Traditional weather observation methods have limitations in detecting low-altitude,small-scale areas and sudden weather events.They often have insufficient coverage,slow response,or high costs.Multi-rotor unmanned aer...Traditional weather observation methods have limitations in detecting low-altitude,small-scale areas and sudden weather events.They often have insufficient coverage,slow response,or high costs.Multi-rotor unmanned aerial vehicles(UAVs),with their strong vertical take-off and landing ability,precise hovering,flexible movement,and ability to carry various small sensors,are gradually becoming key tools to fill these gaps and build three-dimensional weather observation networks.They show important value in medium-and small-scale weather monitoring and emergency weather support.This paper reviews the main sensors for multi-rotor weather drones,their operating modes,and key supporting technologies,summarizes the current state of technology,and provides references for future development.展开更多
This paper shows that the aerodynamic effects can be compensated in a quadrotor system by means of a control allocation approach using neural networks.Thus,the system performance can be improved by replacing the class...This paper shows that the aerodynamic effects can be compensated in a quadrotor system by means of a control allocation approach using neural networks.Thus,the system performance can be improved by replacing the classic allocation matrix,without using the aerodynamic inflow equations directly.The network training is performed offline,which requires low computational power.The target system is a Parrot MAMBO drone whose flight control is composed of PD-PID controllers followed by the proposed neural network control allocation algorithm.Such a quadrotor is particularly susceptible to the aerodynamics effects of interest to this work,because of its small size.We compared the mechanical torques commanded by the flight controller,i.e.,the control input,to those actually generated by the actuators and established at the aircraft.It was observed that the proposed neural network was able to closely match them,while the classic allocation matrix could not achieve that.The allocation error was also determined in both cases.Furthermore,the closed-loop performance also improved with the use of the proposed neural network control allocation,as well as the quality of the thrust and torque signals,in which we perceived a much less noisy behavior.展开更多
A fusion chemical reaction optimization algorithm based on random molecules(RMCRO) is proposed to meet the special demand of power transmission line inspection. This new algorithm improves the shortcomings of chemical...A fusion chemical reaction optimization algorithm based on random molecules(RMCRO) is proposed to meet the special demand of power transmission line inspection. This new algorithm improves the shortcomings of chemical reaction algorithm by merging the idea of repellent-attractant rule and accelerates convergence by using difference algorithm. The molecules in this algorithm avoid obstacles and search optimal path of transmission line inspection by using sensors on multi-rotor unmanned aerial vehicle(UAV). The option of optimal path is based on potential energy of molecules and cost function without repeated parameter adjustment and complicated computation. By compared with an improved particle swarm optimization(IMPSO) in different circumstances of simulation, it can be concluded that the new algorithm presented not only can obtain more optimal path and avoid to trap in local minimum, but also can keep related sensors in a more stable status.展开更多
To fulfill the training requirements for the daily operations of multirotor unmanned aerial vehicles(UAVs)clusters,a UAV cluster collaborative task integrated simulation platform(UAV-TISP)was developed.The platform in...To fulfill the training requirements for the daily operations of multirotor unmanned aerial vehicles(UAVs)clusters,a UAV cluster collaborative task integrated simulation platform(UAV-TISP)was developed.The platform integrates a suite of hardware and software to simulate a range of collaborative UAV cluster operation scenarios.It features modules for collaborative task planning,UAV cluster simulations,and tactical monitoring.The platform significantly reduces training costs by eliminating physical drone dependencies while offering a flexible environment for testing swarm algorithms.UAV-TISP supports both individual UAV and swarm operations,incorporating high-fidelity flight dynamics,real-time communication via user datagram protocol(UDP),and collision avoidance strategies.Utilizing the OSGEarth engine,it enables dynamic 3D environment visualization and scenario customization.Three key task scenarios-route flight,formation reconstruction,and formation transformation-were tested to validate the platform’s efficacy.Results demonstrated robust formation maintenance,adaptive collision avoidance,and seamless task execution.Comparative analysis with Gazebo Sim revealed lower trajectory deviations in UAV-TISP,highlighting its superior accuracy in simulating real-world flight dynamics.Future work will focus on enhancing scalability for diverse UAV models,optimizing swarm networking under communication constraints,and expanding mission scenarios.UAV-TISP serves as a versatile tool for both operational training and advanced algorithm development in UAV cluster applications.展开更多
