Group tracking and group search working mode of group target radar

CN114935755BActive Publication Date: 2025-08-12耿文东
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111614117.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-27
Publication Date
2025-08-12
Estimated Expiration
2041-12-27

Smart Images

  • Figure CN114935755B_ABST
    Figure CN114935755B_ABST
Patent Text Reader

Abstract

Based on group target tracking, the invention includes: the concept and architecture of group target radar; the concept of group search and the method of forming a group search area; the concept and method of tracking the main target group; the working mode of main target group tracking plus group search within the tracking beam, and the working mode of main target group tracking plus group search within the tracking area. Compared with single-target radar and multi-target radar, group target radar has the characteristics of simultaneous front-end group search and terminal group tracking. The significant feature of the group tracking plus group search working mode of the group target radar is that it maintains the characteristics of the phased array radar TAS in tracking the main target and the search area following the movement of the main target, and has the characteristics of precise search, effective search, dynamic search, real-time search and comprehensive search in which the size and shape of the search area are autonomously adjusted in real time according to the group size, and has the characteristics of main target group tracking.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of target detection, and is particularly applicable to the field of target detection of radar, optical system, infrared system and radio passive positioning system. Background Art

[0002] There are many ways to classify radars. For example, according to the antenna scanning method, they can be divided into mechanical scanning and electronic scanning radars; according to the angular tracking method, they can be divided into single pulse, conical scanning and hidden conical scanning radars; according to the transmitting and receiving positions, they can be divided into single base, dual / multi-base, composite multi-base radars and interferometers; according to the signal form, they can be divided into pulse radar, pulse compression radar, pulse Doppler radar, continuous wave radar and quasi-continuous wave radar; according to the loading platform, they can be divided into land-based, sea-based, air-based, space-based, vehicle-mounted and ball-mounted radars.

[0003] Radars are classified by the number of targets they can detect. Based on this classification, radars can be divided into single-target radars and multi-target radars. Single-target radars are radars that can only detect one target, usually single-target tracking radars. Multi-target radars are radars that can detect multiple targets simultaneously, mainly mechanically scanning multi-target radars and electronically scanning multi-target radars. Currently, there are no group-target radars.

[0004] Research on single-target radar technology can be traced back to 1937, with the development of the world's first single-target radar, the SCR-28, marking the beginning of the development of single-target radar tracking. This radar employed gate-type range tracking and conical scanning angle tracking, resulting in weak anti-interference capabilities and unable to meet the growing demands of modern air defense. Consequently, the monopulse radar, a milestone in the development of single-target tracking radar, emerged in the 1950s. Advances in digital technology and estimation theory from the 1960s onward led to the emergence of digital tracking systems, further improving single-target radar technology. However, single-target radars can only track a single target and are unable to meet the demands of tracking multiple targets.

[0005] To address the problem that single-target radars cannot meet the needs of multi-target tracking, research on multi-target radar technology began. The basic concept of multi-target tracking was proposed by Wax in 1955 and mainly includes basic procedures such as track initiation, track maintenance, and track termination. Stir's Bayesian method for optimal data association between multi-target tracking points and tracks, proposed in 1964, became the foundation of multi-target tracking research. In the early 1970s, with the introduction of related theories and Kalman filtering theory into the field of multi-target tracking, Singer proposed the nearest neighbor data association method, and Bar-Shalom proposed the joint probabilistic data association algorithm, which led to the rapid development of multi-target tracking technology. Because the distance between multiple targets is less than or equal to the radar resolution, forming a group target, data association errors can cause mistracking and incorrect tracking. Therefore, multi-target radar can track multiple targets, but it cannot meet the needs of group target tracking.

[0006] To address the inability of multi-target radars to track group targets, research on group target tracking methods has begun. Group target tracking is a tracking algorithm developed based on formation target tracking. In 1977, G. Binias first proposed the concept of formation target tracking based on phased array radar. Since then, through the efforts of scientists such as Taenzer, formation target tracking technology has reached a relatively advanced level. With the increasing density of multiple targets and the diversity of target characteristics, the concept of group targets has emerged, developing and enriching the definition of formation targets. A group target, also known as a target group or cluster target, is a concept developed from formation targets. A group target refers to a collection of all members whose spatial distances and movement directions remain relatively stable over a certain period of time. For group targets, radars face challenges such as uncertainty in the density of the group targets, the location of the distribution area, the size of the distribution area, the shape of the distribution area, the correspondence between the measurement and the number of targets, and the presence of multiple targets within a beam. Therefore, existing radar systems cannot meet the requirements of group target tracking. Based on this, proposing and defining the concept of group target radar on the basis of the concepts of single target radar and multi-target radar has become a new issue that must be solved.

