Aiming at the terminal defense problem of aircraft,this paper proposes a method to simultaneously achieve terminal defense and seize the dominant position.The method employs aλ-return based reinforcement learning alg...Aiming at the terminal defense problem of aircraft,this paper proposes a method to simultaneously achieve terminal defense and seize the dominant position.The method employs aλ-return based reinforcement learning algorithm,which can be applied to the flight assistance decision-making system to improve the pilot’s survivability.First,we model the environment to simulate the interaction between air-to-air missiles and aircraft.Subsequently,we propose aλ-return based approach to improve the deep Q learning network(DQN),deep advantageous actor criticism(A2C),and proximity policy optimization(PPO)algorithms used to train manoeuvre strategies.The method employs an action space containing nine manoeuvres and defines the off-target distance at the end of the scene as a sparse reward for algorithm training.Simulation results show that the convergence speed of the three improved algorithms is significantly improved when using theλ-return method.Moreover,the effect of the fetch value on the convergence speed is verified by ablation experiments.In order to solve the illegal behavior problem in the training process,we also design a backtracking-based illegal behavior masking mechanism,which improves the data generation efficiency of the environment model and promotes effective algorithm training.展开更多
This paper addresses the challenge of accurately and timely determining the position of a train,with specific consideration given to the integration of the global navigation satellite system(GNSS)and inertial navigati...This paper addresses the challenge of accurately and timely determining the position of a train,with specific consideration given to the integration of the global navigation satellite system(GNSS)and inertial navigation system(INS).To overcome the increasing errors in the INS during interruptions in GNSS signals,as well as the uncertainty associated with process and measurement noise,a deep learning-based method for train positioning is proposed.This method combines convolutional neural networks(CNN),long short-term memory(LSTM),and the invariant extended Kalman filter(IEKF)to enhance the perception of train positions.It effectively handles GNSS signal interruptions and mitigates the impact of noise.Experimental evaluation and comparisons with existing approaches are provided to illustrate the effectiveness and robustness of the proposed method.展开更多
In order to solve the problem of limited computational resources of multi-unmanned systems airborne navigation platform,a distributed cooperative positioning method based on confidence evaluation is proposed.Firstly,t...In order to solve the problem of limited computational resources of multi-unmanned systems airborne navigation platform,a distributed cooperative positioning method based on confidence evaluation is proposed.Firstly,the impact of ranging error,priori information,spatial geometric configuration and adjacent nodes count on cooperative positioning performance are analyzed individually.Secondly,a confidence evaluation method for measurement information of adjacent nodes is designed according to the cooperative positioning principle,which comprehensively considers the coupling relationship between influencing factors.Finally,a distributed cooperative navigation filter based on inter-vehicle ranging is designed.Simulation studies show that confidence evaluation method proposed in this paper can effectively characterize the contribution of measurement information to positioning results,and positioning accuracy under the proposed method is improved by more than 15%compared with the traditional screening methods based on optimal geometric configuration and closest distance.展开更多
High-performance positioning,navigation and timing(PNT)service is critical to the safe flight of low-altitude aircraft and the effective management of low altitude traffic.In low-altitude economic sce-narios,the speci...High-performance positioning,navigation and timing(PNT)service is critical to the safe flight of low-altitude aircraft and the effective management of low altitude traffic.In low-altitude economic sce-narios,the specificity of massive unmanned aerial ve-hicle(UAV)flights and the complexity of low-altitude airspace traffic management impose stringent demand on the high-continuity,high-accuracy,real-time,and high-security PNT service.However,the current PNT service,which primarily relies on Global Navigation Satellite System(GNSS),Micro-Electro-Mechanical System Inertial Navigation System(MEMS INS),etc.,is completely inadequate to support the future needs of low-altitude economic development.In order to bridge the huge gap between existing capability and future demand,a three-layer PNT architecture based on the collaboration of space-based,air-based and ground-based PNT systems is proposed for low-altitude econ-omy.The space-based layer consists of high,medium even possible low orbit GNSS constellations,such as BeiDou Navigation Satellite System(BDS),for high-precision,high-security absolute positioning and tim-ing.The air-based layer leverages inter-aircraft links for high-reliability dynamic relative positioning.The ground-based layer includes pseudolite network,as well as 5G-advanced(5G-A)/6G network,for more comprehensive coverage and real-time positioning.To this end,it is imperative to make breakthroughs in key technologies,from systems to airborne terminal,in-cluding but not limited to high-precision anti-jamming GNSS signal processing,high-reliability relative po-sitioning,real-time pseudolite positioning,and high-efficient multi-source information fusion at airborne terminal,etc.Due to the moderate redundancy,hetero-geneous mechanism,and multiple coverage from mul-tiple PNT systems,the proposed layered PNT archi-tecture possesses high robustness and resilient.Addi-tionally,the integration of INS,LiDAR and vision etc.perception technologies can significantly enhance the PNT capability.As a result,the proposed three-layer PNT architecture enable greater autonomy for low-altitude aircraft and intelligent traffic management for massive UAV operations,and promoting the safe and efficient development of the low-altitude economy.展开更多
We introduce a concept for the majorization order on monomials. With the help of this order, we derive a necessary condition on the positive termination of a general successive difference substitution algorithm （KSDS...We introduce a concept for the majorization order on monomials. With the help of this order, we derive a necessary condition on the positive termination of a general successive difference substitution algorithm （KSDS） for an input form f.展开更多
文摘Aiming at the terminal defense problem of aircraft,this paper proposes a method to simultaneously achieve terminal defense and seize the dominant position.The method employs aλ-return based reinforcement learning algorithm,which can be applied to the flight assistance decision-making system to improve the pilot’s survivability.First,we model the environment to simulate the interaction between air-to-air missiles and aircraft.Subsequently,we propose aλ-return based approach to improve the deep Q learning network(DQN),deep advantageous actor criticism(A2C),and proximity policy optimization(PPO)algorithms used to train manoeuvre strategies.The method employs an action space containing nine manoeuvres and defines the off-target distance at the end of the scene as a sparse reward for algorithm training.Simulation results show that the convergence speed of the three improved algorithms is significantly improved when using theλ-return method.Moreover,the effect of the fetch value on the convergence speed is verified by ablation experiments.In order to solve the illegal behavior problem in the training process,we also design a backtracking-based illegal behavior masking mechanism,which improves the data generation efficiency of the environment model and promotes effective algorithm training.
