With the rapid growth of flight flow,the workload of controllers is increasing daily,and handling flight conflicts is the main workload.Therefore,it is necessary to provide more efficient conflict resolution decision-...With the rapid growth of flight flow,the workload of controllers is increasing daily,and handling flight conflicts is the main workload.Therefore,it is necessary to provide more efficient conflict resolution decision-making support for controllers.Due to the limitations of existing methods,they have not been widely used.In this paper,a Deep Reinforcement Learning(DRL)algorithm is proposed to resolve multi-aircraft flight conflict with high solving efficiency.First,the characteristics of multi-aircraft flight conflict problem are analyzed and the problem is modeled based on Markov decision process.Thus,the Independent Deep Q Network(IDQN)algorithm is used to solve the model.Simultaneously,a’downward-compatible’framework that supports dynamic expansion of the number of conflicting aircraft is designed.The model ultimately shows convergence through adequate training.Finally,the test conflict scenarios and indicators were used to verify the validity.In 700 scenarios,85.71%of conflicts were successfully resolved,and 71.51%of aircraft can reach destinations within 150 s around original arrival times.By contrast,conflict resolution algorithm based on DRL has great advantages in solution speed.The method proposed offers the possibility of decision-making support for controllers and reduce workload of controllers in future high-density airspace environment.展开更多
Multi-agent technology has been used in many complex distributed and concurrent systems. A railway system is such a safety critical system and careful inves- tigation of the functional components is very important. St...Multi-agent technology has been used in many complex distributed and concurrent systems. A railway system is such a safety critical system and careful inves- tigation of the functional components is very important. Study of the various functional components in communi- cation-based train control （CBTC） system necessitates a good structural design followed by its validation and ver- ification through a formal modelling technique. The work presented here is the follow up of our multi-agent-based CBTC system for Indian railway designed using the methodology for engineering system of software agents. Behavioural analysis of the designed system involves several operating scenarios that arise during train run, and helps in understanding the reaction of the system to such situations. This validation and verification are very important as it allows the system designer to critically evaluate the desired function of the system and to correct the design errors, if any, before its actual implementation. Modelling, validation and verification of the structural design through Coloured petri net （CPN） are central to this paper. Analysis of simulation results validates the efficacy of the design.展开更多
Racemic R.S-α-arylethylamine was resolved by R (-) thiazolidine-2-thione-4-carboxylic acid, a new resolving agent abbreviated as [R (-) TTCA], by which R (-) TTCA.S(-) arylethylamine salts2a-2e, [α] D 20 =-47.24...Racemic R.S-α-arylethylamine was resolved by R (-) thiazolidine-2-thione-4-carboxylic acid, a new resolving agent abbreviated as [R (-) TTCA], by which R (-) TTCA.S(-) arylethylamine salts2a-2e, [α] D 20 =-47.24° — 64.40° and optically active R(+)-a-arylethylamines3a – 3e, 74. 54%-94. 