To solve the problem of multi-target hunting by an unmanned surface vehicle(USV)fleet,a hunting algorithm based on multi-agent reinforcement learning is proposed.Firstly,the hunting environment and kinematic model wit...To solve the problem of multi-target hunting by an unmanned surface vehicle(USV)fleet,a hunting algorithm based on multi-agent reinforcement learning is proposed.Firstly,the hunting environment and kinematic model without boundary constraints are built,and the criteria for successful target capture are given.Then,the cooperative hunting problem of a USV fleet is modeled as a decentralized partially observable Markov decision process(Dec-POMDP),and a distributed partially observable multitarget hunting Proximal Policy Optimization(DPOMH-PPO)algorithm applicable to USVs is proposed.In addition,an observation model,a reward function and the action space applicable to multi-target hunting tasks are designed.To deal with the dynamic change of observational feature dimension input by partially observable systems,a feature embedding block is proposed.By combining the two feature compression methods of column-wise max pooling(CMP)and column-wise average-pooling(CAP),observational feature encoding is established.Finally,the centralized training and decentralized execution framework is adopted to complete the training of hunting strategy.Each USV in the fleet shares the same policy and perform actions independently.Simulation experiments have verified the effectiveness of the DPOMH-PPO algorithm in the test scenarios with different numbers of USVs.Moreover,the advantages of the proposed model are comprehensively analyzed from the aspects of algorithm performance,migration effect in task scenarios and self-organization capability after being damaged,the potential deployment and application of DPOMH-PPO in the real environment is verified.展开更多
Self-diffusion prerequisite is obtained as the spreading approach of biological populations.Cooperative hunting is a common behavior in predator populations that promotes predation and the coexistence of the prey-pred...Self-diffusion prerequisite is obtained as the spreading approach of biological populations.Cooperative hunting is a common behavior in predator populations that promotes predation and the coexistence of the prey-predator system.On the other side,the Allee effect among prey may cause the system to become unstable.In this paper,a difusive prey predator system with cooperative hunting and the weak Allee effect in prey populations is discussed.The linear stability and Hopf-bifurcation analysis had been used to examine the system's stability.From the spatial stability of the system,the conditions for Turing instability have been derived.The multiple-scale analysis has been used to derive the amplitude equations of the system.The stability analysis of these amplitude equations leads to the formation of Turing patterns.Finally,numerical simulations are used to analyze spatial patterns forming in 1-D and 2-D.The studies indicate that the model can generate a complex pattern structure and that self-diffusion has a drastic impacton species distribution.展开更多
This paper presents a novel solution to the three-dimensional (3D) cooperative hunting ofmultiple drones that deals with surrounding a target simultaneously while navigating aroundobstacles in the cluttered dynamic 3D...This paper presents a novel solution to the three-dimensional (3D) cooperative hunting ofmultiple drones that deals with surrounding a target simultaneously while navigating aroundobstacles in the cluttered dynamic 3D environment. Meanwhile, drones avoid the air°owdownwash force created by the spinning propellers on unmanned aerial vehicles (UAVs) andtheir e®ect on the other UAVs. This solution consists of a 3D Simultaneous Encirclementstrategy, the cooperative hunting objective with a novel revised particle swarm optimization(PSO*) path planning algorithm, a °ocking theory-inspired obstacle avoidance algorithm, and acascade PI controller. Simulation results with varying conditions were carried out to validatethe e®ectiveness of the proposed solution by successfully taking care of the downwash e®ects,and having multiple hunter UAVs hunt and encircle a moving or stationary target in a dynamicor static obstacle-rich cluttered environment.展开更多
We propose and investigate a discrete-time predator-prey system with cooperative hunting in the predator population.The model is constructed from the classical Nicholson-Bailey host-parasitoid system with density depe...We propose and investigate a discrete-time predator-prey system with cooperative hunting in the predator population.The model is constructed from the classical Nicholson-Bailey host-parasitoid system with density dependent growth rate.A sufficient condition based on the model parameters for which both populations can coexist is derived,namely that the predator’s maximal reproductive number exceeds one.We study existence of interior steady states and their stability in certain parameter regimes.It is shown that the system behaves asymptotically similar to the model with no cooperative hunting if the degree of cooperation is small.Large cooperative hunting,however,may promote persistence of the predator for which the predator would otherwise go extinct if there were no cooperation.展开更多
