As one of the major contributions of biology to competitive decision making, evolutionary game theory provides a useful tool for studying the evolution of cooperation. To achieve the optimal solution for unmanned aeri...As one of the major contributions of biology to competitive decision making, evolutionary game theory provides a useful tool for studying the evolution of cooperation. To achieve the optimal solution for unmanned aerial vehicles （UAVs） that are car- rying out a sensing task, this paper presents a Markov decision evolutionary game （MDEG） based learning algorithm. Each in- dividual in the algorithm follows a Markov decision strategy to maximize its payoff against the well known Tit-for-Tat strate- gy. Simulation results demonstrate that the MDEG theory based approach effectively improves the collective payoff of the roam. The proposed algorithm can not only obtain the best action sequence but also a sub-optimal Markov policy that is inde- pendent of the game duration. Furthermore, the paper also studies the emergence of cooperation in the evolution of self-regarded UAVs. The results show that it is the adaptive ability of the MDEG based approach as well as the perfect balance between revenge and forgiveness of the Tit-for-Tat strategy that the emergence of cooperation should be attributed to.展开更多
In order to solve the problem of small objects detection in unmanned aerial vehicle(UAV)aerial images with complex background,a general detection method for multi-scale small objects based on Faster region-based convo...In order to solve the problem of small objects detection in unmanned aerial vehicle(UAV)aerial images with complex background,a general detection method for multi-scale small objects based on Faster region-based convolutional neural network(Faster R-CNN)is proposed.The bird’s nest on the high-voltage tower is taken as the research object.Firstly,we use the improved convolutional neural network ResNet101 to extract object features,and then use multi-scale sliding windows to obtain the object region proposals on the convolution feature maps with different resolutions.Finally,a deconvolution operation is added to further enhance the selected feature map with higher resolution,and then it taken as a feature mapping layer of the region proposals passing to the object detection sub-network.The detection results of the bird’s nest in UAV aerial images show that the proposed method can precisely detect small objects in aerial images.展开更多
Purpose–The purpose of this paper is to investigate time-optimal control problems for multiple unmanned aerial vehicle(UAV)systems to achieve predefined flying shape.Design/methodology/approach–Two time-optimal prot...Purpose–The purpose of this paper is to investigate time-optimal control problems for multiple unmanned aerial vehicle(UAV)systems to achieve predefined flying shape.Design/methodology/approach–Two time-optimal protocols are proposed for the situations with or without human control input,respectively.Then,Pontryagin’s minimum principle approach is applied to deal with the time-optimal control problems for UAV systems,where the cost function,the initial and terminal conditions are given in advance.Moreover,necessary conditions are derived to ensure that the given performance index is optimal.Findings–The effectiveness of the obtained time-optimal control protocols is verified by two contrastive numerical simulation examples.Consequently,the proposed protocolscan successfully achieve the prescribed flying shape.Originality/value–This paper proposes a solution to solve the time-optimal control problems for multiple UAV systems to achieve predefined flying shape.展开更多
The estimation of fractional vegetation cover(FVC) is important for identifying and monitoring desertification, especially in arid and semiarid regions. By using regression and pixel dichotomy models, we present the c...The estimation of fractional vegetation cover(FVC) is important for identifying and monitoring desertification, especially in arid and semiarid regions. By using regression and pixel dichotomy models, we present the comparison of Sentinel-2A(S2) multispectral instrument(MSI) and Landsat 8(L8) operational land imager(OLI) data regarding the retrieval of FVC in a semi-arid sandy area(Mu Us Sandland, China, in August 2016). A combination of unmanned aerial vehicle(UAV) high-spatial-resolution images and field plots were used to produce verified data. Based on a normalized difference vegetation index(NDVI) regression model, the results showed that, compared with that of L8, the coefficient of determination(R2) of S2 increased by 26.0%, and the root mean square error(RMSE) and the sum of absolute error(SAE) decreased by 3.0% and 11.4%, respectively. For the ratio vegetation index(RVI) regression model, compared with that of L8, the R2 of S2 increased by 26.0%, and the RMSE and SAE decreased by 8.0% and 20.0%, respectively. When the pixel dichotomy model was used, compared with that of L8, the RMSE of S2 decreased by 21.3%, and the SAE decreased by 26.9%. Overall, S2 performed better than L8 in terms of FVC inversion. Additionally, in this paper, we develop a verified scheme based on UAV data in combination with the object-based classification method. This scheme is feasible and sufficiently robust for building relationships between field data and inversion results from satellite data. Further, the synergy of multi-source sensors(especially UAVs and satellites) is a potential effective way to estimate and evaluate regional ecological environmental parameters(FVC).展开更多