The downwash airflow field is an important factor influencing the spraying performance of plant protection UAV,and the structural design of rotors directly affects the characteristics of the downwash airflow field.The...The downwash airflow field is an important factor influencing the spraying performance of plant protection UAV,and the structural design of rotors directly affects the characteristics of the downwash airflow field.Therefore,in this study,three-dimensional models of a six-rotor UAV with various inner tilt angles were established to simulate and analyze the influence of the inner tilt angle on the downwash airflow field based on the Reynolds average NS equation,RNG k-εturbulence model,etc..On this basis,a wireless wind speed acquisition system using the TCP server was developed to carry out the test through the marked points with real-time detection.The simulation results show that,the variation of inner tilt angles of the six-rotor UAV did not cause significant difference in the time dimension of the downwash airflow field,and with the change of the inner tilt angle from 0°to 8°,the distribution of downwash airflow field tended to obliquely shrink towards the central axis direction,and the amplitude of linear attenuation of airflow speed was also increased,which the difference of attenuation amplitude was 1 m/s.Besides,under the different inner tilt angle,the airflow velocity in“lead in area”was significantly greater than that in the“lead out area”,and the difference of air velocity distribution in space would affect the uniformity of droplet deposition.Through the calibration test,the measurement accuracy error of the developed system was lower than 0.3 m/s,and the adjusted R2 of the calibration fitting equation was higher than 0.99.The test and simulation values at test points from 0.2-2.3 m below the rotors exhibit the same variation trend,and the average relative error at the height of 1.1-2.3 m below the rotors and 0.2-0.8 m near the ground was within 10%and 20%,respectively.The simulation and test results were highly reliable,which could provide basis and reference for the design and optimization of plant protection drones.展开更多
The application of deep learning for target detection in aerial images captured by Unmanned Aerial Vehicles(UAV)has emerged as a prominent research focus.Due to the considerable distance between UAVs and the photograp...The application of deep learning for target detection in aerial images captured by Unmanned Aerial Vehicles(UAV)has emerged as a prominent research focus.Due to the considerable distance between UAVs and the photographed objects,coupled with complex shooting environments,existing models often struggle to achieve accurate real-time target detection.In this paper,a You Only Look Once v8(YOLOv8)model is modified from four aspects:the detection head,the up-sampling module,the feature extraction module,and the parameter optimization of positive sample screening,and the YOLO-S3DT model is proposed to improve the performance of the model for detecting small targets in aerial images.Experimental results show that all detection indexes of the proposed model are significantly improved without increasing the number of model parameters and with the limited growth of computation.Moreover,this model also has the best performance compared to other detecting models,demonstrating its advancement within this category of tasks.展开更多
The environment of low-altitude urban airspace is complex and variable due to numerous obstacles,non-cooperative aircraft,and birds.Unmanned Aerial Vehicles(UAVs)leveraging environmental information to achieve three-d...The environment of low-altitude urban airspace is complex and variable due to numerous obstacles,non-cooperative aircraft,and birds.Unmanned Aerial Vehicles(UAVs)leveraging environmental information to achieve three-dimension collision-free trajectory planning is the prerequisite to ensure airspace security.However,the timely information of surrounding situation is difficult to acquire by UAVs,which further brings security risks.As a mature technology leveraged in traditional civil aviation,the Automatic Dependent Surveillance-Broadcast(ADS-B)realizes continuous surveillance of the information of aircraft.Consequently,we leverage ADS-B for surveillance and information broadcasting,and divide the aerial airspace into multiple sub-airspaces to improve flight safety in UAV trajectory planning.In detail,we propose the secure Sub-airSpaces Planning(SSP)algorithm and Particle Swarm Optimization Rapidly-exploring Random Trees(PSO-RRT)algorithm for the UAV trajectory planning in law-altitude airspace.The performance of the proposed algorithm is verified by simulations and the results show that SSP reduces both the maximum number of UAVs in the sub-airspace and the length of the trajectory,and PSO-RRT reduces the cost of UAV trajectory in the sub-airspace.展开更多