[0007] Currently, the basic idea of group target tracking methods is to transform dense multi-targets into sparse multi-targets. The goal is to solve the problem of multi-target radar mistracking and misalignment by tracking dense multi-targets as a whole. Techniques related to group target tracking, including group detection, track initiation, single-group target data association, multi-group target data association, track maintenance, group target merging and separation detection, and group track cancellation, have achieved fruitful results. However, it is important to note that group target tracking only tracks all targets within the radar's scan area, and this scan area is a manually defined search range, without considering whether the dense multi-targets are a whole from the perspective of target tracking. The manually defined search area refers to a search area of artificial size and shape. It can be an area set by operator experience, the maximum or minimum search range specified by the radar antenna, information about the target from other sources before it enters the radar's detection range, or a search area generated by calculation based on guidance information sent by other sensors. In other words, the approach to dense multi-target tracking is considered only from the radar's perspective. It fails to consider whether these multiple targets are a single entity, whether the search beam covers all of them, or whether the search area is located at the head, middle, or tail of the target cluster. From the perspective of a target cluster, what is needed is a complete and comprehensive scan of the cluster—that is, full coverage. Therefore, simply implementing cluster target tracking within the radar data processing unit does not truly address issues such as incomplete coverage of all cluster members, unclear tracking locations, and subjective, manually defined search areas. In short, despite significant effort on the radar data processing side, little consideration has been given, or even neglected, to the fact that target tracking is also a problem related to the radar front-end scanning method. Even less consideration has been given to integrating radar terminal tracking algorithms with front-end search methods to address the cluster target problem. This has led to a proliferation of tracking algorithms with poor performance. It should be noted that in this document, radar antenna beam search and radar antenna beam scanning are used interchangeably, as are search area and scanning area. Summary of the Invention

[0008] S1, Concept and Architecture of Swarm Target Radar

[0009] S1.1, Concept of Group Target Radar

[0010] The term "swarm target radar" refers to a radar that uses radar antenna beam electronic scanning technology, assumes a mathematical premise of a many-to-many correspondence between measurements and targets, and possesses both swarm target search and swarm target tracking capabilities. This refers to a radar where the antenna beam at the radar front end performs swarm target search, and the data processing unit at the radar terminal also performs swarm target tracking. This corresponds to the concept of a broad swarm target radar described below. A swarm target radar that simultaneously utilizes both swarm search and swarm tracking capabilities is referred to as a narrow swarm target radar. Measurements are also referred to as traces, and the many-to-many correspondence between measurements and targets means that one measurement can correspond to multiple targets, and one target can correspond to multiple measurements. The aforementioned swarm target search is referred to as swarm search, and swarm target tracking is referred to as swarm tracking. A swarm target refers to a collection of all swarm target members whose spatial distances and directions of motion remain relatively stable over a certain period of time. A swarm target member refers to a single target that constitutes the swarm target. Single targets and multiple targets are special cases of swarm targets.

[0011] A radar whose antenna beam only has a group search function at its front end, but whose radar terminal uses a traditional tracking method, is called a group target search-only radar, or simply a group target search radar. A radar whose data processing unit only has a group tracking function, but whose radar front end uses a traditional search method, is called a group target tracking-only radar, or simply a group target tracking radar. The search methods of existing electronically scanned radars and mechanically scanned multi-target radars are collectively referred to as traditional search methods, and the tracking methods of existing electronically scanned radars and mechanically scanned multi-target radars are collectively referred to as traditional tracking methods. Group target search radars and group target tracking radars, as well as group target radars that only use the group search function and group target radars that only use the group tracking function, are collectively referred to as generalized group target radars. Generalized group target radars are applicable to both mechanically scanned and electronically scanned radars, can expand the functions and enhance the capabilities of traditional radars, and possess the partial characteristics of group target radars. Electronic scanning refers to both two-dimensional electronic scanning of the radar antenna beam in azimuth and elevation, and one-dimensional electronic scanning in azimuth or elevation.

[0012] The system of group target radar is applicable to both narrowband radar and broadband radar, and to both point targets and extended targets. Compared with single-target radar and multi-target radar, group target radar has achieved a generational breakthrough in radar system and is a new generation of radar system after single-target radar system and multi-target radar system.

[0013] S1.2, Architecture of Group Target Radar

[0014] The swarm target radar architecture describes the overall structure of the swarm target radar and the connections between its components. It serves as a structural diagram of the swarm target radar system, guiding the design of the swarm target radar system and its components. The swarm target radar consists of nine units: a transmitting antenna and electromagnetic energy radiation unit, a receiving antenna and target scattered signal receiving unit, an information processing unit, a system control unit, a timing and waveform generation unit, a display and human-computer interaction unit, an open interface, a swarm search and tracking management and control unit, and a swarm target radar data and instruction cache, transmission, exchange, and management network platform. The transmitting antenna and electromagnetic energy radiation unit is referred to as the radiation unit, the receiving antenna and target scattered signal receiving unit is referred to as the receiving unit, the swarm search and tracking management and control unit is referred to as the swarm control unit, and the swarm target radar data and instruction cache, transmission, exchange, and management network platform is referred to as the network platform. The swarm target radar architecture is shown in Figure 1.

