Researcher

My research & work interests are focused particularly on systems dynamics & control, and machine-man interaction. Particularly, I am interested in (semi)autonomous systems, especially any type of vehicle, and in applications that can have an actual impact on this world. Therefore, I care about seeing the results of my work become reality, i.e., see a helicopter robot fly or a robot car drive itself; I am an engineer and working on real-world problems is my primary interest!

Control Systems Laboratory

CSL Logo In September 2008 I had the invaluable opportunity to join the Control Systems Lab as undergraduate research assistant, where I worked towards the completion of my Diploma Thesis under the title "Development of Nonlinear and Linear Dynamic Model for a Small Scale Helicopter" under the supervision of Professor Kostas J. Kyriakopoulos. My work included extensive literature review on the problem of system parameter identification, specifically for small-scale helicopters, mathematical modeling of the aeromechanical system, model simulation, and verification of results on the literature. The ultimate goal was to develop a simulator that could be succesfully used for testing control designs, as well as the identification procedures for the helicopter at hand.
In September 2010, after the short break to TU Delft, I joined the CSL again. My work had a relatively different scope this time, since I participated in a project regarding the effective and efficient cooperation of autonomous underwater vehicles, namely Remotely Operated Vehicles (ROVs) and/or Uninhabited Underwater Vehicles (UUVs). My role as a researcher was to perform extensive literature survey on the topic of mutliple autonomous vehicle cooperation, control, and coordination, and assess the benefits of each algorithm and approach. Additionally, I worked extensively on the programming of one of the CSL ROVs; specifically, I programmed the communication and control between a Linux-based PC-station and the robotic vehicle with the use of C++ and the Internet Communication Engine (Ice), so that advanced control schemes could be implemented.

Dynamics and Control Systems Lab

DCSL Logo In August 2011 I joined the group of Dynamics and Control Systems Lab, under the direction of Professor Panagiotis Tsiotras at The Daniel Guggenheim School of Aerospace Engineering, Georgia Institute of Technology, as a graduate research assistant. Durin my stay I had the opportunity to participate in two different projects, one sponsered by the NSF and the other sponsored by the ARO. My work was focused on the development of advanced control methodologies for vehicle cotnrol, both for stabilization and performance enhancement purposes. The main goal was to perform tracking of a pre-specified trajectory (or even real-time measured trajectory, i.e., road lane) (semi)autonomously. For the autonomous driving case the driver was completely ingoned and there was no need to adjust the vehicle performance to the driver.
On the other hand, for the semi-autonomous driving case the driver needed to be accounted for. Therefore, I performed extensive literature survey on the topic of human-driving and modeling of the human driver, so that the qualitative results could eventually be translated to quantitative results with the use of mathematical modeling. Consequently, I worked on the problem of designing an Advanced Driver Assistance System (ADAS) for lane-tracking by taking into consideration the both the driver existence and the driver-vehicle interaction. In order to achieve that, I used the two-point visual driver model. Eventually, the control design that took into consideration the driver model performed better than the one that was oblivious to the driver existence, and this is a remarkable result, since it might change the way ADAS are designed; so far, all the active safety systems do not take into consideration the driver model.
The control methodologies used to perform the required task were based both on more classical control theories, i.e., output regulator theory, as well as on more advanced theories, such as Model Predictive Control (MPC) (for the simulations and videos from CarSim please visit the Media Files section). Finally, the idea of exploiting the driver as parameter varying system was investigated and theory based on Linear Parameter Varying (LPV) systems was used.

Mitsubishi Electric Research Laboratories

DCSL Logo I joined MERL twice, once in Summer 2012 and once in Summer 2013, as a research intern, and my host was Dr. Stefano Di Cairano. My work at MERL was focused on vehicle dynamics, control, and estimation, as well as investigating the performance of a control design while taking into consideration the driver-vehicle interaction. Different control methodologies were used, with an emphasis on optimization-based control schemes, e.g. MPC. My stay at MERL gave me the great opportunity to experience the work environment outside the university, as well as to collaborate and interact with people of many different research and work backgrounds.

Publications

  1. Spyridon Zafeiropoulos and Stefano Di Cairano, "Vehicle Yaw Dynamics Control by Torque-based Assist Systems Enforcing Driver's Feel Constraints", in American Control Conference, June 2013, Washington DC, USA, pp. 6746-6751.
  2. Spyridon Zafeiropoulos and Panagiotis Tsiotras, "Design of a lane-tracking driver steering assist system and its interaction with a two-point visual driver model”, in IEEE American Control Conference, June 2014, Portland, Oregon, pp. 3911-3917.
  3. Spyridon Zafeiropoulos and Stefano Di Cairano, "Governor-based Control for Rack-Wheel Coordination in Mechanically Decoupled Steering Systems", in Conference on Decision and Control, December 2014 (To appear/Accepted for publications).

How to: Aerospace at GaTech

The classes

There is NO WAY you cannot find the class you want! On the contrary, you might be overwhelmed with the vast pool of classes, that you will want to register for everything. However, be wise and choose classes will actually contribute the maximum to you, while making sure that the classes don't overlap or have crazy prerequisites; and don't forget the other schools at Tech; ISYE, Computer Science, Math, etc.

The workload

You will just have to deal with it. You might become late birds or do overnight studies, but it is fun in the end. Try to gain as much as you can from each class and don't just try to hand over the homework. The purpose is gaining knowledge, not only getting a 4.0...In case you have research to (most probably) think twice before registering for 4 classes (if not 3!) each semester.

Socializing

I have to admit that GaTech is not the best place to socialize. It might have been just my experience, but it is common knowledge that there are some issues...Don't forget, though, that you have a life and you should try and maintain it. Friends are a huge support throughout this process and you're going to need them!

Why?

For everything that you can gain from this experience; the scientific knowledge, the people you meet, the work in a high-tech campus and environment, the CRC, the city, and the opportunity after graduation to get the life and job you dreamt of.