GB2309068A

GB2309068A - Missile guidance system

Info

Publication number: GB2309068A
Application number: GB8601651A
Authority: GB
Inventors: Richard Vincent Lawrence; Graham Harold Ben Jordan; Tom William Ridler
Original assignee: UK Secretary of State for Defence
Current assignee: UK Secretary of State for Defence
Priority date: 1985-01-30
Filing date: 1986-01-23
Publication date: 1997-07-16
Anticipated expiration: 2006-01-23
Also published as: GB8601651D0; GB8502369D0; GB2309068B

Abstract

A strapdown seeker 11 provides measurements of a target sight line look angle, * small Greek psi *, with respect to the missile body axis and measurements of angular diameter, * small Greek theta *, of the target image at the seeker. A state estimator 5, which is preferably a Kalman type filter, estimates sightline spin * small Greek omega * having inputs of * small Greek theta * and * small Greek psi * from the seeker, and includes means 6 for mathematically modelling the dynamics of * small Greek theta * and * small Greek psi *, calculating the difference 9,10 between the measured and estimated values of * small Greek theta * and * small Greek psi * and generating corresponding corrections to the estimated states from the estimator. The estimator provides corrected variables to the flight control system.

Description

MISSILE GUIDANCE SYSTEM The present invention relates to a missile guidance system, and particularly relates to a guidance system for a missile which employs a passive homing seeker.

If a constant velocity missile is travelling on a collision course with a constant velocity target, then the sightline, S, between missile and target will be shortening at a constant rate and its orientation in a fixed reference space will be constant, ie: its angular spin rate in the reference space will be zero. However, if the missile is not on a collision course the orientation of S will not be constant.

The missile can be brought onto a collision course by causing it to accelerate in a direction normal to the sightline at a rate proportional to the rate of turn of the sightline in the reference space. When the missile comes onto a collision course, the sightline spin becomes zero, and no further accelerative manoeuvre is demanded. This method of control is called proportional navigation.

Where either the missile or the target or both are travelling at non-constant velocity, a particular time history of sightline spin will cause a straight flying missile to impact the target. If the missile can be brought onto a straight line trajectory which yields the correct time variation of sightline spin, an impact will be guaranteed. The time history is a function of the missile and target accelerations. The control required in the intercepting missile is a function of sightline spin, and the missile and target accelerations. Therefore, in order to construct the control, these quantities, or state variables, must be estimated.

Other state variables, such as missile - target range and range rate, are aleo important, either because knowledge of them is required to construct the estimates of sightline spin and target acceleration, or because a higher level control law is to be implemented which employs the states in its construction. Knowledge of range is also useful in discriminating against background, computing intercept time, and optimising trajectory.

In the past, sightline spin has been measured using a gimballed dish assembly in the nose of the missile. By means of servo mechanisms, the dish is slaved to the sightline so that its boresight axis is coincident with the sightline.

Sightline rate estimates are obtained by using the outputs of rate gyroscopes mounted on the dish. Provided the dish servos are able to track the motion of the sightline, and slew the dish against violent angular fluctuations of the missile body, the spin estimates are satisfactory for guidance purposes.

The present invention provides a missile guidance system which is particularly suitable for small, highly agile missiles, and which does not rely on the use of servo driven gimballed dish assemblies. The invention employs signal processing techniques to construct estimates of the state variables required, ie: sightline spin, range and range rate.

According to the present invention a missile guidance system includes a strapdown seeker providing measurements of sight line look angle, +, with respect to the missile body axis and providing measurements of angular diameter, 6, of the target image at the seeker, and a state estimator for sightline spin having inputs of 6 and f from the seeker, and including means for mathematically modelling the dynamics of 6 and , calculating the difference between the measured and estimated values of 6 and X, and generating corrections to the estimated states.

The state estimator may be a Kalman type filter.

The strapdown seeker may include an opto-electronic array detector fan providing measurements of angular diameter, 6, of the target image at the seeker.