In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The e...In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The essence of cross-view geo-localization resides in matching images containing the same geographical targets from disparate platforms,such as UAV-view and satellite-view images.However,images of the same geographical targets may suffer from occlusions and geometric distortions due to variations in the capturing platform,view,and timing.The existing methods predominantly extract features by segmenting feature maps,which overlook the holistic semantic distribution and structural information of objects,resulting in loss of image information.To address these challenges,dilated neighborhood attention Transformer is employed as the feature extraction backbone,and Multi-feature representations based on Multi-scale Hierarchical Contextual Aggregation(MMHCA)is proposed.In the proposed MMHCA method,the multiscale hierarchical contextual aggregation method is utilized to extract contextual information from local to global across various granularity levels,establishing feature associations of contextual information with global and local information in the image.Subsequently,the multi-feature representations method is utilized to obtain rich discriminative feature information,bolstering the robustness of model in scenarios characterized by positional shifts,varying distances,and scale ambiguities.Comprehensive experiments conducted on the extensively utilized University-1652 and SUES-200 benchmarks indicate that the MMHCA method surpasses the existing techniques.showing outstanding results in UAV localization and navigation.展开更多
In this paper,we investigate a multi-UAV aided NOMA communication system,where multiple UAV-mounted aerial base stations are employed to serve ground users in the downlink NOMA communication,and each UAV serves its as...In this paper,we investigate a multi-UAV aided NOMA communication system,where multiple UAV-mounted aerial base stations are employed to serve ground users in the downlink NOMA communication,and each UAV serves its associated users on its own bandwidth.We aim at maximizing the overall common throughput in a finite time period.Such a problem is a typical mixed integer nonlinear problem,which involves both continuous-variable and combinatorial optimizations.To efficiently solve this problem,we propose a two-layer algorithm,which separately tackles continuous-variable and combinatorial optimization.Specifically,in the inner layer given one user association scheme,subproblems of bandwidth allocation,power allocation and trajectory design are solved based on alternating optimization.In the outer layer,a small number of candidate user association schemes are generated from an initial scheme and the best solution can be determined by comparing all the candidate schemes.In particular,a clustering algorithm based on K-means is applied to produce all candidate user association schemes,the successive convex optimization technique is adopted in the power allocation subproblem and a logistic function approximation approach is employed in the trajectory design subproblem.Simulation results show that the proposed NOMA scheme outperforms three baseline schemes in downlink common throughput,including one solution proposed in an existing literature.展开更多
Unmanned aerial vehicle(UAV)imagery poses significant challenges for object detection due to extreme scale variations,high-density small targets(68%in VisDrone dataset),and complex backgrounds.While YOLO-series models...Unmanned aerial vehicle(UAV)imagery poses significant challenges for object detection due to extreme scale variations,high-density small targets(68%in VisDrone dataset),and complex backgrounds.While YOLO-series models achieve speed-accuracy trade-offs via fixed convolution kernels and manual feature fusion,their rigid architectures struggle with multi-scale adaptability,as exemplified by YOLOv8n’s 36.4%mAP and 13.9%small-object AP on VisDrone2019.This paper presents YOLO-LE,a lightweight framework addressing these limitations through three novel designs:(1)We introduce the C2f-Dy and LDown modules to enhance the backbone’s sensitivity to small-object features while reducing backbone parameters,thereby improving model efficiency.(2)An adaptive feature fusion module is designed to dynamically integrate multi-scale feature maps,optimizing the neck structure,reducing neck complexity,and enhancing overall model performance.(3)We replace the original loss function with a distributed focal loss and incorporate a lightweight self-attention mechanism to improve small-object recognition and bounding box regression accuracy.Experimental results demonstrate that YOLO-LE achieves 39.9%mAP@0.5 on VisDrone2019,representing a 9.6%improvement over YOLOv8n,while maintaining 8.5 GFLOPs computational efficiency.This provides an efficient solution for UAV object detection in complex scenarios.展开更多
基金supported by the High-Level Talent Foundation of Natural Science Research Funding Project for Ordinary Universities in Jiangsu Province(grant number.25KJD520004)Jinling Institute of Technology(grant number.JIT-B-202413).
