This project presents the design and implementation of a robust two-wheeled self-balancing robot equipped with an Adaptive Proportional-Integral-Derivative (PID) control system optimized for rough terrain navigation. Traditional PID controllers are often insufficient in unpredictable environments due to their fixed parameters. To address this, the robot integrates real-time sensor feedback from an MPU-6050 accelerometer and gyroscope to dynamically tune PID gains, ensuring continuous balance and stability. Powered by an Arduino Nano V3.0 and geared DC motors, the system is further enhanced with obstacle avoidance features using ultrasonic and infrared sensors. The project encompasses mechanical design, electronic integration, software development, and testing across various terrains such as grass, sand, and gravel. The results for this report validate the system’s ability to adapt to changing conditions, maintain stability, and avoid obstacles efficiently. This work contributes to the development of intelligent systems for real-world applications, including warehouse logistics and assistive mobility, by improving robustness, manoeuvrability, and terrain adaptability.