Collaborative Simultaneous Localization and Mapping (CSLAM) is critical to enable multiple robots operate in complex environments.Most CSLAM techniques rely on raw sensor measurement or low-level features such as keyframe descriptors, which can lead wrong loop closures due the lack of deep understanding environment.Moreover, exchange these measurements among requires transmission a significant amount data, limits scalability system.To overcome limitations, we present Multi S-Graphs,...

In recent years, the robotics community has witnessed development of several software stacks for ground and articulated robots, such as Navigation2 MoveIt. However, same level collaboration standardization is yet to be achieved in field aerial robotics, where each research group developed their own frameworks. This work presents Aerostack2, a framework autonomous systems that aims address lack fragmentation efforts field. Built on ROS 2 middleware featuring an efficient modular architecture...

Robotic missions for solar farm inspection demand agile and precise object detection strategies. This paper introduces an innovative keypoint-based framework specifically designed real-time inspections with UAVs. Moving away from conventional bounding box or segmentation methods, our technique focuses on detecting the vertices of panels, which provides a richer granularity than traditional approaches. Drawing inspiration CenterNet, architecture is optimized embedded platforms like NVIDIA AGX...

Abstract In recent years, the use of unmanned aerial vehicles has spread across different fields industry due to their ease deployment and minimal operational risk. Firefighting is a dangerous task for humans involved, in which UAVs presents itself as good first-action protocol rapid response an incipient fire because safety speed action. Current research mainly focused on wildland fires, but fires urban environments are barely mentioned bibliography. To motivate this topic, ICUAS’22...

This paper presents a novel approach for accurate counting and localization of tropical plants in aerial images that is able to work new visual domains which the available data not labeled. Our uses deep learning domain adaptation, designed handle shift between training test data, common challenge this agricultural applications. method source dataset with annotated target without annotations, adapts model trained on using unsupervised alignment pseudolabeling. The experimental results show...

The utilization of autonomous unmanned aerial vehicles (UAVs) has increased rapidly due to their ability perform a variety tasks, including industrial inspection. Conducting testing with actual flights within facilities proves be both expensive and hazardous, posing risks the system, facilities, personnel. This paper presents an innovative reliable methodology for developing such applications, ensuring safety efficiency throughout process. It involves staged transition from simulation...

This paper presents a novel semi-supervised approach for accurate counting and localization of tropical plants in aerial images that can work new visual domains which the available data are not labeled. Our uses deep learning domain adaptation, designed to handle shifts between training test data, is common challenge this agricultural applications. method source dataset with annotated target without annotations adapts model trained on using unsupervised alignment pseudolabeling. The...

This work presents a cost-effective implementation method of photovoltaic (PV) panel digital twin (DT) on low-cost device. The proposed uses non-iterative and explicit parameter extraction the Single Diode Model (SDM) based Lambert W function utilizes an function-based to calculate I-V P-V characteristics curves PV panel. accuracy performance results have been tested, showing low errors in comparison with experimental data obtained by Internet Things (IoT) device that measures electrical...

Collaborative Simultaneous Localization and Mapping (CSLAM) is critical to enable multiple robots operate in complex environments. Most CSLAM techniques rely on raw sensor measurement or low-level features such as keyframe descriptors, which can lead wrong loop closures due the lack of deep understanding environment. Moreover, exchange these measurements among requires transmission a significant amount data, limits scalability system. To overcome limitations, we present Multi S-Graphs,...

The development of collective-aware multi-robot systems is crucial for enhancing the efficiency and robustness robotic applications in multiple fields. These enable collaboration, coordination, resource sharing among robots, leading to improved scalability, adaptability dynamic environments, increased overall system robustness. In this work, we want provide a brief overview research topic identify open challenges.

Agile autonomous drones are becoming increasingly popular in research due to the challenges they represent fields like control, state estimation, or perception at high speeds. When all algorithms computed onboard UAV, computational limitations make task of agile flight even more difficult. One most computationally expensive tasks is generation optimal trajectories. these trajectories must be updated online changes environment uncertainties, this cost may result insufficient time reach...

Recent advances have improved autonomous navigation and mapping under payload constraints, but current multi-robot inspection algorithms are unsuitable for nano-drones due to their need heavy sensors high computational resources. To address these challenges, we introduce ExploreBug, a novel hybrid frontier range bug algorithm designed handle limited sensing capabilities swarm of nano-drones. This system includes three primary components: subsystem, an exploration subsystem. Additionally,...

In recent years, autonomous drone races have become increasingly popular in the aerial robotics research community due to challenges perception, localization, navigation, and control at high speeds, pushing forward state-of-the-art every year. However, racing drones are still far from reaching human pilot performance, a lot of has be done accomplish that. this work, complete architecture system an evaluation method for proposed, based on Aerostack 5.0 open-source framework. To evaluate...

Agile autonomous drones are becoming increasingly popular in research due to the challenges they represent fields like control, state estimation, or perception at high speeds. When all algorithms computed onboard uav, computational limitations make task of agile and robust flight even more difficult. One most computationally expensive tasks is generation optimal trajectories that tackles problem planning a minimum time trajectory for quadrotor over sequence specified waypoints. these must be...