The target acceleration may be modelled in the guidance system as a random noise process.

Thus, the state estimator is driven with measurements obtained from sensors which do not need precise servo control but which cannot make direct measurements of state such as sightline spin, range or range rate.

An embodiment of the invention will nov be described, by way of example only, with reference to the drawings of which: Figure 1 is a schematic view of a missile engagement Figure 2 is the guidance system of the missile of Fig 1 Figure 1 shows a missile which carries a missile guidance system shown in Fig 2, and a target 2 which is on a missile-target sightline S, of length r. The sightline S is at an angle X, the look angle, to the longitudinal axis 3 of the missile. The target subtends a small angle e from the missile.

The sightline S has a spin rate w.

The system measures the value of 6 and f. In a simple case, it is assumed that the missile axis 3 and the sightline S both lie in the plane of the Figure. The scope of the invention is not limited to this simple case, which is used for the purposes of explanation, but can be applied to any general three-dimensional engagement.

In a practical system, measurements of 6 and 4? will be corrupted to some degree by noise in the missile system.

The most important state variable from the control aspect is the sightline spin rate, w. The differential equation for the evolution of the dynamics of the state is given by d# # = dt = - 2# + #asy .... (1) where p - irs the ratio between the range rate and the range, p is the reciprocal of the range, and a is the relative sy acceleration between the missile and the target normal to the sightline. Estimates of , p and asy enable an estimate of # to be propagated. These states will in turn evolve according to their own particular dynamics, and so it is necessary tomodel the dynamics of each state in order to propagate the estimates.

The dynamics of are given by d = dt = # - + #asx, ....(2) where a is the relative acceleration between missile and sx target along the sightline.

The dynamics of p are given by dp - - . .. (3) dt The dynamics of the acceleration terms a and a are sx sy modelled. It is not necessary in practice to provide a model for the missile contribution to the terms a and a , because ax sy the missile accelerations are measured accurately by means of on-board accelerometers. The problem of modelling the unknown target accelerations is difficult, because the dynamics of the target is unknown, and the inputs to it are the unpredictable control demands of the pilot. It is therefore necessary to model the target accelerations as a random noise process, possibly with some correlation between each noise event.

For the purposes of this example only it will be assumed that the target is flying at constant velocity, ie: it is not maneouvering.

The above-mentioned states form a closed set and no further states are required to propagate sightline spin.

However, to generate a prediction of look angle # the relationship between f and the above mentioned states needs to be established.

The dynamics of f is given by d# # = dt = # - #b, ....(4) where wb is the missile body angular rate, which is measured accurately with a rate gyro on board the missile.

The following set of equations describe the dynamics of a closed set of states, 4 - # - #b 3 = - 2# + #asy ....(5) = # - + #asx # = - # Estimates of , w, and p are propagated within the missile by a model based on the above dynamics. If the estimates are in error, the estimated values of f over a period of time will not agree with the measured values, and corrections to the estimates will be generated. However, we note from equation (5) that if there are no accelerations in the system, asx and asy are zero, and so no matter what error is present in the estimate of p, it will not affect the estimate of f. Hence the estimate of p can be corrected only when the missile is accelerating. Detailed examination of this problem shows it might be a serious shortcoming, because the missile control would attempt to put the missile on a zeromaneouvre intercept trajectory. This is inconsistent with the requirement to estimate the inverse range state. However, if the missile is intentionally maneouvred so that p can be estimated, the miss-distance will be seriously degraded. The problem is further complicated by the fact that the state II may also be uncertain, and, as the dynamics equations (5) show, errors in the p estimate have no effect on the prediction of f when the sightline spin is zero. In fact X is zero on a constant velocity intercept course.

The effectiveness of an estimator based on a measurement of look angle alone is not satisfactory in a number of applications, notably those in which both range and range rate are very uncertain at the time of missile launch.

