PhD in Predictive Occupancy World Models for Autonomous Driving

About the role

Join Bosch Research Computer Vision and AI Lab in Hildesheim to develop a generic predictive occupancy model that serves as the core representation of the driving scene. You will work on cutting‑edge AI techniques to enable safe and transparent autonomous driving.

Key responsibilities

• Design and implement architectures that capture geometry and semantics of all objects for 4D world dynamics.
• Research, develop, and evaluate end‑to‑end driving architectures based on predictive occupancy world models.
• Create perception systems that transform raw sensor data into rich spatial representations and driving actions.
• Integrate predictive occupancy models into decision‑making pipelines to ensure spatially grounded actions.
• Test models extensively in simulation (e.g., CARLA) and with real‑world data, demonstrating improvements in safety and performance.
• Publish research results at leading AI and robotics conferences.

Required profile

• Excellent master’s degree in Computer Science, Robotics, AI, Data Science or a related field qualifying you for doctoral studies.
• Strong expertise in deep learning and computer vision, with hands‑on experience using PyTorch and Python.
• Creative, self‑driven researcher who enjoys ownership of challenging topics and collaborative technical discussions.
• Passion for solving fundamental AI challenges and building trustworthy, safe systems.
• Excellent English communication skills; German is a plus.

Required skills

• Deep Learning
• Computer Vision
• PyTorch
• Python programming
• 3D Computer Vision
• Occupancy Networks
• End‑to‑End Driving
• World Models
• Sensor Fusion
• Autonomous driving datasets
• Simulation environments (e.g., CARLA)