A Color-Guided Autonomous Rover with Integrated Localization and Sensor-Fusion-Based Motion Control

Abstract

Accurate navigation in indoor industrial environments remains a major challenge, particularly in scenarios where Global Positioning System (GPS) signals are unavailable and physical factors such as wheel slippage and surface irregularities introduce motion errors. Conventional robotic systems often rely on open-loop control or basic proximity sensing, which lack feedback correction and result in cumulative positional drift over time. To overcome these limitations, this work presents a ColorAssisted Autonomous Rover (CAAR) designed for precise localization within a structured grid-based environment using a multi-sensor fusion approach. The system is built around an ESP32 Microcontroller Unit (MCU) and integrates a Micro-Electro-Mechanical Systems (MEMS)-based Inertial Measurement Unit (IMU) along with quadrature encoders to implement closed-loop Proportional-Integral-Derivative (PID) control, enabling continuous correction of motion errors during both linear and rotational movements. A dual Inter-Integrated Circuit (I2C) communication setup is employed to efficiently manage high-frequency motion data separately from lower-frequency sensor inputs, including data from a TCS34725 Red-Green-Blue (RGB) color sensor. The navigation framework converts Cartesian coordinate inputs into sequential movement commands, allowing the rover to systematically traverse predefined grid points. Upon reaching each location, the system performs color-based validation to confirm positional accuracy, providing an additional layer of environmental feedback. A WebSocket-based interface facilitates real-time communication, supporting both manual operation and autonomous navigation modes. Experimental results demonstrate a significant reduction in drift and improved consistency compared to traditional methods, making the system highly suitable for reliable indoor automation tasks.