用关键区域信息提升赛车局部决策的全局视野
QuayPoints: A Reasoning Framework to Bridge the Information Gap Between Global and Local Planning in Autonomous Racing
- 通过QuayPoints区域传递最优路径的时间优化信息
- 在4条赛道上实现对75%车速对手的稳定超车
- 适合关注自动驾驶决策与规划融合的研究者
自动驾驶赛车需要感知、规划与控制的紧密协同以降低延迟并实现实时决策。标准自动化流程中，全局规划器的上下文信息逐级传递至任务导向的局部规划器时不断丢失。尤其，全局规划器对最优性的理解常被简化为稀疏路点，导致局部规划器在缺乏全局上下文的情况下做出反应式决策。本文探究将额外的全局信息——特别是时间最优性——有效传递给局部规划器是否可行。我们提出一个框架，通过定义QuayPoints区域来保留关键全局知识：这些区域中偏离最优赛道将显著损害整体效率。由此，局部规划器可在偏离赛道时（如超车）做出更具全局意识的决策。我们将该框架集成至现有规划系统，在四条不同赛道上验证，能持续超越速度达自身75%的对手。
原文摘要 · Abstract (English)
Autonomous racing requires tight integration between perception, planning and control to minimize latency as well as timely decision making. A standard autonomy pipeline comprising a global planner, local planner, and controller loses information as the higher-level racing context is sequentially propagated downstream into specific task-oriented context. In particular, the global planner's understanding of optimality is typically reduced to a sparse set of waypoints, leaving the local planner to make reactive decisions with limited context. This paper investigates whether additional global insights, specifically time-optimality information, can be meaningfully passed to the local planner to improve downstream decisions. We introduce a framework that preserves essential global knowledge and conveys it to the local planner through QuayPoints regions where deviations from the optimal raceline result in significant compromises to optimality. QuayPoints enable local planners to make more informed global decisions when deviating from the raceline, such as during strategic overtaking. To demonstrate this, we integrate QuayPoints into an existing planner and show that it consistently overtakes opponents traveling at up to 75% of the ego vehicle's speed across four distinct race tracks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。