Research group Vision for Robotics and Autonomous Systems (VRAS) at the Department of Cybernetics, Faculty of Electrical Engineering, Czech Technical University in Prague, is looking for outstanding Ph.D. students and postdoctoral researchers.
Ph.D. Positions
Ph.D. positions are in the areas of:
- 3D Lidar perception, LiDAR to text embedding, Generative models for LiDAR data
- Physics-based deep learning (Neural symbolic architectures)
- Imagine you have a differentiable “Gazebo” and plug it into a DNN!
- Explainable end-to-end differentiable perception for self-supervised learning in robotics
- More topics possible, e.g. 3D lidar perception/simulation, advanced SLAM for ground robots, control and perception for amphibious (walking/swimming) robots, Search&Rescue robotics.
Requirements: Excellent grades, experience in the relevant fields, good programming skills (Python, C/C++, ROS).
Successful candidates will get a full tuition waiver and a monthly compensation equivalent to about 1400 EUR of net income (the Czech Republic’s cost of living is at 59% of the US price level, according to OECD statistics).
PostDoc Positions
We seek highly motivated applicants with a PhD in robotics, AI, or related fields and a proven track record relevant to the topic - publications in top journals or conferences (e.g. computer vision (CVPR/ICCV/ECCV), machine learning (NeurIPS/ICML), or robotics (ICRA, IROS, RSS, CoRL).
- End-to-end learning with explainability. Self-supervised learning
- More topics possible, e.g. 3D lidar perception/simulation, advanced SLAM for ground robots, control and perception for amphibious (walking/swimming) robots, Search&Rescue robotics.
Prague and the workplace
The group has a strong track record in machine perception and computer vision. More recently, there has been substantial development in robotics, with groups focusing on search and rescue robotics and cooperation with firefighters. We regularly publish at top robotic and CV conferences like ICRA, IROS, ICCV, ECCV (most of our publications are listed here).
The group, together with roboticists from the Department of Computer Science and the group Multi Robotic Systems, form the Center for Robotics and Autonomous Systems. CRAS has succeeded in numerous competitions: the CTU team ranked 3rd overall and 1st among self-funded teams in the Tunnel Circuit round (2019) and the Urban Circuit (2020) of the DARPA SubTerranean challenge and was 2nd overall in the Final Event (2021) in the Virtual challenge.
Hardware infrastructure at the Department includes wheeled, tracked, and legged robots (including 2 Boston Dynamics Spots and 2 Spots with arm) for search and rescue operation equipped with state-of-the-art 3D lidars from Ouster, high-resolution omnivision pack, industrial-grade IMUs and other sensors. Further, there are industrial as well as collaborative or lightweight manipulators (KUKA LBR iiwa, UR10e with Airskin, Kinova Gen3), flying robots (including fleets of UAVs), and humanoid robots (iCub, Pepper, 2x Nao). The Department has access to a high-performance computational cluster dedicated to artificial intelligence research and developments using traditional multi-CPU systems, but also GPUs, including NVidia DGX stations and Graphcore IPUs.
There is currently a very lively international ecosystem in the areas of robotics and AI - at the Department and in Prague in general. See, for example, prg.ai.
CTU has an international student club assisting with study affairs and organising leisure-time events. Prague was TripAdvisor’s #7,#8, and #11 world’s top destinations in the years 2017-19. Beautiful mountain landscapes and historic cities are within reach. Owing to Czechia’s Schengen-area membership, numerous exquisite European destinations can be reached from Prague with no border restrictions or delays. Our teams take great joy in social and savoir-vivre activities.
The administration and project office provides professional support with long experience in managing international projects.
ROS and Gazebo
Our team heavily uses ROS and Gazebo for research. And we also give back to the community by publishing the interesting parts of our work as ROS packages (VRAS github, officially released ROS packages), most notably robot_body_filter or point_cloud_transport. We also gave a talk at ROSCon and helped organizing ROSCon 2022.
We have implemented the first open-source simple (yet plausible) simulation model of tracked robots in both Gazebo Classic and new Gazebo (here’s a comparison to the traditional “workaround” models) . Thanks to that, anyone can play with all our mobile robots in Gazebo!
How to apply
Please send your application or inquiries to Ms. Petra Ivaničová (ivanipet at fel.cvut.cz) with “VRAS - Open positions” in the subject and include the following, preferably in a single PDF file:
- CV
- Full publication list
- PostDoc applicants: Make sure you satisfy the condition of having publications in relevant A* or A conferences like CVPR/ICCV/ECCV or NeurIPS/ICML or ICRA/IROS/RSS/CoRL or in Q1 journals like RA-L, Science Robotics, JFR, Field Robotics
- Motivation letter
- List of at least two referees
- PostDoc applicants: add at least one recommendation letter to the package
- PhD applicants: your most interesting research paper and your master thesis (if applicable)
- PhD applicants: transcript of grades of master/bachelor
The positions are open until filled.
You can also PM me here for more details.
See also: Career - ROBOPROX (search for Svoboda, Zimmermann), https://cyber.felk.cvut.cz/research/groups-teams/vras/ .