Aug 31, 2023 · In this paper we propose D-VAT, a novel end-to-end visual active tracking methodology based on deep reinforcement learning that is tailored to micro aerial ...

In this work, we proposed D-VAT, an end-to-end visual active tracking approach for MAV systems. The D-VAT agent is trained by exploiting an asymmetric actor- ...

Apr 18, 2024 · The D-VAT agent computes the vehicle thrust and angular velocity commands needed to track the target by directly processing monocular camera ...

Apr 8, 2024 · This paper proposes a novel deep reinforcement learning approach called D-VAT for visual active tracking on micro aerial vehicles. XShare on X.

Oct 20, 2023 · The D-VAT system computes the vehicle thrust and angular velocity commands directly from data captured by a single monocular camera. We show ...

D-VAT is proposed, a novel end-to-end visual active tracking methodology based on deep reinforcement learning that is tailored to micro aerial vehicle ...

D-VAT: End-to-End Visual Active Tracking for Micro Aerial Vehicles. A Dionigi, S Felicioni, M Leomanni, G Costante. IEEE Robotics and Automation Letters, 2024.

D-VAT: End-to-End Visual Active Tracking for Micro Aerial Vehicles. IEEE Robotics Autom. Lett. 9(6): 5046-5053 (2024). [j7]. view. electronic edition via DOI ...

This paper studies active object tracking, where a tracker receives an input visual observation and directly outputs the most appropriate control actions.

This paper presents a solution for cognitive visual tracking. The main novelty is a high-level feedback control loop for a set of PTZ cameras.