Height Cooperative Control For Quadrotor Swarm System:Algorithms And Experiments

Comparing to the single agent,such as aircraft,missile and robot,the multi-agent has a great advantage on its performance as well as reliability in cooperative control field.In recent decades,the cooperative control technology of the multi-agent has become a hotspot in the field of control,robotics and communication.Considering that the quadrotor is a simple but reliable platform of the verification of the flying control algorithm,it is of great both engineering and scientific significance to build up the multi-aircraft cooperative system.In light of that height cooperative control is the most basic cooperative model,this thesis concentrates on the height cooperative control problems of the quadrotor,which involves hardware selection,algorithm design,simulation and experimental verification.In particular,the details of the research are as follows.Firstly,this thesis extracts the height dynamic model by analyzing the non-linear model of the quadrotor,and designs model-based algorithms of height cooperative and formation.Aside from that,the convergence condition of the designed algorithms is proved.Some relative cooperative algorithms are also verified by the MATLAB model.Secondly,in order to build up the multi-aircraft cooperative system,the following parts must be done.Selection of the sensors,building up a single quadrotor system and setting up the communication network between the quadrotors.Finally,a test experiment is carried out to ensure that the hardware and software design meets the requirements of the verification of the cooperative algorithm.Thirdly,after researching in the fusion algorithms of the altitude sensors of the quadrotors,the accuracy of the altitude state information is improved.After analyzing the characteristics of the altitude sensors,calibrating the sensors and preprocessing the original data,a fusion algorithm of multi-sensor is designed tested.Meanwhile,the actual effect of the algorithm is evaluated through the/a closed-loop control experiment.Finally,the designed cooperative algorithms are verified by experiments which based on the established platform.Those algorithms include situations like height synchronization with or without a leader and height formation with a leader.