45% e, e., were obtained. Optically active S (-) -α-arylethylamines4a-4e, 72.84%-90.36% e.e., were obtained by the decomposition of2a-2e in basic solutions. The influence of substitutive group of the benzene ring on the basicity of the amino group was studied by semiempirical PM3 method. The structures of the R (-) TTCA.S (-) -α-phenylethylamine salt (2a(R-S) configuration) and R (-) TTCA-R(+)-a-phenylethylamine salt (2a(R-R) configuration) have been established by means of X-ray diffraction. They crystallize in a monoclinic system. Space group isP21. The cell constants of2a(R-S) configuration were obtained as follows: α = 1.387 8(2), b = 0.664 05(101,c = 1.580 O(2) nm; β = 90.844(10)° Z = 4; those obtained for2a(R-R) configuration were α = 1.080 6(2),b = 0.584 80(12),c = 1.2188(2) nm, β= 110.38(3)dg, V = 0.7220nm3,Z = 2. There are intermolecular hydrogen bonds in the crystals of the two kinds of configurations of the amine salt. The hydrogen bond number in the unit cell of R (-) TTCA.S (-)-α-phenylethylamine salt is twice as much as that of R (-) TTCA.R(+)-a-phenylethylamine salt.展开更多
在自由飞行的环境下,为解决飞行冲突探测与解脱(conflict detection and resolution,CDR)问题,提出一种基于高度层、航向和速度调配的综合解脱方法,并将多agent系统(multi-agent system,MAS)的分布式技术与启发式算法相结合,进行问题求...在自由飞行的环境下,为解决飞行冲突探测与解脱(conflict detection and resolution,CDR)问题,提出一种基于高度层、航向和速度调配的综合解脱方法,并将多agent系统(multi-agent system,MAS)的分布式技术与启发式算法相结合,进行问题求解.首先设计了分布式MAS框架结构,然后建立了飞行冲突探测模型,高度层调配模型及航向、速度调配模型,最后,综合运用了基于合同网协议的分布式算法和自适应遗传算法进行问题求解.仿真实验表明,所设计的MAS框架是可行的,同时分布式算法和自适应遗传算法的综合应用能很快找到基于高度层、航向和速度分配的近似最优解,为CDR问题提供了新的解决思路.展开更多
为解决固定航路上飞行冲突探测与解脱(conflict detection and resolution,CDR)的问题,本文提出一种基于高度层分配的解脱方法,并利用分布式多agent系统(multi-agent system,MAS)进行算法求解。首先建立固定航路网络图对管制扇区进行建...为解决固定航路上飞行冲突探测与解脱(conflict detection and resolution,CDR)的问题,本文提出一种基于高度层分配的解脱方法,并利用分布式多agent系统(multi-agent system,MAS)进行算法求解。首先建立固定航路网络图对管制扇区进行建模;然后分析影响高度层分配的主要因素,并建立高度层使用优先权的评价模型;最后,设计基于合同网协议的多agent系统,将目前依靠管制员的集中调配模式,转变为路口agent和航空器agent之间自主进行通信、协商和协作的分布式冲突解脱模式。仿真实验表明:高度层分配方法是可行的,与传统调整航向或速度的方法相比,该方法更接近实际情况,同时设计的分布式多agent系统算法能够快速找到高度层分配的最优解,为CDR问题提供了新的解决思路。展开更多
基金supported by Safety Ability Project of Civil Aviation Administration of China(No.TM 2018-5-1/2)the Open Foundation project of The Graduate Student Innovation Base,China(Laboratory)of Nanjing University of Aeronautics and Astronautics,China(No.kfjj20190720)。
文摘With the rapid growth of flight flow,the workload of controllers is increasing daily,and handling flight conflicts is the main workload.Therefore,it is necessary to provide more efficient conflict resolution decision-making support for controllers.Due to the limitations of existing methods,they have not been widely used.In this paper,a Deep Reinforcement Learning(DRL)algorithm is proposed to resolve multi-aircraft flight conflict with high solving efficiency.First,the characteristics of multi-aircraft flight conflict problem are analyzed and the problem is modeled based on Markov decision process.Thus,the Independent Deep Q Network(IDQN)algorithm is used to solve the model.Simultaneously,a’downward-compatible’framework that supports dynamic expansion of the number of conflicting aircraft is designed.The model ultimately shows convergence through adequate training.Finally,the test conflict scenarios and indicators were used to verify the validity.In 700 scenarios,85.71%of conflicts were successfully resolved,and 71.51%of aircraft can reach destinations within 150 s around original arrival times.By contrast,conflict resolution algorithm based on DRL has great advantages in solution speed.The method proposed offers the possibility of decision-making support for controllers and reduce workload of controllers in future high-density airspace environment.
基金The work is a part of project named ＂＇Multi- Agent based Train Operation in Moving Block Setup＂ funded by Department of Information Technology （DIT）, Ministry of Commu- nications and Information Technology, Government of India, vide Grant Number 2（6）/2010-EC dated 21/03/2011.
