GAD-Generative Learning for HD Map-Free Autonomous Driving

arXiv:2405.00515v1 Announce Type: cross
Abstract: Deep-learning-based techniques have been widely adopted for autonomous driving software stacks for mass production in recent years, focusing primarily on perception modules, with some work extending this method to prediction modules. However, the downstream planning and control modules are still designed with hefty handcrafted rules, dominated by optimization-based methods such as quadratic programming or model predictive control. This results in a performance bottleneck for autonomous driving systems in that corner cases simply cannot be solved …

abstract arxiv autonomous autonomous driving control cs.cv cs.ro driving free generative however map modules optimization perception planning prediction production rules software stacks type work