EP4232771B1 - Interceptor missile and method for steering same - Google Patents

Description

The invention relates to the guidance of an engine-driven steerable interceptor missile for intercepting a moving target, in particular a target missile, during a midcourse phase of the interception, and to such an interceptor missile.

An interceptor missile is launched to defend against a moving target, particularly an approaching target missile. After a launch phase, in which the interceptor missile leaves its launch base and begins its flight roughly in the direction of the target, the midcourse phase of its flight follows. This serves to overcome most of the distance to the target and to get close to it, in particular so close that the interceptor missile's on-board systems are sufficient to hit the target accurately in an endgame following the midcourse phase.

The EP 2 423 774 A1 describes a method for guiding a missile to a target whose possible destination is described by non-Gaussian statistics. From the EN 10 2010 032 281 A1 A method for controlling a guided missile driven by an engine is known, in which a process means of the guided missile calculates a trajectory property of a trajectory to a target point during the flight and controls the flight of the guided missile depending on the trajectory property. When calculating the trajectory property, unguided flight processes that influence the flight speed of the guided missile and are controlled by the process means are taken into account. The inclusion of future or current flight processes controlled by the process means into the flight control based on proportional navigation is possible, but complex. Such incorporation is easier if the process agent uses point-miss navigation instead of proportional navigation, in particular a method called Zero Effort Miss (ZEM) navigation.

The object of the invention is to propose improvements with regard to an interceptor missile or the guidance of an interceptor missile in the midcourse phase of its flight to a moving target.

The object is achieved by a method according to claim 1 for steering a steerable interceptor missile driven by an engine. Preferred or advantageous embodiments of the invention, as well as other categories of invention, emerge from the further claims of the following description and the attached figures.

The interceptor missile or its flight is used to intercept a moving target, in particular a target missile. The guidance process can be referred to as model predictive guiding. It is carried out during a mid-course phase of the interception. The mid-course phase is the phase of the interceptor missile's flight from its launch to entry into the endgame. The endgame begins with the activation of the on-board target search sensors (seeker head). During the mid-course phase, the target data comes in particular from the sensors of the higher-level weapon system and is transmitted to the interceptor via data link. The desired successful impact on the target represents the end of the mission. Alternatively, a mission abort (mission end) occurs if the target cannot be reached or is definitely missed or entry into the endgame is not possible or the interception is aborted or terminated for other reasons. The proposed control process then ends.

The interceptor missile is actually or really guided as follows: At each steering time, the interceptor missile generates real steering commands for itself based on the free control parameters currently available in the interceptor missile at the steering time, which are in the form of a parameter vector. "Real" means that the interceptor missile can actually is steered. Additional values can optionally be included in the steering commands, e.g. (free) parameters that are not part of the parameter vector.

The method assumes that when the interceptor missile enters the mid-course phase, i.e. when it is already in flight, or from this point onwards and preferably until the end of the mid-course phase or even beyond, there is always a current parameter vector of free control parameters for the interceptor missile, from which the real steering commands are then generated. This parameter vector forms in particular a suitable initial value for the steering, and if necessary also for optimization, as explained below.

The free control parameters are optimized permanently and/or repeatedly during the midcourse phase using an optimization process to optimize the control parameters. This optimization or optimization process takes place in parallel with the actual steering. This can be understood to mean that the control parameters in the current parameter vector (the basis of the real steering) initially remain unchanged. In this case, it is not the parameters currently used for steering that are optimized directly, but rather - figuratively speaking - a copy or image of these control parameters or the parameter vector outside of the actual steering process. In this respect, the optimization can also take place independently of the actual steering, which does not initially have to be influenced by the optimization taking place in parallel.

Newly detected information on the movement of the target and/or information on the flight of the interceptor missile are incorporated into the optimization process as soon as they are available. The optimization process can therefore always be based on the most current and best available data on the circumstances of the current mission.

Optimized control parameters are then adopted into the current parameter vector after, and in particular as soon as, they are available as a result of the optimization process. Only when the optimized parameters have been adopted into the current parameter vector, which is the basis for the real steering, do the optimized control parameters influence the actual steering. In the above picture, the optimized copy of the parameter vector is only then integrated, transferred or adopted into the parameter vector actually used for steering.

The invention is based on the following core idea:
An interceptor missile is guided in the midcourse phase in order to hit a moving and in particular (potentially) maneuvering target. For this purpose, the real steering commands are calculated from a vector (parameter vector) of free (control) parameters. These free parameters can be target trajectory angles, but they can also be target values for the lateral acceleration. In addition, in the case of a multi-stage engine, the ignition times of the respective stages can be treated as free parameters or, in the case of a controllable engine, its target thrust or the time-discrete target thrust curve. The invention is based on the idea that these free parameters are permanently and/or repeatedly optimized during the midcourse phase using a search method for parameter optimization. This optimization takes place independently and in parallel to the actual steering. Whenever new information on the target movement or the flight path of the interceptor (interceptor missile) is available, it is included in the optimization to determine better, ideally optimal parameters. The improved parameters are used by the actual steering as soon as they are available.

