Published Paper Details:

The Optimized Path Planning for Last-Meter Drone Delivery Using Deep Q-Networks system is an intelligent autonomous delivery framework designed to improve the efficiency, speed, and reliability of last-meter logistics. Traditional delivery methods are often hampered by traffic congestion, high fuel consumption, increasing labour costs and delays in delivery, especially in congested urban areas and remote areas. To overcome these limitations, the proposed system employs Unmanned Aerial Vehicles (UAVs) with Deep Q-Network (DQN)-based reinforcement learning to achieve optimal path planning and autonomous navigation. The system combines GPS navigation, real-time obstacle detection sensors, flight controllers, wireless communication technologies (RF/Wi-Fi/4G/5G), and a web-based management platform to execute safe and efficient package delivery. Unlike traditional shortest path algorithms, the DQN agent learns the optimal delivery route on the fly through interaction with the environment, taking into account dynamic obstacles, battery constraints, weather conditions, no-fly zones, and delivery priorities. This adaptive learning enables the drone to achieve minimal travel distance, energy consumption, and delivery time, whilst maximising delivery success and operational safety. The proposed architecture is composed of hardware, software and communication layers and modules for user management, order processing, route optimisation, autonomous flight control, payload management, real-time tracking and monitoring. The system is implemented using Python, Flask, HTML, CSS and JavaScript and simulations are used to validate the effectiveness of the DQN based navigation strategy before implementation. Experimental evaluation demonstrates that the proposed approach significantly improves route efficiency, reduces operational costs, decreases human intervention, and enhances delivery reliability compared to traditional ground-based logistics systems. Furthermore, the system is particularly suitable for medical supply transportation, emergency response, food delivery, grocery distribution, and e-commerce logistics, where rapid and contactless delivery is essential. The proposed framework is a scalable, eco-friendly and intelligent solution for future autonomous drone-based logistics, paving the way to smart last-meter delivery systems in modern cities and remote areas.