This paper describes a Mission Definition System and the automated flight process it enables to implement measurement plans for discrete infrastructure inspections using aerial platforms, and specifically multi-rotor drones. The mission definition aims at improving planning efficiency with respect to state-of-the-art waypoint-based techniques, using high-level mission definition primitives and linking them with realistic flight models to simulate the inspection in advance. It also provides flight scripts and measurement plans which can be executed by commercial drones. Its user interfaces facilitate mission definition, pre-flight 3D synthetic mission visualisation and flight evaluation. Results are delivered for a set of representative infrastructure inspection flights, showing the accuracy of the flight prediction tools in actual operations using automated flight control.

The paper describes an approach developed to estimate construction productivity for concrete formwork tasks. The system utilizes artificial neural networks, historical information, and input from experienced superintendents employed by a leading construction general contractor. It also summarizes a study undertaken to determine factors that affect labor productivity, the survey conducted to collect relevant data, and the design, training, and implementation of artificial neural networks at the participating company. A number of alternative neural network structures were investigated, the adopted one was a three-layered network with a fuzzy output structure. It was found that this structure provided the most suitable model since much of the input was subjective. A brief overview of the computer implementations and the overall experience with the system development is also provided. The method was compared to an existing statistical model developed by the same contractor and was found to improve the quality...

Advancements in embedded systems and Artificial Intelligence (AI) have enhanced the capabilities of Unmanned Aircraft Vehicles (UAVs) in computer vision. However, the integration of AI techniques o-nboard drones is constrained by their processing capabilities. In this sense, this study evaluates the deployment of object detection models (YOLOv8n and YOLOv8s) on both resource-constrained edge devices and cloud environments. The objective is to carry out a comparative performance analysis using a representative real-time UAV image processing pipeline. Specifically, the NVIDIA Jetson Orin Nano, Orin NX, and Raspberry Pi 5 (RPI5) devices have been tested to measure their detection accuracy, inference speed, and energy consumption, and the effects of post-training quantization (PTQ). The results show that YOLOv8n surpasses YOLOv8s in its inference speed, achieving 52 FPS on the Jetson Orin NX and 65 fps with INT8 quantization. Conversely, the RPI5 failed to satisfy the real-time processing needs in spite of its suitability for low-energy consumption applications. An analysis of both the cloud-based and edge-based end-to-end processing times showed that increased communication latencies hindered real-time applications, revealing trade-offs between edge (low latency) and cloud processing (quick processing). Overall, these findings contribute to providing recommendations and optimization strategies for the deployment of AI models on UAVs.

This paper summarizes in depth the state of the art of aerial swarms, covering both classical and new reinforcement-learning-based approaches for their management. Then, it proposes a hybrid AI system, integrating deep reinforcement learning in a multi-agent centralized swarm architecture. The proposed system is tailored to perform surveillance of a specific area, searching and tracking ground targets, for security and law enforcement applications. The swarm is governed by a central swarm controller responsible for distributing different search and tracking tasks among the cooperating UAVs. Each UAV agent is then controlled by a collection of cooperative sub-agents, whose behaviors have been trained using different deep reinforcement learning models, tailored for the different task types proposed by the swarm controller. More specifically, proximal policy optimization (PPO) algorithms were used to train the agents’ behavior. In addition, several metrics to assess the performance of the swarm in this application were defined. The results obtained through simulation show that our system searches the operation area effectively, acquires the targets in a reasonable time, and is capable of tracking them continuously and consistently.

Drone operators, institutions (e.g. law enforcement, rescue agencies, preservation organisms, etc.) and companies (logistics, surveillance, inspection, etc.), are constituting themselves as drone operators, at the same time that increase the size of their fleets of drones. Nowadays, each operator independently manages its drones, on service request. Beyond this isolated approach, many operation scenarios would be much better addressed if automated means for coordinated resource allocation and simultaneous drone control were available. In this paper, we describe a microservice-oriented architecture that provides organization and optimization functionalities to orchestrate drone operation. The platform built on the proposed architecture may be used to enable a single operator to optimize the management of its fleet of drones, but also to facilitate coordination of different resources towards a common goal, assuming a drone-as-a-service operation mode. The architecture is validated through a specific use case, for real time allocation of drone resources to enable fast emergency response.

This paper proposes a new paradigm for the formulation of aircraft trajectory computation and prediction processes. Under this paradigm, an aircraft trajectory can be expressed at different levels of detail using a set of novel intent-related formal languages. These languages can be used for describing different aspects of aircraft motion or the resulting predicted trajectory, ranging from high-level restrictions imposed on the trajectory to details on how the aircraft will be operated at any specific instant. After defining the complete hierarchy of these formal languages, the concept of trajectory language processing engine is introduced. Every engine can perform automated modifications over a trajectory specified at different levels or at the same level considering the language hierarchy. Finally, several engine examples are described, and a complete trajectory computation process is built as an engine composed through the smart interconnection of several engines. Examples of executions of these automated processes are also included.