The mathematical model of quadcopter-unmanned aerial vehicle （UAV） is derived by using two approaches： One is the Newton-Euler approach which is formulated using classical meehanics; and other is the Euler-Lagrange...The mathematical model of quadcopter-unmanned aerial vehicle （UAV） is derived by using two approaches： One is the Newton-Euler approach which is formulated using classical meehanics; and other is the Euler-Lagrange approach which describes the model in terms of kinetic （translational and rotational） and potential energy. The proposed quadcopter＇s non-linear model is incorporated with aero-dynamical forces generated by air resistance, which helps aircraft to exhibits more realistic behavior while hovering. Based on the obtained model, the suitable control strategy is developed, under which two effective flight control systems are developed. Each control system is created by cascading the proportional-derivative （PD） and T-S fuzzy controllers that are equipped with six and twelve feedback signals individually respectively to ensure better tracking, stabilization, and response. Both pro- posed flight control designs are then implemented with the quadcopter model respectively and multitudinous simulations are conducted using MATLAB/Simulink to analyze the tracking performance of the quadcopter model at various reference inputs and trajectories.展开更多
吊舱稳定平台在无人直升机电力巡线中发挥着重要作用。为了提高巡线质量和效率,克服自动跟踪单一模式下多源扰动导致的光学载荷视轴对目标的跟踪丢失,提出了一种基于自动/手动混合模式的吊舱稳定平台控制策略。通过组合使用基于POS(Posi...吊舱稳定平台在无人直升机电力巡线中发挥着重要作用。为了提高巡线质量和效率,克服自动跟踪单一模式下多源扰动导致的光学载荷视轴对目标的跟踪丢失,提出了一种基于自动/手动混合模式的吊舱稳定平台控制策略。通过组合使用基于POS(Position and Orientation System)的位置和姿态信息的自动跟踪控制模式和基于人工手柄操作的手动跟踪控制模式,当目标跟踪丢失时利用手动模式及时对自动跟踪进行校正,实现吊舱稳定平台载荷视轴对电力线路的长时高精度稳定跟踪。根据控制策略,设计了基于DSP和FPGA的三环复合伺服控制系统,通过仿真分析得到将手动控制设置在速率回路是最佳方案的结论。通过实际线路飞行实验对控制方法进行了验证,结果表明:基于自动/手动混合模式的平台控制对目标跟踪灵活准确,实现了无人机多传感器系统对电力线路的高效高精度数据采集。展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.61425008,61333004 and 61273054)Top-Notch Young Talents Program of China,and Aeronautical Foundation of China(Grant No.20135851042)
文摘As one of the major contributions of biology to competitive decision making, evolutionary game theory provides a useful tool for studying the evolution of cooperation. To achieve the optimal solution for unmanned aerial vehicles （UAVs） that are car- rying out a sensing task, this paper presents a Markov decision evolutionary game （MDEG） based learning algorithm. Each in- dividual in the algorithm follows a Markov decision strategy to maximize its payoff against the well known Tit-for-Tat strate- gy. Simulation results demonstrate that the MDEG theory based approach effectively improves the collective payoff of the roam. The proposed algorithm can not only obtain the best action sequence but also a sub-optimal Markov policy that is inde- pendent of the game duration. Furthermore, the paper also studies the emergence of cooperation in the evolution of self-regarded UAVs. The results show that it is the adaptive ability of the MDEG based approach as well as the perfect balance between revenge and forgiveness of the Tit-for-Tat strategy that the emergence of cooperation should be attributed to.
基金National Defense Pre-research Fund Project(No.KMGY318002531)。
文摘In order to solve the problem of small objects detection in unmanned aerial vehicle(UAV)aerial images with complex background,a general detection method for multi-scale small objects based on Faster region-based convolutional neural network(Faster R-CNN)is proposed.The bird’s nest on the high-voltage tower is taken as the research object.Firstly,we use the improved convolutional neural network ResNet101 to extract object features,and then use multi-scale sliding windows to obtain the object region proposals on the convolution feature maps with different resolutions.Finally,a deconvolution operation is added to further enhance the selected feature map with higher resolution,and then it taken as a feature mapping layer of the region proposals passing to the object detection sub-network.The detection results of the bird’s nest in UAV aerial images show that the proposed method can precisely detect small objects in aerial images.
基金supported by the National Natural Science Foundation of China under Grant Nos 61602163 and 61471163the Science Fund for Distinguished Young Scholars of Hubei Province under Grant No.2017CFA034the Natural Science Foundation of Hubei Province under 2016CFC735.The Hubei Education Department Science and Technology Research Program for young talents under Grant No.Q20182503.
