Mean_Based Approach to Omnidirectional Pathfinding Algorithm Using Single Ultrasonic Sensor for Robotic Car

Abstract: For design and testing the smart pathfinding system for a robotic car using ultrasonic sensor. In this project we are using Q learning algorithm which means it can be created own path itself. The proposed algorithm works in two phases: normal detection and intensive detection. In the normal detection phase, the sensor scans for obstacles directly in front of the car. In the intensive detection phase, a servo motor moves the sensor to scan a wider area, acting like multiple sensors combined. This method improves obstacle detection and Accuracy. The system is built on an Arduino UNO platform and controls DC motors using an H-W 130 motor shield. The ultrasonic sensor gathers distance data, and the algorithm processes this information to determine the safest path. A mean based approach is used to reduce false detections and improve accuracy. Unlike traditional robotic systems that require multiple sensors, our proposed model achieves 360-degree obstacle detection with single sensor, reducing cost hardware complexity and Energy Consumption. The robotic car was tested in real world conditions, and obstacles were successfully detected and avoided with high accuracy