Research Activities

Research Activities

My Research Interests

Robotics, Mechatronics, Control, Vision for Robotic Systems, and design and control of Unmanned Vehicles

Possible Master of Science Thesis Topics (Tez Konuları):

Bitirme Tasarım Projesi Konuları

(Senior Design Project Topics in Turkish):

· Increasing driving safety of vehicles with vision systems.

· Design and control of novel UAV systems

· Visual-servo control of UAVs

· Vision-based control of unmanned systems (ground vehicles, flying systems, etc.)

· Design and control of mechatronic systems, and industry applications.

· Any thesis topic related to the continuing/completed thesis topics listed below.

· Ufak bir insansız hava aracının tasarımı ve kontrolü

· Mekatronik ve otomasyon eğitiminde kullanılabilecek bir deney düzeneğinin tasarlanması

· İmalat sistemlerinin otomasyon ile geliştirilmesi

· Eğitim amaçlı modüler ve çok algılayıcılı bir gezgin robotun tasarımı ve imalatı

· Aşağıda listelenmiş devam eden/tamamlanmış yüksek lisans tez konuları ile ilgili bitirme çalışması da yapılabilir.

Graduate Students (Graduated, current graduate students)

- C. Bora Yiğit, (continuing thesis studies.)

- Taylan Keleş, (continuing thesis studies.)

- Uğur Acar, (continuing thesis studies.)

- Onurcan Şahin (M.S. Thesis Title: Design and Control of Visual Servomechanism System for Automating Corneal Linking Treatment on Keratoconus Patients, March 2011)

- Zehra Ceren (M.S. Thesis Title: Control of An Unmanned Aerial Vehicle by using Visual Servo Control, Görsel Servo Kontrol Yöntemi Kullanılarak Bir İnsansız Hava Aracının Kontrol Edilmesi (in Turkish), June 2010)

- Ahmet Yörükoğlu (M.S. Thesis Title: Determining The Mass, Horizontal and Angular Position of the Unbalanced Load in Washing Machines, Çamaşır Makinelerinde oluşan dengesiz yükün ağırlığının, yatay ve açısal konumunun belirlenmesi (in Turkish), January 2010)

- Metin Tarhan (M.S. Thesis Title: Control of a Quadrotor Air Vehicle by Vanishing Points in Catadioptric Images, Quadrotor Hava Aracının Katadioptrik İmgelerdeki Kaçış Noktaları Kullanılarak Kontrol Edilmesi (in Turkish), June 2009)

- Burcu Aytekin (M.S. Thesis Title: Camera Based Vehicle Detection and Tracking, December 2008)

- Başar Denizer (M.S. Thesis Title: Real-time Simultaneous Localization and Mapping for Mobile Robots, June 2008)

- Bora Erginer (M.S. Thesis Title: Modeling and Control of a Quadrotor VTOL Vehicle, Quadrotor VTOL Aracının Modellemesi ve Kontrolü (in Turkish), June 2007)

Team MEKAR - Grand Cooperative Driving Challenge Project

In 14 and 15th of May 2011, the first Grand Cooperative Driving Challenge (GCDC) was held in Helmond, the Netherlands. The GCDC is an open competition for between research groups on the topic of cooperative, autonomous driving. Team MEKAR formed by the students and faculty of Istanbul Technical University, Okan University, Arel University, the Ohio University, and Tofaş-Fiat grouped together to represent the only Turkish team in the competition.

Mekar Team’s developed vehicle.

Photo of gcdc event. (Team Mekar’s vehicle is on the right side, just behind the lead vehicle.)

More information available from; GCDC, OU, TUBITAK and see some videos (Reuters, one week before, info).

Design and Control of a Visual Servomechanism for Automating Corneal Cross Linking Treatment on Keratoconus Patients (with graduate student Onurcan Şahin)

In this study, a mechatronic system designed for automated corneal cross linking treatment on keratoconus patients was introduced. Keratoconus is a serious illness if not treated may cause serious distortion of vision. Regular treatment of this disease is performed manually. Developed automated system aims to increase the efficiency in treatment and to eliminate any potential side effects and risks of the treatment. The system consists of an image processing algorithm developed on OpenCV sharp, a servomechanism system consisting of various mechanical, electronic components to track the eye of the patient, and a digital PID controller to eliminate any errors.

