Multimodal Learning and Control for Dexterous Robotic Manipulation

About the Project

The start date: no later than fall 2026

We are offering a PhD position that provides a unique opportunity to develop deep expertise in robotics, machine learning, and control. The project combines rigorous theoretical research with practical, application-driven, and experimental work within a dynamic, international research environment.

The research focuses on robotic manipulation, a key capability enabling robots to interact robustly and autonomously with the physical world. The goal of the project is to develop methods that allow robots to perform complex manipulation tasks by learning continuously from experience. This includes enabling robots to acquire and refine knowledge about the physical and geometric properties of objects, as well as adapting to dynamic and unstructured environments.

The work will leverage rich multimodal sensory data, such as vision and tactile sensing, to support learning-based perception, decision-making, and control. A central challenge is to integrate low-level dexterous manipulation with high-level planning and goal-directed behavior, allowing robots to operate effectively in real-world, human-centric settings.

As a PhD student, you will receive comprehensive academic training and work closely with senior researchers with expertise in robotics, machine learning, control, and optimization. You will be part of an active international research community and have opportunities to contribute to both foundational research and real robotic systems.

Eligibility and Requirements

Applicants should hold a Master’s degree in one of the following or related fields:

• Robotics
• Computer Science
• Electrical and Computer Engineering
• Mechanical Engineering
• Applied Mathematics
• Applied Physics
• Statistics and Optimization

A strong background in robotics, machine learning, and control is essential. Applicants should also demonstrate:

• High academic achievement in relevant undergraduate and graduate courses
• Proficiency in programming (C/C++, Python) and experience with ROS
• Practical experience in hardware and simulation-based implementations
• Fluency in both written and spoken English

Previous experience in robot learning research will be considered a strong merit.