The following embodiments or variants of the invention are conceivable:
The optimization is carried out in particular with regard to an objective function (quality function / criterion / quality value), which in turn is based on a zero effort miss (ZEM) prediction of the trajectories of the target and interceptor. As soon as the closest approach of the target and interceptor (ZEM) is reached, the prediction (modified ZEM method) is aborted. The resulting trajectories are evaluated by the objective function (quality value). For example, it evaluates how close the interceptor comes to the target (ZEM), what speed the interceptor has at the end of its trajectory (maneuverability in the endgame), how long the engagement lasts (Tgo), and at what angle the target and interceptor meet, as well as any other sub-criteria.

The prediction of the target trajectory, i.e. the trajectory of the target, is carried out in particular on the basis of suitable hypotheses. For example, the target can be assumed to continue the current maneuver until it reaches a minimum approach speed and then continue to fly in a straight line with maximum thrust (evasive maneuver). Or there is information about possible attack targets that suggests corresponding target maneuvers. The target can also be assumed to have a ballistic or pseudo-ballistic trajectory. The hypotheses about the target trajectory are based on prior knowledge and the observation of the target trajectory up to the current point in time. There is extensive literature on this. The formation and use of hypotheses is not the subject of this invention.

In predicting the interceptor flight path, the free parameters to be optimized are used in particular, for example by implementing the target trajectory angles or the target values for the lateral acceleration using a behavior model of the missile, in the case of a multi-stage engine, the ignition times of the stages are selected accordingly, and in the case of an adjustable engine, the thrust or thrust curve is adjusted accordingly. In particular, the fuel consumption and mass loss as well as the existing restrictions (minimum and maximum adjustable thrust, no thrust after the fuel has been used up) are taken into account, as are the resistance and gravity in the known ZEM method. A step size control ensures in particular that events such as the ignition of an engine stage or the reaching of the ZEM are calculated precisely in terms of time.

Since the calculation of the objective function always involves the relatively complex step-size-controlled simulation of the control, it is particularly useful to use search methods that achieve the optimum or at least a significant improvement with relatively few iterations. Gradient-based methods are rather unsuitable for this due to the necessary approximation of the gradient using difference quotients. The simplex method according to Nelder Mead has proven to be very robust. This has been known in the state of the art for around 60 years.

According to the invention, improved guidance of the interceptor missile towards the target is achieved due to the continuously updated control parameters and their use for real guidance.

In a preferred embodiment of the method, the following optimization process is carried out, whereby the steps or the process can be terminated or aborted at the end of the midcourse phase. This is followed by guidance of the interceptor missile in the endgame, which - like the launch phase - is not part of the present patent application.

In step a), a predeterminable or predetermined parameter vector is selected as the current candidate of an MPC optimization method (Model Predictive Control). The MPC method is used to potentially determine control parameters that are improved compared to the predetermined parameter vector. The current candidate therefore forms a starting value for optimizing the control parameters using the MPC method.

In a step or process section b), a set of possible candidates (first and further subsequent candidates) for an improved parameter vector is determined as follows in or by carrying out the MPC optimization process; each of the candidates is assigned a quality value, which is also determined as part of the MPC process. The set can contain any number of candidates, although there can also be just one candidate, which is always replaced, for example, when a better candidate is available. The number of candidates to be used is simply a question of the optimization process selected. Of course, powerful processes work with several candidates. For example, the widely used Nelder-Mead process operates with a simplex of n+1 parameter vectors, where n is the length of the parameter vector. However, this is not important for the idea of the MPG or the present invention. Even a "stupid" random search process based on the motto of randomly varying the current parameter vector, evaluating the result and, in the event of an improvement, continuing with the variation as the new current parameter vector, would work. (Apart from the excessive computing time.) The sentence therefore contains in particular 1 to n candidates, where n depends on the optimization method, which, however, is not the subject of the invention.

The procedure section b) comprises steps c1) to c5):
In a step or process section c1), a modified ZEM procedure (Zero-Effort-Miss) is carried out on the current candidate as follows. The modified ZEM procedure comprises steps d1) to d4):
In a step or process section d1), iterative predictions are made at the respective step times as follows; the process section d1) comprises the steps d2) to d4):

In a step d2), a possible intercept trajectory of the interceptor missile is predicted based on the current candidate, taking into account the guidance of the interceptor missile. The guidance is carried out using virtual guidance commands that are only generated as part of the optimization process and are not used for the actual guidance of the interceptor missile. Instead, the guidance commands are used to virtually determine the predicted trajectory. However, the virtual guidance commands can be generated in the same way as the real guidance commands. This creates a realistic simulation of the trajectory.

In a step d3), a possible target trajectory of the target is predicted based on hypothetical maneuvers of the target.

