Autonomous Driving
Autonomous driving research aims to develop vehicles capable of navigating and operating without human intervention, prioritizing safety and efficiency. Current efforts heavily focus on improving perception (using techniques like 3D Gaussian splatting and Bird's-Eye-View representations), prediction (leveraging diffusion models, transformers, and Bayesian games to handle uncertainty), and planning (employing reinforcement learning, large language models, and hierarchical approaches for decision-making). These advancements are crucial for enhancing the reliability and safety of autonomous vehicles, with significant implications for transportation systems and the broader AI community.
Papers
In Rain or Shine: Understanding and Overcoming Dataset Bias for Improving Robustness Against Weather Corruptions for Autonomous Vehicles
Aboli Marathe, Rahee Walambe, Ketan Kotecha
Autonomous Highway Merging in Mixed Traffic Using Reinforcement Learning and Motion Predictive Safety Controller
Qianqian Liu, Fengying Dang, Xiaofan Wang, Xiaoqiang Ren
Transferring Multi-Agent Reinforcement Learning Policies for Autonomous Driving using Sim-to-Real
Eduardo Candela, Leandro Parada, Luis Marques, Tiberiu-Andrei Georgescu, Yiannis Demiris, Panagiotis Angeloudis
Dense Residual Networks for Gaze Mapping on Indian Roads
Chaitanya Kapoor, Kshitij Kumar, Soumya Vishnoi, Sriram Ramanathan
Distributed Learning for Vehicular Dynamic Spectrum Access in Autonomous Driving
Pawe\{l} Sroka, Adrian Kliks
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
Yurong You, Katie Z Luo, Xiangyu Chen, Junan Chen, Wei-Lun Chao, Wen Sun, Bharath Hariharan, Mark Campbell, Kilian Q. Weinberger