Job description:
The goal and main contribution of this project will be to design a sample-efficient control scheme tailored to control of soft robots. You will especially focus on the case in which the robot interacts with an unstructured environment, possibly including humans. To this end, you will investigate techniques that are at the intersection between model based strategies (Model Predictive Control) and data driven techniques (Reinforcement Learning). The resulting solutions will take the best out of the two worlds.
For this project, you will collaborate with researchers from two departments at the faculty of 3mE, namely Cognitive Robotics (CoR) and Delft Center for Systems and Control (DCSC).
Requirements:
You have a PhD degree (or close to finish) in systems and control, robotics, applied mathematics, artificial intelligence, machine learning, or a related subject. You must have strong analytical skills and must be able to work at the intersection of several research domains.
The project is going to be at the intersection of MPC and RL. So, research experience in one of the two and at least a basic understanding of the other is mandatory. Experience with robot control, multi-body dynamics, real robot applications, soft robotics, and/or advanced optimization is a plus.
You must have demonstrated ability to conduct high-quality research according to international standards, as demonstrated by publications in international, high-quality journals. A very good command of the English language is required, as well as excellent communication skills.
For more information about this vacancy, please contact Azita Dabiri, Assistant Professor (email: A.Dabiri@tudelft.nl), and Cosimo Della Santina, Assistant Professor (email: C.DellaSantina@tudelft.nl).
To apply for this position, please visit here