In a step or loop d4), steps d2) to d3) are repeated iteratively until a ZEM approximation of the interceptor flight path and the target flight path is achieved. At the end of the loop, both flight paths (and possibly a corresponding remaining flight time, see below) are available until the ZEM is reached (i.e. the minimum distance between the flight paths, ideally zero if the interceptor missile can actually reach the target according to the prediction). Now continue with step c2) as follows:
In a step c2), based on the results of the ZEM procedure (the results are in particular: trajectories, ZEM, predicted time duration Tgo of the Flight along the flight paths until reaching the ZEM, etc.) a current quality value is determined based on a quality criterion and assigned to the current candidate.

In a step c3), the current candidate is stored as the first or next candidate in the set of candidates. The current quality value is assigned to the candidate as a quality value and is also stored in the set. When step c3) is reached for the first time, a first pair of values consisting of candidate and quality value is stored, the next time it is reached (see below) a second, then a third, and so on, until the MPC process is completed and the set of candidates is available. For example, after ten runs of steps c1) to c4), there are ten candidates with their quality values.

In a step c4), the existence of an end criterion of the optimization or the MPC optimization process is checked. If this has not yet been reached, ie the MPC optimization process has not yet finished its optimization of the current candidates or candidates in the set, the two steps e1) and e2) are carried out:
In step e1), the current candidate is varied to a varied candidate using an MPC search procedure. A second candidate is created from the first candidate, a third from the second, and so on. The search procedure is used to numerically optimize the free parameters.

In step e2), the varied candidate just determined is adopted as the current candidate and the process continues with or returns to step c1).

In a step c5), which represents the alternative to step c4), if the end criterion is reached, the process continues with step f):
In step f), the process returns to step a) and continues from there.

During the entire midcourse phase or the execution of the above-mentioned process steps, (in a certain sense parallel to this) at predefined At correction times, one of the currently available candidates is selected according to a correction criterion and the current parameter vector is replaced by the selected candidate. This means that optimized control parameters are incorporated into the current parameter vector. From this point on, the real steering commands can then be generated on a modified basis, namely on the basis of a modified parameter vector or optimized control parameters that have been improved in relation to the target.

Optionally, in step c3), one or more of the candidates determined in the MPC process are rejected according to a rejection criterion and removed from the set together with their quality values. In this way, the set is kept small and unnecessary candidates are removed.

The "modification" of the ZEM procedure consists in the fact that in a conventional or known or usual ZEM procedure both a virtual guidance of the interceptor missile based on a parameter vector in the form of the current candidate is taken into account, as well as hypothetical maneuvers for the target trajectory of the target.

The procedure begins after launch, i.e. when the interceptor missile is already in flight. At the moment the procedure begins, it can therefore be assumed that a current parameter vector is already available that was used to guide the interceptor missile in the launch phase. The current parameter vector at the end of the launch phase can therefore be selected as a predefinable parameter vector of the procedure.

The method can be terminated when the mid-course phase is finished and the endgame steering begins. Both the MPC and the ZEM method are known in various forms in the state of the art, so that they are not explained in more detail here. Any form of the respective known individual method can be used and combined in embodiments of the invention. In particular, for example, the prediction step size in the ZEM method is controlled according to known procedures, e.g. in such a way that the time steps become smaller as the target is approached. As "Hypothetical maneuvers" of the target are particularly possible: unaccelerated movement (zero effort), ballistic trajectory, known or suspected evasive maneuvers or any other a-prior knowledge about the target. When determining the trajectory of the interceptor missile and/or the target, passive effects such as non-controllable thrust, decreasing weight depending on fuel consumption, air resistance, etc. are taken into account in addition to the free parameters.

According to one embodiment of the invention, zero effort miss guidance (ZEM) is combined with the model predictive control (MPC) approach. The control parameters for shaping the trajectory of the interceptor missile are optimized permanently (correction time) and online (i.e. during the midcourse phase, by the interceptor missile itself) using prediction models for the target (step d3), predicted course of the trajectory based on suspected maneuvers, etc.) and interceptor missile (interceptor, step d2, predicted course of the trajectory based on the guidance model, etc.).

The procedure can therefore also be called "Model Predictive Guidance (MPG)".

The method makes it possible to use an estimate of the target acceleration (hypothetical maneuvers) to guide the interceptor missile. The proposed method provides a starting point for the guidance of a missile that approaches over long distances and has a controllable thrust profile (according to a free parameter), such as an interceptor missile based on a ramjet propulsion system (ramjet interceptor).

According to one embodiment of the invention, a modified MPC is applied in the field of missile guidance. According to one embodiment of the invention, the combination of the MPC and ZEM prediction modified and working together in this way results in the guidance of an interceptor missile.

