Abstract:Wireless sensor networks (WSNs) with ultra-dense sensors are crucial for several industries, such as smart agricultural systems deployed in the fifth generation (5G) and beyond 5G Open Radio Access Networks (O-RAN). The WSNs employ multiple unmanned aerial vehicles (UAVs) to collect data from multiple sensor nodes (SNs) and relay it to the central controller for processing. UAVs also provide resources to SNs and extend the network coverage over a vast geographical area. The O-RAN allows the use of open standards and interfaces to create a wireless network for communications between the UAVs and ground SNs. It enables real-time data transfer, remote control, and other applications that require a reliable and high-speed connection by providing flexibility and reliability for UAV-assisted WSNs to meet the requirements of smart agricultural applications. However, the limited battery life of UAVs, transmission power, and shortage of energy resources SNs make it difficult to collect all the data and relay it to the base station, resulting in inefficient task computation and resource management in smart agricultural systems. In this paper, we propose a joint UAV task scheduling, trajectory planning, and resource-sharing framework for multi-UAV-assisted WSNs for smart agricultural monitoring scenarios that schedule UAVs' charging, data collection, and landing times and allow UAVs to share energy with SNs. The main objective of our proposed framework is to minimize the UAV energy consumption and network latency for effective data collection within a specific time frame. We formulate the multi-objective, which is a non-convex optimization problem, and transform it into a Markov decision process (MDP) with a multi-agent deep reinforcement learning (MADRL) algorithm. The simulation results show that the proposed MADRL algorithm reduces the energy consumption cost when compared to deep Q-network, Greedy, and mixed-integer linear program (MILP) by 61.92%, 68.02%, and 69.9%, respectively.

Multi-Agent Collaborative Framework for Automated Agriculture

An analysis of the robustness of UAV agriculture field coverage using multi-agent reinforcement learning

MULTI-AGENT PATH PLANNING OF ROBOTIC SWARMS IN AGRICULTURAL FIELDS

An intelligent WSN-UAV-based IoT framework for precision agriculture application

Coupling an Autonomous UAV With a ML Framework for Sustainable Environmental Monitoring and Remote Sensing

A Framework for Multi-Event Data Collection Using Unmanned Aerial Vehicle Aided Internet of Things in Smart Agriculture

Immersive Human-Machine Teleoperation Framework for Precision Agriculture: Integrating UAV-based Digital Mapping and Virtual Reality Control

UAV Path Planning Model Leveraging Machine Learning and Swarm Intelligence for Smart Agriculture

A Collaborative Framework for 3D Mapping Using Unmanned Aerial Vehicles

Sustainable Task Offloading in Secure UAV-assisted Smart Farm Networks: A Multi-Agent DRL with Action Mask Approach

Building an Aerial-Ground Robotics System for Precision Farming: An Adaptable Solution

Challenges and Opportunities for Autonomous Micro-UAVs in Precision Agriculture.

Cooperative Multi-Agent Planning Framework for Fuel Constrained UAV-UGV Routing Problem

Placement and drone flight path mapping of agricultural soil sensors using machine learning

Solving Vehicle Routing Problem for unmanned heterogeneous vehicle systems using Asynchronous Multi-Agent Architecture (A-teams)

A Multi Moving Target Localization in Agricultural Farmlands by Employing Optimized Cooperative Unmanned Aerial Vehicle Swarm

Multi-agent Deep Reinforcement Learning-based Task Scheduling and Resource Sharing for O-RAN-empowered Multi-UAV-assisted Wireless Sensor Networks

Multi-Agent Collaboration for Building Construction

Experimental Evaluation of a Hierarchical Operating Framework for Ground Robots in Agriculture

Multi-agent systems for quadcopters

A Hierarchical Framework for Long-term and Robust Deployment of Field Ground Robots in Large-Scale Farming