r2auto_nav

Tánkyu 2310i
Group 2, Studio 2
Ang Kai Jun, David Chong Joon Wei, Ha Zhe Li, Rayner Lim Fang Yuh, Wang Yanxiao Austin

EG2310 - Fundamentals of Systems Design, AY 21/22,
Innoovation and Design Programme, National University of Singapore (NUS).

Description

The Tánkyu 2310i, developed by NUS IDP students from EG2310, is a modified Robotis Co. Turtlebot 3. It is augmented with LIDAR, NFC-detection, IR-detection and a flywheel firing system to accomplish its set objectives - Autonomous navigation within a closed and connected maze, locating a loading bay demarcated by NFC tags, and firing a ping-pong ball at a target with high IR-signature.

In this repository, we are using Ubuntu 20.04.4, ROS2 Foxy and Python3.6 to explore a closed and connected maze using a left-wall following algorithm. You may check out our project report under the Documentations folder for more details on our robot and its mission.

File Organisation

• Documentations folder contain all the documentation of the Tánkyu 2310i, this includes the project report, assembly manual, full software setup guide and end user documentation.
• navigation.py code contains the main system and wall-following logic. This code also communicates with the mission code to allow the Tánkyu 2310i to engage on the ir signature when the conditions are met.
• mission.py code initiates all the i2c connections and contain the firing algorithm. This code performs all necessary logic processing with the input from the NFC and IR detection systems and communicates the information to the navigation code to allow the Tánkyu 2310i to engage on the ir signature when the conditions are met.
• Ubuntu_Files folder is an archive of the miscellaneous code that was tested but not implemented into the final system. The code are functional independently but requires some edits to integrate it into the final system.
• RPi_Files folder contains the factory acceptance test codes, all required packages and other test codes for each individual subsystem.
• Original_Files folder is an archive of the forked repository from shihchengyen's r2auto_nav_repository and is not necessary for the Tankyu 2310i's operations.

Calibration and configuration

Navigation Code

Under 'Adjustable variables to calibrate wall follower', you may experiment with different parameters to calibrate the navigation algorithm to suit your needs.

Variable name	Description	Recommended value
d	Distance from wall	0.35
fd	Front distance when approaching a merging wall	d + 0.1
reverse_d	Distance to reverse when too close to front wall	0.20
speedchange	Linear speed	0.17
turning_speed_wf_fast	Fast rotate speed, used when avoiding obstacle in front	0.75
turning_speed_wf_slow	Slow rotate speed, used when reversing, finding wall or too close to wall	0.40
snaking_radius	Distance from wall before correcting drift	d - 0.07
cornering_speed_constant	Coefficient of speedchange during cornering to prevent over/under steer	0.5

Mission Code

Under 'Adjustable variables to calibrate targeting' you may experiment with different parameters to calibrate the targeting algorithm to suit your needs.

Variable name	Description	Recommended value
rotatechange	Angular speed	0.35
speedchange	Linear speed	0.20
min_temp_threshold	IR live feed minimum value reference (Appears blue)	30.0
max_temp_threshold	IR live feed maximum value reference (Appears red)	35.0
detecting_threshold	Minimum temperature to identify target as a "Hot target"	32.0
firing_threshold	Acts as a second check after centering target	35.0
ir_offset	Offset angle for each IR sensor orientation (0°→0, 45°→31, 90°→69)	31.0

Operating Instructions

Installation

Tested on Ubuntu 20.04 - ROS2 Foxy
For beginners, follow our step-by-step software setup guide in our documentations folder.
For M1 MacBook users, refer to our video guide for Ubuntu setup and installation.

1. Install Ubuntu 20.04 and ROS 2 Foxy using the instructions in Section 3.1 of the Robotis e-manual (Ensure you click on the "Foxy" tab).
2. Using Ubuntu, complete the RPi in Section 3.2
3. Using Ubuntu, complete the OpenCR setup in Section 3.3

Set Up

1. Clone this repository onto your Ubuntu ros2 directory:
   

cd colcon_ws/src/auto_nav
git clone https://github.com/hahaha2002/r2auto_nav.git 
cd colcon_ws
colcon build

2. Copy mission.py and the RPi_files folder to the RPi:
   

scp -r <path to r2auto_nav directory>/mission.py ubuntu@<RPi IP address>:~/turtlebot_ws/src 
scp -r <path to r2auto_nav directory>/RPi_files ubuntu@<RPi IP address>:~/turtlebot_ws/src 

3. Build the package on RPi:
   

cd turtlebot3_ws 
colcon build 

4. Install the pigpio package on RPi and automate the daemon on boot:

sudo apt install pigpio
sudo systemctl enable pigpiod

Running Instructions

1. Ssh into the RPi, ssh ubuntu@<RPi IP Address>
2. On the RPi, initiate the bring up `ros2 launch turtlebot3_bringup robot.launch.py'
3. On RPi, in the RPi_files directory, Start the targeting code python3 mission.py.
4. On Ubuntu, in the r2auto_nav directory, Start the navigation code python3 navigation.py.