RU106760U1 - OPTICAL ELECTRONIC SUPPORT SYSTEM - Google Patents

Abstract

 1. An optical-electronic tracking system comprising a programmed motion assignment unit, a coordinate prediction device, an optical-electronic unit in series, a strobe coordinate determination unit, a switch, a raster coordinate formation unit and a guidance and stabilization device, as well as a disturbance sensor connected in series and a shaper of inertial tracking logic, while the first output of the program move job block is connected to the second input of the switch, the output of the block coordinate in the raster is connected to the second input of the coordinate determination unit in the strobe and the first input of the coordinate prediction device, the output of the inertial tracking logic generator is connected to the third input of the coordinate determination unit in the strobe, characterized in that an additional motion sensor is introduced, while the program task unit the movement is additionally connected to the second input of the coordinate prediction device, the output of the media motion sensor is connected to the third input of the predictive device coordinates, the output of which is connected to the input of the guidance and stabilization device. ! 2. The system according to claim 1, characterized in that the coordinate prediction device is based on a series-connected first adder, a trajectory filter, a second adder and a constant part filter, the first input of the coordinate prediction device is connected to the second input of the first adder, the output of which is connected to the input trajectory filter, the output of the latter through the first delay block is connected to the first input of the first adder, t

Description

The utility model relates to the field of optoelectronic control systems designed for automatic tracking of moving objects from a moving base.

An analogue of the proposed solution is the circuit of optoelectronic systems (ECO) (USSR copyright certificate No. 1107340, publ. 08/07/1984). The ECO contains an optical-electronic device (OEP), which generates a video signal from the object, connected to the coordinate determination unit in the strobe (BOX), which determines the signal mismatch of the object image relative to the center of the raster. From the output of the BOX, the signal enters the block forming the coordinates in the raster (BFKR). The output of the BFKR is connected to the second input of the BOX for forming the coordinates of the object to the next frame (half frame). The output of the BFKR is also connected to the input of the guidance and stabilization device (ONS), which includes a block of corrective filters (BKF) and a device for stabilizing the field of view (CSS) (Fig. 1). In case of interruption of optical visibility, a zero signal is formed at the output of the BOX, and at the output of the BFKR, which is usually an adder, the coordinates of the strobe are frozen (Fig. 2).

The disadvantage of such an extrapolation of coordinates is the “freezing” of their estimates for the time that there are no reliably measured coordinates, which increases their fluctuation measurement error. This is ineffective, especially with prolonged lack of visibility.

The closest analogue to the proposed technical solution is an optoelectronic tracking system containing serially connected OEP, BOX, the first switch (Kom1), BFKR, the second switch (Kom2) and ONS, as well as the perturbation sensor (DV) connected in series, the inertial logic driver tracking (FLIS) and coordinate prediction device (CPC). In this case, the first, second, third, fourth and fifth outputs of the BOX are connected to the first, second inputs of the programmed motion assignment block (BZPP), second and third im KOM1 inputs and the second input Fleece. The third input of the FLIS is an external control input, and the second and third outputs of the FLIS are connected to the second input of the BOX and the second input of Com2. The output of the BFKR is connected to the third input of the BOX and the second input of the CPC. The third input of the CPC is connected to the second output of the ONS, and the output of the CPC is connected to the third input of Com2, whose first and second outputs are connected to the first and second inputs of the ONS (RF patent No. 2191407, publ. September 20, 1998).

The introduction of the DV, FLIS, UPC blocks, as well as Kom1 and Kom2, allows one to take into account a priori information about the loss of optical visibility. In this case, the DW, which forms the sign of the transition to extrapolation according to a priori data, is connected to the FLIS. The FLIS is connected to the input of the BOX and forcibly interrupts the measurement of coordinates, and is also connected to the input of the control device, providing in the extrapolation mode the formation of extrapolated control signals at the input of the ONS. The switches Kom1 and Kom2, respectively, according to signals from the BZPP and FLIS, make the transition from the auto tracking mode to the extrapolation mode, while the control signals are “frozen” at the output of the BFKR, and the control signals are sent to the ONS from the control panel. For example, if information about the transition to the extrapolation mode is not received from the DV, but the criterion function in the BOX does not exceed the threshold level, then from BZPP to Com1 a signal is received to interrupt communication with the BOX, which is equivalent to “freezing” the signals from the output of the BFKR.