[0015] The radiating unit completes the generation and radiation of electromagnetic energy under the control of the system control unit; the receiving unit completes the collection, frequency domain transformation, envelope detection and multi-stage matched filtering of the target scattered electromagnetic energy through the timing and time-space frequency matching of the system control unit and sends the video signal to the network platform; the network platform sends the video signal to the information processing unit, which consists of two sub-units, signal processing and data processing. The signal processing part completes the processing of the target video signal, and the data processing part completes the target point extraction and sends it to the network platform for other units to call; the system control unit is responsible for the generation of system-wide instructions, the control of other units and the scheduling management of system-wide resources. It is the control center of the group target radar; the network platform connects all units into a whole, realizes the management, caching, transmission, exchange, interface management of radar data and control instructions and the access of radar-related units, and is the transit station for group target radar data and instructions; the group control unit realizes group search and group tracking functions, supports the regeneration of group target point under the same type of multiple radar data, and different types of radars The generalized point trace generation under data, and the function of system control unit backup, is the soul of the group target radar and also the symbol of the group target radar; the timing and waveform generation unit realizes the unification of time, frequency and phase of the whole system, and generates waveforms under different working modes, which is the benchmark of the group target radar; the display and human-computer dialogue unit realizes the display of group target tracks, group situation display, and the issuance and reception of human-computer dialogue instructions; the open interface completes the sending of local radar data, the random reception of external data and the exchange of send and receive data instructions; the architecture of the group target radar is networked, open, scalable and has the characteristics of random access to multiple radar data. All unit data and instructions realize network connection. The video signals, point traces and tracks of external radars can be randomly sent to the network platform through cache management without time alignment. The external data sent through the open interface can be either the same type of data or different types of data, which expands the function of the local radar. Among them, these radars can be either group target radars or other types of radars.

[0016] The group target radar is not simply a phased array radar with group search and tracking functions added. This is because: first, the mathematical assumptions of the group target data association method are different from those of other traditional data association methods. Traditional data association methods all assume that there is a one-to-one correspondence between measurements and targets, while the group target data association method assumes a many-to-many correspondence between measurements and targets, that is, one measurement may correspond to multiple targets, and one target may correspond to multiple measurements. This is the inevitable result of the target multipath effect between group target members and the occlusion between target members. Therefore, the traditional data association method cannot meet the needs of the group target radar. Furthermore, the mathematical assumption of a non-one-to-one correspondence between traces and targets essentially means that group target radars no longer require target scattering to be point targets, expanding the scope of application of traditional radar resolution theory. Secondly, phased array radars employ both TWS and TAS modes for multi-target detection. TWS stands for Track-While-Scanning, which refers to the radar tracking multiple targets while scanning the search space. TAS stands for Track-And-Search, which independently completes tracking and search tasks using time-alternating or simultaneous multi-beam modes. This unique mode of operation fully utilizes the radar's ability to rapidly change beam pointing. The TAS mode selects a primary target and performs single-target tracking. A search area is artificially defined around the primary target, and multiple targets within this area are tracked using the TWS mode. The search stops within the primary target tracking time and the search time. This mode does not form a group target according to the target formation criteria for the multiple targets within the search area. In particular, targets outside the search area that are part of the group target are not merged and are therefore artificially discarded. Because the TAS operating mode of phased array radar has no concept of group size, nor does it associate group size with the size and shape of the search area, it does not have the ability to self-manage the size and shape of the search area. Radar operators also call the TAS operating mode a scene-based tracking mode. Third, the group target radar has the ability to integrate target tracking and situational awareness, while traditional phased array radars do not have this ability. This is because when using generalized point traces, it can also group targets according to their characteristics. Generalized point traces refer to point traces that contain target scale information and one or more target characteristic information. The target characteristic information refers to the target radar characteristics and target optical characteristics exhibited by the target under the action of electromagnetic waves, as well as the target infrared characteristics and the signal characteristics of the target's own transmitted signal. Finally, because the group target tracking radar adopts the mathematical assumption of multi-correspondence between measurements and targets in the data association algorithm, and can track equivalent measurements to achieve overall tracking, the echoes generated by the target multipath effect are treated as useful echoes. Therefore, it has a certain ability to resist target multipath interference.In short, group target radar is compatible with multi-target radar and single-target radar, but multi-target radar is not compatible with group target radar, just as multi-target radar is compatible with single-target radar, but single-target radar is not compatible with multi-target radar.