文摘Traditional weather observation methods have limitations in detecting low-altitude,small-scale areas and sudden weather events.They often have insufficient coverage,slow response,or high costs.Multi-rotor unmanned aerial vehicles(UAVs),with their strong vertical take-off and landing ability,precise hovering,flexible movement,and ability to carry various small sensors,are gradually becoming key tools to fill these gaps and build three-dimensional weather observation networks.They show important value in medium-and small-scale weather monitoring and emergency weather support.This paper reviews the main sensors for multi-rotor weather drones,their operating modes,and key supporting technologies,summarizes the current state of technology,and provides references for future development.
文摘This paper shows that the aerodynamic effects can be compensated in a quadrotor system by means of a control allocation approach using neural networks.Thus,the system performance can be improved by replacing the classic allocation matrix,without using the aerodynamic inflow equations directly.The network training is performed offline,which requires low computational power.The target system is a Parrot MAMBO drone whose flight control is composed of PD-PID controllers followed by the proposed neural network control allocation algorithm.Such a quadrotor is particularly susceptible to the aerodynamics effects of interest to this work,because of its small size.We compared the mechanical torques commanded by the flight controller,i.e.,the control input,to those actually generated by the actuators and established at the aircraft.It was observed that the proposed neural network was able to closely match them,while the classic allocation matrix could not achieve that.The allocation error was also determined in both cases.Furthermore,the closed-loop performance also improved with the use of the proposed neural network control allocation,as well as the quality of the thrust and torque signals,in which we perceived a much less noisy behavior.
基金the Fund of the Aeronautics Science(No.20162852031)the National Natural Science Foundation of China(No.61473144)
文摘A fusion chemical reaction optimization algorithm based on random molecules(RMCRO) is proposed to meet the special demand of power transmission line inspection. This new algorithm improves the shortcomings of chemical reaction algorithm by merging the idea of repellent-attractant rule and accelerates convergence by using difference algorithm. The molecules in this algorithm avoid obstacles and search optimal path of transmission line inspection by using sensors on multi-rotor unmanned aerial vehicle(UAV). The option of optimal path is based on potential energy of molecules and cost function without repeated parameter adjustment and complicated computation. By compared with an improved particle swarm optimization(IMPSO) in different circumstances of simulation, it can be concluded that the new algorithm presented not only can obtain more optimal path and avoid to trap in local minimum, but also can keep related sensors in a more stable status.
文摘To fulfill the training requirements for the daily operations of multirotor unmanned aerial vehicles(UAVs)clusters,a UAV cluster collaborative task integrated simulation platform(UAV-TISP)was developed.The platform integrates a suite of hardware and software to simulate a range of collaborative UAV cluster operation scenarios.It features modules for collaborative task planning,UAV cluster simulations,and tactical monitoring.The platform significantly reduces training costs by eliminating physical drone dependencies while offering a flexible environment for testing swarm algorithms.UAV-TISP supports both individual UAV and swarm operations,incorporating high-fidelity flight dynamics,real-time communication via user datagram protocol(UDP),and collision avoidance strategies.Utilizing the OSGEarth engine,it enables dynamic 3D environment visualization and scenario customization.Three key task scenarios-route flight,formation reconstruction,and formation transformation-were tested to validate the platform’s efficacy.Results demonstrated robust formation maintenance,adaptive collision avoidance,and seamless task execution.Comparative analysis with Gazebo Sim revealed lower trajectory deviations in UAV-TISP,highlighting its superior accuracy in simulating real-world flight dynamics.Future work will focus on enhancing scalability for diverse UAV models,optimizing swarm networking under communication constraints,and expanding mission scenarios.UAV-TISP serves as a versatile tool for both operational training and advanced algorithm development in UAV cluster applications.
基金supported by National Natural Science Foundation of China(Grant No.31801783).
