Scholarlink

This study proposes a vision-enhanced heterogeneous unmanned system framework integrating unmanned aerial vehicles (UAVs) and unmanned surface vehicles (USVs). The proposed approach consists of two key innovations: First, a UAV vision-guided cooperative navigation method is developed to achieve real-time global situational awareness and formation posture perception, enabling dynamic path renormalization for USV formations. Second, a COLREGs (Convention on the International Regulations for Preventing Collisions at Sea) compliant cooperative obstacle avoidance strategy is formulated through systematic behavior rule encoding, establishing a hierarchical decision library that enhances the safety and regulatory compliance of multi-USV collaborative operations. Experimental validation conducted on the CoppeliaSim platform demonstrates the framework’s effectiveness in high-dynamic scenarios. This work advances marine swarm intelligence by bridging the perception-action gap through cross-domain UAV-USV coordination.

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

Cooperative Autonomous Obstacle Avoidance for UAV-USV Heterogeneous Unmanned Systems Based on UAV Visual Assistance

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Tingyi Wu,
  • Yong Tang,
  • Tianning Xu

摘要

This study proposes a vision-enhanced heterogeneous unmanned system framework integrating unmanned aerial vehicles (UAVs) and unmanned surface vehicles (USVs). The proposed approach consists of two key innovations: First, a UAV vision-guided cooperative navigation method is developed to achieve real-time global situational awareness and formation posture perception, enabling dynamic path renormalization for USV formations. Second, a COLREGs (Convention on the International Regulations for Preventing Collisions at Sea) compliant cooperative obstacle avoidance strategy is formulated through systematic behavior rule encoding, establishing a hierarchical decision library that enhances the safety and regulatory compliance of multi-USV collaborative operations. Experimental validation conducted on the CoppeliaSim platform demonstrates the framework’s effectiveness in high-dynamic scenarios. This work advances marine swarm intelligence by bridging the perception-action gap through cross-domain UAV-USV coordination.