Drone-to-Drone Detection, Tracking & Follow System on Raspberry Pi

@sajjadtaghvaeifr

Hello, I trust you're doing well. I am well experienced in machine learning algorithms, with nearly a decade of hands-on practice. My expertise lies in developing various artificial intelligence algorithms, including the one you require, using Matlab, Python, and similar tools. I hold a doctorate from Tohoku University and have a number of publications in the same subject. My portfolio, which showcases my past work, is available for your review. Your project piqued my interest, and I would be delighted to be part of it. Let's connect to discuss in detail. Warm regards. please check my portfolio link: https://www.freelancer.com/u/sajjadtaghvaeifr

@shadabkhan92

As an experienced AI engineer with a penchant for edge systems, I have successfully integrated machine learning models into real-life scenarios on several projects much like yours. My proficiency in Python and OpenCV is complemented by a solid understanding of YOLO/Object Detection that will be invaluable for detecting and tracking drones through the camera feed, ensuring smooth, autonomous follow behavior as desired. Expertise in Raspberry Pi, including deployment of edge AI models using specialized accelerators, will bring out the best in your project by optimizing performance for low latency and high FPS on the platform. My comfort with drone systems, navigation, and project management adds to my arsenal of skills making me an ideal fit for this task. Having completed similar projects in the past, notably those that involved drone vision and autonomous tracking systems, think of me as the missing piece to make your drone-to-drone vision system complete. Let's combine your Raspberry Pi 5 and 40 TOPS AI Accelerator with my experience and build an efficient solution for your needs!

@amirrmusavifr

Hi, I’m an AI expert with professional experience in computer vision, with a proven track record of working on complex image processing and AI/ML model development. With skill sets: • Algorithm Development: Strong understanding of computer vision algorithms and techniques, including convolutional neural networks (CNNs), object detection, image segmentation and feature extraction. • Model Training & fine-tuning: Develop and train machine learning models tailored for image analysis and visual data interpretation. I have worked on some well-known models like YOLO, RCNN, U-Net, Deeplab, ViT etc. • AI Integration: Implement and integrate AI models into existing software and hardware systems, ensuring high performance and scalability. • Data Analysis: Analyze and process large datasets of images and video feeds to identify patterns, trends, and insights. • Data Handling: Experience in handling and processing large datasets, including image and video data. Familiarity with data augmentation techniques and synthetic data generation. • Performance Optimization: Optimize algorithms and models for real-time processing and ensure they can handle large-scale data efficiently. • Programming Skills: Proficient in programming languages such as Python. Experience with deep learning frameworks like TensorFlow, PyTorch, or Keras. • Tools & Libraries: Proficiency with OpenCV, scikit-image, and other relevant libraries. Experience with version control systems like Git.

@quan9824

https://www.freelancer.com/projects/raspberry-pi/Powered-Monitoring-Prototype-Development/reviews ## EXPERT ##(Python and Raspberry PI, Machine learning) Hello, How are you? I’ve completed several raspberry pi projects before successfully. Recently, I developed Smart Dashcam using Raspberry PI and USB Dongle in Netherland. You can check this in my portfolio. I can upload my previous works too.. I am sure and I can start immediately. Awarding me will be the fastest way to complete your task with the best rates possible. THANK YOU.

@vishal1145

Your YOLO model will fail at 30+ FPS if you don't implement frame skipping and async inference on the AI accelerator. Most developers run detection on every frame, which bottlenecks the pipeline and causes tracking lag during fast drone movements. Before architecting the solution, I need clarity on two things: What's your target detection range (50m vs 200m changes camera resolution requirements), and are you using GPS fusion or pure vision-based tracking? This affects whether we need Kalman filtering or just optical flow. Here's the architectural approach: - YOLO + HAILO AI ACCELERATOR: Deploy YOLOv8-nano with INT8 quantization to hit 60 FPS at 640x480 resolution, using the 40 TOPS accelerator's NPU instead of CPU inference. - BYTETRACK + OPENCV: Implement multi-object tracking with motion prediction to maintain lock during occlusions or when the target exits frame temporarily. - RASPBERRY PI 5 OPTIMIZATION: Use hardware-accelerated H.264 encoding for the camera feed and DMA buffers to reduce memory copy overhead between camera and AI pipeline. - PID CONTROL LOOP: Generate pan/tilt commands based on target centroid offset, with velocity compensation to prevent oscillation during aggressive maneuvers. - EDGE CASE HANDLING: Add confidence thresholding and re-acquisition logic to prevent false positives from birds or debris triggering follow behavior. I've built 3 real-time vision systems for autonomous drones, including a 2-drone formation flight project that maintained sub-100ms tracking latency at 45 FPS. Let's schedule a 15-minute technical call to discuss your camera specs and flight envelope before I scope the full implementation.

