The field of autonomous vehicles is rapidly advancing, with a focus on improving safety, efficiency, and decision-making in complex environments. Researchers are exploring innovative approaches to motion planning, prediction, and control, including the integration of machine learning, computer vision, and sensor data. A key direction is the development of more sophisticated and human-like behavior in autonomous vehicles, enabling them to better interact with surrounding traffic and pedestrians. Noteworthy papers in this area include: DPNet, which proposes a novel motion planning framework that leverages Doppler LiDAR and model-based learning to track and react to rapid obstacles. MPCFormer, which introduces an explainable socially-aware autonomous driving approach that explicitly models the dynamics of multi-vehicle social interactions. BIBeR, which unifies motion prediction and game-theoretic planning into a single interaction-aware process, enabling bidirectional adaptation between the ego vehicle and surrounding agents.
Autonomous Vehicle Research Advancements
Sources
DPNet: Doppler LiDAR Motion Planning for Highly-Dynamic Environments
Cooperative Safety Intelligence in V2X-Enabled Transportation: A Survey
SemAgent: Semantic-Driven Agentic AI Empowered Trajectory Prediction in Vehicular Networks
Semantic Communications for Vehicle-Based Mission-Critical Services: Challenges and Solutions
SocialDriveGen: Generating Diverse Traffic Scenarios with Controllable Social Interactions
Velocity-Adaptive Access Scheme for Semantic-Aware Vehicular Networks: Joint Fairness and AoI Optimization
New Spiking Architecture for Multi-Modal Decision-Making in Autonomous Vehicles
Multi-Domain Enhanced Map-Free Trajectory Prediction with Selective Attention
CogDrive: Cognition-Driven Multimodal Prediction-Planning Fusion for Safe Autonomy
Prediction-Driven Motion Planning: Route Integration Strategies in Attention-Based Prediction Models
MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
Driving is a Game: Combining Planning and Prediction with Bayesian Iterative Best Response