The disadvantage of the closest analogue is, as in the analogue, the "freezing" of coordinates in the mode of rare (short in duration) interruptions in optical visibility, which is not an effective way of extrapolation and leads to an additional fluctuation tracking error. This can reduce tracking accuracy, especially in the long-path extrapolation mode.

Another disadvantage of this solution to the synthesis problem of the ECO control loop in the regime of interruption of optical visibility is that information from other motion sensors, for example, from the ECO carrier, is not fully taken into account. So, the input perturbation α _in (t) of the ECO control loop can be represented as the sum of the angular displacement of the object α _o (t) and the support α _n (t), arbitrarily moving in space, in the form

In this case, the second term (1), due to the motion of the carrier, can be measured independently of the optical visibility of the object. In addition, the second term of expression (1) has a wider spectrum of perturbations. For example, even if the accompanied object is maneuvering air, then the maximum width of the spectrum of the input disturbance α _o (t) (see Singer R. Evaluation of the characteristics of the optimal filter for tracking a manned target // Foreign Radio Electronics. - 1971. - No. 8) is equal to

The maximum width of the spectrum of the input disturbance α _n (t), due to the movement of the air carrier (see p. 232, Barsukov F.I., Velichkin A.I., Sukharev A.D. Television systems of aircraft. - M: Sov. Radio , 1979), taking into account (see p. 44, p. 74 Mikhalev I.A. et al. Automatic airplane control systems. - M: Mechanical Engineering, 1971), is

which is much wider than the perturbation spectrum of the trajectory of an air object (2).

If the object is ground-based, then the spectrum of perturbations of its trajectory does not exceed the spectrum of perturbations of the trajectory of an air object (2). In the simplest case, an object can be completely motionless, and if the carrier’s own motion were taken into account, optical visibility would not be necessary at all.

Thus, in the solution of the closest analogue, “frozen” values of sample samples are used, which leads to a significant increase in the fluctuation error of target tracking. In addition, information about the carrier’s own movement is not taken into account, which leads to a large dynamic error in tracking objects.

The task to which the proposed technical solution is directed is to increase the noise immunity and to maintain the line of sight on simultaneously tracked objects during the maneuver of the carrier and objects.

The technical result consists in a significant increase in the allowable duration of natural and organized interference, against which the detection and tracking of objects is performed, as well as in an increase in the number of simultaneously tracked objects.

The problem is solved, and the technical result is achieved due to the fact that the optoelectronic tracking system contains serially connected OEP, BOKS, Com, BFKR, UPC and ONS, as well as serially connected DV and FLIS. In this case, the first output of the BZPP is connected to the second input of the Com. The output of the BFKR is connected to the second input of the BOX and the first input of the CPC. The output of the FLIS is connected to the third input of the BOXING.

The novelty lies in the fact that an additional motion sensor has been introduced (media). In this case, the BZPP is additionally connected to the second input of the CPC. The output of the DDN is connected to the third output of the CPC, the output of which is connected to the input of the ONS.

The essence of the proposed solution is that

- firstly, the CPC provides the closed-loop ECO transmission coefficient of the optimal trajectory filter both in the mode of short and long interruptions in optical visibility, which provides a significant reduction in the fluctuation error of the auto tracking circuit;

- secondly, the carrier motion sensor (BOT) can significantly reduce the range of input disturbances of the ECO circuit and, as a result, significantly reduce the dynamic tracking error.

When using the proposed solutions, the total dynamic and fluctuation error of object tracking is significantly reduced, which is especially important in the mode of extrapolating the trajectory of the movement of objects.

