Semantic Sensor Fusion for Unitree Go1 (WIP)

Welcome to the repository for the Semantic Sensor Fusion for Unitree Go1 project, developed by the Real World Robot Informatics Lab at the University of Tokyo.

This project aims to enhance the robot's ability to associate its surroundings with objects, thereby imbuing them with meaning, and subsequently applying SLAM (Simultaneous Localization and Mapping) techniques.

This work is part of the research conducted by PhD student Jiaxu Wu.

Real World Robot Informatics Lab's Website

Qi An's Website

Code

The robot operates using ROS1 Kinetic. If you're not using Ubuntu 16, try to create a Docker image. There's one based on the OSRF ROS-Desktop-Full Kinetic environment. I'll publish a template of it soon.

To use my packages you need to install in addition to ROS Kinetic:

sudo apt-get update && apt-get install -q -y --no-install-recommends \
	ros-kinetic-velodyne \
	ros-kinetic-realsense2-camera \
	ros-kinetic-realsense2-description \
    && rm -rf /var/lib/apt/lists/*

cd ~/catkin_ws/src/
git clone https://github.com/ros-drivers/velodyne.git
cd ..
rosdep install --from-paths src --ignore-src --rosdistro kinetic -y
catkin_make

LiDAR VLP16 sensor

To configure the LiDAR, you have to declare the ip port, verify with wireshark what are yours and then run:

sudo ifconfig enp2s0f2 192.168.3.100
sudo route add 192.168.1.201 enp2s0f2

Using terminator, in the first terminal run:

roslaunch velodyne_pointcloud VLP16_points.launch

In a second one, run:

rosnode list
rostopic echo /velodyne_points

Stop it and then run (the arg is to put velodyne as a fixed frame):

rosrun rviz rviz -f velodyne

Add "cloudpoint2", and then select the topic "/velodyne_points". If you see the cloud point, it means everything is working correctly.

RealSense Camera

The configuration is automatically done. To test it:

roslaunch realsense2_camera demo_pointcloud.launch

If you see the cloud point, it means everything is working correctly

Calibration (in implementation)