Aleksandar (Alex) Vakanski

 

 

The full Curriculum Vitae is available for download here.

Education:

Ph.D. in Mechanical and Industrial Engineering, Toronto Metropolitan University, Toronto, Canada, 2013

M.A.Sc. in Mechanical Engineering, Ss. Cyril and Methodius University, Skopje, Macedonia, 2003

B.Eng. in Mechanical Engineering, Ss. Cyril and Methodius University, Skopje, Macedonia, 1998

Research Experience:

University of Idaho (Idaho Falls, USA) Assistant/Associate Professor

Human movement modeling in medical applications: The research employs a hierarchical neural network architecture for modeling and evaluating movements by patients undertaking physical rehabilitation therapy. The research also formulated metrics for assigning quality scores to patient performance. The team created the UI-PRMD dataset (University of Idaho – Physical Rehabilitation Movement Dataset) to facilitate model validation.

Breast cancer detection in ultrasound images: The project employs machine learning for development of robust and explainable models for detection and segmentation of breast tumors in ultrasound images. Addressed challenges include integrating prior knowledge of visual saliency for tumor segmentation, design of explainable models for cancer diagnosis using the BI-RADS lexicon, detection and segmentation of small-size tumors in breast images, and segmentation of histopathology images for breast cancer detection.

Crop health assessment in precision agriculture: The project developed a novel neural network-based method for image-processing of aerial images of crop fields collected with an unmanned aerial system. The set of collected field images with RGB and multispectral sensors was annotated and posted for public access. The goal of the project is discrimination of healthy and stressed plants, and early diagnosis of crop health symptoms.

Robotic visual learning: The project is a continuation of my previous work on robotic learning systems, focusing on improving object tracking, pose estimation, and the graphical user interface, and employing ROS software for robot control and interfacing the system components.

Toronto Metropolitan University (Toronto, Canada) Postdoctoral Fellow

Commercialization of a system for image-based robotic learning: Sponsored by the Scientists and Engineers in Business program, during the fellowship I co-founded a start-up company Visual Learning Robotics, which developed a prototype system for robotic learning from visual observations. The system is based on a patented approach that I developed during the Ph.D. studies.

Toronto Metropolitan University (Toronto, Canada) Research Assistant

Robot programming by demonstration: The objective is to teach robots new tasks from observation of human demonstrations. I developed approaches based on Hidden Markov Models and Conditional Random Fields for stochastic task modeling at a trajectory level of task abstraction, and for generation of a generalized trajectory for task reproduction by the robot learner. Other aspects of the problem which I investigated are: formulation of the learning process as a constrained optimization problem for direct learning in the image space of a camera sensor, and using dynamical movement primitives for learning cyclic motions.

Visual servoing of robots: I worked on developing a hybrid robot control method for simultaneous optimization of trajectories of scene features projected in the image space of a camera and the trajectory of a robot’s end-point in the Cartesian space.

Visual perception of workspace: Project for analysis of the visual perception of human workers using visual acuity and motion resolvability as metrics for improved ergonomics with regards to the visual perceptibility of the workspace.

Theses:

A. Vakanski, Statistical Modeling, Planning, and Control for Vision-Based Robot Programming by Demonstration, Ph.D. Thesis, Department of Mechanical and Industrial Engineering, Toronto Metropolitan University, Toronto, Canada. Supervisors: Dr. Farrokh Janabi-Sharifi, Dr. Iraj Mantegh.

A. Vakanski, Digital Dynamic Control of Read-Write Systems in a Disk Drive, M.A.Sc. Thesis, Faculty of Mechanical Engineering, Ss. Cyril and Methodius University, Skopje, Macedonia. Supervisor: Dr. Atanasko Tuneski.

A. Vakanski, Control of Robots in Interactive Environment, B.Eng. Diploma Work, Faculty of Mechanical Engineering, Ss. Cyril and Methodius University, Skopje, Macedonia. Supervisor: Dr. Atanasko Tuneski.

Awards:

ASNT Faculty Grant Award (2018) awarded by the American Society for Nondestructive Testing.

Scientists and Engineers in Business Fellowship (2013) awarded by the Federal Economic Development Agency for Southern Ontario and Toronto Metropolitan University.

Italian Machine Tool Technology Award (2011) awarded by the Italian Trade Commission.

Certification:

Professional Engineer Faculty Restricted Licence (2017) granted by the Board of Licensure of Professional Engineers and Professional Land Surveyors of Idaho.

Certified Technology Manager (2017) granted by the Association for Technology, Management, and Applied Engineering.

Remote Pilot Airman Certificate (2017) granted by the Federal Aviation Agency.