[0017] The introduction of the swarm target radar concept marks the birth of a new generation of radar systems, following the single-target and multi-target radar systems. This concept, nearly 50 years in the making, represents a generational breakthrough. Swarm target radar is applicable not only to point targets but also to extended targets. It features precise, efficient, dynamic, and real-time search, along with comprehensive coverage of the target area. It also enables integrated group tracking of dense, difficult-to-resolve multiple targets and integrated target tracking and situational awareness for sparse, resolvable multiple targets. Compared to single-target and multi-target radars, swarm target radar represents a generational breakthrough, representing a new generation of radar systems, following the single-target and multi-target systems.

[0018] S2. Concept of group search and method of forming group search area

[0019] S2.1, the concept of group search

[0020] Cluster search refers to the use of a defined cluster target formation criterion to form a cluster target and calculate the cluster size. The search range is adjusted in real time over time based on the cluster target radar beamwidth and beam scanning arrangement, and at a set data rate. The search range's boundaries are autonomously adjusted based on the cluster size and shape. Traditional search methods are a special case of cluster search. The cluster target formation criterion refers to the threshold used to determine whether a target belongs to the cluster target status. The cluster size refers to the spatial region of a certain size and shape occupied by all members of the cluster target. Compared to traditional search methods, cluster search offers the characteristics of precise, efficient, dynamic, real-time, and comprehensive search. Precise and efficient search refers to searching only areas where targets exist. Dynamic and real-time search refers to the dynamic adjustment of the cluster search area's size and shape in real time with the radar cycle. Comprehensive search ensures that all members of the cluster target are searched, eliminating the issue of not knowing which part of the cluster target has been searched or how many members of the cluster target have been found.

[0021] S2.1, Group Search Area Formation Method

[0022] The group search area formation method, referred to as the collective term for the methods and steps involved in generating the search area boundaries for the next radar cycle using a filter prediction method based on the size of the group target radars in the current radar cycle, and using the position and motion direction of the primary target in the current radar cycle as constraints, is characterized by comprehensive coverage of all group target members, close tracking of the primary target, and high search efficiency. The primary target is the target of highest interest and greatest interest.

[0023] First, when the main target's motion direction is positive, the area surrounded by the maximum and minimum values of the azimuth and pitch angles of the main target tracking area of this radar cycle is the main target angle tracking area of this radar cycle. After prediction calculation, the predicted maximum and minimum values of the azimuth and pitch angles of the main target angle tracking area of the next radar cycle are given. The area surrounded by these predicted values is the main target predicted angle tracking area, among which the maximum value of the azimuth angle of the main target predicted angle tracking area is called the front boundary, the minimum value of the azimuth angle is called the rear boundary, the maximum value of the pitch angle is called the upper boundary, and the minimum value of the pitch angle is called the lower boundary. When the main target's motion direction is negative, the above situation is the opposite, among which the main target motion direction is negative. A positive direction means that the main target's direction of motion is consistent with the positive direction specified by the radar, and a negative direction of motion means that the main target's direction of motion is opposite to the positive direction specified by the radar; secondly, when the main target's direction of motion is positive, the main target's direction of motion is divided into two search areas, front and rear, based on the main target's direction of motion and based on the predicted main target angle tracking area boundary. The search area in front of the main target's direction of motion is called the front search area, and the search area behind the main target's direction of motion is called the rear search area. The group target radar calculates the front search area group size and the rear search area group size in this radar cycle, where the front search area group size and the rear search area group size are referred to as the front group size and the rear group size, respectively. When the main target's direction of motion is negative, The above situation is the opposite; thirdly, when the main target motion direction is positive, extract the two measurements corresponding to the maximum values of the azimuth and elevation angles of the front group scale in this radar cycle, and extract the two measurements corresponding to the minimum values of the azimuth and elevation angles of all members in the rear group scale. The predicted values of the four measurements of the front group scale and the rear group scale in the next radar cycle are obtained through filter prediction. Among them, when the main target motion direction is negative, the above situation is the opposite; fourthly, when the main target motion direction is positive, the two measurement values corresponding to the maximum values of the predicted azimuth and elevation angles of the front group scale, and the area formed by the combination of the front boundary and the upper boundary corresponding to the main target prediction tracking area are the front prediction search area; the main target prediction tracking area The area formed by the rear boundary and lower boundary corresponding to the domain, and the combination of the two measured values corresponding to the minimum value of the azimuth and pitch angle prediction of the rear group scale is the rear prediction search area, among which the front prediction search area refers to the prediction area of the front group scale, and the rear prediction search area refers to the prediction area of the rear group scale. When the main target motion direction is negative, the above situation is opposite; fifth, tracking the motion direction of the main target will form four situations: azimuth positive + pitch positive, azimuth positive + pitch negative, azimuth negative + pitch positive, and azimuth negative + pitch negative; this search area formation method is simple to calculate, and has clear physical and mathematical meanings. It has the characteristics that the size of the search area is synchronized with the group scale in time, and the shape of the search area is consistent with the group scale on the boundary;

[0024] The following describes a method for group searching when only the area where the main target is located can be identified but it is not possible to confirm which target is the main target within the area. Group tracking is performed on all targets within the area, and the relationship between the main target and the group target radar is positive in azimuth and positive in elevation.

