交通运输工程与信息学报2025，Vol.23Issue(3)：115-129,15.DOI:10.19961/j.cnki.1672-4747.2025.06.019
ICAROUS框架下多旋翼无人机应急自主着陆轨迹优化
Trajectory optimization for autonomous emergency landing of multirotor UAVs under the ICAROUS framework
摘要
Abstract
[Background]With the increase in the frequency of low-altitude flight activities and high failure rate of unmanned aerial vehicle(UAV)systems,ensuring high-reliability emergency landing capabilities has become crucial.[Objective]To address the safety challenges of autonomous emer-gency landing for multirotor UAVs in complex urban airspaces in unexpected scenarios,this study proposes a hierarchical trajectory optimization approach for emergency decision making within the ICAROUS framework,with the aim of achieving efficient,autonomous,and safe emergency land-ings.[Methods]The proposed approach adopts an integrated optimization model that combines high-level path planning with low-level velocity profile optimization.At the high level,the A*algorithm is utilized along with climb cost and flight risk assessment to generate an initial emergency path,which is smoothed and reconstructed using Bézier curves.At the low level,a Markov decision process model is employed to define the state space based on path segments and velocities,with acceleration as the action.A suitable reward function and state transition probabilities are designed,and the opti-mal velocity profile is derived using the value iteration algorithm.[Conclusions]Simulation results show that,compared with the path planning algorithm in ICAROUS,the proposed emergency plan-ning algorithm better balances the distance,risk,and climb cost among multiple candidate landing sites and reduces the risk metric by 28.8%.Furthermore,it selects the optimal landing point and gen-erates paths with controllable risk.Compared with sequential quadratic programming and particle swarm optimization,the generated velocity profiles exhibit improved smoothness and energy effi-ciency with 10.7%and 26.8%lower flight times,respectively.The average velocity fluctuation re-duces by more than 60%,thereby significantly enhancing the flight safety and energy management performance.[Applications]The proposed method can be applied to various low-altitude urban flight scenarios for UAV emergency planning.It effectively supports UAVs in making rational deci-sions under emergency conditions and provides technical support for flight safety assurance in urban air-traffic-management systems.关键词
航空运输/轨迹优化/马尔可夫决策过程/多旋翼无人机/应急着陆/A*算法Key words
air transportation/trajectory optimization/markov decision process/multirotor UAVs/emergency landing/A*algorithm分类
航空航天引用本文复制引用
廖亿福,汤新民,顾俊伟,鲁其兴,汤盛家..ICAROUS框架下多旋翼无人机应急自主着陆轨迹优化[J].交通运输工程与信息学报,2025,23(3):115-129,15.基金项目
国家自然科学基金项目(52072174) （52072174）
高端外国专家引进计划项目(G2023202003L) （G2023202003L）
天津市科技计划项目(24JCZDJC00090) （24JCZDJC00090）