According to one embodiment of the invention, during the midcourse phase, two processes in particular run alongside each other or in parallel and to a certain extent independently of each other: The first process is the generation of Steering commands based on a parameter vector that is currently available (at the moment the steering command is generated). The second process is the optimization of the parameter vector. Using the MPC method, a set of possible alternative parameter vectors is generated and each of these parameter vectors is given a quality value. A modified ZEM prediction is used within the MPC method. Using the second process, if one is found, an optimized parameter vector (e.g. better quality value than the first parameter vector that corresponds to the current one from the steering command generation) is selected and the current parameter vector in the first process is replaced by the optimized parameter vector. The first process then generates the steering commands on the basis of the improved, replaced parameter vector.

Both processes run independently of each other in that a certain number of steering commands are generated from one and the same parameter vector before the parameter vector from the second process is replaced at a later point in time. The reason for this is, for example, that the MPC process takes a certain amount of time before an improved parameter vector is found, but in the meantime steering commands continue to be generated at shorter intervals.

In a preferred embodiment - as already explained above - in step a) the predefinable parameter vector is specified by selecting the last current parameter vector of a start phase preceding the midcourse phase as the predefinable parameter vector. In an alternative embodiment, a parameter vector is selected that corresponds to the prediction of a direct approach to the target. In particular, two MPC evaluations are carried out based on these two different first candidates and the parameter vector that leads to the better quality value (quality, quality measure) is selected as the first candidate. This ensures good starting conditions for the MPC process in the midcourse phase.

In a preferred embodiment, the correction time is chosen to be the reaching of the end criterion in step c5) and - as correction criterion - the candidate selected from the set to which the best quality value is assigned. The end of the optimization is therefore waited for and only then is the parameter vector actually used for steering replaced. This new parameter vector is the best (best quality value) for route guidance that could be determined using the optimization.

In a preferred embodiment, step c3) is selected as the correction time and in step c3) the current candidate (which has just been or is being saved as a candidate in the set together with its quality value) is additionally adopted as the current parameter vector as a correction criterion if its assigned quality value is the best of all quality values previously present in the set. This means that the current parameter vector, i.e. the one used for the actual generation of steering commands, is not updated until after the optimization process has been completed (step c5)), but already during its processing. Optimizations are therefore incorporated earlier into the steering behavior and thus the flight path of the interceptor missile.

This strategy of replacing the current parameter vector according to the mentioned alternatives can also be varied for different iterations of the MPC procedure.

In a preferred embodiment, in step e1) the variation towards a further or varied candidate is carried out at least partially based on the previous candidates and their quality values. One or several or all of the candidates/quality values determined so far in the MPC method or set are thus used in the search method in order to enable an improved determination of a next potential candidate.

In a preferred embodiment, the quality criterion contains at least as a sub-criterion: a minimum deviation from the target (ZEM, closest approach to/distance of the interceptor missile from the target) and/or a maximum final speed when hitting the target and/or a minimum remaining flight time to the target and/or a desired angle of impact on the target. The corresponding sub-criteria or their result values are in particular assigned evaluation factors in order to ultimately generate a quality value. All of these sub-criteria are those which are ultimately decisive for a successful or even as effective approach to / combating the target.

In a preferred embodiment, the method is designed for an interceptor missile whose engine is a solid fuel booster or a two-pulse engine or a controllable engine. In the case of a solid fuel booster, the method takes into account in particular the remaining burn time for the mid-course phase in step d2). In the case of a two-pulse engine, the ignition time for the ignition of the second engine stage is taken into account as a free parameter in the parameter vector and is also optimized in particular within the framework of the M PC method. In the case of a controllable (or adjustable) engine, e.g. a ramjet, the thrust control value or the course of the thrust control value over time is taken into account as a free control parameter in the parameter vector and is also optimized in particular. For all three variants, the weight of the interceptor missile, which decreases with fuel consumption, is taken into account in particular in step d2).

In a preferred embodiment, in step a) a currently predicted remaining flight time of the interceptor missile until its mission end is also determined. The end of the mission is in particular when it hits the target or reaches a minimum distance from the target (ZEM). This remaining flight time (also "Tgo") can also optionally be used as a free control parameter and/or as a sub-criterion for the quality criterion (e.g. the shortest possible remaining flight time) and/or for determining step sizes in the ZEM process. As part of the ZEM prediction, a current remaining flight time can then also be determined, namely as the time at which the ZEM is reached.

In a preferred variant of this embodiment, the current parameter vector and thus in particular also the predefinable parameter vector and/or the current candidate, etc., is one in which at least one of the free parameters is a value based on the remaining flight time or a sequence of partial values based on the remaining flight time. A corresponding value is, for example, the ignition time mentioned above for a two-pulse engine. A sequence of partial values is used, for example, for the thrust control value (as a free parameter) of a controllable engine as follows: the remaining flight time to the target in an optimization process is divided into n, e.g. n=5, in particular equal-length time periods, and a specific thrust control value is constantly assigned to each time period as a partial value. In the optimization process, a thrust curve that runs over time in n or five steps (corresponding to the time periods) is used and optimized in step d2) for predicting the flight path of the interceptor missile. In particular, a variable thrust curve provides a free control parameter to enable the interceptor missile to react particularly well to highly agile evasive maneuvers by the target. For this process variant, the most current remaining flight time of the interceptor missile to the target is always predicted.