[0025] Assume that the group target radar obtains M valid measurements in the tracking area at time k, and the i-th measurement is expressed as

[0026] Y i (k)=(R i , α i , β i )(i=1,2,…,M) (1)

[0027] In formula (1): R i , α i , β i are the distance, azimuth, and elevation values of the i-th measurement respectively. Let:

[0028]

[0029]

[0030]

[0031] In formulas (2), (3) and (4), R max (k), α max (k), β max (k) represents the maximum value of distance, azimuth and elevation at time k; R min (k), α min (k), β min (k) represents the minimum value of distance, azimuth and elevation at time k respectively.

[0032] Extract the measurements corresponding to the maximum and minimum values of azimuth, elevation, and distance. Assuming that the maximum and minimum corresponding measurements of azimuth, elevation, and distance do not overlap, we can obtain:

[0033]

[0034] Use the standard Kalman filter to predict the six sets of measurements at time k, and get the predicted measurements at time k+1 as:

[0035]

[0036] Assume that the group target radar obtains N1 valid echoes in the first group size area within the search area at time k, then the i1th measurement is expressed as

[0037]

[0038] In formula (7): are the distance, azimuth, and elevation values of the i1th measurement respectively.

[0039]

[0040]

[0041]

[0042] In formulas (8), (9), and (10), R max1 (k), α max1 (k), β max1 (k) represents the maximum value of distance, azimuth and elevation at time k; R min1 (k), α min1 (k), β min1 (k) represents the minimum value of distance, azimuth and elevation at time k respectively.

[0043] Extract the measurements corresponding to the maximum and minimum values of azimuth, elevation, and distance. Assuming that the maximum and minimum corresponding measurements of azimuth, elevation, and distance do not overlap, we can obtain:

[0044]

[0045] The six sets of measurements at time k are predicted by the filter in one step, and the predicted measurements at time k+1 are obtained as follows:

[0046]

[0047] According to formula (6), we can get the tracking area and According to formula (12), we can get and According to the above group size search area construction method, the previous group size in azimuth and elevation directions is:

[0048]

[0049] Where A 前群规模 (k+1) and E 前群规模 (k+1) are the azimuth and elevation ranges of the front group, A 前群规模 (k+1) and E 前群规模 The area enclosed by (k+1) is the size of the front group, that is,

[0050] Front group size = A 前群规模 (k+1)×E 前群规模 (k+1) (14)

[0051] Assume that the group target radar obtains N2 valid echoes in the group size area within the search area at time k, then the i2th measurement is expressed as

[0052]

[0053] In formula (15): are the distance, azimuth, and elevation values of the i2th measurement respectively.

[0054]

[0055]

[0056]

[0057] In formulas (16), (17), and (18), R max2 (k), α max2 (k), β max2 (k) represents the maximum value of distance, azimuth and elevation at time k; R min2 (k), α min2 (k), β min2 (k) represents the minimum value of distance, azimuth and elevation at time k respectively.

[0058] Extract the measurements corresponding to the maximum and minimum values of azimuth, elevation, and distance. Assuming that the maximum and minimum corresponding measurements of azimuth, elevation, and distance do not overlap, we can obtain:

[0059]

[0060] The six sets of measurements at time k are predicted by the filter in one step, and the predicted measurements at time k+1 are obtained as follows:

[0061]

[0062] According to formula (6), we can get the tracking area and According to formula (20), we can get and According to the above method of constructing the group scale search area, the group scale in azimuth and elevation directions is:

[0063]

[0064] Where A 后群规模 (k+1) and E 后群规模 (k+1) are the azimuth and elevation ranges of the rear group, A 后群规模 (k+1) and E 后群规模 The area enclosed by (k+1) is the size of the posterior group, i.e.

