.png){kind=icon}
Return
Advanced collaborative system for UAV navigation in human interactive environments
.png){kind=icon}
PRE
AI
DOI：10.1016/j.aei.2026.104815.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Unmanned aerial vehicles (UAVs) have demonstrated exceptional capabilities in environmental sensing and autonomous navigation across diverse real-world scenarios. However, in highly dynamic and human-interactive environments, UAVs still require effective human-robot collaboration to achieve shared autonomy (SA), balancing safety, efficiency, and operational awareness. Current SA approaches fail to adequately integrate human control intent while maintaining computational efficiency and user-friendly interfaces. This paper presents an Advanced Collaborative System (ACS) that addresses these challenges through two complementary innovations. First, we propose a dual-stage SA policy that integrates human control intent into both path planning and trajectory optimization. A human intent based potential field (HIPF) guides front-end path search, while a human-robot control intent consistency evaluation function (HICEF) ensures trajectory-intent alignment during back-end optimization. Second, we develop a portable Mixed Reality (MR) interface integrated with Large Language Model (LLM)-powered speech recognition, enabling intuitive UAV status visualization and natural language command interpretation. Comprehensive simulations and real-world experiments demonstrate that the proposed ACS achieves 96.63% lateral intent consistency success rate while maintaining real-time performance (7.34 ms average planning time). The system successfully enables context-aware UAV navigation in cluttered environments, ensuring trajectory compliance with human intent.
Keywords:
Motion planning
Mixed reality
Human-robot shared autonomy
Journal
IF:
9.9
Papers:
4.4K
Citations:
1.7W
Organization
Cited Papers
No cited papers available