INVOLI’s Post

🚀 Excited to Share My Latest Project: Real-Time Drone Detection Using YOLOv11! 🛩️ I recently completed a project focused on developing a real-time drone detection system leveraging the power of the YOLOv11 (You Only Look Once) architecture. This project addresses the growing concern of drone usage in various sectors, from security to environmental monitoring. 🔍 Project Highlights: Real-Time Performance: Implemented YOLOv11 to achieve fast and accurate drone detection in live video streams. Robust Training: Trained the model on a diverse dataset, ensuring it can accurately identify drones in various conditions and environments. Application Potential: This system has significant implications for security surveillance, air traffic management, and wildlife protection. The integration of deep learning with computer vision has opened up exciting possibilities in drone detection and monitoring. I'm thrilled to explore further enhancements and applications in this field! 💡 Looking Forward: If you have any suggestions or insights on improving real-time object detection systems or experiences with YOLOv11, I’d love to connect and learn more! #DeepLearning #ComputerVision #YOLO #DroneDetection #AI #MachineLearning #RealTimeDetection

📃Scientific paper: UAV-enabled Collaborative Beamforming via Multi-Agent Deep Reinforcement Learning Abstract: In this paper, we investigate an unmanned aerial vehicle (UAV)-assistant air-to-ground communication system, where multiple UAVs form a UAV-enabled virtual antenna array (UVAA) to communicate with remote base stations by utilizing collaborative beamforming. To improve the work efficiency of the UVAA, we formulate a UAV-enabled collaborative beamforming multi-objective optimization problem (UCBMOP) to simultaneously maximize the transmission rate of the UVAA and minimize the energy consumption of all UAVs by optimizing the positions and excitation current weights of all UAVs. This problem is challenging because these two optimization objectives conflict with each other, and they are non-concave to the optimization variables. Moreover, the system is dynamic, and the cooperation among UAVs is complex, making traditional methods take much time to compute the optimization solution for a single task. In addition, as the task changes, the previously obtained solution will become obsolete and invalid. To handle these issues, we leverage the multi-agent deep reinforcement learning (MADRL) to address the UCBMOP. Specifically, we use the heterogeneous-agent trust region policy optimization (HATRPO) as the basic framework, and then propose an improved HATRPO algorithm, namely HATRPO-UCB, where three techniques are introduced to enhance the performance. Simulation results demonstrate that the proposed algorithm can learn a better strategy compared with other methods. Moreover, e... Continued on ES/IODE ➡️ https://etcse.fr/PU0cb ------- If you find this interesting, feel free to follow, comment and share. We need your help to enhance our visibility, so that our platform continues to serve you.

✨Highlights for this Hot Article 1. We propose a multi-UAV multicast system assisted by IRSs to improve the system throughput and reduce the impact of the NLoS link because of complicated terrestrial infrastructures. 2. We utilize the K-medoids algorithm to solve the multicast grouping problem, and the simulation results show that the multicast communication method can improve the average throughput. 3. The MADRL framework and MADDPG algorithm are employed to efficiently obtain the optimal 3D trajectory and power allocation scheme and eliminate the non-stationarity of multi-agent training, which can reduce the energy consumption of UAVs effectively compared to traditional DRL methods. 🗞️Three-Dimensional Trajectory and #Resource Allocation #Optimization in Multi-Unmanned Aerial Vehicle Multicast System: A Multi-Agent Reinforcement Learning Method 👨🏫 by Dongyu Wang et al 👉 https://lnkd.in/g-bi5f4b #UAS #3D

This article investigates the evolution of existing state-of-the-art multi-agent actor-critic frameworks based on their internal neural network structure that is employed to solve various joint optimization problems in UAV swarm networks along with their advantages and limitations. Additionally, open issues and potential solutions are also addressed. We believe this article will assist researchers in enhancing network performance in UAVSNs by employing data-driven MA-DRL-based actor-critic frameworks. Article title: Variations in Multi-Agent Actor-Critic Frameworks for Joint Optimizations in UAV Swarm Networks: Recent Evolution, Challenges, and Directions

From my Ph.D studies, I am working on designing physics informed neural networks to control the trajectory of unmanned aerial vehicles. The virtual force generated by Reynolds flocking model are fed into the neural networks to produce the acceleration of vehicles following the Newton's second law of motion (F=ma). After returning to BD, I am working to design hybrid loss function to adhere the physics laws of motion and training the neural networks. Few days ago I wrote this survey and posted at researchgate archive. The first research direction of this paper (Section III-A) addressed the potentiality to explore Physics Informed Neural Networks (PINN) for Trajectory Control of UAVs. Yesterday very glad to know that Prof. John Hopfield and Geoffrey Hinton won the nobel prize for tranining neural network using physics. It inspires me more to work on this topic.