The proposed technical solution is illustrated by the following graphic materials:

Figure 1 - Functional diagram of an analog of an optical-electronic system for tracking objects Figure 2 - Functional diagram of an analog of an optoelectronic tracking system of objects with “freezing” strobe coordinates Figure 3 - Functional diagram of the optoelectronic system for tracking the objects of the proposed solution Figure 4 - Functional diagram of a coordinate prediction device Figure 5 - Block diagram of an optoelectronic system for tracking objects 6 - Collapsed block diagram of an optoelectronic object tracking system 7 - Modules of transmission coefficients of optoelectronic systems of the prototype and the proposed solution in the absence of rare interruptions in optical visibility (T _ek = T _d )

Fig. 8 - Modules of transmission coefficients of the optoelectronic systems of the prototype and the proposed solution in the absence of rare interruptions in optical visibility (T _ek = 2T _d ) Fig.9 - Comparative analysis of the effectiveness of auto tracking modes and prolonged extrapolation in the ECO control loops of the prototype and the proposed solution

The system contains series-connected OEP, BOX, Com, BFKR, UPC and ONS, as well as series-connected DV and FLIS. The system also contains BZPP and DDN. In this case, the first output of the BZPP is connected to the second input of the Com. The output of the BFKR is connected to the second input of the BOX and the first input of the CPC. The output of the FLIS is connected to the third input of the BOXING. The DDN output is connected to the third input of the control device, the second input of which is connected to the second output of the BZPP, and the first input of the control device is connected to the output of the BFKR. The output of the CPC is connected to the input of the ONS (figure 3).

The coordinate prediction device (CPC) contains series-connected Sum1, TF, Sum2 and low-pass filters. The input of the CPC is connected to the second input Sum1. The output Sum1 is connected to the input of the TF. The output of the TF through the delay unit BZ1 is connected to the first input Sum1. The third inverting input Sum1 through the delay unit BZ2 is connected to the third input of the CPC. Moreover, the third input of the CPC is also connected to the second input Sum2, the first input of which is connected to the output of the TF. The output Sum2 is connected to the input of the low-pass filter, the output of which, as indicated earlier, is connected to the input of the ONS (figure 4). The CPC is made with the possibility of setting the duration of the delay by blocks BZ1 and BZ2 and the duration of extrapolation to the TF for a time T _EC based on the signals received at the second input of the CPC.

The operation of the optoelectronic system is carried out as follows (for simplicity, as in the closest analogue, the description is given for one coordinate). During auto tracking of an object, an optoelectronic device (OED) generates a video signal that is fed to the input of the BOX. BOXING determines the correspondence functional between the reference image recorded in its memory and the current images in the strobe. The image closest to the reference image is taken as the image of the object, the coordinates of which are received through the Com switch on the BFKR. The output coordinate of the BFKR obtained, for example, by summing the previous coordinate of the BFKR with the current coordinate of the BOX (see RF patent No. 2191407, publ. September 20, 1998), is supplied to the BOX for setting the coordinates in the next frame (half frame). The same signal is fed to the control device, at the output of which a signal is generated corresponding to the extrapolated ONS control signal. If the circuit is in auto tracking mode, the signal is extrapolated for the time of coordinate measurement in BOX T _ek = T _s . If the OES circuit is in extrapolation mode, then control signals are generated at the second output of the BZPP in accordance with the extrapolation time T _ek > T _s . Switching the extrapolation durations to the CPC is carried out from the second output of the BZPP.

Unlike the closest analogue, in the CCP control signals are generated optimally not only in the extrapolation mode, but also in the auto tracking mode in accordance with the selected optimality criterion (for example, least squares, maximum likelihood, etc.). For example, in the work (see p.148-152, Kuzmin SZ, Fundamentals of Designing Systems for Digital Processing of Radar Information. - M: Radio and Communication, 1986.), a quasi-optimal trajectory filter is considered by the criterion of the most probable assessment.

In contrast to the closest analogue in the mode of rare interruptions in optical visibility, the first input of the CCP does not receive a “frozen” signal from the output of the BFKR, but forms its own coordinate value, extrapolated from the previous values. As noted earlier, to reduce the dynamic tracking error, the input perturbation at the input of the CCP decreases by the amount of perturbation caused by the carrier.

