Research On Finite-time Convergent Guidance Control In Sea Clutter

With the advancement of science and technology,the form of war has changed a lot.The destructive power of weapons is increasing,and its ability to hit the target accurately is also improving The battlefield environment is also increasingly complex.The continuous development of missile technology has extended its use from land to sea.It becomes a very important weapon in the ocean battlefield.But there are all kinds of uncertainties in the ocean,such as wind and waves.This makes it much more difficult to identify,track and strike targets accurately.Moreover,the relative velocity of terminal guidance phase is very fast.During this period,only a very short time can be used for precise guidance.Therefore,it is required to be able to complete the control requirements in a limited time.In this paper,The interception process of ballistic missile in sea clutter background is taken as the research object.This paper mainly studies the detection and recognition of targets,tracking filtering algorithm and finite time convergence guidance law.Specific research contents include:The characteristics and modeling methods of sea clutter are studied and several typical statistical models are established.The target detection algorithm is studied.Firstly,the fixed threshold detection method is simply analyzed.Then the adaptive threshold detection method is studied in this paper.CA-CFAR has good detection performance under the background of uniform noise.But there are target shadowing effects in detection.And False alarm and missed alarm will occur at the clutter edge under the background of non-uniform noise.SO-CFAR can successfully solve the problem of target shadowing effects and missed alarm at clutter edge,but it also increases the probability of false alarm at clutter edge.GO-CFAR can solve the problem of false alarm at clutter edge,but it can’t solve the target shadowing effects.OS-CFAR can solve the target shadowing effects of the adjacent target,and it does not raise the threshold value on both sides of the target like SO-CFAR.The tracking filter of target is studied.Firstly,a three-dimensional system model with relative position,relative velocity and target acceleration as state variables is established.Kalman filter is used to estimate the state of this system.When the target acceleration is constant,the estimation error is small.However,when the target does snake maneuver,although it can also estimate the state of the system,there is a large lag.Then,a four-dimensional system model with relative distance,relative velocity,target acceleration and target acceleration as state variables is established.The Kalman filter designed on thisbasis has much better estimation effect on target acceleration.All the filters designed above default that the maneuvering frequency of the target can be accurately obtained,but it is difficult to accurately measure this information in practical engineering applications.For this reason,a five-dimensional system model is established by adding target maneuvering frequency as state variable.At this time,the system model is no longer linear.The acceleration,acceleration and maneuvering frequency of the target can be estimated smoothly by using EKF.But its robustness is poor.Continue to improve,using IMM-KF for filtering.Each model in the algorithm is a four-dimensional Kalman filter.This method can match the exact model quickly.It makes the estimation of target maneuvering frequency converge to its true value.And the estimated value of the target maneuver frequency is always very close to the true value.The finite-time convergence control theory is studied.On this basis,a finite-time convergent guidance law in two-dimensional plane is designed and extended to three-dimensional space.There are three main situations: the guiding law in the case where the target is not maneuverable;the guidance law in the case where the target is maneuverable and the target effective information can be obtained;the guidance law in the case where the target is maneuver but cannot obtain the target effective information.Compared with the proportional guidance law,it can make the line-of-sight angular rate converge to zero in limited time.So it has high accuracy.