This paper studies the moving path following(MPF)problem for fixed-wing unmanned aerial vehicle(UAV)under output constraints and wind disturbances.The vehicle is required to converge to a reference path moving with re...This paper studies the moving path following(MPF)problem for fixed-wing unmanned aerial vehicle(UAV)under output constraints and wind disturbances.The vehicle is required to converge to a reference path moving with respect to the inertial frame,while the path following error is not expected to violate the predefined boundaries.Differently from existing moving path following guidance laws,the proposed method removes complex geometric transformation by formulating the moving path following problem into a second-order time-varying control problem.A nominal moving path following guidance law is designed with disturbances and their derivatives estimated by high-order disturbance observers.To guarantee that the path following error will not exceed the prescribed bounds,a robust control barrier function is developed and incorporated into controller design with quadratic program based framework.The proposed method does not require the initial position of the UAV to be within predefined boundaries.And the safety margin concept makes error-constraint be respected even if in a noisy environment.The proposed guidance law is validated through numerical simulations of shipboard landing and hardware-in-theloop(HIL)experiments.展开更多
This paper presents a Fuzzy Preference Function-based Robust Multidisciplinary Design Optimization(FPF-RMDO) methodology. This method is an effective approach to multidisciplinary systems, which can be used to designe...This paper presents a Fuzzy Preference Function-based Robust Multidisciplinary Design Optimization(FPF-RMDO) methodology. This method is an effective approach to multidisciplinary systems, which can be used to designer experiences during the design optimization process by fuzzy preference functions. In this study, two optimizations are done for Predator MQ-1 Unmanned Aerial Vehicle(UAV):(A) deterministic optimization and(B) robust optimization. In both problems, minimization of takeoff weight and drag is considered as objective functions, which have been optimized using Non-dominated Sorting Genetic Algorithm(NSGA). In the robust design optimization, cruise altitude and velocity are considered as uncertainties that are modeled by the Monte Carlo Simulation(MCS) method. Aerodynamics, stability and control, mass properties, performance, and center of gravity are used for multidisciplinary analysis. Robust design optimization results show 46% and 42% robustness improvement for takeoff weight and cruise drag relative to optimal design respectively.展开更多
This paper deals with the robust stabilization problem for a class of nonlinear systems with structural uncertainty. Based on robust control Lyapunov function, a sufficient and necessary condition for a function to be...This paper deals with the robust stabilization problem for a class of nonlinear systems with structural uncertainty. Based on robust control Lyapunov function, a sufficient and necessary condition for a function to be a robust control Lyapunov function is given. From this condition, simply sufficient condition for the robust stabilization （robust practical stabilization） is deduced. Moreover, if the equilibrium of the closed-loop system is unique, the existence of such a robust control Lyapunnv function will also imply robustly globally asymptotical stabilization. Then a continuous state feedback law can be constructed explicitly. The simulation shows the effectiveness of the method.展开更多
