Pedestrian/Cyclist Motion Prediction for Self-Driving Vehicles

SAVE board

  • PhD position
  • Delft

Website TU Delft

The Intelligent Vehicles group at TU Delft is seeking a PhD candidate with an interest in performing cutting edge research in the area of self-driving vehicles. This PhD position is part of the project “Safe interaction of automated vehicles with vulnerable road users” (SafeVRU). The project will develop sensing, intent recognition, and vehicle control strategies for the interaction of automated vehicles with vulnerable road users (VRU) such as pedestrians and cyclists. Several Dutch industry partners participate in this project (2getthere, TNO Automotive, NXP); beyond these, multiple collaborations with the European automotive industry exist.
The Ph.D. position involves developing motion/behavior models for pedestrians/cyclists, that can accurately predict their future trajectories. These models are to be learned/parametrized using training data and will take into account sensor measurements (e.g. position, velocity, pose, gestures, and traffic environment) and prior knowledge (physical constraints, map data). A challenge is to deal with the uncertainty derived from imprecise sensor measurements and from highly maneuverable targets that can change their velocities abruptly (e.g. change of direction, start/stop walking). Several motion model types will be considered initially, based on explicit dynamics, trajectory learning and planning (agents). In particular, deep learning for motion prediction will be investigated.

To apply for this job please visit intelligent-vehicles.org.