Christian Smith

I am an associate professor in Computer Science - Robotics and Autonomous Systems. I specialize in human-centric robotics in general, and manipulation under different types of uncertainty in particular. My work so far has covered robot control, design, and teleoperation, human tracking and motion modelling, reinforcement learning, and both physical and social aspects of HRI. I tend to be a bit of an experimentalist, and do a lot of hands-on real-world implementations. Application areas range from social robotics to small-scale industrial manufacturing and medical and health care robotics.

Some of my current research projects

In collaboration with industry (see my students list) I am currently exploring how to combine supervised and unsupervised learning for mobile manipulation and assembly in small-scale manufacturing. The goal is to find methods that enable fast and flexible set up for automation for small batch sizes, building on existing expertise from human operators.

Our research on human-robot interaction is currently focusing on human-robot handovers, trying to learn effective policies for generating handover motions and controlling object transfer. The goal is to find methods that can be adapted to varying needs of different users and situations. In the longer perspective, this should be seen as an integrated part of a multimodal dialogue.

Teaching

I teach two second cycle robotics courses, with the goal to give students a chance to work with the broad multidisciplinary field of robotics as a whole, as contrasted to the narrower courses giving in-depth focus to the different building blocks, e.g. control, signal theory, estimation, programming, mechanics, electronics, etc. The first course (DD2410) is a broad introduction, and the second (DD2411) invites students into our research labs so that they can take part in cutting edge research in long-term projects.

I have been teaching programming at KTH in one way or another since 2002 when I started as a TA, and try to put my experience into use when designing new courses. My philosophy is that large amounts of practice and exposure to realistic problems are fundamental when learning programming, so my courses tend to have a large focus on the students' own programming work, and providing the students with assignments as close to realistic applications as possible.

Background

I received a MSc in Engineering Physics from KTH in 2005, and a PhD in Computer Science from KTH in 2009. Since 2018, I am a docent in Computer Science - Robotics and Autonomous Systems. I have spent two longer visits at the Intelligent Robotics and Communication Laboratory at ATR in Japan, in 2004, and 2010-2011, respectively.

An upper bound on my Erdős number is 3 (H.I Christensen -> P. Tetali -> P. Erdős).