In a preferred variant of this embodiment (in the alternative or variant of a sequence of partial values), as already explained above by way of example, the predicted remaining flight time is taken into account in step d2) in such a way that it is divided into a predeterminable number of time periods in the ZEM method, and a different partial value is taken into account for each time period. As explained above, the parameter vector therefore contains a free parameter, which in turn is formed from a sequence of partial values and represents, for example, a thrust curve in 5 stages/time periods.

In a preferred variant of this embodiment - as already explained above - the value or the partial values are an ignition point, which depends on the remaining flight time or is determined by reaching a certain point in time, or several ignition points of a respective first or further combustion stage of one or more engines of the interceptor missile. This embodiment is suitable for interceptor missiles that contain one or more single- or multi-stage engines, whereby such a stage of the engine can be assigned its own ignition point to be optimized.

In a preferred variant of this embodiment - as already explained above - at least one of the values or partial values is a thrust control value dependent on the remaining flight time for an engine of the interceptor missile that is controllable with regard to its thrust. Here, the dependency exists, for example, in a partial or continuous variation of the thrust during the remaining flight time.

In a preferred variant of this embodiment, at least one of the partial values is a control value for a transverse acceleration element of the interceptor missile that depends on the remaining flight time. The control of a corresponding transverse acceleration element leads to a transverse acceleration, i.e. a change in direction of the interceptor missile. The statements made above regarding a controllable thrust curve apply accordingly to the control of a corresponding transverse acceleration according to a time curve.

In a preferred embodiment, the current parameter vector (in particular also a predeterminable candidate, see above) is selected as one that contains at least two trajectory angles for the flight path of the interceptor missile as two free parameters. This results in a particularly simple optimization problem for the MPC method. This also enables a particularly fast reaction to highly agile targets, so that the interceptor missile can follow them particularly well.

The object of the invention is also achieved by an interceptor missile according to claim 15. The interceptor missile is at least temporarily powered by its engine (at least one) during its flight and can be steered using real steering commands. The method can also continue to be used after all engines have burned out. For this purpose, it has a steering device, e.g. controllable rudders or transverse acceleration means, e.g. control nozzles, which are actuated using steering commands and serve to steer the flying interceptor missile. The interceptor missile also serves to intercept a target. The interceptor missile contains a current parameter vector of free control parameters for the interceptor missile, on the basis of which, as explained above, real steering commands for the steering are generated. The interceptor missile also contains a control and evaluation unit. The control and evaluation unit is set up to carry out the method according to the invention.

The interceptor missile and at least some of its embodiments as well as the respective advantages have already been explained in connection with the method according to the invention.

The control and evaluation unit is set up, adapted or configured to carry out the method according to the invention. "Set up" / "adapted" / "configured" is to be understood in such a way that the control and evaluation unit is not only suitable for carrying out the relevant steps/functions, but rather has been specially designed for this purpose. The control and evaluation unit is "set up" accordingly, in particular by programming a computing device contained therein or by hardwiring it.

The invention is based on the following findings, observations and considerations and also has the following embodiments. The embodiments are sometimes referred to as "the invention" for the sake of simplicity. The embodiments can also contain parts or combinations of the above-mentioned embodiments or correspond to them and/or possibly also include embodiments not previously mentioned.

New types of hypersonic weapons such as an HGV (hypersonic glide vehicle) or an HCM (hypersonic cruise missile) represent a new type of threat against which conventional interceptor missiles can hardly be used successfully.

The invention is based on the idea of using an interceptor missile, e.g. a so-called "Ramjet Interceptor" (RJI), namely a multi-stage missile based on a ramjet drive, against such hypersonic targets, i.e. against hypersonic weapons. The invention is also based on the idea of creating a guidance concept for the midcourse phase of such an interceptor missile. While various concepts exist for the endgame, most of which are based on recognizing the target maneuver and directly switching to the guidance of the interceptor missile, the midcourse phase of existing concepts is limited to the best possible determination of the encounter point (predicted intercept point = PIP) and the trajectory to it. For the new target class of potentially highly maneuverable hypersonic glide vehicles (HGV) or hypersonic cruise missiles (HCM), the PIP approach is not sufficient because the approach of the interceptor missile takes a comparatively long time and the target can develop large deviations from an originally suitable PIP during this time.

To date, guidance in the midcourse phase has mostly been based on the PIP approach. This is determined using all available a priori knowledge of the respective weapon system and the interceptor missile only has the task of flying to this PIP and ensuring a handover from the weapon system's guiding sensor (radar) to the on-board sensor (seeker).