System Block diagram

Developed system

edtt

Experimental system

uvfin Screen shot 2011-02-25 at 02

Lab. Experiments

Project: Vision-based Control of Unmanned Aerial Vehicles

This work was supported in part by the Scientific and Technological Research Council of Turkey (TÜBİTAK)

A Unmanned Aerial Vehicles (UAVs) can be simply defined as vehicles that fly without any pilot onboard. UAVs include unmanned aircrafts, helicopters, blimps and other flying vehicles. They come various sizes, small insects to full-size planes. An autonomous UAV brings enormous benefits and is suitable for applications like search and rescue, surveillance, remote inspection, military applications, therefore saving human pilots from dangerous flight conditions. UAVs are especially useful when (i) the working environment is inaccessible or hard to reach (planetary flier), (ii) dangerous (due to war, environment conditions, etc.), (iii) flight is monotonous, (vi) flight time is extended (atmospheric observations, data relay, etc.), (v) flight is not possible even by a skilled pilot (e.g. movie making, flight of experimental vehicles, etc.).

With the advancements in controls, propulsions, advanced materials UAVs are now more frequently used for civilian purposed and for armed forces around the world. The future of the skies lies at the UAV technology.

Our research at the Istanbul Technical University is on modeling and control of helicopter based, gas and electrical powered Unmanned Aerial Vehicles.

Experiment set-ups for rotorcraft based UAV control research.

standb

Developed experimental flight test apparatus.

Developed vision systems on board the experimental set-up.

The use of catadioptric cameras on UAV systems is very promising.

The images obtained from the catadioptric camera can be used to stabilize the UAV. The following pictures show the estimation of the parallel lines in a scene, and estimation of the vanishing lines from those lines.

lines

Attitude estimation steps on the images.

Final result of the attitude estimation.

Related Video

Estimating the pose of a flying vehicle with vision for landing and take-off

Estimation of moving vehicles on a scene can be easily performed from a UAV.

Tracking ground vehicles using a Pan Tilt Zoom Camera System using Vision

Increasing Driving Safety with a Multiple Vehicle Detection and Tracking System using Ongoing Vehicle Shadow Information

Due to the accidents, driving safety became an important topic. The work has mainly been focused on detecting and tracking vehicles viewed from inside a vehicle the camera mounted in daylight conditions. The vision sensor can be used to track the traffic and warn the driver in case of a collision. Unlike previous work, this approach uses vehicle shadow clues and vehicle edge information to obtain cost effective and fast estimation.

Papers about this topic: SMC2010, TOK'09, IJVD	Related Video

Project: Determining the Mass and Angular Position of the Unbalanced Load in Horizontal Washing Machines (supported by ARÇELİK)

Unbalanced load in washing machines limit the performance, and therefore accurate estimation of unbalanced loads, including their magnitudes and locations in washing drum, is required. In this work, we propose an approach to evaluate the angular position and mass of the unbalanced load.

Washing machine with unbalanced loads.

Papers about this topic: AIM2009

Project: Design of a mobile robot for map building Tasks

In this Project, a mobile robot has been designed and manufactured for map building applications. Map building by a mobile robot is useful for many different scenarios, such as earthquake destroyed or contaminated buildings, or mines.

Papers about this topic: IECON2009, TOK2009		Related Video

Control of a Quadrotor helicopter using Fuzzy Logic (with student Bora Erginer)

In this work, modeling and control of the Quadrotor using Fuzzy logic has been studied. Various simulations performed in MATLAB Simulink show the effectiveness of the fuzzy logic approach. Also, various experiments have been performed to verify the results. A MEMs IMU unit was used as a sensing device. The quadrotor has been fixed on a test stand and the control has been performed using the MATLAB Simulink model.

Quadrotor on a test stand

FuzzyController

Papers about this topic: TOK’07