ETH Zurich Robotic Systems Lab

ETH Zurich Robotic Systems Lab
Name	Robotic Systems Lab
Established	1990s
Director	Roland Siegwart
Location	Zurich, Switzerland
Affiliation	ETH Zurich

ETH Zurich Robotic Systems Lab is a research group at ETH Zurich focusing on autonomous mobile robots, perception, control, and human-robot interaction. It operates within ETH Zurich's Institute for Dynamic Systems and Control and maintains active programs in field robotics, aerial systems, and service robots. The lab contributes to European research frameworks and Swiss technology initiatives while training students at bachelor, master, and doctoral levels.

History

The lab was founded during the expansion of robotics research at ETH Zurich and grew alongside groups led by Roland Siegwart, Oussama Khatib (as a collaborator), Hugh Durrant-Whyte (through international links), and Seth Hutchinson through shared conferences such as ICRA and IROS. Early milestones included participation in the DARPA Grand Challenge ecosystem and cooperative projects with Swiss Federal Institute of Technology in Lausanne researchers like Luca Benini and Davide Scaramuzza. The lab's timeline intersects with initiatives from European Research Council grants, Swiss National Science Foundation funding, and collaborations with industrial partners such as ABB Group, Siemens, and Bosch. Key leadership transitions involved partnerships with groups at Carnegie Mellon University, Massachusetts Institute of Technology, and ETH Zurich departments including Laboratory for Information and Inference Systems and Institute of Robotics and Intelligent Systems.

Research Areas

The lab's core research spans autonomous navigation, state estimation, mapping, and motion planning with ties to work by Hector Geffner, Richard M. Murray, Nikos Vlassis, Sven Behnke, and J. Andrew Bagnell. Perception research connects to studies at Max Planck Institute for Intelligent Systems, Google DeepMind, and Microsoft Research on visual SLAM and sensor fusion. Control and planning research references algorithms popularized at Stanford University, University of Pennsylvania, and University of California, Berkeley, while multi-robot coordination refers to paradigms from University of Cambridge and ETH Zurich collaborators such as Raffaello D’Andrea. Human-robot interaction work links to projects at Toyota Research Institute, Shadow Robot Company, and Honda Research Institute. The lab addresses field robotics challenges seen in NASA and European Space Agency contexts and applies machine learning approaches inspired by Yann LeCun and Geoffrey Hinton.

Teaching and Education

Courses taught by the lab contribute to ETH Zurich curricula alongside modules from Institute for Dynamic Systems and Control, Department of Mechanical and Process Engineering, and Department of Computer Science. Student supervision draws on networks with supervisors at EPFL, Imperial College London, TU Munich, KTH Royal Institute of Technology, and University of Oxford. The lab hosts project semesters linking to industrial projects with ABB Robotics, Schindler Group, and Google. It organizes workshops at conferences including NeurIPS, CVPR, ECCV, and RSS, and supports student teams entering competitions like RoboCup and DARPA Subterranean Challenge.

Facilities and Labs

The lab maintains indoor and outdoor testbeds comparable to facilities at ETH Zurich centers, equipped with motion capture systems similar to those at AMSL and Stanford Intelligent Systems Laboratory. Hardware includes unmanned ground vehicles, aerial platforms akin to designs from DJI Innovations prototypes, and manipulation rigs used by groups at Georgia Tech and University of Tokyo. Sensor suites reflect state-of-the-art components from Velodyne, SICK AG, Intel RealSense, and NVIDIA compute clusters. Collaboration spaces link to shared facilities at Laboratory of Intelligent Systems, Autonomous Systems Lab and high-performance computing resources such as those used by CERN researchers.

Notable Projects and Publications

The lab has produced influential publications in venues like IEEE Transactions on Robotics, Science Robotics, Nature Communications, ICRA, and IROS. Notable projects include long-range autonomous navigation efforts comparable to work at ETH Zurich by Davide Scaramuzza and multi-modal perception studies related to research from University of Oxford and Caltech. Publications cover SLAM, visual-inertial odometry, and learning-based perception referencing contributions from Michael Kaess, Juan D. Tardos, Andrew J. Davison, and Mourad Oussalah. Field trials mirror deployments by Boston Dynamics and research campaigns coordinated with Swiss Federal Laboratories for Materials Science and Technology and Swisscom-backed initiatives.

Collaborations and Industry Partnerships

The lab engages with academic partners including ETH Zurich, EPFL, Imperial College London, TU Delft, University of Washington, and KTH Royal Institute of Technology, and with corporate partners such as ABB, Siemens, Google, Microsoft, Bosch, and NVIDIA. It participates in European projects funded by Horizon 2020, projects coordinated with European Space Agency, and bilateral programs with Swiss National Science Foundation. Industry consortia include collaborations with Schindler, SBB (Swiss Federal Railways), Swisscom, and startups spun out in the ETH Zurich ecosystem like firms incubated at ETH Zurich Foundry and Kickstart Accelerator.

Category:Robotics research institutes