According to a more flexible approach, the interceptor missile sets its own PIP. In any case, significant target maneuvers in the mid-course phase result in a shift in the PIP and thus a deviation from the original optimal trajectory. Therefore, novel methods are based on classifying non-significant target maneuvers as such and avoiding unnecessary shifting of the PIP and the associated loss of energy. A concept for the explicit treatment of maneuvering targets in the mid-course phase is the aim of the present invention. As mentioned above, there are a large number of known solutions regarding endgame guidance (terminal guidance).

The MPC approach in embodiments of the present invention consists in not only predicting the target and interceptor trajectories (flight paths of the target and interceptor missile) with a ZEM predictor, but also optimizing suitable control parameters using numerical, real-time search methods, e.g. in every nth steering cycle. The prediction of target and interceptor movement, in particular based on the zero-effort approach, is a well-known concept. When predicting the target trajectory, the estimated target accelerations can be used using game theory hypotheses. In one conceivable implementation, for example, the goal can be to minimize the approach speed of the interceptor with the current maneuver (evasive maneuver). The method is based on the idea of free control parameters to be optimized. In a first approach, these can be the trajectory angles of the interceptor at the current time. If the optimization has determined the optimal trajectory angles, then These are interpreted as target path angles (changed current parameter vector) and commanded in the form of path steering (generation of the real steering commands).

The optimization can use a similar cost function (quality) as in an offline process (determination of the trajectory by a guidance system outside the interceptor missile). Criteria such as minimum offset, maximum terminal speed, minimum remaining flight time (time to go = Tgo), as well as geometric requirements such as certain impact angles can be used in the cost function. In other implementations, for example, the ignition time of a second engine pulse can be used as a parameter to be optimized. Finally, it is possible to discretize the controllable thrust curve of a jet, ramjet or gel engine in time (sequence of partial values) and to determine it optimally iteratively in the sense of the MPC.

Figur 1

ein Prinzipdiagramm des erfindungsgemäßen Verfahrens.

Further features, effects and advantages of the invention emerge from the following description of a preferred embodiment of the invention, as well as the attached figures. Each of these shows a schematic principle sketch:

Figure 1

a principle diagram of the method according to the invention.

The method is used to guide an interceptor missile 2 that is driven by an engine 4 - here the thrust of which is controllable - and is steerable, here by the control of the tail units (not shown in detail). The steering is carried out by the implementation of real steering commands 6 in the interceptor missile 2 to the engine 4 and the tail units (not explained in detail). The interceptor missile 2 is used to intercept a target 8. The method is carried out exclusively in a midcourse phase PM of the flight of the interceptor missile 2, i.e. the interception of the target 8.

The guidance is based on a current parameter vector 10. The parameter vector 10 contains a series, here three, of free control parameters SP1-3 for the interceptor missile 2. The control parameters SP1 and SP2 are trajectory angles, the control parameter SP3 is a thrust control value for the engine 4, which comprises a total of five partial values SP3a-e. The respective remaining flight time Tgo of the interceptor missile 2 until impact with the target 8 is divided into five equal time periods. In each of these time periods, the engine 4 is controlled in turn by a corresponding thrust control value SP3a-e.

At the beginning of the procedure, the launch phase of the interceptor missile 2 has just ended and the midcourse phase PM begins. When entering the midcourse phase PM, a current parameter vector 10 is available. At the respective steering times, here every 10 ms, a respective real steering command 6 is generated from the parameter vector 10 and the interceptor missile 2 is guided based on these steering commands 6.

The method begins with a step a) in which a predefinable parameter vector 15 is selected as the current candidate 12 of an MPC optimization method 14. In the present case, the specification is made in such a way that the current parameter vector 10 available from the end of the start phase is used as the predefinable parameter vector 15. The MPC optimization method 14 is used to determine an improved parameter vector that is to replace the current parameter vector 10.

Now the MPC optimization process 14 begins. Within this process (step or loop b)), a set of 16 possible candidates 18a-c, here three in the example, each with assigned quality values 20a-c is determined. Each of the candidates 18a-c is a possible parameter vector that could replace the parameter vector 10 if it would promise an improved mission success than the currently actually available parameter vector 10.

Based on the current candidate 12, a modified ZEM method 22 is now carried out in a step c1): In a step or loop d1), the following steps are carried out iteratively at respective step times t1, 2, 3,...:
In a step d2), a possible intercept trajectory 24 of the interceptor missile 2 is predicted. For this purpose, virtual steering commands 7 (corresponding to the real steering commands 6) are determined based on the current candidate 12 at the respective step times t1, 2, 3,..., so that respective predicted locations (circles in the figure) of the interceptor missile 2 result. The trajectory 24 results from the temporal or spatial sequence of the locations. In other words, it is iteratively simulated how the interceptor missile 2 would move if the current candidate 12 were used as parameter vector 10 for its guidance.

Furthermore, in a step d3), locations corresponding to the step times t1, 2, 3,... and thus iteratively a target flight path 28, i.e. a flight path of the target 8, are predicted, but here taking into account a respective hypothetical flight maneuver 26 of the target 8. For example, it is assumed that the target 8 flies a certain assumed evasive curve in order to escape the interceptor missile 2.

