467 Final Project - SimBot

Authors

• Laasya Chukka - lchukka@umich.edu
• Ansh Mehta - anshm@umich.edu
• Christian Vega - email@example.com
• Wendi Zhang - zwendi@umich.edu

Prerequisites

• Mbot
• Python 3
• Terminal
• Wifi Router (Please be sure that both Local Machine and MBot are on the same network on the Wifi Router)

Hand Gesture Model Specifics

• mediapipe 0.8.1 (if using Mac, please download mediapipe 0.10.9)
• OpenCV 3.4.2 or Later
• Tensorflow 2.3.0 or Later
• tf-nightly 2.5.0.dev or later (Only when creating a TFLite for an LSTM model)
• scikit-learn 0.23.2 or Later (Only if you want to display the confusion matrix)
• matplotlib 3.3.2 or Later (Only if you want to display the confusion matrix)

April Tag

• AprilTag repo is not specifically in the repository, you will need to clone it locally to your machine (This is due to dependancies in CMake that are specific to your machine)
• Please Refer to april_tag/README.md for details
• Add templates/ directory & receive_stream.py from the support_files directory to the AprilTag/scripts/ directory

  cp -r support_files/ AprilTag/scripts/

Demo of Final Project

April Tag Recognition

Setup and Usage

1. Open 3 terminals - 1 Locally & 2 MBot

In Terminal 1 (Mbot):

• Navigate to the camera_stream/ directory:

  cd camera_stream

• Run the following command:

  python3 camera_final.py

In Terminal 2 (Locally):

• Navigate to the april_tag/scripts/ directory:

  cd april_tag/scripts/

• Run the following command:

  python3 receive_stream.py

In Terminal 3 (Mbot):

• Navigate to the teleop_gesture/python/ directory:

  cd teleop_gesture/python/

• Run the following command:

  python3 data_delivery.py

Teleop Gesture

Setup and Usage

1. Reflash the *.uf2 file into Mbot.

2. Open three separate terminals - 3 MBot

In Terminal 1 (Mbot):

• Navigate to the teleop_gesture/shim_timesync_binaries/ directory:

  cd teleop_gesture/shim_timesync_binaries/

• Give the shim file permission to execute:

  chmod +x shim

• Run the following command:

  ./shim

In Terminal 2 (Mbot):

• Navigate to the teleop_gesture/shim_timesync_binaries/ directory:

  cd teleop_gesture/shim_timesync_binaries/

• Run the following command:

  ./timesync

In Terminal 3 (Mbot):

• Navigate to the teleop_gesture/python/ directory:

  cd teleop_gesture/python/

• Run the following command:

  python3 teleop_gesture_v#.py

  (Replace # with the appropriate version number)

3. Run the commands in the following order:
   • Terminal 1
   • Terminal 2
   • Terminal 3

Ensure that the terminals are running in the background to maintain the teleoperation functionality.

Hand Gesture Model

Setup and Usage

1. Open 1 terminal - 1 Locally

In Terminal 1 (Locally):

• Navigate to the hand_gesture_recognition_mediapipe_main/ directory:

 cd hand_gesture_recognition_mediapipe_main/

• Run the following command:

  python3 app.py

SLAM

1. Ensure that the LIDAR is plugged in and turned on

2. Open three separate terminals - 3 MBot

In Terminal 1 (Mbot):

• Navigate to the bot_lab/bin/ directory:

cd bot_lab/bin/

• Run the following command:

./rplidar_driver

In Terminal 2 (Mbot):

• Navigate to the bot_lab/bin/ directory:

cd bot_lab/bin/

• Run the following command:

./slam

• Navigate to the bot_lab/ directory:

cd bot_lab/

• Run the following command:

source setenv.sh

• Navigate to the bin/ directory:

cd bin/

• Run the following command: Useful Commands if using ssh connection

./botgui

Note

• Replace teleop_gesture_v#.py with the appropriate version of the Python script.
• Make sure to follow the order of commands as mentioned above for proper execution.
• Locally means to open terminals on your local computer, NOT on the Mbot's Raspberry Pi
• The following require that you update the IP address from the Mbot ifconfig use the wlan0 IP address:
  • camerafinal.py; line 26
  • app.py; line 47
  • teleop_gesture_v3.py; line 42
  • receive_stream.py; line 31

Useful Commands

To allow UI's to show up on VNC Viewer

ssh -X pi@[insert Mbot IP Address here]

once in the Raspberry Pi, run this command:

export DISPLAY=:0.0

To link receive_stream.py; allowing changes in your personal cloned AprilTag's receive_stream.py to support/receive_stream.py

ln AprilTag/scripts/receive_stream.py support_files/receive_stream.py

You might need to move support_files/receive_stream.py out of the it's directory momentarily for this to work (after you added a copy into AprilTag/scripts)

References

• hand-gesture-recognition-using-mediapipe
• AprilTag