@mobasherreza10

With a robust background in Computer Vision, Deep Learning, and Machine Learning (ML), I am perfectly positioned to elevate your drone-to-drone detection and tracking system using the Raspberry Pi 5 and an AI Accelerator. My extensive experience includes developing real-time AI models deployed on devices with low latency requirements, such as the Raspberry Pi. I am particularly adept at working with object detection and tracking algorithms like YOLO and ByteTrack. What sets me apart is my familiarity with drone systems and navigation alongside my Python and OpenCV expertise. To maximize the efficiency of Preprocessing camera images for your AI inference, I can design a tailored pipeline that ensures optimal object detection and adapts smoothly to the guidance commands generation for autonomous follow behavior. Furthermore, my skills in drone vision projects and actual autonomous tracking systems also favor a comprehensive understanding of the challenges this project poses and equipped me to develop effective solutions.

@KrisuTechkul

Hi, Krishna here from Delhi. We are a team of 20+ Engineers and have completed 300+ projects with 4.7+ rating. Would love to connect with you to discuss the project. Our demonstrated expertise in Python, OpenCV, YOLO/Object Detection, Raspberry Pi, Edge AI Deployment undeniably aligns with what you're seeking. In fact, effective data preprocessing for AI inference and Low-latency & High FPS optimization on Raspberry Pi remain core elements of our domain proficiency. As an AI specialist with a deep-rooted passion for leveraging cutting-edge strategies for real-world challenges, I believe my team's experience in drone systems and navigation will prove invaluable for your project. In addition to that, we have a significant understanding of autonomous tracking systems, real-time AI inference on embedded devices as well as Raspberry Pi AI accelerator deployment - all key necessities for the success of your venture. What sets us apart is not only our technical prowess but also our commitment to repurposing innovative technologies to address complex challenges effectively. Our work often has tangible effects - improving customer experiences through conversational AI, strengthening cybersecurity using fraud detection systems, automating quality control operations with computer vision; the list goes on. Let us step into this project together and redefine innovation in real-time object tracking through drone-to-drone detection.

@Vsion2

Hi, I'm Alema, Python developer with 10 years of experience. My main passion and specialization is working with data extraction. I have worked with many sites and wrote a lot of bots which can extract data in any required format, like Excel, CSV, Json or save data to a DB. You can see completed projects: https://www.freelancer.com/projects/excel/Looker-Studio-Performance-Dashboard/proposals https://www.freelancer.com/projects/french-translator/ation-Agent-Polyvalent/reviews https://www.freelancer.com/projects/ubuntu/Ubuntu-Price-Tracking-Suite-Deployment/proposals https://www.freelancer.com/projects/documentation/Python-Security-Audit-Report/proposals https://www.freelancer.com/projects/beautifulsoup/Data-from-https-www-gsmarena/proposals If you're looking for a reliable Python backend developer to help with your project, feel free to reach out. Your faithfully. Eng. Alema Akter

@Visharaa2024

Hello, Your project combines computer vision, edge AI, and autonomous navigation an exciting challenge that aligns well with my experience in AI-powered vision systems and real-time inference pipelines. I can help develop and optimize a complete drone-to-drone tracking solution using Raspberry Pi 5 and the 40 TOPS AI Accelerator, including object detection, real-time tracking, target localization, and guidance command generation for autonomous follow behavior. Using Python, OpenCV, YOLO-based detection models, and advanced tracking algorithms such as DeepSORT, I will build an optimized pipeline capable of detecting and tracking target drones with low latency and high frame rates. The solution will include image preprocessing, model training and fine-tuning, inference optimization, target position estimation, and integration with guidance logic to keep the target centered within the camera view. Special focus will be placed on maximizing performance on embedded hardware through model quantization, accelerator optimization, and efficient resource utilization. My goal is to deliver a robust, scalable, and real-time tracking system suitable for deployment in dynamic environments while providing clear documentation, testing support, and development guidance throughout the project. I would be happy to discuss your hardware setup, target specifications, and performance requirements to recommend the most effective architecture for achieving reliable autonomous drone tracking.

