| At present,there are a lot of researches on the consensus of the multi-agent system,but most of them assume that the energy of each agent in the multi-agent system is infinite by default.In the practical situation,the energy of each agent is provided by the battery,solar energy,fuel and so on,and the energy they can carry is limited.Therefore,when the multi-agent system performs a task,some agents may run out of energy,which will lead to the system crash,and finally the task cannot be executed successfully.Based on the above fact,in order to extend the running time of the system and ensure its successful implementation of the established tasks,this thesis makes an in-depth study on the basis of existing theories.The main researches are as follows:Firstly,aiming at the traditional potential field function collision avoidance problem,a novel collision avoidance function is proposed in this thesis which can make the agent consume less energy.In addition,in view of the current situation that the collision avoidance methods are all equal,a new collision avoidance strategy is proposed,unequal collision avoidance.According to the remaining energy of the neighboring agents,one agent performs the collision avoidance action,and the other agent does not perform the collision avoidance action.Combined with the above novel collision avoidance function and the new collision avoidance strategy,an unequal collision avoidance consensus controller is designed.Compared with the traditional equal collision avoidance strategy and consensus controller,the controller designed in this thesis can prolong the running time of the system.Finally,the simulation results verify the effectiveness and superiority of the controller designed in this thesis.Then,the consensus and distributed convex optimization of a class of the multi-agent system with time-varying objective function is studied,and the chattering problem of the control input caused by the sign function in the controller is solved.Firstly,we use the sign function to design the controller to improve the convergence rate of the multi-agent system.In addition,we design the distributed optimization term for the more complex time-varying functions to ensure that the agent consumes less energy when the system achieves consensus;Secondly,on the basis of the designed controller,considering the input chattering problem caused by the sign function,this thesis improves the controller,which can effectively solve the chattering problem.The simulation results show the correctness of the theoretical design and the superiority of saving the agent energy.Finally,in the above studies,it is assumed that the communication between neighboring agents is always maintained by default.However,due to the complexity of the practical environment,communication is difficult to maintain.Therefore,a communication maintenance function is designed in this thesis to ensure that the communication between neighboring agents is maintained at all time.Secondly,in order to save the consumed energy in the process of achieving consensus of the multi-agent system,an optimization item is designed in the controller to ensure that the system has a longer running time.Finally,the sign function is used to ensure that the multi-agent system can achieve consensus in a fixed time.The simulation results show the superiority of the designed controller in saving energy. |
