PhD Position Multi-robot coordination under uncertainty

Autonomous robots (such as cars and UAVs) will be widespread in our daily lives. These robots will operate near humans in complex dynamic environments. To safely coordinate, these robots will benefit from exchanging a large amount of information, for example for perceiving and understanding a complex dynamic environment. However, exchanging large amounts of data is not always possible due to bandwidth constraints, faults and delays.

This project will allow you, as a PhD candidate, to design coordination strategies for the robots to limit the amount of information exchanged among the robots, while preserving safety. To achieve these objectives, you will investigate how to combine reinforcement learning (RL) and game theory to devise a multi-agent RL framework to safely coordinate the robots with limited communication exchanges. For example, the proposed method might decide with whom, when and what information to share for efficient perception-aware and communication-aware distributed motion planning.

You will be able to test your algorithms in practice in our experimental facilities, designing experiments on real autonomous robots (e.g., UAVs or ground robots) to demonstrate the effectiveness of the proposed design.

For your PhD you will be embedded within the Autonomous Multi-robots Lab in the Department of Cognitive Robotics at TU Delft. For more information of our ongoing research see https://www.autonomousrobots.nl .

Starting date is flexible between 1st September 2021 and 31st December 2021.