Resistant training in radial basis function (RBF) networks is the topic of this paper. In this paper, one modification of Gauss-Newton training algorithm based on the theory of robust regression for dealing with outli...Resistant training in radial basis function (RBF) networks is the topic of this paper. In this paper, one modification of Gauss-Newton training algorithm based on the theory of robust regression for dealing with outliers in the framework of function approximation, system identification and control is proposed. This modification combines the numerical ro- bustness of a particular class of non-quadratic estimators known as M-estimators in Statistics and dead-zone. The al- gorithms is tested on some examples, and the results show that the proposed algorithm not only eliminates the influence of the outliers but has better convergence rate then the standard Gauss-Newton algorithm.展开更多
The problem of robust L 1 filtering with pole constraint in a disk for linear continuous polytopic uncertain systems is discussed. The attention is focused on design a linear asymptotically stable filter such that th...The problem of robust L 1 filtering with pole constraint in a disk for linear continuous polytopic uncertain systems is discussed. The attention is focused on design a linear asymptotically stable filter such that the filtering error system remains robustly stable, and has a L 1 performance constraint and pole constraint in a disk. The new robust L 1 performance criteria and regional pole placement condition are obtained via parameter-dependent Lyapunov functions method. Upon the proposed multiobjective performance criteria and by means of LMI technique, both full-order and reduced-order robust L 1 filter with suitable dynamic behavior can be obtained from the solution of convex optimization problems. Compared with earlier result in the quadratic framework, this approach turns out to be less conservative. The efficiency of the proposed technique is demonstrated by a numerical example.展开更多
为更精确地预测航班过站时间,将全国机场按照规模差异及不同地理位置所导致的客流量差异和天气差异对航班过站时间造成的不同影响进行分类,基于各类机场航班数据,构建混合轻量级梯度提升机算法(LightGBM)模型对航班过站时间分类预测。...为更精确地预测航班过站时间,将全国机场按照规模差异及不同地理位置所导致的客流量差异和天气差异对航班过站时间造成的不同影响进行分类,基于各类机场航班数据,构建混合轻量级梯度提升机算法(LightGBM)模型对航班过站时间分类预测。引入自适应鲁棒损失函数(adaptive robust loss function,ARLF)改进LightGBM模型损失函数,降低航班数据中存在离群值的影响;通过改进的麻雀搜索算法对改进后的LightGBM模型进行参数寻优,形成混合LightGBM模型。采用全国2019年全年航班数据进行验证,实验结果验证了方法的可行性。展开更多
当前基于深度学习的地表覆盖分类方法依赖于大规模且标注精准的训练样本。但受限于成本和技术因素,训练样本不可避免地混入噪声标签,导致分类精度降低。因此,本文提出了融合协同学习和抗噪损失的含噪地表覆盖分类方法。该方法以协同学...当前基于深度学习的地表覆盖分类方法依赖于大规模且标注精准的训练样本。但受限于成本和技术因素,训练样本不可避免地混入噪声标签,导致分类精度降低。因此,本文提出了融合协同学习和抗噪损失的含噪地表覆盖分类方法。该方法以协同学习机制为主体架构,首先利用参数非共享的双支卷积网络分别提取影像初分类特征;然后,基于双支网络的影像分类概率建模干净数据和噪声数据,构建基于信息熵的噪声可信度评价指标;最后,以可信度评价指标代替人工设定权重的方式,提出基于噪声可信度的自适应主动被动损失函数,引导协同学习网络关注噪声样本。试验表明,该方法在公开数据集GID(模拟噪声)和众源数据OSM(实际噪声)上的平均交并比(mean intersection over union,mIOU)分别提升3.85%~23.3%和5.89%~6.73%,说明该方法具有更好的抗噪性能。展开更多
基金supported in part by the National Natural Science Foundations of China(62173016,62073019)the Fundamental Research Funds for the Central Universities(YWF-23-JC-04,YWF-23-JC-02)。
文摘This paper studies the moving path following(MPF)problem for fixed-wing unmanned aerial vehicle(UAV)under output constraints and wind disturbances.The vehicle is required to converge to a reference path moving with respect to the inertial frame,while the path following error is not expected to violate the predefined boundaries.Differently from existing moving path following guidance laws,the proposed method removes complex geometric transformation by formulating the moving path following problem into a second-order time-varying control problem.A nominal moving path following guidance law is designed with disturbances and their derivatives estimated by high-order disturbance observers.To guarantee that the path following error will not exceed the prescribed bounds,a robust control barrier function is developed and incorporated into controller design with quadratic program based framework.The proposed method does not require the initial position of the UAV to be within predefined boundaries.And the safety margin concept makes error-constraint be respected even if in a noisy environment.The proposed guidance law is validated through numerical simulations of shipboard landing and hardware-in-theloop(HIL)experiments.
