Diego Mercado-Ravell

Hector Tovanche-Picon, Javier Gonzalez-Trejo, Angel Flores-Abad et al.

Safe autonomous landing for Unmanned Aerial Vehicles (UAVs) in populated areas is a crucial aspect for successful urban deployment, particularly in emergency landing situations. Nonetheless, validating autonomous landing in real scenarios is a challenging task involving a high risk of injuring people. In this work, we propose a framework for real-time safe and thorough evaluation of vision-based autonomous landing in populated scenarios, using photo-realistic virtual environments. We propose to use the Unreal graphics engine coupled with the AirSim plugin for drone's simulation, and evaluate autonomous landing strategies based on visual detection of Safe Landing Zones (SLZ) in populated scenarios. Then, we study two different criteria for selecting the "best" SLZ, and evaluate them during autonomous landing of a virtual drone in different scenarios and conditions, under different distributions of people in urban scenes, including moving people. We evaluate different metrics to quantify the performance of the landing strategies, establishing a baseline for comparison with future works in this challenging task, and analyze them through an important number of randomized iterations. The study suggests that the use of the autonomous landing algorithms considerably helps to prevent accidents involving humans, which may allow to unleash the full potential of drones in urban environments near to people.

Julio de la Torre-Vanegas, Miguel Soriano-Garcia, Israel Becerra et al.

Landing safely in crowded urban environments remains an essential yet challenging endeavor for Unmanned Aerial Vehicles (UAVs), especially in emergency situations. In this work, we propose a risk-aware approach that harnesses semantic segmentation to continuously evaluate potential hazards in the drone's field of view. By using a specialized deep neural network to assign pixel-level risk values and applying an algorithm based on risk maps, our method adaptively identifies a stable Safe Landing Zone (SLZ) despite moving critical obstacles such as vehicles, people, etc., and other visual challenges like shifting illumination. A control system then guides the UAV toward this low-risk region, employing altitude-dependent safety thresholds and temporal landing point stabilization to ensure robust descent trajectories. Experimental validation in diverse urban environments demonstrates the effectiveness of our approach, achieving over 90% landing success rates in very challenging real scenarios, showing significant improvements in various risk metrics. Our findings suggest that risk-oriented vision methods can effectively help reduce the risk of accidents in emergency landing situations, particularly in complex, unstructured, urban scenarios, densely populated with moving risky obstacles, while potentiating the true capabilities of UAVs in complex urban operations.

Miguel S. Soriano-García, Diego A. Mercado-Ravell

ViVa-SAFELAND is an open source software library, aimed to test and evaluate vision-based navigation strategies for aerial vehicles, with special interest in autonomous landing, while complying with legal regulations and people's safety. It consists of a collection of high definition aerial videos, focusing on real unstructured urban scenarios, recording moving obstacles of interest, such as cars and people. Then, an Emulated Aerial Vehicle (EAV) with a virtual moving camera is implemented in order to ``navigate" inside the video, according to high-order commands. ViVa-SAFELAND provides a new, safe, simple and fair comparison baseline to evaluate and compare different visual navigation solutions under the same conditions, and to randomize variables along several trials. It also facilitates the development of autonomous landing and navigation strategies, as well as the generation of image datasets for different training tasks. Moreover, it is useful for training either human of autonomous pilots using deep learning. The effectiveness of the framework for validating vision algorithms is demonstrated through two case studies, detection of moving objects and risk assessment segmentation. To our knowledge, this is the first safe validation framework of its kind, to test and compare visual navigation solution for aerial vehicles, which is a crucial aspect for urban deployment in complex real scenarios.

Javier González-Trejo, Diego Mercado-Ravell, Israel Becerra et al.

Autonomous landing of Unmanned Aerial Vehicles (UAVs) in crowded scenarios is crucial for successful deployment of UAVs in populated areas, particularly in emergency landing situations where the highest priority is to avoid hurting people. In this work, a new visual-based algorithm for identifying Safe Landing Zones (SLZ) in crowded scenarios is proposed, considering a camera mounted on an UAV, where the people in the scene move with unknown dynamics. To do so, a density map is generated for each image frame using a Deep Neural Network, from where a binary occupancy map is obtained aiming to overestimate the people's location for security reasons. Then, the occupancy map is projected to the head's plane, and the SLZ candidates are obtained as circular regions in the head's plane with a minimum security radius. Finally, to keep track of the SLZ candidates, a multiple instance tracking algorithm is implemented using Kalman Filters along with the Hungarian algorithm for data association. Several scenarios were studied to prove the validity of the proposed strategy, including public datasets and real uncontrolled scenarios with people moving in public squares, taken from an UAV in flight. The study showed promising results in the search of preventing the UAV from hurting people during emergency landing.

Javier Antonio Gonzalez-Trejo, Diego Alberto Mercado-Ravell

In the context of crowd counting, most of the works have focused on improving the accuracy without regard to the performance leading to algorithms that are not suitable for embedded applications. In this paper, we propose a lightweight convolutional neural network architecture to perform crowd detection and counting using fewer computer resources without a significant loss on count accuracy. The architecture was trained using the Bayes loss function to further improve its accuracy and then pruned to further reduce the computational resources used. The proposed architecture was tested over the USF-QNRF achieving a competitive Mean Average Error of 154.07 and a superior Mean Square Error of 241.77 while maintaining a competitive number of parameters of 0.067 Million. The obtained results suggest that the Bayes loss can be used with other architectures to further improve them and also the last convolutional layer provides no significant information and even encourage over-fitting at training.