Autonomous Interception of Aerial Targets
Project Rundown
The Interceptor is a lightweight, software-driven aerial platform built around a small fixed-wing UAV, designed for autonomous or semi-autonomous interception of aerial threats. At its core is a vision-based targeting module powered by neural networks, enabling real-time detection, classification, and continuous optical tracking of fast-moving targets – even in GPS-denied or electronically contested environments.
The system integrates a Raspberry Pi-based onboard computer and a digital camera to perform detection and guidance tasks independently of GNSS. Enhanced with deep learning and sensor fusion, the computer vision engine filters out false positives and decoys. Once a target is visually acquired, the UAV autonomously adjusts its trajectory for interception. While tracking and guidance are fully automated, final engagement remains under human authorization, providing a critical layer of control and accountability.
Currently integrated with Ardupilot, the modular software stack is adaptable to a variety of airframes with appropriate calibration. This architecture delivers a balanced, high-performance counter-drone solution suitable for dynamic defense scenarios with minimal operator input.
Technical Specifications
- Visual Target Detection
- Maneuverable Target Tracking
- Autonomous Lock-on
- Dynamic Trajectory Adjustment
- Radar-cued Initial Guidance
- Target Engagement Range: Up to 500 m
- Optical / Digital Zoom: 4× optical, 15× digital
- Sensor Resolution: 640×512 (thermal) or 4K (daylight)
- Frame Rate: ≥30 fps for smooth, real-time tracking
Implementation Overview
Vision Systems:
- 12 MP daylight camera for high-detail visual tracking
- High-resolution LWIR night vision camera for low-light environments
- Thermal camera for accurate heat-based target detection
- Neural network-based object detection (powered by YOLO architecture)
Processing Unit:
- Onboard low-powered dedicated AI processor
- Low-latency control loop for effective interception
Sensor Fusion Module:
- Combines vision and inertial sensors to enhance situational awareness
- Supports MAVLink and UART communication with flight controller
Guidance & Navigation:
- Preliminary target alignment using commands from tactical radars
- Detection of maneuvering targets, visual tracking, and terminal guidance
Communication:
- Supports both analog and digital links between the interceptor and ground control
- Additional functions controlled via standard RC transmitter
Applications
- Reconnaissance Drones: Disrupts intelligence-gathering operations by visually detecting and tracking low-observable UAVs.
- Strike Drones: Engages and intercepts weaponized drones before they reach critical targets.
- Loitering Drones: Eliminates slow-moving or hovering UAVs that evade radar and are often used for target acquisition or delayed attacks.