@Retrytechteam

This is exactly the kind of edge-AI and real-time vision challenge I enjoy working on. Your stack—Raspberry Pi 5 + 40 TOPS AI accelerator—is more than capable of running a high-performance drone tracking pipeline when the detection, tracking, and inference stages are properly optimized. I can help build the complete workflow: camera preprocessing, YOLO-based drone detection, multi-frame tracking (ByteTrack/DeepSORT), target centering logic, and guidance command generation for autonomous follow behavior. My approach focuses on keeping latency low while maintaining stable tracking. That includes model training and fine-tuning on drone-specific datasets, inference optimization for embedded deployment, frame pipeline optimization with OpenCV, and hardware acceleration tuning to maximize FPS on the Raspberry Pi platform. I’m comfortable working across: • Python & OpenCV • YOLO model training and deployment • Real-time object tracking • Edge AI optimization • Raspberry Pi AI accelerators • Guidance and navigation logic • Performance profiling and bottleneck elimination Beyond simply detecting a drone, the goal is reliable target lock, smooth tracking, and consistent command generation under real-world conditions such as varying lighting, distance, motion blur, and background clutter. I can also assist with testing methodology, dataset preparation, model benchmarking, and deployment validation to ensure the system performs reliably outside the lab. .

@NEHABHAT92

Being an established web and mobile app development company, my team and I have a deep understanding of the required skills for this project, including Python, OpenCV and Raspberry Pi. We thrive on turning ideas into reality, and your project aligns perfectly with our mission. With 9+ years of experience under our belt, we have successfully delivered complex projects requiring similar qualities as yours: object detection, tracking, and AI inference in real-time. Additionally, our familiarity with edge AI deployment will enable us to optimize performance without compromising latency or FPS. Furthermore, while we may not have direct drone-specific experience in our portfolio presently, our expertise in mobile apps includes work with GPS-based navigation systems. We can easily apply that knowledge to your drone tracking system wherein we train and optimize object detection models to guide and keep the target centered. For this assignment, you require not just technical skills but also an appetite for innovative thinking- both of which my team possess. Trust us to deliver what you need with excellence, creativity, and cost-efficiency!

@sgwtechnology

Hi — Mukesh here. Your project is very aligned with my experience in real-time computer vision systems, especially on edge devices where latency, tracking stability, and inference optimization are critical. I have worked on Python-based vision pipelines using OpenCV and YOLO for object detection and tracking systems, including optimization for low-power hardware setups. My focus in such systems is maintaining stable FPS, reducing inference lag, and ensuring reliable object persistence during motion and partial occlusion. For your drone-to-drone vision system, I would structure the solution around a lightweight detection model (YOLO variant optimized for edge), combined with a real-time tracking algorithm such as ByteTrack or DeepSORT, and a control layer that translates visual offsets into smooth guidance commands for autonomous follow behavior. I also have experience in preprocessing camera streams for inference efficiency and tuning models for Raspberry Pi-class hardware with accelerators. A few questions to better understand your setup: Q1 – Which AI accelerator are you using on the Raspberry Pi 5 (Coral, Hailo, or other)? Q2 – What camera module and resolution/FPS are you targeting? Q3 – Do you already have labeled drone datasets, or should data collection and training be included? Looking forward to your response. Best regards, Mukesh

@igoichsanp

I see you are deploying a Raspberry Pi 5 paired with a 40 TOPS accelerator (likely the Hailo-8 NPU). I know exactly how to architect the end-to-end pipeline to utilize that specific silicon efficiently. Here is my architectural blueprint for your Drone-to-Drone system: Detection & Tracking Pipeline: I will deploy a lightweight but highly accurate model like YOLOv11 optimized specifically for your 40 TOPS accelerator. We will pair this with ByteTrack (preferred over DeepSORT for edge devices due to lower CPU overhead) to maintain a persistent ID lock on the target drone, even with camera jitter. Vision-to-Flight Translation (The Hard Part): Most computer vision developers fail here. I will write the logic to calculate the pixel error (the difference between the center of the camera frame and the target's bounding box). I will then implement a Custom PID Controller script to translate that X/Y error into smooth yaw, pitch, and altitude MAVLink velocity commands (using DroneKit or MAVSDK) to keep the target dead center. Edge Optimization: I will handle the C++ / Python wrappers to ensure the video stream is preprocessed directly in memory with minimal latency, securing the high FPS required for high-speed tracking.

@serviceaxisgroup

Hi, I can help with a Raspberry Pi 5 drone-to-drone vision system that detects, tracks, and autonomously follows another drone. I’ll start by scoping your camera stream, AI accelerator setup (40 TOPS), and defining latency/FPS targets, then build a minimal detection+tracking loop with OpenCV and YOLO. To reduce risk, we’ll validate inference speed on the Pi early, use a known tracker (ByteTrack/DeepSORT), and tune preprocessing plus guidance outputs in small checkpoints. Which camera module and video resolution/FPS are you using? Do you want guidance as pixel-center offsets or converted flight/control commands? If you share these, we can plan the first prototype and timeline right away.

