Currently, I am a Ph.D. student at ETH Zurich, Robotic Systems Lab. I graduated from my master’s degree Robotics, Systems and Control at ETH Zurich, Switzerland. Before that, I received my bachelor’s degree in Mechanical Engineering with first place in my cohort and I recieved my second bachelor degree in Control Engineering and Automation.
My research mainly focuses on incorporating state-of-the-art optimization and machine learning approaches into robust localization and mapping frameworks. Specifically, I specialize in robust localization and perception in degenerate and challenging environments utilizing optimization and learning based methods. Additionally, I am highly interested in multi-modal fast volumetric mapping and efficient and consistent local and global representations.
For more information, please refer to my Curriculum Vitae and Project Portfolio.
MSc in Robotics, Systems and Control
ETH Zurich
BSc in Control and Automation Engineering
Istanbul Technical University
BSc in Mechanical Engineering
Istanbul Technical University
In this work we propose i) a robust multi-category (non-)localizability detection module, and ii) a localizability-aware constrained ICP optimization module and couples both in a unified manner. The proposed localizability detection is achieved by utilizing the correspondences between the scan and the map to analyze the alignment strength against the principal directions of the optimization as part of its multi-category LiDAR localizability analysis. In the second part, this localizability analysis is then tightly integrated into the scan-to-map point cloud registration to generate drift-free pose updates along well-constrained directions.
This article presents a team of legged robots for exploration missions in challenging planetary analog environments. We equipped the robots with an efficient locomotion controller, a mapping pipeline for online and post-mission visualization, semantic and instance segmentation to highlight scientific targets, and advanced payloads for remote and in-situ scientific investigation.
In this work, we present Simulation and prediction focuced DNNs for non-linear system identification problem. Furthermore, we provide recommendations on how deep AEs and LSTMs should be utilized to end up with efficient Prediction-focused (Pf) and Simulation-focused (Sf) DNNs for time series and system identification problems.
ALMA is an ANYmal that is equipped with a robotic arm designed for articulated locomotion and manipulation.
ANYmal is a unique legged robot that provides reliable industrial solution with autonomous robot inspection.
ANYmal-D is the next generation ANYmal equipped with 6 D435 depth cameras, 2 Wide-angle RGB cameras, a IMU and a VLP-16 LiDAR in addition to the joint encoders.
Folly is a self-foldable and self-deployable drone designed and produced during my BSc Thesis in Mechanical Engineering which is under patent protection. Folly is a ~ 1 kg quadcopter with 18 minute battery life that is designed for confined space applications.
SuperMegaBot is a wheeled automonous ground robot for eduation, and research in autonomous inspection. The current mechanical design is a product of my work at ASL as a HiWi.