Acme Robotics Inc. aims to explore new realms in defense robotics by developing a quadrotor drone capable for exploring an unknown user-defined land patch and collect intelligence (number of hostages & terrorists) all the while developing a map of the explored area. The scope of this project is to develop the end-to-end software package for the above-mentioned drone operation.
-
Venkatraman Narayanan: Currently pursuing Masters in Robotics at the University of Maryland. Interests include Motion Planning, Machine Learning, drones, SLAM algorithms and deep learning.
-
Amrish Baskaran: Robotics enthusiast interested in automation and control. Has 2 years experience in hobby CNC manufacturing and control software programming. Has an Undergrad degree in Mechanical Engineering from VIT University, India currently pursuing his Master's degree in Robotics at University of Maryland College Park.
License file can be found here
MIT License
Copyright (c) 2018 Venkatraman Narayanan, Amrish Baskaran
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
This module will be developed using the Solo Iterative Process(SIP), Test Driven Development and agile development in a 3 week sprint method. The spreadsheet for the Product log, iteration backlog, work log and sprint details can be found in this link-Agile Development Spreadsheet
Notes from the sprint review sessions can be found in the link-Sprint review Doc
Demo video for the installation and execution of the provided package can be seen below-
The corresponding slides for the demo can be found here.
- ROS Kinetic
- Ubuntu 16.04
- Catkin
- Gtest
- Rostest
- Travis CI Documentation
- Coveralls Documentation
- tf
- Rosbag
- OpenCV
- Rviz
- Gazebo 7.0.0
- tum_Simulator
- ardrone_autonomy
- LSD SLAM
- industrial_ci
sudo apt install libsuitesparse-dev libqglviewer-dev-qt4 ros-kinetic-libg2o ros-kinetic-opencv3
sudo ln -s /usr/lib/x86_64-linux-gnu/libQGLViewer-qt4.so /usr/lib/x86_64-linux-gnu/libQGLViewer.so
Go to Catkin workspace and clone the lsd slam repository
cd ~/catkin_ws/
git clone https://github.com/vijay4313/lsd_slam.git
cd ..
catkin_make
If Catkin worspace is not available or needs to be created-
mkdir -p ~/catkin_ws/src
cd ~/catkin_ws/
catkin_make install
source devel/setup.bash
After creating the catkin workspace move to the src/ folder.
cd src/
git clone --recursive https://github.com/vijay4313/intelli_bot.git
cd ..
catkin_make
source devel/setup.bash
Go to Catkin Workspace
cd ~/catkin_ws/
catkin_make run_tests intelli_bot_test
The demo can be run using the commandline instructions bellow
cd ~/catkin_ws/
source devel/setup.bash
roslaunch intelli_bot intelli_bot_demo.launch
- Recording rosbag with launch file. It is turned off by default
roslaunch intelli_bot intelli_bot_demo.launch record:=true
The intelli_bot_bag.bag file is saved in the results directory
- To change the world environment
roslaunch intelli_bot intelli_bot_demo.launch world_name:=<Location of World.world file>
-
r: Reset, will clear all displayed data. -
w: Print the number of points / currently displayed points / keyframes / constraints to the console. -
p: Write currently displayed points as point cloud to file lsd_slam_viewer/pc.ply, which can be opened e.g. in meshlab. Use in combination with sparsityFactor to reduce the number of points.
The simulation is achieved using Gazebo simulator.The demo consists of an environment with buildings and people(not friendly) scaterred around. The drone will take off from a known position and traverse through the area following a rectangular path. During this it will detect the Human objects, draw a bounding box around them, publish the image for viewing and record their position that will be displayed as blocks in rviz. This is accomplished using the onboard monocular camera of resolution 640 x 480. It also uses a Monocular SLAM package called LSD SLAM to map the area and display a point cloud in a viewer to the user.
Example of Human detection using the camera :
Example of the point cloud generated by LSD SLAM using the monocular slam :
roslaunch intelli_bot_demo.launch record:=1
recorded bag file will be stored in the results folder and records all except camera topics
To record for a specific time use the following command, as an example records 20 seconds.
roslaunch intelli_bot_demo.launch record:=1 seconds:=20
Navigate to the results folder and run the following command-
rosbag play intelli_bot_bag.bag
The doxygen documentation can be generated manually using the following commands
- Installation
sudo apt-get install doxygen
sudo apt-get install doxygen-gui
- Open Doxywizard and follow the instructions to generate the required files
doxywizard
- Occasional tracking loss due to sudden translation or rotational motion
- Usage of 3rd party non-ROS packages in project directories leading to reduction in coveralls percentage. Inclusion in project directory was necessary to avoid travis build errors.