Advancements in Autonomous Vehicle Decision Making

The field of autonomous vehicles is moving towards more sophisticated decision-making systems, with a focus on real-time autonomous racing, trust-aware lane-changing, and safe merging on highways. Researchers are developing innovative frameworks and models that incorporate game theory, machine learning, and human factors to improve the safety, efficiency, and cooperation of autonomous vehicles in complex traffic environments. Notable papers in this area include:

  • IteraOptiRacing, which presents a unified planning-control strategy for autonomous racing,
  • TGLD, which proposes a trust-aware game-theoretic lane-changing decision framework,
  • SMART-Merge Planner, which introduces a lattice-based motion planner for safe and comfortable highway on-ramp merging.

Sources

IteraOptiRacing: A Unified Planning-Control Framework for Real-time Autonomous Racing for Iterative Optimal Performance

TGLD: A Trust-Aware Game-Theoretic Lane-Changing Decision Framework for Automated Vehicles in Heterogeneous Traffic

SMART-Merge Planner: A Safe Merging and Real-Time Motion Planner for Autonomous Highway On-Ramp Merging

Collaborative Trustworthiness for Good Decision Making in Autonomous Systems

HCOMC: A Hierarchical Cooperative On-Ramp Merging Control Framework in Mixed Traffic Environment on Two-Lane Highways

A Cellular Automata Approach to Donation Game

MR-LDM -- The Merge-Reactive Longitudinal Decision Model: Game Theoretic Human Decision Modeling for Interactive Sim Agents

Built with on top of