The UAV industry is developing rapidly and drones are increasingly used for monitoring industrial facilities. When designing such systems, operating companies have to find a system configuration of multiple drones that is near-optimal in terms of cost while achieving the required monitoring quality. Stochastic influences such as failures and maintenance have to be taken into account. Model-based systems engineering supplies tools and methods to solve such problems. This paper presents a method to model and evaluate such UAV systems with coloured Petri nets. It supports a modular view on typical setup elements and different types of UAVs and is based on UAV application standards. The model can be easily adapted to the most popular flight tasks and allows for estimating the monitoring frequency and determining the most appropriate grouping and configuration of UAVs, monitoring schemes, air time and maintenance periods. An important advantage is the ability to consider drone maintenance processes. Thus, the methodology will be useful in the conceptual design phase of UAVs, in monitoring planning, and in the selection of UAVs for specific monitoring tasks.
9 October 2024 in Aerospace Science and Technology
Modern vehicles are characterized by a high degree of environmental perception and automation. This is made possible by the development of new motion control and driver assistance systems as well as new methods for sensor-based detection of driving parameters. However, the functionality of commonly used visual and wave-based sensor systems, such as cameras, radars, etc., can be limited when installed in a vehicle as some obstacles fall into the blind zone due to the road profile or other road users. It is also difficult to detect road-specific parameters such as gradient, slope, or surface conditions with onboard vehicle sensors. A method to improve the vehicle’s detection capabilities can be proposed by extending the information channels with additional sensors by mounting them on an assistant drone. This article presents a framework for drone-to-vehicle collaboration that enables the integration of data from drone sensors into the vehicle motion control. The article gives an overview of the system, the technology created for the implementation, and the challenges to be overcome by integrating the different components. The hardware implementation of the drone and the integration of the data into the vehicle control unit (VCU) are presented. The proposed approach allows the vehicle to view the road from different angles and detect changes in road conditions from a distance. It extends the capabilities of the vehicle motion control. The effectiveness of the developed framework is experimentally validated with an electric vehicle’s velocity control and traction control on low-friction surfaces.