Patric Jensfelt

I am a professor of computer science at the division of Robotics, Perception and Learning (RPL) at the School of Electrical Engineering and Computer Science (EECS). My interest is mainly centered around robotics and autonomous systems. I like to put theory into practice and most of the time I am driven by problems rather than methods. Even after 20+ years in the field, a robot that actually does something on its own makes me excited. I have spent quite some energy into taken the methods from academia out into the real world, mainly through the company Intelligent Machines.

During the last decade my research has been focused on spatial cognition. I am interested in providing robotics with the tools to "understand" the world, for lack of a better world. I spent quite some time working on high level approaches to this, but I am now back where I started and build ground up instead. This means starting from the sensor data and modeling the world. The recent leaps in mainly visual perception means that what I worked on at a high, mainly symbolic, level now is getting close to come together with the lower levels. A real world result from this strand of work, that started as a quest to gather data from people's home, is the company Volumental. Volumental builds 3D foot scanners sold across the world that allows companies to give their customer a new experience and better fitting shoes.

I am involved in the courses

• DD2410 Introduction to Robotics given in the fall and providing a broad overview of the field of robotics.
• DD2419 Project Course in Robotics and Autonomous Systems given in the spring. We assume that you have taken DD2410 or similar and in particular that you are familiar with ROS. We work with drones and you get to combine the ability to localize, recognize objects, plan, etc into an integrated system. You are likely to see that the building blocks that you have learned in many previous courses is just the start of a long journey to build an autonomus system where anything can happen and you need to program the system to handle that which you do not know.

• DD2414 Engineering project in Robotics, Perception and Learning which will be given for the first time in the fall of 2020. Here we go deeper into building robot systems in tailored projects defined by you and some of your fellow students and me or some other faculty. It starts if and when there are interesting projects.

I also supervise Master's degree projects, mainly under the course codes DA236X, DA233X, and DA231X. If you want to do a thesis project with me, it will be much easier to convince me if you took my courses and showed interest there.

I offer a similar course to the project courses above by for phd stdudents

• FDD3356 System Integration for Robotics

Much of my energy on the doctoral education level goes into the work in the WASP Graduate School. The graduate school accepts roughly 40 students every second year at five+ different universities in Sweden. I teach the course

• Autonomous Systems that runs every year in the fall.

The courses gives a broad overview of the vast area of autonomous systems. The aim is to give all the doctoral students in the graduate program a common language and understanding and appreciation for the problems in other research areas.  The emphasis is on doing rather than proving.

Finally a word on internships. We do accept a few interns every year but we cannot offer a salary. If you want a reply to your request, show me that you have read at least this far about me.

Keywords: Service robotics, localization, mapping, SLAM, navigation, system integration, spatial modeling, spatial cognition, spatial priors, exploration, drones