According to a step or loop d4), steps d2) and d3) are repeated iteratively for as many times t1, 2, 3,... until a ZEM approximation 30 of intercept trajectory 24 and target trajectory 28 is reached. This terminates the ZEM method 22.

The results 32 of the ZEM procedure 22 in the example are the achievable ZEM approach 30, an updated remaining flight time Tgo, the impact speed and the impact angle of the interceptor missile 2 at the target 8, etc.

In a step c2), a current quality value 33 is determined for the respective candidate 12 on the basis of these results 32 and assigned to it. The assignment is made based on a quality criterion 36.

In a step c3), the current candidate 12 is stored together with its determined goods value 33 in the set 16 as candidate 18a-c with quality value 20a-c. In the first run, the quality value 20a is assigned to candidate 18a, in subsequent runs the quality value 20b is assigned to candidate 18b and stored in the set 16, etc.

In a step c4), an end criterion 38 for the optimization process 14 is now checked. If this is not achieved, in a step e1), the current candidate 12 is varied using an MPC search process 40 to a varied candidate 42. This is adopted as the current candidate 12 in a step e2), and the MPC Optimization procedure 14 is started again with the now optimized or modified candidate 12.

In the example, the optimization process 14 is run three times, resulting in three candidates 18a-c with assigned quality values 20a-c. Then the end criterion 38 is reached, namely the fixed number of three process runs.

Since the end criterion 38 is reached, we return to step a) to calculate a new set 16.

The procedure ends or is aborted when the midcourse phase PM is completed.

During the duration of the procedure, one of the candidates 18a-c is selected at a respective predeterminable correction time TK according to a correction criterion 44 and is used from then on as the current parameter vector 10 for the actual guidance of the interceptor missile 2. In the example, the correction time TK is always the reaching of the end criterion 38. The correction criterion 44 is the selection of the candidate 18a-c from the set 16 to which the best quality value 20a-c is assigned in the respective current set 16.

An alternative possibility is to choose step c3) as the correction time TK and (from the second test / determination of the quality value) to make the best of the candidates 18a-c tested so far the parameter vector 10. The best is present if its quality value 20b-c is better than the quality values 20a-c of the candidates 18a-c previously present in set 16.

In the present case, in step a), a currently predicted remaining flight time Tgo of the interceptor missile 2 to the target 8 is also determined in order to have a time basis for the use of the control parameters SP3a-e in step d2). An updated remaining flight time Tgo is also available as part of the result 32 at the end of each run of the ZEM method 22 and can be used from then on.

The current parameter vector 10 is always present in the interceptor missile 2. The interceptor missile 2 also contains a control and evaluation unit 50, here a central computer, which is set up to carry out the method according to the invention. The "setup" here is carried out by means of appropriately powerful hardware and programming in order to implement the method.

List of reference symbols

2

Interceptor missiles

4

Engine

6

Steering command (real)

7

Steering command (virtual)

8

Goal

10

Parameter vector (current)

12

Candidate (current)

14

MPC optimization method

15

Parameter vector (specifiable)

16

Sentence

18a-c

candidate

20a-c

Quality value

22

ZEM procedure

24

Intercept trajectory

26

Flight maneuvers (hypothetical)

28

Target trajectory

30

ZEM approach

32

Results

33

Quality value (current)

36

Quality criterion

38

End criterion

40

MPC search procedure

42

Candidate (varies)

44

Correction criterion

50

Control and evaluation unit

SP

Control parameters

Tgo

Remaining flight time

PM

Midcourse Phase

t1, 2, 3,...

Step time

TK

Correction time

Claims (15)

1. Method for steering a steerable interceptor missile (2) powered by an engine (4) for intercepting a moving target (8) during a midcourse phase (PM) of the interception,

   wherein the interceptor missile (2) is steered by means of real steering commands (6), which are generated at respective steering times on the basis of free control parameters (SP), which are available in the form of a current parameter vector (10),

   wherein the free control parameters (SP) are constantly and/or repetitively optimized in the course of the midcourse phase (PM) by means of an optimization procedure (14) for optimizing the control parameters (SP),

   wherein the optimization procedure (14) takes place in parallel with the actual steering,

   wherein optimized control parameters (SP) are taken into the current parameter vector (10) once they are available from the optimization procedure (14), characterized in that newly detected information about the movement of the target and/or information about the flight of the interceptor missile are included in the optimization procedure as soon as they are available.
2. Method according to Claim 1,
   characterized in that

   during the midcourse phase (PM) the following optimization procedure (14) is performed:

   a) a predeterminable parameter vector (15) is selected as the current candidate (12) of a model predicted control (MPC) optimization procedure (14) to determine improved control parameters (SP);

   b) in the MPC optimization procedure (14), a set (16) of possible candidates (18a-c) for an improved parameter vector (10) with associated quality values (20a-c) is determined as follows:

   c1) based on the current candidate (12), a modified zero effort miss (ZEM) procedure (22) is performed as follows:
   d1) at each step time (t1-3) iterative predictions are made as follows:

   d2) a possible interceptor trajectory (24) of the interceptor missile (2), taking into account virtual steering commands (7) of the interceptor missile (2) based on the current candidate (12), d3) a possible target trajectory (28) of the target (8) based on hypothetical manoeuvres (26) of the target (8),

   d4) steps d2) to d3) are repeated iteratively until a ZEM approach (30) of the interceptor trajectory (24) and the target trajectory (28) is achieved,

   c2) on the basis of the results (32) of the ZEM procedure (22), a current quality value (33) is determined on the basis of a quality criterion (36) and is assigned to the current candidate (12),

   c3) the current candidate (12) is successively placed in the set (16) as the first (18a) or further candidate (18b-c) together with the current quality value (33) as an assigned quality value (20a-c),

   c4) if an end criterion (38) of the optimization has not yet been reached:

   e1) an MPC search procedure (40) is used to vary the current candidate (12) to a varied candidate (42),

   e2) the varied candidate (42) is henceforth adopted as the current candidate (12) and the procedure continues with step c1),

   c5) if the end criterion (38) is achieved, the procedure proceeds as follows:

   (f) it returns to step (a),

   wherein during the midcourse phase (PM) at predeterminable correction times (TK) one of the candidates (18a-c) is selected according to a correction criterion (44) and the current parameter vector (10) is replaced by the selected candidate (18a-c) in order to transfer optimized control parameters (SP) into the current parameter vector (10) as a result.
3. Method according to Claim 2,
   characterized in that
   achievement of the end criterion (38) in step c5) is selected as the correction time (TK) and the candidate (18a-c) from the set (16) to which the best quality value (20a-c) is assigned is selected as the correction criterion (44).
4. Method according to Claim 2 or 3,
   characterized in that
   step c3) is selected as the correction time (TK) and additionally in step c3) the current candidate (12) just stored as a candidate (18a-c) is also adopted as the current parameter vector (10) as the correction criterion (44) if its assigned quality value (20a-c) is the best of all quality values (20a-c) available in the set (16) so far.
5. Method according to any one of Claims 2 to 4, characterized in that
   in step e1) the variation to a varied candidate (42) is carried out at least partially based on the candidates (18a-c) so far and the quality values (20a-c) thereof.
6. Method according to any one of Claims 2 to 5, characterized in that
   the quality criterion (36) contains at least as a subcriterion: a minimum deviation from the target (8), a maximum final speed when hitting the target (8), a minimum remaining flight time to the target (8), and a desired angle of impact on the target (8).
7. Method according to any one of Claims 2 to 6, characterized in that
   in step a) a currently predicted remaining flight time (Tgo) of the interceptor missile (2) until the end of the mission thereof is additionally determined.
8. Method according to Claim 7,
   characterized in that
   a parameter vector for which at least one of the free parameters (SP) is a value oriented to the remaining flight time (Tgo) or a sequence of subvalues (SP3a-e) oriented to the remaining flight time is used as the current parameter vector (10).
9. Method according to Claim 8,
   characterized in that
   for the variant of a sequence of subvalues (SP3a-e):
   in step d2) the predicted remaining flight time (Tgo) is taken into account in such a way that it is divided into a predeterminable number of time periods in the ZEM procedure (22), and for each time period a respective different one of the subvalues (SP3a-e) is taken into account
10. Method according to any one of Claims 7 to 9,
    characterized in that
    at least one of the values or subvalues (SP3a-e) is an ignition time of a respective first or further combustion stage of one or more engines (4) of the interceptor missile (2), said ignition time being dependent on the remaining flight time (Tgo).
11. Method according to any one of Claims 7 to 10, characterized in that
    at least one of the values or subvalues (SP3a-e) is a thrust control value for an engine (4) of the interceptor missile (2) controllable with respect to its thrust, said thrust control value being dependent on the remaining flight time (Tgo).
12. Method according to any one of Claims 7 to 11, characterized in that
    at least one of the subvalues (SP3a-e) is a control value for a lateral acceleration element of the interceptor missile (2), said control value being dependent on the remaining flight time (Tgo).
13. Method according to any one of Claims 2 to 12, characterized in that
    a parameter vector which contains at least two trajectory angles for the trajectory of the interceptor missile (2) as two free parameters (SP1,2) is selected as the current parameter vector (10).
14. Method according to any one of the preceding claims, characterized in that
    the method is designed for an interceptor missile (2), the engine (4) of which is a solid booster or a dual-pulse engine or a steerable engine.
15. Interceptor missile (2) which is propelled by an engine (4) and is steerable by means of real steering commands (6) and is used to intercept a target (8), and which contains a respective current parameter vector (10) of free control parameters (SP) for the interceptor missile (2) and contains a control and evaluation unit (50) which is set up to carry out the method according to any one of Claims 1 to 14.

Images