A5017713206- UAV Applications and Optimization
- Robotic Path Planning Algorithms
- Distributed Control Multi-Agent Systems
- Advanced SAR Imaging Techniques
- Wireless Signal Modulation Classification
- Robotics and Sensor-Based Localization
- Blockchain Technology Applications and Security
- Machine Fault Diagnosis Techniques
Özyeğin University
2022-2024
In this work, we propose and analyze multi-drone path planners for multi-target search connectivity. The goal of the unmanned aerial vehicle (UAV) mission is to an unknown area detect, connect monitor multiple randomly distributed targets ground control station (GCS) while maintaining connectivity UAVs GCS. To end, use two types UAVs: relay. drones scan via onboard sensors, whereas relay provide We three different responses target detection with increasing adaptability: (i) follow...
Recently, the usage of blockchain on swarm UAV applications has gained popularity due to simplicity and security designed frameworks. Applications drones precision architecture have also been a prominent research topic for years. However, architectures that use as service (DAAS) such are still lacking some detailed analysis. This paper addresses known problems challenges networks, availability, route de-confliction, confidentiality authenticity, energy consumption. It presents conceptual...
We propose a dynamic path planner that uses multi-agent reinforcement learning (MARL) model with novel reward functions for multi-drone search and rescue (SAR) missions. design mission environment where team covers an area to detect randomly distributed targets inform the ground base station (BS) by continuously forming relay chains between BS. The training procedure of agents includes convolutional neural network (CNN) images which represent trajectory histories connectivity states each...
As the number of radar waveforms in cognitive electronic warfare applications increases, individual detection and classification performances each waveform vary furthermore due to their different characteristics. To provide a supervised signal an efficient framework, we propose multi-stage system, where multiple modular blocks are combined classify 18 waveforms. In first stage, transform signals into time-frequency images (TFIs) using Fourier-based Synchrosqueezing Transform (FSST)...