文摘This paper presents a Fuzzy Preference Function-based Robust Multidisciplinary Design Optimization(FPF-RMDO) methodology. This method is an effective approach to multidisciplinary systems, which can be used to designer experiences during the design optimization process by fuzzy preference functions. In this study, two optimizations are done for Predator MQ-1 Unmanned Aerial Vehicle(UAV):(A) deterministic optimization and(B) robust optimization. In both problems, minimization of takeoff weight and drag is considered as objective functions, which have been optimized using Non-dominated Sorting Genetic Algorithm(NSGA). In the robust design optimization, cruise altitude and velocity are considered as uncertainties that are modeled by the Monte Carlo Simulation(MCS) method. Aerodynamics, stability and control, mass properties, performance, and center of gravity are used for multidisciplinary analysis. Robust design optimization results show 46% and 42% robustness improvement for takeoff weight and cruise drag relative to optimal design respectively.
基金Sponsored by the Natural Science Foundation of Zhejiang Province in China(Grant No. Y105141).
文摘This paper deals with the robust stabilization problem for a class of nonlinear systems with structural uncertainty. Based on robust control Lyapunov function, a sufficient and necessary condition for a function to be a robust control Lyapunov function is given. From this condition, simply sufficient condition for the robust stabilization （robust practical stabilization） is deduced. Moreover, if the equilibrium of the closed-loop system is unique, the existence of such a robust control Lyapunnv function will also imply robustly globally asymptotical stabilization. Then a continuous state feedback law can be constructed explicitly. The simulation shows the effectiveness of the method.
文摘Resistant training in radial basis function (RBF) networks is the topic of this paper. In this paper, one modification of Gauss-Newton training algorithm based on the theory of robust regression for dealing with outliers in the framework of function approximation, system identification and control is proposed. This modification combines the numerical ro- bustness of a particular class of non-quadratic estimators known as M-estimators in Statistics and dead-zone. The al- gorithms is tested on some examples, and the results show that the proposed algorithm not only eliminates the influence of the outliers but has better convergence rate then the standard Gauss-Newton algorithm.
文摘The problem of robust L 1 filtering with pole constraint in a disk for linear continuous polytopic uncertain systems is discussed. The attention is focused on design a linear asymptotically stable filter such that the filtering error system remains robustly stable, and has a L 1 performance constraint and pole constraint in a disk. The new robust L 1 performance criteria and regional pole placement condition are obtained via parameter-dependent Lyapunov functions method. Upon the proposed multiobjective performance criteria and by means of LMI technique, both full-order and reduced-order robust L 1 filter with suitable dynamic behavior can be obtained from the solution of convex optimization problems. Compared with earlier result in the quadratic framework, this approach turns out to be less conservative. The efficiency of the proposed technique is demonstrated by a numerical example.
文摘为更精确地预测航班过站时间,将全国机场按照规模差异及不同地理位置所导致的客流量差异和天气差异对航班过站时间造成的不同影响进行分类,基于各类机场航班数据,构建混合轻量级梯度提升机算法(LightGBM)模型对航班过站时间分类预测。引入自适应鲁棒损失函数(adaptive robust loss function,ARLF)改进LightGBM模型损失函数,降低航班数据中存在离群值的影响;通过改进的麻雀搜索算法对改进后的LightGBM模型进行参数寻优,形成混合LightGBM模型。采用全国2019年全年航班数据进行验证,实验结果验证了方法的可行性。
文摘当前基于深度学习的地表覆盖分类方法依赖于大规模且标注精准的训练样本。但受限于成本和技术因素,训练样本不可避免地混入噪声标签,导致分类精度降低。因此,本文提出了融合协同学习和抗噪损失的含噪地表覆盖分类方法。该方法以协同学习机制为主体架构,首先利用参数非共享的双支卷积网络分别提取影像初分类特征;然后,基于双支网络的影像分类概率建模干净数据和噪声数据,构建基于信息熵的噪声可信度评价指标;最后,以可信度评价指标代替人工设定权重的方式,提出基于噪声可信度的自适应主动被动损失函数,引导协同学习网络关注噪声样本。试验表明,该方法在公开数据集GID(模拟噪声)和众源数据OSM(实际噪声)上的平均交并比(mean intersection over union,mIOU)分别提升3.85%~23.3%和5.89%~6.73%,说明该方法具有更好的抗噪性能。