Fetching from low light - no comms for autonomous inspection! Boston Dynamics + Emesent is a fantastic tool for BVLOS inspection and mapping for mining! #robotics #mining #lidar #hovermap

Hein Matthee, Pre-Sales Lead at Nearby, and Jorge Muñoz, Co-Founder and CCO at Unmanned Life, delve deeply into #automatedrobots 🤖 and the benefits of #NearbyOne in robot orchestration. 🚀Discover the details of this interview 👇🏻

🚀 Exciting Channel Partnership Announcement 🚀 I am thrilled to share that Exyn Technologies has welcomed a new U.S.-based partner, Keystone Precision Solutions! This partnership marks a significant milestone in expanding our channel partner program. Key Highlights: 🔹 Strategic Collaboration: Keystone Precision Solutions, a leader in providing innovative solutions for surveying and construction, joins us to enhance our cutting-edge autonomous drone technology's reach and impact. 🔹 Enhanced Capabilities: This partnership will leverage Keystone's extensive industry expertise and Exyn's state-of-the-art autonomous aerial robots to deliver unparalleled data collection and analysis capabilities to a wider range of customers. 🔹 Market Expansion: Together, we aim to revolutionize how industries such as construction and infrastructure inspection approach data gathering and operational efficiency. Stay tuned for more updates. #ExynTechnologies #KeystonePrecisionSolutions #AutonomousDrones #Innovation #Surveying #Construction #Mining #DataAnalysis #Partnership Read the full press release for more details: Exyn Technologies Welcomes First U.S.-Based Partner, Keystone Precision Solutions (https://lnkd.in/dkvXHfYx)

This letter proposes using #deep #learning (#DL) for #intelligent #reflecting #surface (#IRS) element selection to reduce the bit error rate (BER) of #unmanned #aerial #vehicle (#UAV) communications affected by imperfect phase estimation and compensation. In the presence of phase errors, increasing the number of elements does not necessarily reduce the BER. In contrast, muting certain IRS elements can improve BER. However, solving the optimization problem heuristically is extremely complex because it requires evaluating BER expressions numerically for an enormous number of cases. Consequently, a #long #shortterm #memory (#LSTM)-based #element #selection (#ES) technique is proposed to reduce the substantial complexity inherent in the conventional solution. A supervised learning approach with offline training is adopted where the decision of ES is made based on the phase estimation error parameter j. ---- Sobia Jangsher, Arafat Al-Dweik, Emad Alsusa More details can be found at this link: https://lnkd.in/ezUN9JGf

🚀 Exciting Advances in UAV Communication Networks! 🚀 I am thrilled to share my recent work on "Centralized Deep Reinforcement Learning and Optimization in UAV Communication Networks Towards Enhanced User Coverage" as part of my Master of Science thesis at Miami University. 🔍 What’s the Big Idea? Unmanned Aerial Vehicle (UAV) based communication networks represent a revolutionary approach to providing connectivity, especially in remote or challenging areas where traditional base stations fall short. My research tackles the critical challenge of optimizing UAV positions to maximize user coverage using a centralized deep reinforcement learning approach. 📈 Key Contributions: Developed a Multi-Agent Deep Q Learning (MA-DQL) Algorithm: Enabled individual UAVs to make autonomous decisions, improving user coverage in dynamic environments. Proposed a Centralized Deep Q Learning (C-DQL) Approach: Centralized control of UAVs for coordinated decision-making, overcoming the complexities of decentralized systems. Optimization Algorithm: Enhanced the centralized approach to address time delay issues, utilizing industrial optimization solvers for optimal UAV positioning. 🌐 Why It Matters? This work has significant implications for disaster management, remote area connectivity, and emergency response, providing a robust framework for deploying UAVs as temporary mobile towers in complex scenarios. Interestingly, industry leaders like AT&T and Verizon have started implementing UAV-based solutions to enhance network coverage, validating the potential of this technology. 📊 Impact: Improved network connectivity and user coverage. Reduced decision-making complexity in multi-UAV systems. Enhanced the reliability and efficiency of UAV communication networks. I am eager to bring this innovative approach to industry applications and collaborate with forward-thinking teams to push the boundaries of what's possible in UAV communication technology. 🔗 Let's Connect! If you're interested in UAVs, communication networks, or looking for collaboration opportunities, let's connect and discuss how we can drive technological advancements together! #UAV #DeepLearning #ReinforcementLearning #Optimization #TechInnovation #Engineering #AI #MachineLearning