Single-Equipment with Multiple-Application for an Automated Robot-Car Control System

Saleem Ullah, Zain Mumtaz, Shuo Liu, Mohammad Abubaqr, Athar Mahboob, Hamza Ahmad Madni2019 · DOI

TL;DR

This paper presents a low-cost, versatile robot car control system that integrates hand-gesture recognition and an Android mobile application for intuitive control. The system aims to simplify car operation for non-expert users and offers a foundation for more advanced robotics and automotive applications.

What Problem Does This Paper Address?

Controlling complex machinery like vehicles can be challenging, especially for users unfamiliar with traditional remote controls or switches. While gesture control and mobile applications offer more intuitive interfaces, existing solutions often focus on a single control method or require expensive, specialized hardware. This paper addresses the need for a unified, affordable, and user-friendly system that combines multiple control modalities (hand gestures, touch, and voice commands via a mobile app) for a robot car, making advanced control accessible and adaptable.

Methods

The proposed system utilizes an Arduino microcontroller as the central processing unit. Two primary control modes are implemented:

  • Hand-Gesture Control: This mode employs an MPU-6050 accelerometer integrated into a glove worn by the user. The accelerometer detects hand movements and their orientation, converting these into angle values. These values are then transmitted wirelessly via an RF transmitter to an Arduino Uno on the robot car, which interprets the gestures to control the car's movement (forward, backward, left, right, brake).
  • Android Mobile Application Control: This mode leverages a Bluetooth module (HC-05) to connect a smartphone to the robot car. The Android application offers two sub-modes:
  • Touch Buttons: Users can control the car using on-screen arrow buttons or simple function buttons.
  • Voice Recognition: The application utilizes Google Speech-to-Text to convert spoken commands (forward, backward, left, right, brake) into text, which is then sent to the Arduino for execution.

An additional feature is an automatic obstacle detection system using sensors placed at the front of the car to enhance safety by stopping the car before it collides with an obstacle.

Datasets

Not discussed in the paper. The system relies on real-time sensor input and predefined gesture/command mappings rather than training on large datasets.

Key Insights and Ideas

  • Unified Control Platform: The core innovation is the integration of multiple control methods (hand gestures, touch, voice) into a single, cohesive system, offering flexibility and ease of use.
  • Low-Cost Accessibility: By using readily available and affordable components like Arduino, accelerometers, and RF/Bluetooth modules, the system democratizes advanced robotics control.
  • Gesture Interpretation: The paper demonstrates a practical approach to translating accelerometer data into meaningful car movements, mapping specific hand tilts and rotations to directional commands.
  • Mobile App Integration: The seamless integration of an Android app, offering both touch and voice control, significantly enhances user interaction and accessibility.
  • Safety Enhancement: The inclusion of an obstacle detection system highlights a crucial aspect for real-world applications, improving the robustness and safety of the robot car.

Key Results

The paper presents a lab-scale prototype that successfully demonstrates the functionality of both the hand-gesture and mobile application control modes.

  • Hand-Gesture Control: The system accurately translates hand movements into corresponding car movements (forward, backward, left, right, brake) with a defined range of angles.
  • Mobile Application Control: Both touch-button and voice-recognition modes effectively control the robot car's actions. The voice recognition is limited to five commands, and the touch interface offers various button layouts.
  • Obstacle Detection: The obstacle detection system successfully stops the car when an obstacle is sensed, preventing collisions.
  • Overall System: The experimental results validate the system's efficiency, accuracy, and affordability, confirming its ease of use and assembly.

Limitations and Future Work

Limitations:

  • Range: The Bluetooth module has a limited range (10-15m), which can lead to dropped connections and increased response delays.
  • Noise Sensitivity: Voice recognition can be affected by ambient noise, potentially leading to errors.
  • Gesture Set: The presented system supports a limited set of gestures.
  • Lab-Scale Prototype: The current implementation is a lab-scale prototype, and real-world implementation may require further refinement.

Future Work:

  • Increased Accuracy and Gestures: Expanding the gesture recognition capabilities and improving accuracy.
  • GPS Integration: Adding GPS for location tracking and navigation.
  • Alternative Control Methods: Exploring head motion, eye blink, and voice-activated power.
  • Extended Wireless Range: Utilizing Wi-Fi or GSM modules for greater wireless range and scalability.
  • Onboard Camera: Integrating a camera for live streaming and remote control from afar.
  • Dynamic Driving Conditions: Improving accuracy for dynamic automobile applications by assessing driving conditions.
  • Real-Scale Implementation: Adapting the system for smart cars, personal robots, and other advanced robotic platforms.

Summary updated

BibTeX
@misc{ullah2019single,
  author = {Saleem Ullah and Zain Mumtaz and Shuo Liu and Mohammad Abubaqr and Athar Mahboob and Hamza Ahmad Madni},
  title = {{Single-Equipment with Multiple-Application for an Automated Robot-Car Control System}},
  year = {2019},
  doi = {10.3390/s19030662},
}
Figure