Making robot airplanes
My name is Bilal Kaddouh and I am currently in the third year of my PhD at the University of Manchester. I have completed my BEng (Hons) with distinction in Electrical and Computer Engineering at the American University of Beirut in 2010, and then decided to concentrate on Robotics and Control, hence I did a MSc (Hons) in Robotics Engineering at King’s College London where I graduated with distinction in 2011. I am currently a Doctorate Candidate at the University of Manchester in the field of Aerospace Engineering. My main research area is concerned with Unmanned Aerial Vehicles (UAVs), in particular system and mission management, resources allocation, collaborative control and efficient planning.
I have worked for a year with Cummins Power Generation as a project application engineer which gave me an insight to real life work problems as well as a practical experience in applying my engineering knowledge to solve those problems. I was also responsible for delivering technical training to distributors all over Europe and the Middle East, this gave me a practical experience in teaching and conveying knowledge to students.
Through my research I aim to design a method for efficiently managing multi UAV resources in the civil airspace under temporal and dynamic constraints. In simple words, given a set of required tasks that needs to be completed within a certain time window, I am creating a system of rules which allows a group of UAVs to decide what each UAV is going to be doing at each point in time so that all the required tasks are completed in the most efficient way while the UAVs are flying in a safe condition all the time.
What is a UAV?
UAVs are airplanes without a pilot onboard. Their computational capabilities vary from simple remotely piloted airplanes to highly sophisticated autonomous flying platforms. They are essentially flying robots, and the aim of my research is to let the robots decide what to do to efficiently achieve various goals. UAVs can carry different sensors onboard, like cameras, infrared sensors, CO2 sensors, laser scanners, radars and so on. Due to current advancement in electronics UAVs possess an increasing level of computational power onboard for performing real time processing and decision making.
Why multiple UAVs?
UAVs are being used in various civilian applications such as remote sensing, aerial photography, crop health monitoring, emergency response, firefighting, atmospheric studies and many more. Many applications in the civilian world involve multiple teams working on the ground together in real time to accomplish a certain mission such as disaster management and relief, large event management security protection and crowd control, emergency services, firefighting ... A Multi User Multi UAV system is important for real time data gathering, in particular for live aerial imagery. When talking about a multi user application we are not considering single task multi users we are focusing on multi task multi users which gives users different task options to choose from.
Currently all commercial UAV operations models are built around one user flying one UAV. People are now slowly introducing UAVs into various applications for the added value it brings to any operation. Current trend of research has been focusing on moving from multiple operators managing one UAV to one operator managing many UAVs and therefore we find contributions in the operator situational awareness systems, in task allocation systems and in real time data processing. We will probably get to a point where UAVs are allowed to fly autonomous missions under certain rules and regulations enforced by the appropriate aviation authority. When we get to that stage, systems allowing one user to control multiple UAVs would be desirable.
What is the problem?
As a UAV operator, there are a lot of decisions that need to be made in terms of what sensors to install and how to plan and execute the required mission safely and efficiently. The problem gets complicated when multiple versatile UAVs are to be used especially when deciding on which ones to use and what factors to consider and so on. Therefore, the workload faced by the operator is overwhelming. With the flexibility and diversity available in a multi UAV system, it becomes impossible for an operator to take all those decisions in a timely manner and in an efficient way. Computerized automatic resource management systems are designed to answer those questions.
What is my approach?
Technology is developing fast and many advancements are not yet accessible to the public. Effective management systems of multiple UAVs will allow this cutting-edge technology to be utilized by everyone. Instead of having to own and learn how to control a UAV yourself and having limited resources on your particular machine, soon you will be able to benefit from the numerous services of a UAV simply by using an app on your mobile phone or by visiting a website. The key for succeeding in a UAV resource sharing system is an efficient resource allocation system, and that’s where my research comes in.
For more information about UoM UAV Research Group: http://uavs.mace.manchester.ac.uk/
For more information about aerospace system group: http://www.mace.manchester.ac.uk/our-research/research-themes/aerospace-engineering/specialisms/aerospace-systems/
For more information about studying aerospace: http://www.mace.manchester.ac.uk/study/undergraduate/courses/aerospace-engineering/meng-aerospace-engineering-4years/
Some ted talks about UAVs:
A video indicating the simplicity and important usages of UAVs: