Autonomous Driving Research

Autonomous driving research aims to develop safe and reliable self-driving systems through advancements in perception, planning, and control. Current efforts focus on improving model robustness in challenging scenarios (e.g., cyclist interactions, rare events, and inclement weather) using techniques like reinforcement learning to fine-tune agent behavior models and vision-language models for enhanced scene understanding. This research is crucial for advancing the safety and reliability of autonomous vehicles, impacting both the scientific understanding of complex multi-agent systems and the development of practical, real-world applications.