Institute of Robotics and Intelligent Systems D-MAVT

Open Opportunities

Advancement of Magnetically Shape-Morphed Microrobots for Biomedical Applications ETH Zurich Multiscale Robotics Lab This project aims to explore the dynamic capabilities of microrobots to undergo a shape-change through dipole-dipole interactions in magnetic fields. The project encompasses three main phases: simulation, fabrication, and application, with a specific focus on biomedical scenarios. Engineering and Technology Master Thesis, Semester Project

Teleoperated Bimanual Dexterous Manipulation using the faive Robotic Hand ETH Zurich Soft Robotics Lab Develop and integrate a teleoperation setup for two dexterous, human-like robotic hand on robotic arms. Develop control strategies to solve real-world manipulation challenges with this system, to demonstrate the benefits of two-handed dexterous manipulation in mobile robotics. Electrical Engineering, Interdisciplinary Engineering, Mechanical Engineering Bachelor Thesis, Master Thesis, Semester Project

Learning Robust Agile Flight via Adaptive Curriculum University of Zurich Robotics and Perception This project focuses on developing robust reinforcement learning controllers for agile drone navigation using adaptive curricula. Commonly, these controllers are trained with a static, pre-defined curriculum. The goal is to develop a dynamic, adaptive curriculum that evolves online based on the agents' performance to increase the robustness of the controllers. Engineering and Technology Master Thesis, Semester Project

Deep Policy Adaptation for Dexterous Rope Manipulation ETH Zurich Robotic Systems Lab The innumerable uses of rope, including climbing, sailing, circus acts, and wrangling livestock, suggest that it could be used as an universal robotic manipulator. When combined with a mobile base, e.g. a cowboy on a horse, rope manipulation can be an effective tool for all manner of robot-environment interaction challenges. In this project, we seek to realize a control framework for ANYmal such that it can use a rope launcher to manipulate various objects. Intelligent Robotics, Robotics and Mechatronics, Systems Theory and Control Master Thesis, Semester Project

How does an injured ANYmal walk? ETH Zurich Robotic Systems Lab When quadrupedal animals injure their limbs, they adapt their gait to compensate for the injury. Conversely, when a quadruped robot’s leg breaks, its functionality is severely curbed, and in many cases, entirely compromised. This project will investigate the interplay of disabled/dismembered quadrupedal robot legs, locomotion performance, and control strategy, with the goal of converging on effective policies that abstract across different kinds of “injuries.” Intelligent Robotics, Robotics and Mechatronics, Systems Theory and Control Master Thesis

Continuous Skill Learning with Generative Latent Dynamics ETH Zurich ETH Competence Center - ETH AI Center Other organizations: Robotic Systems Lab In recent years, advancements in reinforcement learning have achieved remarkable success in teaching robots discrete motor skills. However, this process often involves intricate reward structuring and extensive hyperparameter adjustments for each new skill, making it a time-consuming and complex endeavor. This project proposes the development of a skill generator operating within a continuous latent space. This innovative approach contrasts with the discrete skill learning methods currently prevalent in the field. By leveraging a continuous latent space, the skill generator aims to produce a diverse range of skills without the need for individualized reward designs and hyperparameter configurations for each skill. This method not only simplifies the skill generation process but also promises to enhance the adaptability and efficiency of skill learning in robotics. Engineering and Technology, Information, Computing and Communication Sciences Master Thesis

Instrumented campus board: Stand-alone performance analysis tool for climbers ETH Zurich Sensory-Motor Systems Lab A campus board was instrumented to measure forces applied to the rungs. Now, the entire setup should be transferred to climbing gyms. Biomedical Engineering, Interdisciplinary Engineering, Mechanical and Industrial Engineering Bachelor Thesis, Semester Project

Learning Object Affordances from Large-Scale Human Interactions with Diffusion Models ETH Zurich ETH Competence Center - ETH AI Center Other organizations: Robotic Systems Lab This project explores unsupervised learning using extensive videos from the internat that capture human interactions with objects. By harnessing advanced generative AI models, the focus is on understanding object affordances, such as identifying interaction points and predicting post-grasp trajectories. Computer Vision, Intelligent Robotics ETH Zurich (ETHZ), Master Thesis, Semester Project

Learning Object Manipulation from Demonstrations using Vision and Haptic Feedback ETH Zurich Robotic Systems Lab We aim to develop a method to incorporate fine-grained tactile and visual feedback into our haptic teleoperation setup and investigate their effectiveness with state-of-the-art imitation learning methods. Intelligent Robotics Master Thesis, Semester Project

Minimal: Learning-based design optimization for robots over challenging terrain ETH Zurich Robotic Systems Lab Minimal is a mostly 3D-printed, highly reconfigurable robot. Using state-of-the-art reinforcement learning, we will explore novel and highly advanced hardware design possibilities that will be coupled with design optimization through learning. This will enable the next generation of robots to be a lot faster, stronger and agile. Engineering and Technology Master Thesis, Semester Project