Modelling to take micro air vehicles to new levels
KACST researchers are developing computational technology to accurately simulate the flight dynamics of micro air vehicles with broad scientific and engineering applications
12 May 2019
Insect-sized, remote-controlled flying robots are the latest developments in the field of micro air vehicles (MAVs), popularly known as drones. Such small devices, however, face specific aerodynamic challenges, and experimental testing of different vehicle designs and constructions is impractical and costly. To make testing easier, the Center of Excellence for Aeronautics and Astronautics (CEAA) at KACST has developed software that simulates the aerodynamic performance of MAVs to inform their design.
The CEAA is a collaborative project between KACST and Stanford University in the United States, and was established to advance space and aeronautical technology. The project is managed by Charbel Farhat, CEAA’s director and professor of aircraft structures at Stanford, and is supervised by Yaser Alahmadi, assistant professor of mechanical engineering, and the centre’s co-director. The main objective of the project is to develop and validate a high-fidelity computational technology that allows accurate simulation of the aerodynamics and inflight behaviour of UAVs and MAVs.
“The Kingdom of Saudi Arabia seeks to become a regional leader in critical areas of the aeronautics sector, which is characterized by accelerating technological advancement and increasing competition,” explains Alahmadi. “Very light aircraft, UAVs and MAVs are critical components of this sector, as they are used in many applications ranging from national security and border surveillance, to atmospheric monitoring, agricultural inspection, weather prediction and surveying of natural resources.”
Most people now think of drones as recreational devices that are fun to fly or for taking aerial photographs. However, UAVs were first developed as military tools, and their recent commercial popularity is part of a wider expansion of their capabilities and increasing scientific and regulatory applications. MAVs are a specific subset of UAVs defined as being smaller than 15 centimetres in all dimensions.
“MAVs operate in the lower speed regime and tend to have lightweight, flexible, flapping wings,” explains Alahmadi. “Their unsteady and turbulent aerodynamics are closely linked to their structural dynamics, which feature large motions and deformations. Their flight characteristics are affected by environmental factors, such as wind gusts. These factors make it difficult to conduct practical experiments on these types of vehicles, and numerical simulation techniques are considered to be a more reliable alternative.”
Alahmadi and colleagues aim to model the dynamics of UAVs and MAVs to determine how their flight performance is affected by complex environments like cities, and the structural properties of the vehicles, such as wing flexibility.
The starting point for this simulation technology is a method for computational fluid dynamics called FIVER, recently developed by Farahat and his colleagues. This technique will be expanded upon, and combined with computational models of wing structures and algorithms that simulate environmental situations that UAVs must be able to withstand, such as gusts of wind. To ensure maximum accuracy, the computational technology needs to incorporate appropriate models of turbulence and models of the materials that make up the surfaces of MAVs, such as Mylar, Capran-coated nylon films and carbon-fibre-reinforced polymers. The final technology must combine all of this, and simulate all possible interactions between the materials and the dynamic air around them. Processing power is essential, and super computers will be needed to run the software.
“Development of a computational technology for physics-based, multi-disciplinary analysis of the dynamic behavior of UAVs and MAVs in complex environments is a formidable challenge,” says Alahmadi. He adds that it is challenging to find a method that takes into account the fluid-structure interaction (the movable or deformable structure and the internal or surrounding fluid flow) when analyzing the dynamics of UAVs and MAVs. “In general, these problems are too complex to be solved analytically, so they have to be analyzed by experimental testing of numerical simulation,” he adds.
The second objective of the project is the application of the computational technology to gain new insights into UAVs and MAVs, and the ways in which the performance of these vehicles is affected by their structural flexibility and atmospheric turbulence. Design configurations and environmental scenarios that are difficult to test experimentally will be simulated to inform design of the vehicles and improve flight dynamics. In addition, the researchers will explore the possibility of the vehicles harnessing energy from the turbulence they encounter, improving their range and efficiency.
Although the project has a specific focus on MAVs, the principles to which the computational technology will apply are important in other engineering systems and beyond, so the final product will have broader applications.
“The fluid–structure interaction is a crucial consideration for modern aircraft, high-performance cars, underwater systems and high-rise buildings, to name only a few,” says Alahmadi. “The computational numerical methods developed here are also relevant to biomedical applications, such as blood flow in arteries and veins, blood coagulation and speech modelling.”
Other work at the CEAA also focuses on UAV design testing. The aim of the other project, titled “Autonomous UAVs using machine learning for system identification and control,” is to approximate aerodynamic forces through experimental testing and numerical techniques to inform the development of entirely autonomous UAVs. The techniques being used in the two projects differ, but they complement one another in their intended applications of improving UAV design and efficiency.
The project on MAVs has been running since the CEAA was formed in 2014, and is set to continue until the end of 2020.