针对同时具有虚假数据注入(false data injection,FDI)攻击与执行器故障下的多无人系统编队协同跟踪问题,提出了一种基于FDI攻击检测机制与故障观测器的韧性容错协同控制新方法。首先,以二阶非线性固定翼无人机模型作为多无人系统研究对...针对同时具有虚假数据注入(false data injection,FDI)攻击与执行器故障下的多无人系统编队协同跟踪问题,提出了一种基于FDI攻击检测机制与故障观测器的韧性容错协同控制新方法。首先,以二阶非线性固定翼无人机模型作为多无人系统研究对象;其次,构建了带有概率约束的贝叶斯概率检测模型对FDI攻击进行检测,作为韧性容错协同控制器的辅助系统;之后,设计了包含执行器故障补偿的韧性容错协同跟踪控制器,并利用Lyapunov稳定性理论证明系统的渐进稳定;最后,通过仿真验证了设计的控制器针对FDI攻击与执行器故障的安全性与可靠性以及编队跟踪能力。展开更多
This paper presents a scenario of forest fire suppression using UAVs （Unmanned Aerial Vehicles） and addresses task assignment algorithm to coordinate UAVs. Forest fires are a major problem in many nations and fast e...This paper presents a scenario of forest fire suppression using UAVs （Unmanned Aerial Vehicles） and addresses task assignment algorithm to coordinate UAVs. Forest fires are a major problem in many nations and fast extinguishing forest fires brings a lot of ecological advantages so proper use of firefighting resources is very critical. In this sense, multi UAVs forest fire suppression system can be effective way to prevent fire outbreaks. In multi agent system, an appropriate task assignment according to the SA （Situational Awareness） is the most essential to conduct mission. We should consider real time re-planning or re-scheduling of multi UAVs team because environmental situations such as wind are changeable and that changes affect the forest fire spreading. Furthermore, we have to think about convergence to a consistent SA because it may take too much time. CBBA （Consensus-Based Bundle Algorithm） is robust decentralized task assignment tool so it can be implemented in real time re-planning application. A simulation model which is the main topic in this paper shows that multi UAVs can be properly operated to suppress forest fires even if there are unpredictable random factors and partial disconnection. The simulation model includes concrete operating scenarios and recursive task re-assign algorithm until fires in the whole area are suppressed.展开更多
文摘针对同时具有虚假数据注入(false data injection,FDI)攻击与执行器故障下的多无人系统编队协同跟踪问题,提出了一种基于FDI攻击检测机制与故障观测器的韧性容错协同控制新方法。首先,以二阶非线性固定翼无人机模型作为多无人系统研究对象;其次,构建了带有概率约束的贝叶斯概率检测模型对FDI攻击进行检测,作为韧性容错协同控制器的辅助系统;之后,设计了包含执行器故障补偿的韧性容错协同跟踪控制器,并利用Lyapunov稳定性理论证明系统的渐进稳定;最后,通过仿真验证了设计的控制器针对FDI攻击与执行器故障的安全性与可靠性以及编队跟踪能力。
文摘This paper presents a scenario of forest fire suppression using UAVs （Unmanned Aerial Vehicles） and addresses task assignment algorithm to coordinate UAVs. Forest fires are a major problem in many nations and fast extinguishing forest fires brings a lot of ecological advantages so proper use of firefighting resources is very critical. In this sense, multi UAVs forest fire suppression system can be effective way to prevent fire outbreaks. In multi agent system, an appropriate task assignment according to the SA （Situational Awareness） is the most essential to conduct mission. We should consider real time re-planning or re-scheduling of multi UAVs team because environmental situations such as wind are changeable and that changes affect the forest fire spreading. Furthermore, we have to think about convergence to a consistent SA because it may take too much time. CBBA （Consensus-Based Bundle Algorithm） is robust decentralized task assignment tool so it can be implemented in real time re-planning application. A simulation model which is the main topic in this paper shows that multi UAVs can be properly operated to suppress forest fires even if there are unpredictable random factors and partial disconnection. The simulation model includes concrete operating scenarios and recursive task re-assign algorithm until fires in the whole area are suppressed.
