PhD Pitches

Time: 15:00 - 16:00

1. Real-time retina modelling for enhanced robot-assisted eye surgery

Presenter: Ir. Yanick Douven
Research group: Control Systems Technology
Department: Mechanical Engineering
University: Eindhoven University of Technology

Vitreoretinal surgery addresses sight-threatening conditions on the back of the eye. Since structures on the back of the eye are often only a few micrometers, the technical demands placed on the surgeon by these procedures are very high. Robotic systems can provide a distinct and positive alteration in patient outcome when they can show to really enhance the surgeon’s capability by providing both physical and mental assistance.

The back of the eye is viewed through a stereo-microscope, which provides a three dimensional image to the physician. Still, determining distance between instrument-tip and retina proves to be very hard. Optical Coherence Tomography is a sensor solution based on interference between two light paths, a reference path and a path that reflects on the to be measured sample. An interference spectrum is measured and processed to obtain a one-dimensional depth image of the sample. This depth image is used to measure the distance between the instrument tip and the back of the eye. Using this distance, forbidden regions and cognitive assistance is provided to the surgeon.

2. Collaborative robotics in precision farming

Presenter: Ir. Roy Cobbenhagen
Research group: Control Systems Technology
Department: Mechanical Engineering
University: Eindhoven University of Technology

The objective of precision farming is to device a control strategy that optimizes the quality and quantity of the yield whilst reducing used resources such as water, fertilizer and pesticides. A rapidly increasing amount of robots is used in precision farming in order to increase the resolution of both sensing and actuating.

Our work focuses on 1) how we can non-destructively sense and then 2) identify the state of the farmland such that 3) we can optimize the precision farming objective with a higher resolution and then 4) synthesize control laws for the multi-agent system from this global optimum. In this talk I will present an overview of this research. 

3. Electrodermal Activity: Explorations in the Psychophysiology of Engagement with Social Robots in Dementia

Presenter: Giulia Perugia MSc 
Research group: Designed Intelligence 
Department: Industrial Design
University: Eindhoven University of Technology

The study of engagement is central to improve the quality of care and provide people with dementia with meaningful activities. Current assessment techniques of engagement for people with dementia rely exclusively on behavior observation. However, novel unobtrusive sensing technologies, capable of tracking psychological states during activities, can provide us with a deeper layer of knowledge about engagement. We compared the engagement of persons with dementia involved in two playful activities, a game-based cognitive stimulation and a robot-based free play, using observational rating scales and electro dermal activity (EDA). Results highlight significant differences in observational rating scales and EDA between the two activities and several significant correlations between the items of observational rating scales of engagement and affect, and EDA features. 

4. Self-Supervised Single-Sensor Data Fusion for Automated Driving

Presenter: Ir. Willem Sanberg
Research group: Video Coding & Architectures
Department: Electrical Engineering
University: Eindhoven University of Technology

Advanced Driver Assist Systems for intelligent vehicles require environment perception modules that are both robust and efficient. In our work, we explore how to exploit stereo camera vision for that goal, with a focus on free-space detection. We run different algorithms in parallel and experiment with fusing results over cues and over time in an efficient and elegant manner. Our strategy shows that the algorithms can boost each other's performance, even though they rely on the same data stream.

5. Orchestration of collaborative agents in dynamic manufacturing processes

Presenter: Jonnro Erasmus MSc.
Research group: Information Systems
Department: Industrial Engineering & Innovation Sciences
University: Eindhoven University of Technology

Factories look towards smart robotics and versatile operators to meet the demand for mass customised products. A factory floor populated by heterogeneous technology and versatile agents presents a new operations management challenge. The manufacturing processes are subject to frequent change and control of the production agents must be scalable to accommodate distributed decision-making. Business process management is a proficient instrument to manage dynamic processes populated by a diverse set of agents. We aim to demonstrate the use of business process management for manufacturing operations. This demonstration consists of three parts: the use of business process notation for manufacturing, a mechanism to allocate production agents to tasks, and the detection of events which indicate that process adjustment is needed. These three developments will be built into an information system which will be deployed at three factories for testing and evaluation.