July 22, 2026 · Wednesday2026 年 7 月 22 日 · No. 18 · updated更新于 02:06 () SIGNALS信号EVENTS日历ARCHIVE归档ACCESS接入ABOUT关于

AUTOSIGNAL

Human-curated. Expert-annotated. Every signal traced to source. 人工精选 · 专家点评 · 每条信号可溯源

← All signals← 返回全部信号

Research & IP研究与专利

Integrity-Gated Eco-CACC Adds Real-Time Model-Trust Monitoring to Signalized-Intersection Cooperative Driving完整性门控Eco-CACC框架为信号灯路口协同驾驶引入实时模型可信度监测

S 1.7 T1 1 sources1 个来源 R7-research
  1. In an arXiv paper (2607.18565v1, submitted July 20, 2026) on Eco-Cooperative Adaptive Cruise Control (Eco-CACC), authors Lyes Saad Saoud and Moussa Ayyash propose an "Integrity-Gated" framework that explicitly checks whether a vehicle's internal world model is still valid before using it for energy-optimal control at signalized intersections.
  2. The system computes a unified integrity metric from positional innovation, observability loss, and semantic inconsistency between the world model and sensor data, then uses the resulting trust score to switch control authority between eco-driving optimization and a conservative safety fallback.
  3. This targets a key real-world deployment gap for CACC at intersections: existing eco-driving controllers assume their localization, signal-timing, and interaction models stay valid, so they can keep optimizing on stale or corrupted premises during sensing outages, whereas the gated approach preserves fuel/energy savings in normal operation while automatically falling back to safety-first behavior when model integrity degrades.
  1. 一篇提交于2026年7月20日的arXiv论文(编号2607.18565v1)针对生态型协同自适应巡航控制(Eco-CACC),由作者Lyes Saad Saoud和Moussa Ayyash提出"完整性门控"框架,在信号灯路口用于能耗优化控制前,先明确检验车辆内部世界模型是否仍然有效。
  2. 该系统通过位置创新(positional innovation)、可观测性损失和语义不一致性构建统一的完整性指标,衡量世界模型与传感器数据的一致程度,再依据由此得到的信任评分,在生态驾驶优化与保守安全回退模式之间切换控制权。
  3. 该方法针对路口场景下Eco-CACC实际部署的关键短板:现有生态驾驶控制器默认定位、信号灯时序和交互模型始终有效,一旦传感中断或语义失配也可能继续基于失效前提做优化;而完整性门控方案能在正常状态下保留节能效果,并在模型可信度下降时自动切换为安全优先的保守响应。