IEEE Case 2025 

Daniel Fuertes, researcher at the Grupo de Tratamiento de Imágenes (GTI), presented his latest work at the prestigious IEEE International Conference on Automation Science and Engineering (CASE), held in Los Angeles, California.

His research, entitled “Multi-stage Planning for Multi-target Surveillance using Aircrafts Equipped with Synthetic Aperture Radars Aware of Target Visibility”, addresses one of the key challenges in modern aerial surveillance: how to generate efficient flight trajectories for aircraft equipped with Synthetic Aperture Radars (SAR). These systems must simultaneously ensure high-quality imaging, respect terrain constraints, and maximize visibility across multiple targets.

To tackle this, Daniel designed an innovative multi-stage planning system. The method first determines the optimal sequence of waypoints to cover all targets. It then leverages a novel deep reinforcement learning model to predict straight-flight segments that maximize target visibility in relation to the 3D terrain and aircraft orientation. Finally, these segments are connected through an optimization framework based on 3D Dubins curves, yielding trajectories that are both feasible and efficient.

This integrated approach provides robust, real-time solutions for multi-target SAR missions, demonstrating significant improvements in adaptability and imaging quality compared to traditional trajectory planning methods.