文摘The downwash airflow field is an important factor influencing the spraying performance of plant protection UAV,and the structural design of rotors directly affects the characteristics of the downwash airflow field.Therefore,in this study,three-dimensional models of a six-rotor UAV with various inner tilt angles were established to simulate and analyze the influence of the inner tilt angle on the downwash airflow field based on the Reynolds average NS equation,RNG k-εturbulence model,etc..On this basis,a wireless wind speed acquisition system using the TCP server was developed to carry out the test through the marked points with real-time detection.The simulation results show that,the variation of inner tilt angles of the six-rotor UAV did not cause significant difference in the time dimension of the downwash airflow field,and with the change of the inner tilt angle from 0°to 8°,the distribution of downwash airflow field tended to obliquely shrink towards the central axis direction,and the amplitude of linear attenuation of airflow speed was also increased,which the difference of attenuation amplitude was 1 m/s.Besides,under the different inner tilt angle,the airflow velocity in“lead in area”was significantly greater than that in the“lead out area”,and the difference of air velocity distribution in space would affect the uniformity of droplet deposition.Through the calibration test,the measurement accuracy error of the developed system was lower than 0.3 m/s,and the adjusted R2 of the calibration fitting equation was higher than 0.99.The test and simulation values at test points from 0.2-2.3 m below the rotors exhibit the same variation trend,and the average relative error at the height of 1.1-2.3 m below the rotors and 0.2-0.8 m near the ground was within 10%and 20%,respectively.The simulation and test results were highly reliable,which could provide basis and reference for the design and optimization of plant protection drones.
文摘The application of deep learning for target detection in aerial images captured by Unmanned Aerial Vehicles(UAV)has emerged as a prominent research focus.Due to the considerable distance between UAVs and the photographed objects,coupled with complex shooting environments,existing models often struggle to achieve accurate real-time target detection.In this paper,a You Only Look Once v8(YOLOv8)model is modified from four aspects:the detection head,the up-sampling module,the feature extraction module,and the parameter optimization of positive sample screening,and the YOLO-S3DT model is proposed to improve the performance of the model for detecting small targets in aerial images.Experimental results show that all detection indexes of the proposed model are significantly improved without increasing the number of model parameters and with the limited growth of computation.Moreover,this model also has the best performance compared to other detecting models,demonstrating its advancement within this category of tasks.
基金supported by the National Key R&D Program of China(No.2022YFB3104502)the National Natural Science Foundation of China(No.62301251)+2 种基金the Natural Science Foundation of Jiangsu Province of China under Project(No.BK20220883)the open research fund of National Mobile Communications Research Laboratory,Southeast University,China(No.2024D04)the Young Elite Scientists Sponsorship Program by CAST(No.2023QNRC001).
文摘The environment of low-altitude urban airspace is complex and variable due to numerous obstacles,non-cooperative aircraft,and birds.Unmanned Aerial Vehicles(UAVs)leveraging environmental information to achieve three-dimension collision-free trajectory planning is the prerequisite to ensure airspace security.However,the timely information of surrounding situation is difficult to acquire by UAVs,which further brings security risks.As a mature technology leveraged in traditional civil aviation,the Automatic Dependent Surveillance-Broadcast(ADS-B)realizes continuous surveillance of the information of aircraft.Consequently,we leverage ADS-B for surveillance and information broadcasting,and divide the aerial airspace into multiple sub-airspaces to improve flight safety in UAV trajectory planning.In detail,we propose the secure Sub-airSpaces Planning(SSP)algorithm and Particle Swarm Optimization Rapidly-exploring Random Trees(PSO-RRT)algorithm for the UAV trajectory planning in law-altitude airspace.The performance of the proposed algorithm is verified by simulations and the results show that SSP reduces both the maximum number of UAVs in the sub-airspace and the length of the trajectory,and PSO-RRT reduces the cost of UAV trajectory in the sub-airspace.