@shubhankar1967

Hi, I am an IITian with 10+ years of experience to develop similar projects ,I will implement YOLOv8-nano fine-tuned on drone imagery with TensorFlow Lite or ONNX Runtime optimized for the Hailo-8 accelerator (40 TOPS), paired with BoT-SORT for multi-object tracking at 30+ FPS. Camera feed preprocessing via OpenCV with GStreamer hardware encoding, MAVLink or UART serial commands for gimbal/flight control, deployed in Python 3.11 with threading or asyncio for parallel inference and telemetry loops on Raspberry Pi OS. Kindly click on the chat button so I can share you my relavant projects. Lets connect

@Geetam00

As a seasoned developer with a significant background in machine learning and computer vision, coupled with my proficiency in object detection, tracking, and OpenCV, I am uniquely equipped to tackle your challenging drone project. What sets me apart is not just my expertise but a consistent track record of innovation and problem-solving that has seen me deliver successful projects worldwide. My work in real-time AI inference on embedded devices like Raspberry Pi, specifically with AI accelerators, matches the requirement for your current project and would undoubtedly provide an added advantage. Moreover, having handled drone vision projects in the past, I have the unique insight needed to optimize performance for low latency and high FPS on Raspberry Pi while maintaining stability. Clients who have worked with me praise not only my technical expertise but also my excellent communication skills – an important aspect in collaborative projects like this one. I'll bring a timely response factor of fewer than 3 hours which gives us an edge over other freelance developers based abroad. With my dedication and focus on delivery as well as added services like Performance Optimization and Cloud Deployment familiarities— I'm the perfect partner in realizing your Raspberry Pi vision.

@hitoshin

Hi, I’m excited about your drone-to-drone vision system project. I have experience developing real-time object detection and tracking systems using Python, OpenCV, YOLO, and ByteTrack/DeepSORT, and deploying them on embedded devices like Raspberry Pi with AI accelerators. I’ve worked on drone vision and autonomous tracking projects that required low-latency AI inference and high FPS performance, including preprocessing camera feeds and generating guidance commands. I can help you design, train, and optimize the models, and ensure smooth real-time tracking and autonomous follow behavior. I’m happy to share examples of similar projects I’ve completed. Looking forward to collaborating and bringing this system to life. Best regards, Hitoshi

@mitra36

As a technologist whose expertise aligns perfectly with your project requirements, I sincerely believe I am the person you’re looking for. With an advanced understanding and experience in Python, OpenCV, YOLO/Object Detection, and Raspberry Pi deployment - I have all the necessary skillsets which this project demands. Moreover, my proficiency extends towards applying AI/ML solutions on edge devices like Raspberry Pi, ensuring high-performance results. Additionally, I have a keen understanding of drone systems and navigation coupled with successful past projects in drone-vision and autonomous tracking systems. I can confidently say that I’ll provide extensive support throughout the project lifecycle - from idea inception to architecture development, deployment and necessary ongoing support. My team at Mitraa Technology and I feel strongly about maintaining clear communication channels with our clients - ensuring that we understand your vision and goals for the project perfectly. With my agile approach to work combined with my commitment to on-time delivery within budget confines, you can rely on me to turn your vision into a reality. Looking forward to working with you!

@ishtiaq49

Hi there, Your project aligns perfectly with my hands-on experience in edge AI deployment and autonomous systems. I specialize in bridging high-performance computer vision with real-time flight control. Technical Approach: AI & Edge Optimization: I will train a lightweight YOLOv8 model for drone-to-drone detection and compile it for your 40 TOPS NPU to guarantee high-FPS, low-latency inference on the Raspberry Pi 5 (8GB). I will decouple camera preprocessing using multi-threading to prevent I/O bottlenecks. Tracking: I will implement ByteTrack for robust, lightweight tracking, relying on IoU rather than computationally heavy Re-ID, saving vital edge compute. Guidance & Control: By calculating target offsets from the camera center, I will design a tuned PID controller. This will translate visual data into smooth MAVLink velocity commands via ROS/MAVROS to maintain stable autonomous follow behavior. Relevant Experience: I have extensive experience deploying real-time vision pipelines on edge devices using NCNN and customized hardware toolchains. Recently, I developed a vision-based autonomous system and am currently designing a heavy-lift autonomous drone for target detection in GPS-denied environments utilizing the exact stack of Raspberry Pi 5, YOLOv8, ROS, and MAVROS. I would love to discuss your specific AI accelerator and flight controller setup. Best regards, Ishtiaq Ahmad