To explain the processes in the ECO, we can consider equivalent structural diagrams of the system (Fig. 5, Fig. 6). Figure 5 presents the low-pass filter with the transmission coefficient of the CPC , the value of which is determined taking into account (see p.691, 692 Tsypkin Y.Z. Theory of linear impulse systems. - M: FIZMATGIZ, 1963) by the ratio

Signals α _n (t) from the DDN reduce the input signal range α _in (t) (1), changing the disturbance band from (3) to (2) (Fig. 5). In view of the foregoing and expression (4), the structural diagram of the ECO circuit is converted to a trajectory filter circuit (Fig. 6). In this case, the signal at the output of the trajectory filter, which is optimal, for example, according to the criterion of maximum likelihood, is formed as follows:

where N _o is the sample size of the trajectory filter, which optimally takes into account the dynamics of disturbances and measurement errors of coordinates, for example, in the minimax case; - the number of extrapolated samples of the trajectory of the object; T _ek and T _d - time extrapolation and sampling, respectively; h (k, N _eq ) is the impulse response of the optimal trajectory filter, extrapolating the motion path (see p.148-152, Kuzmin SZ. Fundamentals of designing systems for digital processing of radar information. - M: Radio and communications, 1986).

Usually, the duration of the impulse response N _{o of the} trajectory filter (5) is selected inversely to the input disturbance bandwidth (see Singer R. Evaluation of the characteristics of the optimal filter for tracking a manned target // Foreign Radio Electronics. - 1971. - No. 8). Therefore, the gain in fluctuation error of the proposed filtration path in comparison with the prototype can be estimated as

For a typical ratio of disturbance bands (2) and (3), the gain is

In the simplest case, when the object is motionless, then , and the gain in fluctuation error (6) is asymptotically infinite. In practice, this means that stable tracking of a stationary object requires only extremely rare observations of the object, which can be used even with a very rare appearance of its optical visibility.

The reduction of dynamic errors in the proposed solution is achieved using DDN. In DDN information can be used, for example, from a rangefinder, gyroscope and accelerometer that measures the speed of movement of the medium. Typically, the distance to the object is large, so almost without compromising the accuracy of the measurement does not require frequent measurement of range.

In addition, the gain in tracking accuracy can be significantly greater if the carrier does not have a gyrostabilization system. Then DDN should provide information that takes into account not only plane-parallel movement of the carrier relative to the target, but also the angular oscillations of the carrier in an inertial coordinate system (for example, using gyro sensors).

Consider an example showing a significant gain in the proposed solution compared to the closest analogue in both fluctuation and dynamic errors.

The fluctuation error at the output of the ECO is determined by the formula

where σ ₀ is the standard deviation of the uncorrelated measurement error of the coordinates ξ (t) in the BOX; f _d - sampling rate.

Both in the closest analogue and in the proposed solution, the transfer coefficient K _z.zh (f, T _eq ) of the closed ECO control loop in the auto tracking mode in (8) is represented as

where K _r.zh - the desired transmission coefficient of the open loop. In the closest analogue, this coefficient is usually selected (see p. 933, Besekersky VA, Popov EP Theory of automatic control systems. - M: Energy, 1966) in the form

In the proposed technical solution, taking into account (see p. 691, 692 Y. Tsypkin, Theory of linear impulse systems. - M: FIZMATGIZ, 1963), this coefficient is

Here K _Tr (p, T _ek ) is the transfer coefficient of the optimal trajectory filter, which for the case of the most plausible assessment of the trajectory (5) can be represented as

The quantity D (h (k, N _eq )) is a discrete Laplace transform. Taking into account expressions (9) and (10) for the closest analogue, we obtain

and for the proposed solution we have

For T _ec = T _d = 0.04 s in Fig. 7 the dashed line shows the frequency dependence of the modulus of the transmission coefficient (12) of the optimal transmission coefficient (14) for the proposed solution, and the solid line for the closest analogue for which the parameters in ( 10) are selected on the basis of the equality to zero of the standard deviation of fluctuation errors (8).