All of the above difficulties can be overcome by making use of seeker measurements of image diameter, 0. By modelling the dynamics of 6 in the estimator, an estimate of 6 can be propagated, and compared with measurements as they become available. The dynamics of 6 is e d6 di ....(6) dt Therefore, regardless of the missile trajectory, errors in the estimates of 6 and p will cause a measurement residual to appear, allowing the 6 and p estimates to be corrected. The important result is that the p estimate converges on its true value no matter what trajectory the missile follows. The implications for estimation of the inverse range p can be seen by considering the full set of dynamics equations: 4, - we web t - - 2# + pasy = # - + #asx ... (7) b I - l P e - - where # and 6 are measured by a seeker.

In the zero acceleration case, p does not couple to either of the measured states. Therefore errors in the p estimate cannot be corrected. In order to converge the p estimate there must be accelerations in the system. However, since the estimate converges independently, quite small missile manoeuvres, consistent with those which are necessary to place the missile on a collision course, are sufficient to cause the p estimate to converge. Computer simulations show that with convergence of the estimate, the accelerations due to missile boost and drag are sufficient to cause convergence.

Referring to Fig 2 a seeker system 11 on the missile measures the missile-target state variables including f and 6. f and 6 are measured by cadmium mercury telluride array detectors. Target accelerations are modelled as random noise.

The missile components of a and a sy' and ubs are measured by sx on-board accelerometers and an angular rate gyro respectively.

The measured values of f and 8 are compared with estimated values of f and 6 from a state estimator 6, implemented in a microprocessor 5, in comparators 9 and 10, respectively, and the differences between real and estimated values of f and 6 (termed measurement residuals) are input to the state estimator 6 via a set 5 of optimal weighting gains (termed the Kalman gains). The state estimator includes a dynamic model of the missile and target dynamics and propagates estimates of , w, , p and 6.

The estimated states are fed back into the seeker processor to steer the seeker beam.

Estimates of range r may be useful in permittleg the seeker to differentiate the target against background and counte. Xeasures.

Claims

1. A missile guidance system including a strapdown seeker providing measurements of sight line look angle, X, with respect to the missile body axis and providing measurements of angular diameter, 8, of the target image at the seeker, and a state estimator for sightline spin having inputs of 6 and f from the seeker, and including means for mathematically modelling the dynamics of 6 and X, calculating the difference between the measured and estimated values of 6 and X, and generating corresponding corrections to estimated states from the estimator.

2. A missile guidance system as claimed in claim 1 wherein the state estimator a Kalman type filter.

3. A missile guidance system as claimed in claim 1 on claim 2 wherein the strapdown seeker inclues an opto-electronic array detector for providing measurements of 6 and f.

4. A missile guidance system as claimed in any preceding claim wherein target accelerations are modelled as a random noise process.

5. A missile guidance system substantially as herein described with reference to the drawings.

6. A missile having a guidance system as claimed in any pecedi claim.

6. A missile having a guidance system as claimed in any preceding claim.

Amendments to the claims have been filed as follows 1. A missile guidance system including a strapdown seeker providing measurements of sightline look angle, 9, with respect to the missile body axis and providing measurements of angular diameter, 0, of the target image at the seeker, and a computer including: a state estimator for generating updated estimates of state variables required for missile control; means for comparing said updated estimates of * and 6 with the latest values measured by the strapdown seeker; means for applying the resulting residuals as correction terms to the state estimator to generate a new updated estimate of Wand 6; and means for calculating updated values of spate variables ip and proportional range rate p derived frown 6 2. A missile guidance system as claimed in claim 1 wherein he means for applying the resulting residuals as correction erms to the state esuimator is a Kalman type filter.

3. A missile guidance system as claimed in claim 1 or clai 2 wherein he strapdown seeker includes an opto-electronic array detector for providing measurements of 6 and $ 4 A missile guidance system as claimed in any preceding claim wherein target accelerations are modelled as a random noise process 5. A missile guidance system substantially as herein described with reference to the drawings.