Active Classification of Moving Targets With Learned Control Policies

Álvaro Serra-Gómez*, Eduardo Montijano, Wendelin Böhmer, Javier Alonso-Mora

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

1 Downloads (Pure)


In this paper, we consider the problem where a drone has to collect semantic information to classify multiple moving targets. In particular, we address the challenge of computing control inputs that move the drone to informative viewpoints, position and orientation, when the information is extracted using a “black-box” classifier, e.g., a deep learning neural network. These algorithms typically lack of analytical relationships between the viewpoints and their associated outputs, preventing their use in information-gathering schemes. To fill this gap, we propose a novel attention-based architecture, trained via Reinforcement Learning (RL), that outputs the next viewpoint for the drone favoring the acquisition of evidence from as many unclassified targets as possible while reasoning about their movement, orientation, and occlusions. Then, we use a low-level MPC controller to move the drone to the desired viewpoint taking into account its actual dynamics. We show that our approach not only outperforms a variety of baselines but also generalizes to scenarios unseen during training. Additionally, we show that the network scales to large numbers of targets and generalizes well to different movement dynamics of the targets.
Original languageEnglish
Pages (from-to)3717-3724
JournalIEEE Robotics and Automation Letters
Issue number6
Publication statusPublished - 2023

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.


  • Machine learning for robot control
  • reactive and sensor-based planning
  • surveillance robotic systems


Dive into the research topics of 'Active Classification of Moving Targets With Learned Control Policies'. Together they form a unique fingerprint.

Cite this