Rare loss of optical visibility can increase the equivalent delay in the control loop, in the extreme case, to the values of T _ek = 2 · T _d = 0.08 s. In this case, the fluctuation error (8) almost does not increase in the control loop of the proposed ECO (dashed line in Fig. 8), and, on the contrary, sharply increases in the contour of the closest analogue (dashed line in Fig. 8). In addition, in the analogue, there may even be a loss of stability (and the error can increase to infinity) due to a decrease in phase margin at the cutoff frequency of the control loop with the transmission coefficient (13).

Thus, the “freezing” of coordinates in the closest analogue leads to an increase in the phase delay, a sharp increase in the fluctuation error, especially at the cutoff frequency, and can even lead to a loss of stability at a very probable ratio T _ec > 2 · T _d . At the same time, an increase in the delay time practically does not increase the fluctuation error and almost does not reduce the phase reserves in the proposed circuit with the transmission coefficient of the optimal filter (14).

To simulate the processes of reducing the dynamic error, the input disturbance (1) of the form was considered taking into account the values (2) and (3)

where A _o , f _o , A _n , f _n are the amplitudes and frequencies of the input disturbances caused by the object and the carrier, respectively; while taking into account (2), f _o = 0.2 Hz was selected, and taking into account (3) f _n = 1.0 Hz. At time t = t _{0, the} control loops switched from the auto tracking mode to the extrapolation mode. The simulation results are presented in Fig. 9, in which the solid line corresponds to the input perturbation (15), and the small dashed line and the large dashed line correspond to the perturbation at the output of the circuit of the closest analogue and the proposed ECO, respectively.

As can be seen from Fig. 9, both in the auto tracking mode and in the extrapolation mode, the dynamic tracking error in the proposed solution is much smaller due to the fact that the input of the equivalent trajectory filter of the proposed solution is a narrow-band process corresponding to the first term of the expression (15.

Thus, the accuracy of the proposed ECO tracking contour is, firstly, equivalent to the accuracy of the optimal trajectory filter both in tracking mode and in extrapolation mode, which ensures a small fluctuation tracking error; secondly, taking into account a priori information about the carrier’s own motion can significantly reduce the dynamic tracking error, especially in extrapolation mode. The proposed technical solution can significantly increase the accuracy of tracking objects, which can provide both an increase in the number of simultaneously tracked objects and the stability of the transition from extrapolation mode to auto tracking mode. In addition, when accompanied by obviously inactive (stationary) objects, the sensitivity of the ECO can be significantly increased by interframe filtering of image signals.

Claims (3)

1. An optical-electronic tracking system comprising a programmed motion assignment unit, a coordinate prediction device, an optical-electronic unit in series, a strobe coordinate determination unit, a switch, a raster coordinate formation unit and a guidance and stabilization device, as well as a disturbance sensor connected in series and a shaper of inertial tracking logic, while the first output of the program move job block is connected to the second input of the switch, the output of the block coordinate in the raster is connected to the second input of the coordinate determination unit in the strobe and the first input of the coordinate prediction device, the output of the inertial tracking logic generator is connected to the third input of the coordinate determination unit in the strobe, characterized in that an additional motion sensor is introduced, while the program task unit the movement is additionally connected to the second input of the coordinate prediction device, the output of the media motion sensor is connected to the third input of the predictive device coordinates, the output of which is connected to the input of the guidance and stabilization device. 2. The system according to claim 1, characterized in that the coordinate prediction device is based on a series-connected first adder, a trajectory filter, a second adder and a constant part filter, the first input of the coordinate prediction device is connected to the second input of the first adder, the output of which is connected to the input trajectory filter, the output of the latter through the first delay unit is connected to the first input of the first adder, the third inverting input of which through the second delay unit is connected to the third input the coordinate prediction device, while the third input of the coordinate prediction device is also connected to the second input of the second adder, the first input of which is connected to the output of the trajectory filter, the output of the second adder is connected to the input of the filter of the constant part, the output of which is connected to the input of the guidance and stabilization device. 3. The system according to claim 2, characterized in that the coordinate prediction device is configured to set the extrapolation duration for external signals supplied to its second input.