[0065] Post-group size = A 后群规 Modulo (k+1)×E 后群规模 (k+1) (22)

[0066] S3, group target radar main target tracking method

[0067] The main target group tracking of the group target radar refers to the tracking of the main target using the group tracking method under the constraint of the group target formation criterion. The group tracking of the main target is divided into two situations. One is that when the main target has been identified, the group tracking is implemented on the main target and all the targets around it that meet the group target formation criterion in the tracking beam; the other is that when only the area where the main target is located can be identified but it is impossible to confirm which one is the main target in the area, the group tracking is implemented on all the targets that meet the group target formation criterion in the area. These two situations are collectively referred to as the main target group tracking of the group target radar, abbreviated as the main target group tracking, among which a single target is a special case of the group target. The group tracking method refers to a group target overall tracking method that forms a group target according to the group target formation criterion, forms equivalent measurement with the measurement of all members of the group target as elements, and uses the equivalent measurement of the group target as the measurement. Among them, the equivalent measurement is a virtual measurement obtained by weighting and normalizing the measurement of all members of the group target; the main target group tracking has the characteristics of implementing overall group tracking for dense multi-targets that are difficult to distinguish, and also has the characteristics of realizing target tracking and situation awareness integration for distinguishable sparse multi-targets. Compared with the main target tracking method in the tracking and search working mode of the traditional phased array radar, the main target group tracking method of the group target radar has the characteristics of resisting interference from other target echoes entering the same distance wave gate as the main target, interference from other targets blocking the main target, interference from echoes generated by the multipath effect of the target on the main target, and interference caused by the main target itself being split into multiple targets.

[0068] Among the M valid measurements set in formula (1), let any two measurements be: Y i (k)=(R i , α i , β i ) and Y j (k)=(R j , α j , β j )(i, j = 1, 2, ..., M) where R i , α i , β i With R j , α j , β j are the distance, azimuth, and elevation values of the i-th and j-th measurements respectively, then:

[0069]

[0070] Equation (23) is the spatial distance between any two measurements.

[0071] The cluster detection matrix based on the spatial distance between any valid measurements is:

[0072]

[0073] In formula (24):

[0074]

[0075] K0 in formula (25) is the group target formation criterion, that is, the threshold value that needs to be met to form a group target.

[0076] Assuming that M measurements are obtained from equation (1), we have:

[0077]

[0078] In formula (26), β j (k) is the weight of each measurement. The weight of each measurement to the equivalent measurement is

[0079]

[0080] In formula (27), ρ i (k)

[0081] ρ i (k) = 1 - g i (k) / K0 (28)

[0082] Formula (28)g i (k) is the statistical distance between each measurement and the equivalent measurement, calculated using the same method as equation (23). The tracking area is determined by equations (1) to (6) and will not be repeated here.

[0083] According to the Kalman filter (Kalman and other filtering technologies are already mature technologies), the equivalent measurement Y from time k to time k+1 is e The predicted value of (k) for:

[0084]

[0085]

[0086] Where H(k+1) is the measurement matrix of the measurement space, is the state prediction value of the group target from time k to time k+1, Φ(k+1 / k) is the state transfer matrix from time k to time k+1, and X(k) is the state at time k.

[0087] After obtaining the equivalent measurement and the predicted value of the equivalent measurement, track maintenance can be performed through data association according to the selected filter to keep tracking the group target.

[0088] S4, the working mode of group search and group tracking of group target radar

[0089] The group tracking plus group search working mode of the group target radar refers to that the group target radar takes the main target as the benchmark and the group target formation criterion as the constraint, divides the area of interest outside the main target tracking area into two front and rear search areas with real-time size adjustment and autonomous shape adjustment according to the movement direction of the main target, and adopts an alternating and continuous working mode of group search in the front search area, group tracking of the main target, and group search in the rear search area in sequence, which is called the group tracking plus group search working mode of the group target radar, wherein alternating means that the search and tracking are carried out in time, and continuous means that the search and tracking are continuously cycled in the order of search, tracking, and search.

[0090] The group tracking plus group search working mode of the group target radar includes two working modes. One is the group tracking plus group search working mode of the main target located in the tracking beam when the main target has been identified. The other is the group tracking plus group search working mode of the main target located in the tracking area when only the area where the main target is located can be identified but the main target in the area cannot be confirmed. Among them, the working mode of using group tracking for the main target and using the traditional search mode for the search areas before and after the main target, and the working mode of using the traditional tracking mode for the main target and using the group search mode for the search areas before and after the main target, are special cases of the group tracking plus group search working mode of the group target radar. The main target in the tracking beam refers to the target that has been identified as being tracked and is located in the tracking beam. The main target in the tracking area refers to the target that has been identified as being located in the tracking area but the main target has not been confirmed. Among them, the traditional tracking mode refers to a closed-loop tracking mode in which the radar beam points to and follows a single target.

[0091] The present invention has the following beneficial effects:

[0092] First, the main target adopts a swarm tracking method, which has strong anti-interference capabilities, is not easily lost, and has a wide range of applications. Second, the front and rear search areas are constrained by the main target's position, ensuring that the search area does not deviate from the main target. Third, the size and shape of the front and rear search areas match the size of the swarm, achieving accurate, effective, dynamic, and real-time full coverage of the search area. The swarm target radar's swarm tracking plus swarm search mode is backward compatible with single-target tracking and multi-target tracking. It is applicable to both radar systems and optical systems, single radars and multi-radar networks, narrowband radars and broadband radars, and point targets and extended targets.

