Trajectory Capture and Visualization Package

Overview

ROS2 package for capturing and visualizing robot trajectories. Records odometry data and provides visualization tools.

Installation

Clone the Repository

git clone https://github.com/SuperGops7/nav.git

Install required dependencies and build the package

cd /nav/nav_ws
rosdep update
rosdep install --from-paths src --ignore-src -r -y

colcon build --symlink-install
source install/setup.bash

Directory Structure

The entire workspace consists of three packages:

turtlebot3_simulations

Forked from the official ROBOTIS/turtlebot3 repository. We use this as the AMR with SLAM and Navigational capabilities to save the trajectory of. This helps in testing and valdiating the trajectory capturing capability of the solution.

To get the gazebo simulation of the Turtlebot3 Waffle up and running, run the following

ros2 launch turtlebot3_gazebo turtlebot3_world.launch.py

To launch rviz to set Nav2 goals:

ros2 launch turtlebot3_navigation2 navigation2.launch.py use_sim_time:=True map:=<path-to-map>/map.yaml

Replace with the path to the map parameters file (can be found in /nav_ws/src/trajectory_capture/maps).

trajectory_capture

The trajectory_Capture package houses two nodes trajectory_capture - to capture the trajectory information trajectory_publisher - to visualize the captured trajectories.

To capture, run the following:

ros2 run trajectory_capture trajectory_capture

To visualize, run the following:

ros2 launch trajectory_capture map_launch.py
ros2 run trajectory_capture trajectory_publisher

These nodes have been written with very minimal overhead, to ensure quick and fast execution.

trajectory_interfaces

To allow for the user to specify the name of the filename to save to, and the duration of recording, we create our custom service, which is defined in this package.

The custome service takes the following structure:

Request

string filename - name of file, is also used to determine the type of file (.csv, .json or .yaml) and parse the trajectory points as per the format

double duration - specify the duration of recording.

Response

bool success - specify whether any errors were met with. string message - to respond with error codes, if any.

To call the service, use the following command. The user can change the filename and duration as per their need.

ros2 service call /trajectory_capture_service turtlebot_interfaces/srv/TrajectoryRequest "{filename: "test.yaml",  duration: 40.0}"

A custom struct has also been create to capture trajectory points, with the following structure:

struct TrajectoryStruct {
  double x;
  double y;
  double z;
  int32_t sec;
  uint32_t nsec;
};

This is done to ensure only the required details to visualize are captured, avoiding storing unwanted details such as twist information, saving memory.

Further improvements

1. Dubugging the issue where map does not load during visualization
2. Dockerizing the entire workspace in a container to facilitate better deployability
3. Add more jobs to the current Github Action workflow to impove code quality and CI/CD compliance.

Demo

A video run of the entire workflow can be found here

Contact

For any clarifications or to report any issues, please reach out to me.