基金supported by the National Natural Science Foundation of China(Nos.12072027,62103052,61603346 and 62103379)the Henan Key Laboratory of General Aviation Technology,China(No.ZHKF-230201)+3 种基金the Funding for the Open Research Project of the Rotor Aerodynamics Key Laboratory,China(No.RAL20200101)the Key Research and Development Program of Henan Province,China(Nos.241111222000 and 241111222900)the Key Science and Technology Program of Henan Province,China(No.232102220067)the Scholarship Funding from the China Scholarship Council(No.202206030079).
文摘In global navigation satellite system denial environment,cross-view geo-localization based on image retrieval presents an exceedingly critical visual localization solution for Unmanned Aerial Vehicle(UAV)systems.The essence of cross-view geo-localization resides in matching images containing the same geographical targets from disparate platforms,such as UAV-view and satellite-view images.However,images of the same geographical targets may suffer from occlusions and geometric distortions due to variations in the capturing platform,view,and timing.The existing methods predominantly extract features by segmenting feature maps,which overlook the holistic semantic distribution and structural information of objects,resulting in loss of image information.To address these challenges,dilated neighborhood attention Transformer is employed as the feature extraction backbone,and Multi-feature representations based on Multi-scale Hierarchical Contextual Aggregation(MMHCA)is proposed.In the proposed MMHCA method,the multiscale hierarchical contextual aggregation method is utilized to extract contextual information from local to global across various granularity levels,establishing feature associations of contextual information with global and local information in the image.Subsequently,the multi-feature representations method is utilized to obtain rich discriminative feature information,bolstering the robustness of model in scenarios characterized by positional shifts,varying distances,and scale ambiguities.Comprehensive experiments conducted on the extensively utilized University-1652 and SUES-200 benchmarks indicate that the MMHCA method surpasses the existing techniques.showing outstanding results in UAV localization and navigation.
基金supported by Beijing Natural Science Fund–Haidian Original Innovation Joint Fund(L232040 and L232045).
文摘In this paper,we investigate a multi-UAV aided NOMA communication system,where multiple UAV-mounted aerial base stations are employed to serve ground users in the downlink NOMA communication,and each UAV serves its associated users on its own bandwidth.We aim at maximizing the overall common throughput in a finite time period.Such a problem is a typical mixed integer nonlinear problem,which involves both continuous-variable and combinatorial optimizations.To efficiently solve this problem,we propose a two-layer algorithm,which separately tackles continuous-variable and combinatorial optimization.Specifically,in the inner layer given one user association scheme,subproblems of bandwidth allocation,power allocation and trajectory design are solved based on alternating optimization.In the outer layer,a small number of candidate user association schemes are generated from an initial scheme and the best solution can be determined by comparing all the candidate schemes.In particular,a clustering algorithm based on K-means is applied to produce all candidate user association schemes,the successive convex optimization technique is adopted in the power allocation subproblem and a logistic function approximation approach is employed in the trajectory design subproblem.Simulation results show that the proposed NOMA scheme outperforms three baseline schemes in downlink common throughput,including one solution proposed in an existing literature.
文摘Unmanned aerial vehicle(UAV)imagery poses significant challenges for object detection due to extreme scale variations,high-density small targets(68%in VisDrone dataset),and complex backgrounds.While YOLO-series models achieve speed-accuracy trade-offs via fixed convolution kernels and manual feature fusion,their rigid architectures struggle with multi-scale adaptability,as exemplified by YOLOv8n’s 36.4%mAP and 13.9%small-object AP on VisDrone2019.This paper presents YOLO-LE,a lightweight framework addressing these limitations through three novel designs:(1)We introduce the C2f-Dy and LDown modules to enhance the backbone’s sensitivity to small-object features while reducing backbone parameters,thereby improving model efficiency.(2)An adaptive feature fusion module is designed to dynamically integrate multi-scale feature maps,optimizing the neck structure,reducing neck complexity,and enhancing overall model performance.(3)We replace the original loss function with a distributed focal loss and incorporate a lightweight self-attention mechanism to improve small-object recognition and bounding box regression accuracy.Experimental results demonstrate that YOLO-LE achieves 39.9%mAP@0.5 on VisDrone2019,representing a 9.6%improvement over YOLOv8n,while maintaining 8.5 GFLOPs computational efficiency.This provides an efficient solution for UAV object detection in complex scenarios.