[0093] The invented group target radar and its group search + group tracking working mode have revolutionary significance for the design, application and development of modern radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 is an architectural diagram of a group target radar;

[0095] Figure 2 shows the two-dimensional theoretical tracks of four targets with four clutters;

[0096] Figure 3 shows the simulation results of radar detection of group targets containing clutter. DETAILED DESCRIPTION

[0097] Taking the three-coordinate narrow group target radar as an example, without loss of generality, four uniformly variable speed moving targets are designed, and the simulation scenarios and parameter designs implemented are as follows.

[0098] The simulation scenario design implemented is as follows: target 1 is designed as the main target, and single targets 3 and 4 gradually approach a group target 1 consisting of targets 1 and 2 with a distance of 10 meters. That is, target 2 and main target 1 are in the same tracking area; after the 4th second, the two single targets 3 and 4 are 20 meters apart, and at the 5th second, the distance is 15 meters to form group target 2, but it does not affect the main target 1; at the 10th second, group target 2 begins to separate and forms two separated single targets 3 and 4 again. Group target 1 including main target 1 continues to maintain its original flight status.

[0099] Implementation of simulation parameter design: The initial value of the track is defined as X1(0) = [10000, 25, 10100, 12.5, 10000, 0], X2(0) = [10000, 25, 1009012.5, 10000, 0], X3(0) = [10000, 25, 9960, 25, 10000, 0], X4(0) = [10000, 25, 9815, 50, 10000, 0]. In order to more realistically approach the actual working scenario of the actual group target radar, four uniformly distributed clutters are added in the simulation process. The theoretical track is shown in Figure 2. The target flight time is 25 seconds, the sampling period is T = 0.01s; the group target radar beam width is 1°, and the distance error σ R = ±6m, angular error σ θ =σ β = ±0.25 mrad; target model noise is ±9 m in the X, Y, and Z directions; X-direction speed is 25 m / s, Y-direction speed is 50 m / s maximum and 12.5 m / s minimum, and Z-direction speed is 0. Simulation data is generated and collected in real time under these parameters, and the group target spacing criterion K0 is 20 m.

[0100] The implementation process of group search + group tracking of the group target radar is shown in Equations (1) to (30). Standard Kalman filtering is used for data association and track maintenance. The detection results of the group target radar for four targets are shown in Figure 3.

[0101] The simulation results of group target radar detection show that: first, the target spacing criterion determines that main target 1 and target 2 meet the group target criterion, forming group target 1, which is tracked continuously in a group manner. Second, at the fourth second, two single targets 3 and 4 meet the target track spacing criterion, forming group target 2. These two single targets 3 and 4 are tracked in a group search using TWS mode, but this does not affect the tracking of main target 1. Third, targets 3 and 4 begin to separate at the tenth second. Finally, tracking ends at the 25th second. Furthermore, clutter suppression is excellent during the above process.

[0102] Through the above implementation process, it can be seen that: in the process of tracking all targets, there was no mistracking or loss of tracking, which verified the correctness and effectiveness of the group target radar and its group search + group tracking working mode, and it has good anti-clutter capability.

