.1016/j.compag.2026.

@article{RoSA_SLAM_2026,
title = {Consistent lidar-only SLAM for legged agricultural robots in arboreal environments via robust dimensionality reduction},
journal = {Computers and Electronics in Agriculture},
volume = {247},
pages = {111687},
year = {2026},
issn = {0168-1699},
doi = {https://doi.org/10.1016/j.compag.2026.111687},
url = {https://www.sciencedirect.com/science/article/pii/S0168169926002826},
author = {Paola Nazate-Burgos and Miguel Torres-Torriti and Shoudong Huang and Fernando {Auat Cheein}},
}