Emergency Merge

该数据集描述了一个场景，其中汽车接近一个与救护车合并的路段，这引发了一个关于速度与碰撞风险的难题。该场景突显了代理能力的不对称性以及由相互冲突的激励所造成的困境。规模上，数据集包含了n-1辆汽车以及一辆救护车。任务则是多代理强化学习。This dataset depicts a scenario where vehicles approach a road segment requiring merging with an ambulance, creating a challenging problem related to speed and collision risk. This scenario highlights the asymmetry of agent capabilities and the dilemmas caused by conflicting incentives. In terms of scale, the dataset contains n-1 regular cars and one ambulance. The corresponding task is multi-agent reinforcement learning.