Jackal Rescue Robot

This project is an implementation of a rescue operation using a swarm of Clearpath Jackal robots.

Overview

Search and rescue is one of the most widely implemented problems in real-life robotic applications. The task of rescue can be generalized and used for moving an object from a source to any desired destination. In this project, we propose to develop a system that will deploy a swarm of 20 robots in a simulated gazebo environment and help rescue an object successfully to the defined end goal. The robot we propose to use is the Jackal unmanned ground vehicle. Agents in the rescue swarm will spawn at a pre-set start position and then navigate towards the object to be rescued. Once they reach the object contact point the agents will push the object synchronously and start moving towards the goal location.

Specifications:

Number of agents in swarm : We propose to implement 20 agents and scale it to 30 in final sprint if time permits

Arena : The Gazebo world will will a closed area with wall surroundings and will contain immobile obstacles

Robot : The dimensions of the jackal robot can be seen in the figure below

Object : The object to be rescued is a rectangle of dimensions approimately equal to ~ 6m X 6m X 1m

Jackal Robot interesting facts

Clearpath Jackal	Turtlebot
20 Kg payload	> 1.3x more than Turtlebot
Run time 40 hrs	> 1.6x more than Turtlebot
Max speed 2m/s	> 7.7x more than Turtlebot
Widely used in industries

source

Authors

1. Kavyashree Devadiga (117398045)
Computer buff, interested in smart autonomous systems. Has a Bachelors degree in Computer Science and is currently pursuing Masters of Engineering in Robotics at University of Maryland College Park.

2. Aswath Muthuselvam (118286204)
Holds interest in Autonomous Mobile Agents, Computer Vision, Simultaneous Localization and Mapping, AI, Real-Time systems, and controls. Has a Bachelors's degree in Electrical, Electronics, and Communications Engineering and is currently pursuing Masters of Engineering in Robotics at the University of Maryland College Park.

UML

Activity diagram

Class diagram

Quad chart

The Quad chart is as follows. You can access the pdf version here

Presentation

Youtube link of our presentation can be found here
The link to the slides can be found here

Pushing using one jackal and lighter obstacle

Jackal2.mp4

Swarm trying to form pattern

Jackal1.mp4

Assumptions:

Robots are not aware of its neighbours.
Centralized communication, there is a common node(rosmaster) through which all robots communicates with each other.
Swarms only operate in 2D plane.

Tested on:

Ubuntu 20.04 and Ubuntu 18.04
OMPL library version: 1.5.2

Development process

The project will be developed using industry-grade agile methodologies. The agile method being adaptive in nature will quickly adapt to software requirements and changes due to challenges faced if any during the project development cycle. For software development, pair programming strategies will be used which will ensure a robust, bug-free package. We have hierarchical test suites, which will test the submodules by implementing unit test cases and the validity of the whole system. To simulate real-world challenges we plan to use different Gmock test cases and confirm the correctness of the system.

Sprint planning notes and review

AIP Spreadsheet

Installations

It is assumed that the system has Ubuntu 18.04 and above with ROS Melodic/Noetic installed. If not, install Ubuntu from here and ROS from here

First step is to install Jackal robot. Download the setup bash file from here.

Execute the downloaded file using following command

sh setup.sh

Install dependencies

OMPL

sudo sh -c 'echo "deb http://packages.ros.org/ros/ubuntu `lsb_release -sc` main" > /etc/apt/sources.list.d/ros-latest.list'
wget http://packages.ros.org/ros.key -O - | sudo apt-key add -

sudo apt-get update
sudo apt-get install ros-`rosversion -d`-ompl

octomap_mapping

sudo apt-get install ros-<ros-version>-octomap ros-<ros-version>-octomap-mapping
rosdep install octomap_mapping
rosmake octomap_mapping

Clone git repository

cd swarm_robots/src
git clone --recursive https://github.com/kavyadevd/swarm_robots.git

Execute program

catkin_make
roslaunch swarm_robots main.launch

On separate terminal run the following command

rosrun swarm_robots swarm_robots_node

Different options provided with the launch file are:

Launch Rviz : rviz:=y
Launch rqt topic list : rqt_topic:=y
Launch rqt graph : rqt_graph:=y
Launch tf tree : rqt_tf_tree:=y
Launch rosbag record node : rosbag_yn:=y

To launch the demo

catkin_make
roslaunch swarm_robots demo.launch