Complex Driving
Complex driving scenarios, characterized by dense traffic, adverse weather, and unpredictable human behavior, pose significant challenges for autonomous vehicle development. Current research focuses on improving perception and decision-making through advanced models like vision-language models (VLMs), incorporating risk assessment and driving priors into reinforcement learning frameworks, and developing efficient data processing techniques such as coreset selection to handle large datasets. These advancements aim to enhance the safety and reliability of autonomous vehicles by enabling them to navigate and react appropriately in complex and dynamic environments, ultimately contributing to safer and more efficient transportation systems.
Papers
October 17, 2024
September 25, 2024
July 31, 2024
June 18, 2024
June 6, 2024
May 9, 2024
December 6, 2023
November 9, 2023
September 18, 2023
August 2, 2023
June 8, 2023
March 13, 2023
December 21, 2022
December 2, 2022
November 21, 2022