Congratulations to Mr. Kadircan Kara on the successful defense of his MSc thesis. The thesis is entitled “Multi-UAV Cooperative Search under Sensing and Communication Limitations” and it addresses different problems regarding multi-UAV coordination incorporating multi-objective optimization for coverage and multi-target search and rescue missions.
Our paper entitled “Sensing-Aware Cooperative Multi-UAV Search with Reinforcement Learning” is accepted for publication in IEEE PIMRC 2025.
This paper presents a novel reinforcement learning framework for multi-UAV cooperative search missions with sensing imperfections. We introduce a scalable multi-agent envi-
ronment that effectively models sensor uncertainty while enabling decentralized information merging between UAVs. Using Proximal Policy Optimization
(PPO), we create a training framework that achieves consistent
performance across varying numbers of UAVs and targets.
Our papers entitled “Joint Optimization of Connectivity, Coverage, and Revisit Time in Multi-UAV Path Planning” and “Adaptive Multi-UAV Coordination for Heterogeneous Target Search and Connect Missions Using Proximal Policy Optimization” accepted for publication at IEEE VTC 2025.
First, paper incorporates revisiting covered areas into a multi-UAV search mission path to overcome sensing limitations or enable prioritized area search.
Second paper proposes a reinforcement learning based multi-UAV path planner for a multi-target search mission.
Congratulations to Mr. Islam Guven, Mr. Hamid Balanji and Ms. Atefeh Molaei on the successful defense of their MSc thesis. The theses address different problems regarding multi-UAV coordination incorporating multi-objective optimization, reinforcement learning and successive optimization for multi-target search and rescue missions and network performance analysis for a UAV swarm.
Thesis are entitled as follows:
Islam Guven, “Multi-UAV Path Planning for Joint Coverage and Connectivity using Reinforcement Learning,” Ozyegin University, December 2024.
Hamid Balanji, “Search and Rescue with Multiple UAVs: \\[0.5\baselineskip]Target detection and connectivity,” Ozyegin University, December 2024.
Atefeh Molaei, “Network Provisioning using Multiple UAVs in \\[0.5\baselineskip] Search and Rescue Missions,” Ozyegin University, December 2024.
Our paper entitled “Dynamic Multi-UAV Path Planning for Multi-Target Search and Connectivity” is accepted for publication in IEEE TVT.
In this work, we propose and analyze multi-drone path planners for multi-target search and connectivity. The goal of the unmanned aerial vehicle (UAV) mission is to search an unknown
area to detect, connect and monitor multiple randomly distributed targets to the ground control station (GCS) while maintaining the connectivity of the UAVs to GCS. To this end, we propose to use two types of UAVs: search and relay. The search drones scan the
area via onboard sensors, whereas relay UAVs provide connectivity.
We propose three different responses to target detection with increasing adaptability: (i) follow pre-planned paths and inform GCS when possible, (ii) follow pre-planned paths and inject new UAVs to monitor the detected targets, (iii) assign a search UAV to monitor
target, and re-plan remaining UAV paths. Furthermore, we implement multi-objective optimization-based planners for single-type UAVs, where the paths are optimized in terms of total coverage time and percentage connectivity.
Our paper entitled “Multi-objective path planning for multi-UAV connectivity and area coverage” is accepted for publication in Ad Hoc Networks journal.
In this paper, we propose a multi-drone path planner that jointly optimizes area coverage time and connectivity among the drones. We propose a novel connectivity metric that includes not only percentage connectivity of the drones to GCS, but also the maximum duration of consecutive time that the drones are disconnected from the GCS. To solve this optimization formulation, we propose a multi-objective evolutionary algorithm with novel operations. We use our solver to test single, two and many objective path planning problems and compare our Pareto-optimal solutions to benchmark weighted-sum based solutions.
Our paper entitled “Network Analysis of Connectivity Optimized Multi-UAV Path Planners” is accepted for publication in 17th IEEE INFOCOM Workshop on Networked Robotics and Communication Systems (ex WISARN).
In this work, we analyze the network performance of several connectivity -optimized multi-UAV path planners. We analyze jointly optimized as well as relay assisted UAV networks. Our results show that topologically connected multi-UAV paths do not necessarily lead to acceptable network performance in terms of packet delivery rates and throughput. Mobile relay assisted scenarios perform as well as static relay scenarios with less than half the number of relay nodes. Jointly optimized schemes perform well with low number of UAVs when the transmission ranges are high or the number of nodes is low, where hub nodes are less likely to occur.
Our paper entitled “Priority-based Dynamic Multi-UAV Positioning for Multi-Target Search and Connectivity” is accepted for publication in IEEE WCNC 2024.
In this work, we propose an event-driven algorithm that integrates a novel connectivity-based prioritization of targets into dynamic positioning and path planning for multi-UAV systems in search and rescue (SAR) missions. Two distinct groups of UAVs are deployed. While search UAVs sense the area of interest as fast as possible, relay UAVs provide connectivity to search UAVs as well as targets depending on their priority level. 
Our paper entitled “Maintaining connectivity for multi-UAV multi-target search using reinforcement learning” is accepted for publication in ACM DiVANET 2023.
In this paper, we propose a dynamic path planner that uses a multi-agent reinforcement learning (MARL) model with novel reward functions for multi-drone search and rescue (SAR) missions. The training procedure of the agents includes a convolutional neural network (CNN) that uses images which represent trajectory histories and connectivity states of each environment entity such as drones, targets, BS. Agents take actions and get feedback from the environment until the mission is completed. The model is trained with multiple missions with randomized target locations.
We have organized a special session on Digital Technologies in Disaster Management within SIU 2023. The session was held on 7/7/2023, with 5 presentations. The program and titles of the presentations can be found here. The topics of discussion included applications of machine learning techniques for disasters, network proposals in case of disasters, etc.
