r2auto_nav

ROS2 auto_nav code for EG2310, sem2 AY21/22

Welcome to group4's repo for EG2310, and this is all our code for our Beta Gasoline robot.

Please follow our step by step guide to start up your system. We are assuming that you are using the exact same hardware as us, such as our launcher and loading system, and our electrical connections.

Preparation

Do the following things to setup your laptop and turtlebot

1. Follow the instructions here to setup your laptop and turtlebot https://emanual.robotis.com/docs/en/platform/turtlebot3/quick-start/
2. Follow this for instructions on setting up i2c communication on the turtlebot's raspberry pi https://ask.wingware.com/question/3/i2c-problem-with-remote-raspberry-pi/
3. Follow this for instructions on setting up gpiozero and pigpio pin factory (so that the servo used will not spasm) https://gpiozero.readthedocs.io/en/stable/remote_gpio.html
4. Check our our /Hardware folder for the schematics of the pcbs that we use to replicate our robot system. IMPORTANT! FOR OUR 12V SUPPLY AS STATED IN OUR RPI HAT PCB, WE USED 5V INSTEAD DUE TO 12V SPINNING OUR MOTORS TOO FAST AND CAUSING OSCILLATIONS ON OUR ROBOT

File allocation and building of workspaces

clone our repository into your Home directory.

git clone https://github.com/Magmanat/r2auto_nav.git

Then shift the entire folder hardware_bringup into your turtlebot3_ws using scp, then build the workspace, for example

scp -r path_to_r2auto_nav/turtlebot3_ws/src/hardware_bringup ubuntu@(ip-address-of-pi):~/turtlebot3/src
ssh ubuntu@(ip-address-of-pi)
cd turtlebot3_ws
colcon build

cd into colcon_ws and colcon build the workspace to setup the ros package on your laptop

cd path_to_r2auto_nav/colcon_ws
colcon build

Your should be done setting up after this, now its time to have fun with our package

System check

Firstly, we will use the factory_test package to ensure that everything is running properly, we will ssh into the turtlebot and startup all the hardware publishers and subscribers.

For your Turtlebot

In one terminal

ssh ubuntu@(ip-address-of-pi)
roslaunch turtlebot3_bringup turtlebot3_robot.launch

In another terminal

ssh ubuntu@(ip-address-of-pi)
ros2 launch hardware_bringup hardware.launch.py

Finally, we will run the script in your laptop to check that the whole system is working

For your Laptop

In one terminal

ros2 run auto_nav factory_test

Follow the instructions printed on your terminal, and if everything works out fine, it means the system is ready to go.

Running auto_nav in a maze

Now we will get to try the robot in an actual maze, and see it complete its mission, prepare two terminals for your Turtlebot and prepare 3 terminals for your Laptop. Similarly, we will start all the hardware in the Turtlebot first.

For your Turtlebot

In one terminal

ssh ubuntu@(ip-address-of-pi)
roslaunch turtlebot3_bringup turtlebot3_robot.launch

In another terminal

ssh ubuntu@(ip-address-of-pi)
ros2 launch hardware_bringup hardware.launch.py

Finally, we will run the commands to start the auto navigation

For your Laptop

First terminal

ros2 launch turtlebot3_cartographer cartographer.launch.py

Ensure that you see rviz open up, laserscan and map data will be able to be seen on the rviz window.

Second terminal

ros2 run auto_nav map2base

Ensure that the transforms are properly being published by observing the terminal for any error messages that print.

Third terminal

ros2 run auto_nav wall_follow

Finally, after running this command, the robot should move around and perform its mission autonomously, moving around the maze and stopping at an nfc loading bay, which it will then stop and wait to be loaded and button to be pressed to continue with its mission. After moving one entire round around the maze, and also being loaded, it will then search for a thermal target, which when found, the turtlebot would start centering towards the target and finally when the target is approximately 15 cm away from the turtlebot, it would stop and start the firing sequence.

Parameters that affect operation of the robot

From lines 35 to 57 of /colcon_ws/src/auto_nav/auto_nav/r2_wall_follow.py

The explanation of all the parameters are as commented below. These are the values which we found work the best for us in our use case. Please change these values if you require a different behavior of the robot.

#variables affecting targeting
TARGETspeedchange = 0.05 #forward speed for targeting
TARGETrotatechange = 0.1 #rotation speed for targeting
TARGET_hotthreshhold = 31 #target temperature
TARGETshoot_distance = 0.45 #distance from center of lidar to front before shooting
TARGET_front_angle = 3 #angle of front to measure distance
TARGET_moveres = 0.2 #time for sleep when moving
TARGET_target_not_detected_threshhold = 2 #How long in seconds before target considered as not detected


# variables affecting navigation
fastspeedchange = 0.2 #speed when bot moving fast linear speed
slowspeedchange = 0.12 #speed when bot moving slow linear speed

turning_speed_wf_fast = 1.1 #fast turning speed for bot
turning_speed_wf_medium = 0.75 #medium turning speed for bot
turning_speed_wf_slow = 0.5 #slow turning speed for bot

front_d = 0.45 #threshhold front distance that robot should not enter
side_d = 0.45 #threshold side distance that robot should not enter

target_count_threshhold = 5 # (how long you want hot_target to be spotted before activating firing) / hot_timer_delay = target_count_threshhold
hot_timer_delay = 0.1 # in seconds