Research On UAV Conflict Resolution Method Based On Improved Sparrow Search Algorithm And Other Algorithm

With the increasing flow of airspace,the risk of collision between UAV and other aircraft also increases.UAVs lack the pilot’s artificial avoidance in terms of avoidance principle,and need to rely on avoidance technology to make UAVs accomplish autonomous obstacle avoidance,which is different from the principle of “see – avoid” of manned vehicle.The conflict avoidance algorithm is the core of UAVs’ autonomous avoidance.Through the research of the risk avoidance algorithm,the ability of UAV’s autonomous decision-making and conflict avoidance can be improved,conflict resolution can be achieved,and the flight safety of UAVs can be further guaranteed.This paper improves the traditional avoidance algorithm and intelligent algorithm,achieve conflict resolution draw support from the UAV model.The main research contents are as follows:Firstly,proposing the Global and Local Sparrow Search Algorithm.Because the traditional Sparrow Search Algorithm is easy to fall into local optimum and has poor global search ability,the improved Circle chaotic map is used to make the value distribution of solution space initialization more uniform to obtain higher quality initial value.The adoption of spiral search strategy,dynamic step strategy and adaptive quantity strategy is conducive to the global search in the early stage and local search in the later stage,which improves the disadvantages of poor global search ability and easy to fall into local optimum,and accelerates the convergence speed in the later stage.The algorithm before and after improvement are applied to the conflict resolution of UAV respectively.Two and four UAVs conflict resolution in two dimension and three dimension are simulated.The results show that the algorithm after improvement is better than the algorithm before improvement in the conflict resolution.Secondly,according to the Minimum Interval Method,the distance function between UAVs is equal to the longitudinal interval,and the time consumed to reach this distance is calculated.That is to say,the conflict resolution is completed.According to the speed and track azimuth parameters contained in the formula,the conflict resolution can be achieved by changing the speed or track azimuth.After the conflict resolution is completed,the track recovery strategy is implemented to restore the speed or track azimuth to the initial state,Restore to the initial flight path,use the non-cooperative game to solve the optimal release strategy suitable for both sides,and take the release time as the payment function to achieve simulation verification.Finally,the Model Predictive Control is used to predict and solve the optimization problem based on the dynamic model of the system.Linearized and discretized the dynamics and kinematics model of the UAV,then changed the Model Predictive Control to incremental type,set the constraint conditions of input,output and control quantity and added the distance constraint,and improved the objective function.For the conflict resolution of multiple UAVs,the concept of “nucleolus solution” in cooperative game is introduced to improve the fairness of UAV relief,and at the expense of the overall interests,the cost of achieving relief for UAVs with high priority is minimal.The simulation results show that the priority UAV has the lowest cost of relief,and the introduction of “nucleolus solution” for multi-UAV conflict resolution improves the fairness of relief to a certain extent.