Published Paper Details:

Autonomous drones are increasingly being deployed in applications such as surveillance, disaster management, agriculture, and delivery services. However, most existing navigation systems rely heavily on cloud computing, which introduces latency, connectivity issues, and high energy consumption. To address these limitations, this research proposes an Edge AI-based navigation system using ESP32, a lightweight yet powerful microcontroller capable of running TinyML models. The system integrates onboard sensors including a camera, IMU, ultrasonic sensor, and GPS to enable real-time obstacle detection, path planning, and stable flight control without dependency on cloud servers. A convolutional neural network (CNN) optimized with TensorFlow Lite Micro is deployed on the ESP32 for efficient object recognition and obstacle avoidance. Experimental evaluations demonstrate that the proposed system achieves low-latency decision-making with reduced power consumption while maintaining accurate navigation performance. The findings highlight the potential of ESP32-based Edge AI as a cost-effective and scalable solution for autonomous drone navigation, opening opportunities for future applications in smart cities, defense, and industrial monitoring.