Integrating Julia for Perception Tasks in Autonomous Driving

Explore the integration of Julia for perception tasks in autonomous driving in this 11-minute conference talk from JuliaCon 2024. Discover how Julia can process millions of LIDAR points in real-time for future autonomous vehicles, covering ground segmentation, semantic occupancy grid creation, and object detection. Learn about the algorithmic development process compared to PyTorch, training phase experiences, integration into C++ and Python environments, and considerations for parallelism and runtime issues. Gain insights from the EU project AI4CSM (ai4csm.eu) and understand how Julia can be effectively utilized in the development of machine learning algorithms for real-world autonomous driving applications.