基金supported by the National Natural Science Foundation of China(Nos.61925302,62273027)the Beijing Natural Science Foundation(L211021).
文摘This paper addresses the challenge of accurately and timely determining the position of a train,with specific consideration given to the integration of the global navigation satellite system(GNSS)and inertial navigation system(INS).To overcome the increasing errors in the INS during interruptions in GNSS signals,as well as the uncertainty associated with process and measurement noise,a deep learning-based method for train positioning is proposed.This method combines convolutional neural networks(CNN),long short-term memory(LSTM),and the invariant extended Kalman filter(IEKF)to enhance the perception of train positions.It effectively handles GNSS signal interruptions and mitigates the impact of noise.Experimental evaluation and comparisons with existing approaches are provided to illustrate the effectiveness and robustness of the proposed method.
基金supported in part by National Natural Science Foundation of China(Nos.62073163,62103285,62203228)National Defense Basic Research Program(No.JCKY2020605C009)+1 种基金Aeronautic Science Foundation of China(Nos.ASFC-2020Z071052001,202055052003)Foundation Strengthening Program Technology 173 Field Fund(No.2021-JCJQ-JJ-0308)。
文摘In order to solve the problem of limited computational resources of multi-unmanned systems airborne navigation platform,a distributed cooperative positioning method based on confidence evaluation is proposed.Firstly,the impact of ranging error,priori information,spatial geometric configuration and adjacent nodes count on cooperative positioning performance are analyzed individually.Secondly,a confidence evaluation method for measurement information of adjacent nodes is designed according to the cooperative positioning principle,which comprehensively considers the coupling relationship between influencing factors.Finally,a distributed cooperative navigation filter based on inter-vehicle ranging is designed.Simulation studies show that confidence evaluation method proposed in this paper can effectively characterize the contribution of measurement information to positioning results,and positioning accuracy under the proposed method is improved by more than 15%compared with the traditional screening methods based on optimal geometric configuration and closest distance.
基金supported in part by the National Key R&D Program of China under Grant 2021YFA0716600 and 2024ZD1300100in part by the National Natural Science Foundation of China under Grant 42274018,42425401,62371029,62271285 and U2233217.
文摘High-performance positioning,navigation and timing(PNT)service is critical to the safe flight of low-altitude aircraft and the effective management of low altitude traffic.In low-altitude economic sce-narios,the specificity of massive unmanned aerial ve-hicle(UAV)flights and the complexity of low-altitude airspace traffic management impose stringent demand on the high-continuity,high-accuracy,real-time,and high-security PNT service.However,the current PNT service,which primarily relies on Global Navigation Satellite System(GNSS),Micro-Electro-Mechanical System Inertial Navigation System(MEMS INS),etc.,is completely inadequate to support the future needs of low-altitude economic development.In order to bridge the huge gap between existing capability and future demand,a three-layer PNT architecture based on the collaboration of space-based,air-based and ground-based PNT systems is proposed for low-altitude econ-omy.The space-based layer consists of high,medium even possible low orbit GNSS constellations,such as BeiDou Navigation Satellite System(BDS),for high-precision,high-security absolute positioning and tim-ing.The air-based layer leverages inter-aircraft links for high-reliability dynamic relative positioning.The ground-based layer includes pseudolite network,as well as 5G-advanced(5G-A)/6G network,for more comprehensive coverage and real-time positioning.To this end,it is imperative to make breakthroughs in key technologies,from systems to airborne terminal,in-cluding but not limited to high-precision anti-jamming GNSS signal processing,high-reliability relative po-sitioning,real-time pseudolite positioning,and high-efficient multi-source information fusion at airborne terminal,etc.Due to the moderate redundancy,hetero-geneous mechanism,and multiple coverage from mul-tiple PNT systems,the proposed layered PNT archi-tecture possesses high robustness and resilient.Addi-tionally,the integration of INS,LiDAR and vision etc.perception technologies can significantly enhance the PNT capability.As a result,the proposed three-layer PNT architecture enable greater autonomy for low-altitude aircraft and intelligent traffic management for massive UAV operations,and promoting the safe and efficient development of the low-altitude economy.
基金Supported by the National Key Basic Research Project of China(Grant No.2011CB302402)the Fundamental Research Funds for the Central Universities,Southwest University for Nationalities(Grant No.12NZYTH04)
文摘We introduce a concept for the majorization order on monomials. With the help of this order, we derive a necessary condition on the positive termination of a general successive difference substitution algorithm （KSDS） for an input form f.
