Autonomous Tracking and Terminal Guidance of Moving Targets for Fixed-Wing UAVs

2026-07-14 · Source: Artificial Intelligence · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning · Depth: Expert, quick

Summary

A unified control framework for fixed-wing unmanned aerial vehicles (UAVs) fitted with a pan-tilt (PT) camera is introduced, designed for end-to-end missions from target detection to accurate terminal engagement. The system employs a three-phase strategy: vision-based target acquisition, NMPC-based tracking, and terminal guidance. During tracking, an Unscented Kalman Filter (UKF) fuses YOLO-based visual detections with inertial measurements for robust target state estimation. A constraint-aware Nonlinear Model Predictive Control (NMPC) strategy, incorporating Control Barrier Functions (CBFs), explicitly prevents UAV self-occlusion. Upon meeting terminal engagement conditions, control transitions to a quaternion-based Biased Proportional Navigation Guidance (BPNG) law, enforcing precise impact angle constraints. High-fidelity simulations demonstrate stable, robust tracking and accurate terminal interception while respecting vehicle dynamic limits and camera field-of-view constraints.

Key takeaway

For Robotics Engineers designing autonomous fixed-wing UAV systems for moving target engagement, consider integrating this unified control framework. Its three-phase strategy, combining vision-based acquisition, NMPC-based tracking with self-occlusion prevention via CBFs, and quaternion-based Biased Proportional Navigation Guidance, ensures stable tracking and precise terminal interception. This approach can enhance mission reliability and accuracy while respecting critical flight constraints.

Key insights

A unified control framework enables fixed-wing UAVs to autonomously track and intercept moving targets using vision, NMPC, and BPNG.

Principles

• Fuse visual detections with inertial measurements for robust state estimation.
• Use CBFs with NMPC to prevent UAV self-occlusion during tracking.
• Employ quaternion-based BPNG for precise impact angle control.

Method

The system transitions from vision-based acquisition to NMPC-based tracking using UKF and CBFs, then to quaternion-based BPNG for terminal guidance with impact angle constraints.

In practice

• Integrate YOLO with UKF for real-time target state estimation.
• Implement NMPC with CBFs for constrained UAV tracking.
• Apply BPNG for accurate terminal interception with impact control.

Topics

• Fixed-Wing UAVs
• Target Tracking
• Nonlinear Model Predictive Control
• Control Barrier Functions
• Proportional Navigation Guidance
• Unscented Kalman Filter

Best for: Research Scientist, Robotics Engineer, AI Scientist, Machine Learning Engineer

Related on AIssential

• See-and-Reach: Precise Vision-Language Navigation for UAVs within the Field of View — Isolating target-visible navigation and using dynamic 3D cues improves UAV precision reaching.
• End-to-End Learning of Safe Optimal Feedback Control in High Dimensions with Control Barrier Function Layers — Scalable end-to-end learning of safe high-dimensional controllers is achieved by integrating CBFs with JFB and operator splitting.
• Edge-Aware Thermal Infrared UAV Swarm Tracking — An Adaptive Kinematic Kalman Filter (AKKF) pipeline enables efficient, robust thermal infrared UAV swarm tracking on edge devices.
• Isaac Sim-to-Real: Reinforcement Learning based Locomotion for Quadrupeds — The framework uses Isaac Sim/Lab for zero-shot sim-to-real quadruped locomotion, matching integrated controllers and improving disturbance recovery.
• Motion Planning of Cooperative Nonholonomic Mobile Manipulators — A two-step motion planning approach enables nonholonomic MMRs to cooperatively transport objects amidst complex obstacles.
• From Distances to Trajectories: Real-Time Signed Distance Function Mapping and Distance-Accelerated Motion Planning for UAVs — Co-designing mapping and planning with Signed Distance Functions (SDFs) enhances autonomous UAV navigation.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.