Project sTanka is an innovative initiative by the Roboclub at the Technical University of Sofia, focusing on the development of a hand-gesture recognition system to control a tank-like robot. This project seamlessly integrates computer vision and IoT technologies to offer an intuitive and responsive user experience.
The primary objective of Project sTanka is to enable users to maneuver a robotic tank using specific hand gestures. This is achieved through a combination of:
- Hand-Gesture Recognition: Utilizing a computer vision system trained to identify predefined gestures, which are then translated into movement commands for the robot.
- IoT Communication: Implementing the SmartConfig feature of the ESP32 microcontroller to establish a seamless connection between the laptop and the robot over Wi-Fi.
- Hardware Integration: Employing an ESP32 microcontroller on the robot to process incoming commands and control its movements accordingly.
- Microcontroller Programming: The ESP32 microcontroller is programmed using the Arduino platform, facilitating efficient control of the robot's hardware components.
- Hand-Gesture Recognition: The system leverages Python, Jupyter Notebook, TensorFlow, and OpenCV to develop and train the gesture recognition model, ensuring accurate and real-time gesture detection.
- Real-Time Gesture Control: Immediate response to user gestures, ensuring smooth and natural interaction with the robot.
- Wireless Communication: Reliable and fast data exchange over Wi-Fi, facilitated by the ESP32's SmartConfig feature, which simplifies network provisioning.
- Customizable Gestures: The system can be trained to recognize a variety of gestures, allowing for personalized control schemes.