Research On Multi-UAV Supervisory Control Method

Nowadays,the supervisory control of multiple unmanned aerial vehicles(UAVs)by the single operator is an important and difficult research in military field in the world.To achieve the vision that single operator can supervise and control multiple UAVs simultaneously,how to allocate the functions in the supervisory control system between the operator and the automation in a proper way is one of the difficult problems.Because the static function allocation method cannot respond to the change of the operator's state during the system operation,the dynamic human-machine function allocation research should be carried out.This thesis presents a context-aware based dynamic function allocation method used for multiple unmanned combat aerial vehicles supervisory control.The main research work and results are as follows:Emergencies that may occur during the flight and mission of an UAV are summarized.Considering that unexpected events may affect the UAVs' flight or the execution of tasks during the mission of UAVs,two categories of typical emergencies,including the impact of environment and UAV health changes,are summarized.Further,the environmental impact is divided into two categories: heavy weather and obstacles.And the health status of an UAV consists of four categories: body health,payloads health,communication link health,and energy health.And the impact of the six categories of emergencies on the UAV flight or the mission of the UAV is described respectively.And emergency reactions are given to each type of emergencies.The problem of dynamic function allocation in multiple UAVs supervisory control system is studied.Firstly,the architecture of centralized control is briefly described,and control center is ground control station.And then the intension of multi-UAV supervisory control and the hierarchical control loops of multiple UAVs are analyzed.Then,the problem of function allocation between the human and the computer in multi-UAV supervisory control system is presented.And the requirement of utilizing dynamic function allocation to response to operator fatigue is analyzed,and dynamic function allocation methods commonly used are introduced briefly.For levels of automation,except for management by consent and management by exception,this thesis adds management of multiple options,to smoothly transit from manual control to management by consent.Finally,the operation in each level of automation for task reallocation and path replanning is described.A context based multi-UAV supervisory control method is presented.At first,the factors that cause operator fatigue are analyzed,and unexpected events are one of the main factors.Therefore,in order to lessen the operator fatigue caused by the emergencies,function reallocation between the human and the computer should be triggered when there are emergencies.Then a context based multi-UAV supervisory control method is described,which describes the relationship between emergencies and function reallocation.After that,a context-aware model for supervisory control of multiple UAVs is established.Finally,some examples are given to verify the context based multi-UAV supervisory control method.The results show that the dynamic function allocation method is effective.