为了提高围捕系统的围捕效率,提出一种基于融合蛇优化算法的多AUV协同围捕算法(Multi-AUV Cooperative Hunting Algorithm based on Fusion Snake Optimization algorithm,MACHA_FSO)。MACHA_FSO改进随机目标搜索策略,采用莱维飞行策略...为了提高围捕系统的围捕效率,提出一种基于融合蛇优化算法的多AUV协同围捕算法(Multi-AUV Cooperative Hunting Algorithm based on Fusion Snake Optimization algorithm,MACHA_FSO)。MACHA_FSO改进随机目标搜索策略,采用莱维飞行策略设置搜索目标,就近原则变更围捕AUV工作区域,保证围捕AUV的搜索效率。MACHA_FSO构建围捕系统的整体能耗模型,采用最小化围捕距离策略建立围捕联盟,提出融合蛇优化算法合理规划围捕AUV的围捕路径,有效降低围捕AUV能耗。仿真结果表明:相较于CPGBNN,RIGBNN和PRACO围捕算法,MACHA_FSO能够合理设置围捕AUV的搜索目标与围捕路径,且围捕系统平均能量消耗降低41%,围捕逃逸目标平均用时降低32%,围捕逃逸目标平均数量提高1倍,围捕系统平均生存时间提高15%。展开更多
基金financial support from National Natural Science Foundation of China(Grant No.61601491)Natural Science Foundation of Hubei Province,China(Grant No.2018CFC865)Military Research Project of China(-Grant No.YJ2020B117)。
文摘To solve the problem of multi-target hunting by an unmanned surface vehicle(USV)fleet,a hunting algorithm based on multi-agent reinforcement learning is proposed.Firstly,the hunting environment and kinematic model without boundary constraints are built,and the criteria for successful target capture are given.Then,the cooperative hunting problem of a USV fleet is modeled as a decentralized partially observable Markov decision process(Dec-POMDP),and a distributed partially observable multitarget hunting Proximal Policy Optimization(DPOMH-PPO)algorithm applicable to USVs is proposed.In addition,an observation model,a reward function and the action space applicable to multi-target hunting tasks are designed.To deal with the dynamic change of observational feature dimension input by partially observable systems,a feature embedding block is proposed.By combining the two feature compression methods of column-wise max pooling(CMP)and column-wise average-pooling(CAP),observational feature encoding is established.Finally,the centralized training and decentralized execution framework is adopted to complete the training of hunting strategy.Each USV in the fleet shares the same policy and perform actions independently.Simulation experiments have verified the effectiveness of the DPOMH-PPO algorithm in the test scenarios with different numbers of USVs.Moreover,the advantages of the proposed model are comprehensively analyzed from the aspects of algorithm performance,migration effect in task scenarios and self-organization capability after being damaged,the potential deployment and application of DPOMH-PPO in the real environment is verified.
文摘Self-diffusion prerequisite is obtained as the spreading approach of biological populations.Cooperative hunting is a common behavior in predator populations that promotes predation and the coexistence of the prey-predator system.On the other side,the Allee effect among prey may cause the system to become unstable.In this paper,a difusive prey predator system with cooperative hunting and the weak Allee effect in prey populations is discussed.The linear stability and Hopf-bifurcation analysis had been used to examine the system's stability.From the spatial stability of the system,the conditions for Turing instability have been derived.The multiple-scale analysis has been used to derive the amplitude equations of the system.The stability analysis of these amplitude equations leads to the formation of Turing patterns.Finally,numerical simulations are used to analyze spatial patterns forming in 1-D and 2-D.The studies indicate that the model can generate a complex pattern structure and that self-diffusion has a drastic impacton species distribution.
文摘This paper presents a novel solution to the three-dimensional (3D) cooperative hunting ofmultiple drones that deals with surrounding a target simultaneously while navigating aroundobstacles in the cluttered dynamic 3D environment. Meanwhile, drones avoid the air°owdownwash force created by the spinning propellers on unmanned aerial vehicles (UAVs) andtheir e®ect on the other UAVs. This solution consists of a 3D Simultaneous Encirclementstrategy, the cooperative hunting objective with a novel revised particle swarm optimization(PSO*) path planning algorithm, a °ocking theory-inspired obstacle avoidance algorithm, and acascade PI controller. Simulation results with varying conditions were carried out to validatethe e®ectiveness of the proposed solution by successfully taking care of the downwash e®ects,and having multiple hunter UAVs hunt and encircle a moving or stationary target in a dynamicor static obstacle-rich cluttered environment.
文摘We propose and investigate a discrete-time predator-prey system with cooperative hunting in the predator population.The model is constructed from the classical Nicholson-Bailey host-parasitoid system with density dependent growth rate.A sufficient condition based on the model parameters for which both populations can coexist is derived,namely that the predator’s maximal reproductive number exceeds one.We study existence of interior steady states and their stability in certain parameter regimes.It is shown that the system behaves asymptotically similar to the model with no cooperative hunting if the degree of cooperation is small.Large cooperative hunting,however,may promote persistence of the predator for which the predator would otherwise go extinct if there were no cooperation.
文摘为了提高围捕系统的围捕效率,提出一种基于融合蛇优化算法的多AUV协同围捕算法(Multi-AUV Cooperative Hunting Algorithm based on Fusion Snake Optimization algorithm,MACHA_FSO)。MACHA_FSO改进随机目标搜索策略,采用莱维飞行策略设置搜索目标,就近原则变更围捕AUV工作区域,保证围捕AUV的搜索效率。MACHA_FSO构建围捕系统的整体能耗模型,采用最小化围捕距离策略建立围捕联盟,提出融合蛇优化算法合理规划围捕AUV的围捕路径,有效降低围捕AUV能耗。仿真结果表明:相较于CPGBNN,RIGBNN和PRACO围捕算法,MACHA_FSO能够合理设置围捕AUV的搜索目标与围捕路径,且围捕系统平均能量消耗降低41%,围捕逃逸目标平均用时降低32%,围捕逃逸目标平均数量提高1倍,围捕系统平均生存时间提高15%。
