PhD Positions in human-robot cooperation & learning from experience, KTH Royal Institute of Technology, Sweden (2016)

The PhD positions are connected to the project FACT which is financed by SSF – the Swedish Foundation for Strategic Research during the period 2016-2020 with a total of approximately 2.5 million Euro. The outcomes of FACT will provide novel functionalities and design principles for the future factories and will be applicable both to small and large scale frameworks. The focus is on providing safe and flexible feedback in unforeseen situations, enhancement of human-robot cooperation and learning from experience. FACT research contributes to scientific and industrial communities while also having a clear social dimension, providing technologies that contribute to an increased quality in work environments.

Three different aspects on human robot cooperation and interaction will be treated.

1. Learning from Collaboration in Object Handling and Manipulations Tasks:
Our main goal in this area is to develop methods that allow robot to observe a human performing various types of object manipulation tasks (pick and place, put together, insert), learn a structured representation of it, learn how it execute it itself based on multiple sensory feedback and used these to act as a third hand in cases human worker needs it.

2. Natural and Embodied Interaction:
In this area the we will study how verbal and non-verbal information can be used to realize efficient and user friendly interaction with a robot by combining speech, posture shifts and head movements. Furthermore, we will study incremental speech processing to make the interaction faster, and allow for the possibility of clarification interruptions that makes the interaction more natural and robust. It also makes it possible for the system to be more responsive and adapt its output to changes in the shared environment.


3. Interacting via Mobility:
In this area methods for modeling the motion of human operators will be developed. The aim is to predict the need for assistance before the human has communicated it explicitly and to control the motion of the robot to optimize the process flow by being at the right place at the right time. Furthermore, methods to achieve safe mobility will be studied.

A suitable background for this position would be a Master of Science or equivalent in computer science, engineering physics, electrical engineering or similar. Good mathematical and programming skills are needed and experience in dialogue systems, speech processing, computer vision, machine learning and robotics is are regarded as advantageous qualifications. The candidate must have a good knowledge of written and spoken English.

Applicants must be strongly motivated for doctoral studies, possess the ability to work independently and perform critical analysis and also possess good levels of cooperative and communicative abilities.

Further details:
http://academicpositions.eu