文摘Multi-agent technology has been used in many complex distributed and concurrent systems. A railway system is such a safety critical system and careful inves- tigation of the functional components is very important. Study of the various functional components in communi- cation-based train control （CBTC） system necessitates a good structural design followed by its validation and ver- ification through a formal modelling technique. The work presented here is the follow up of our multi-agent-based CBTC system for Indian railway designed using the methodology for engineering system of software agents. Behavioural analysis of the designed system involves several operating scenarios that arise during train run, and helps in understanding the reaction of the system to such situations. This validation and verification are very important as it allows the system designer to critically evaluate the desired function of the system and to correct the design errors, if any, before its actual implementation. Modelling, validation and verification of the structural design through Coloured petri net （CPN） are central to this paper. Analysis of simulation results validates the efficacy of the design.
文摘Racemic R.S-α-arylethylamine was resolved by R (-) thiazolidine-2-thione-4-carboxylic acid, a new resolving agent abbreviated as [R (-) TTCA], by which R (-) TTCA.S(-) arylethylamine salts2a-2e, [α] D 20 =-47.24° — 64.40° and optically active R(+)-a-arylethylamines3a – 3e, 74. 54%-94. 45% e, e., were obtained. Optically active S (-) -α-arylethylamines4a-4e, 72.84%-90.36% e.e., were obtained by the decomposition of2a-2e in basic solutions. The influence of substitutive group of the benzene ring on the basicity of the amino group was studied by semiempirical PM3 method. The structures of the R (-) TTCA.S (-) -α-phenylethylamine salt (2a(R-S) configuration) and R (-) TTCA-R(+)-a-phenylethylamine salt (2a(R-R) configuration) have been established by means of X-ray diffraction. They crystallize in a monoclinic system. Space group isP21. The cell constants of2a(R-S) configuration were obtained as follows: α = 1.387 8(2), b = 0.664 05(101,c = 1.580 O(2) nm; β = 90.844(10)° Z = 4; those obtained for2a(R-R) configuration were α = 1.080 6(2),b = 0.584 80(12),c = 1.2188(2) nm, β= 110.38(3)dg, V = 0.7220nm3,Z = 2. There are intermolecular hydrogen bonds in the crystals of the two kinds of configurations of the amine salt. The hydrogen bond number in the unit cell of R (-) TTCA.S (-)-α-phenylethylamine salt is twice as much as that of R (-) TTCA.R(+)-a-phenylethylamine salt.
文摘在自由飞行的环境下,为解决飞行冲突探测与解脱(conflict detection and resolution,CDR)问题,提出一种基于高度层、航向和速度调配的综合解脱方法,并将多agent系统(multi-agent system,MAS)的分布式技术与启发式算法相结合,进行问题求解.首先设计了分布式MAS框架结构,然后建立了飞行冲突探测模型,高度层调配模型及航向、速度调配模型,最后,综合运用了基于合同网协议的分布式算法和自适应遗传算法进行问题求解.仿真实验表明,所设计的MAS框架是可行的,同时分布式算法和自适应遗传算法的综合应用能很快找到基于高度层、航向和速度分配的近似最优解,为CDR问题提供了新的解决思路.
文摘为解决固定航路上飞行冲突探测与解脱(conflict detection and resolution,CDR)的问题,本文提出一种基于高度层分配的解脱方法,并利用分布式多agent系统(multi-agent system,MAS)进行算法求解。首先建立固定航路网络图对管制扇区进行建模;然后分析影响高度层分配的主要因素,并建立高度层使用优先权的评价模型;最后,设计基于合同网协议的多agent系统,将目前依靠管制员的集中调配模式,转变为路口agent和航空器agent之间自主进行通信、协商和协作的分布式冲突解脱模式。仿真实验表明:高度层分配方法是可行的,与传统调整航向或速度的方法相比,该方法更接近实际情况,同时设计的分布式多agent系统算法能够快速找到高度层分配的最优解,为CDR问题提供了新的解决思路。