文摘Purpose–The purpose of this paper is to investigate time-optimal control problems for multiple unmanned aerial vehicle(UAV)systems to achieve predefined flying shape.Design/methodology/approach–Two time-optimal protocols are proposed for the situations with or without human control input,respectively.Then,Pontryagin’s minimum principle approach is applied to deal with the time-optimal control problems for UAV systems,where the cost function,the initial and terminal conditions are given in advance.Moreover,necessary conditions are derived to ensure that the given performance index is optimal.Findings–The effectiveness of the obtained time-optimal control protocols is verified by two contrastive numerical simulation examples.Consequently,the proposed protocolscan successfully achieve the prescribed flying shape.Originality/value–This paper proposes a solution to solve the time-optimal control problems for multiple UAV systems to achieve predefined flying shape.
基金National Natural Science Foundation of China(No.41301451,41541008)Fundamental Research Funds for the Central Universities(No.2452018144)
文摘The estimation of fractional vegetation cover(FVC) is important for identifying and monitoring desertification, especially in arid and semiarid regions. By using regression and pixel dichotomy models, we present the comparison of Sentinel-2A(S2) multispectral instrument(MSI) and Landsat 8(L8) operational land imager(OLI) data regarding the retrieval of FVC in a semi-arid sandy area(Mu Us Sandland, China, in August 2016). A combination of unmanned aerial vehicle(UAV) high-spatial-resolution images and field plots were used to produce verified data. Based on a normalized difference vegetation index(NDVI) regression model, the results showed that, compared with that of L8, the coefficient of determination(R2) of S2 increased by 26.0%, and the root mean square error(RMSE) and the sum of absolute error(SAE) decreased by 3.0% and 11.4%, respectively. For the ratio vegetation index(RVI) regression model, compared with that of L8, the R2 of S2 increased by 26.0%, and the RMSE and SAE decreased by 8.0% and 20.0%, respectively. When the pixel dichotomy model was used, compared with that of L8, the RMSE of S2 decreased by 21.3%, and the SAE decreased by 26.9%. Overall, S2 performed better than L8 in terms of FVC inversion. Additionally, in this paper, we develop a verified scheme based on UAV data in combination with the object-based classification method. This scheme is feasible and sufficiently robust for building relationships between field data and inversion results from satellite data. Further, the synergy of multi-source sensors(especially UAVs and satellites) is a potential effective way to estimate and evaluate regional ecological environmental parameters(FVC).
基金supported by the National Natural Science Foundation of China(Nos.61673209,61741313,61304223)the Aeronautical Science Foundation(Nos.2016ZA52009)+1 种基金the Jiangsu Six Peak of Talents Program(No.KTHY-027)the Fundamental Research Funds for the Central Universities(Nos.NJ20160026,NS2017015)
文摘The mathematical model of quadcopter-unmanned aerial vehicle （UAV） is derived by using two approaches： One is the Newton-Euler approach which is formulated using classical meehanics; and other is the Euler-Lagrange approach which describes the model in terms of kinetic （translational and rotational） and potential energy. The proposed quadcopter＇s non-linear model is incorporated with aero-dynamical forces generated by air resistance, which helps aircraft to exhibits more realistic behavior while hovering. Based on the obtained model, the suitable control strategy is developed, under which two effective flight control systems are developed. Each control system is created by cascading the proportional-derivative （PD） and T-S fuzzy controllers that are equipped with six and twelve feedback signals individually respectively to ensure better tracking, stabilization, and response. Both pro- posed flight control designs are then implemented with the quadcopter model respectively and multitudinous simulations are conducted using MATLAB/Simulink to analyze the tracking performance of the quadcopter model at various reference inputs and trajectories.
文摘吊舱稳定平台在无人直升机电力巡线中发挥着重要作用。为了提高巡线质量和效率,克服自动跟踪单一模式下多源扰动导致的光学载荷视轴对目标的跟踪丢失,提出了一种基于自动/手动混合模式的吊舱稳定平台控制策略。通过组合使用基于POS(Position and Orientation System)的位置和姿态信息的自动跟踪控制模式和基于人工手柄操作的手动跟踪控制模式,当目标跟踪丢失时利用手动模式及时对自动跟踪进行校正,实现吊舱稳定平台载荷视轴对电力线路的长时高精度稳定跟踪。根据控制策略,设计了基于DSP和FPGA的三环复合伺服控制系统,通过仿真分析得到将手动控制设置在速率回路是最佳方案的结论。通过实际线路飞行实验对控制方法进行了验证,结果表明:基于自动/手动混合模式的平台控制对目标跟踪灵活准确,实现了无人机多传感器系统对电力线路的高效高精度数据采集。
