Scholarlink

<p>Navigating safely in environments populated by dynamic and multi-modal agents poses significant challenges to risk assessment and management for autonomous vehicles (AV). In particular, we consider urban scenarios involving Personal Light Electric Vehicles (PLEVs) that are known to exhibit varying maneuvers and velocity profiles. We propose a Fusion of Predictive Inter-Distance Profile (F-PIDP) to predict the inter-distances and capture the uncertainties in the motion of the AV and the PLEV’s multiple trajectories. Building upon this, a multi-priority-based trajectory sampling algorithm is developed for collision avoidance. This strategy is generalized to assess and prioritize multiple agents with varying motion across several driving lanes, enabling the system to identify and respond to the most dangerous agents. The proposed method is then integrated into a Model Predictive Control (MPC) trajectory tracking framework to perform risk management for safe and adaptive control of the AV’s motion. The proposed method is validated through a series of simulations with critical driving scenarios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multi-Priority-Based Strategy for Risk Assessment and Management in the Presence of Multiple Personal Light Electric Vehicles

  • 期刊论文

  • 发表日期：

    2025年12月17日

查看全文
  • Emmanuel Alao,
  • Lounis Adouane,
  • Philippe Martinet

摘要

Navigating safely in environments populated by dynamic and multi-modal agents poses significant challenges to risk assessment and management for autonomous vehicles (AV). In particular, we consider urban scenarios involving Personal Light Electric Vehicles (PLEVs) that are known to exhibit varying maneuvers and velocity profiles. We propose a Fusion of Predictive Inter-Distance Profile (F-PIDP) to predict the inter-distances and capture the uncertainties in the motion of the AV and the PLEV’s multiple trajectories. Building upon this, a multi-priority-based trajectory sampling algorithm is developed for collision avoidance. This strategy is generalized to assess and prioritize multiple agents with varying motion across several driving lanes, enabling the system to identify and respond to the most dangerous agents. The proposed method is then integrated into a Model Predictive Control (MPC) trajectory tracking framework to perform risk management for safe and adaptive control of the AV’s motion. The proposed method is validated through a series of simulations with critical driving scenarios.