Claims

1. Group target radar refers to a radar that uses radar antenna beam electronic scanning technology as its basis, takes the many-to-many correspondence between measurement and target as its mathematical premise, and has both group target search and group target tracking functions. In other words, the antenna beam at the front end of the radar has the function of group target search, and the data processing unit at the radar terminal has the function of group target tracking. A group target radar that uses both group search and group tracking functions is called a narrow sense group target radar, in which: Measurement is also called point trace. The many-to-many correspondence between measurement and target means that one measurement can correspond to multiple targets, and one target can correspond to multiple measurements. Group target search is referred to as group search, and group target tracking is referred to as group tracking. A group target refers to the set of all members of a group target whose spatial distance and movement direction remain relatively stable within a certain period of time. A group target member refers to a single target that constitutes a group target. Single targets and multiple targets are special cases of group targets. A radar that only has a group search function in the antenna beam at the front end of the radar, but the radar terminal uses traditional tracking methods, is called a group target search-only radar, or simply a group target search radar. A radar that has a group tracking function only in the data processing unit of the radar terminal but adopts a traditional search method at the radar front end is called a group target tracking radar, or simply a group target tracking radar. Among them, the search method of the existing electronic scanning radar and the search method of the mechanical scanning multi-target radar are collectively referred to as the traditional search method, and the tracking method of the existing electronic scanning radar and the tracking method of the mechanical scanning multi-target radar are collectively referred to as the traditional tracking method; group target search radar and group target tracking radar, as well as group target radars that only use the group search function and group target radars that only use the group tracking function are collectively referred to as generalized group target radars, generalized group target radars It is applicable to both mechanically scanned radars and electronically scanned radars, and can expand the functions and enhance the capabilities of traditional radars. It has the local characteristics of group target radars. Electronic scanning refers to both the two-dimensional electronic scanning of the radar antenna beam in azimuth and elevation, and the one-dimensional electronic scanning of azimuth or elevation. The group target radar system is applicable to both narrowband radars and broadband radars, and to both point targets and extended targets. Compared with single-target radars and multi-target radars, the group target radar has achieved a generational breakthrough in radar systems and is a new generation of radar systems following the single-target radar system and the multi-target radar system. The group target radar architecture refers to the description of the overall structure of the group target radar and the relationship between its constituent units. It is a structural diagram of the group target radar system, which is used to guide the design of the group target radar system and its constituent units. The group target radar consists of a transmitting antenna and electromagnetic energy radiation unit, a receiving antenna and target scattered signal receiving unit, an information processing unit, a system control unit, a timing and waveform generation unit, a display and human-computer dialogue unit, a development interface, a group search and group tracking management control unit, and a group target radar data and instruction cache, transmission, exchange and management network platform, a total of 9 units. Among them, the transmitting antenna and electromagnetic energy radiation unit are referred to as the radiation unit, the receiving antenna and target scattered signal receiving unit are referred to as the receiving unit, the group search and group tracking management control unit are referred to as the group control unit, and the group target radar data and instruction cache, transmission, exchange and management network platform are referred to as the network platform. The radiation unit completes the generation and radiation of electromagnetic energy, and the receiving unit The element completes the collection, transformation and detection of target scattered electromagnetic energy and sends out video signals. The information processing unit is composed of two sub-units, signal processing and data processing, which complete the processing of target video signals and target point extraction. The system control unit is responsible for the command control of the entire system and the scheduling management of the entire system resources. The network platform connects all units into a whole, realizing the management, caching, transmission, exchange and interface management of radar data and control instructions. The group control unit realizes the group search and group tracking functions, and also has the function of system control unit function backup. The timing and waveform generation unit completes the generation of the timing and waveform of the entire system. The display and human-computer dialogue unit realizes the display of group target tracks, group situation display, and the issuance and reception of human-computer dialogue instructions. The open interface completes the sending of local radar data, the reception of external data and the exchange of data transmission and reception instructions. The architecture of the group target radar has the characteristics of networking, systematization, openness, scalability and random access to multiple radar data. The group search refers to the use of a determined group target formation criterion to form a group target and calculate the group size. According to the radar beam width and beam scanning arrangement, the size of the group scale area search range is adjusted in real time over time according to the set data rate, and the boundary of the search range is autonomously adjusted according to the group size and shape. Among them, the traditional search method is a special case of group search. The group target formation criterion refers to the threshold value for determining whether a target belongs to the group target state. The group size refers to the spatial area of a certain size and shape occupied by all members of the group target. Compared with the traditional search method, the group search has the characteristics of precise search, effective search, dynamic search, real-time search and comprehensive search. The group search area formation method refers to a general term for the method and steps of generating the search area boundary of the next radar cycle and determining the size of the search area based on the group size of the group target radar in the current radar cycle and the position and movement direction of the main target in the current radar cycle by using the filter prediction method. It has the characteristics of full coverage of all members of the group target, close follow-up of the main target, and high search efficiency. Among them, the main target refers to the target with the highest attention. First, when the movement direction of the main target is positive, the area enclosed by the maximum and minimum values of the azimuth and pitch angles of the main target tracking area in the current radar cycle is the main target angle tracking area of the current radar cycle. The main target angle of the next radar cycle is given through prediction calculation. The predicted maximum and minimum values of the azimuth and pitch angles in the tracking area, the area surrounded by these predicted values is the main target predicted angle tracking area, among which the maximum value of the azimuth angle of the main target predicted angle tracking area is called the front boundary, the minimum value of the azimuth angle is called the rear boundary, the maximum value of the pitch angle is called the upper boundary, and the minimum value of the pitch angle is called the lower boundary. When the main target's motion direction is negative, the above situation is the opposite, among which the positive motion direction of the main target means that the motion direction of the main target is consistent with the positive direction specified by the radar, and the negative motion direction of the main target means that the motion direction of the main target is opposite to the positive direction specified by the radar; secondly, when the main target's motion direction is positive, the predicted main target angle tracking area boundary is drawn based on the motion direction of the main target. It is divided into two search areas, the front search area and the rear search area, the search area in front of the main target movement direction is called the front search area, and the search area behind the main target movement direction is called the rear search area. The group target radar calculates the front search area group size and the rear search area group size in this radar cycle, wherein the front search area group size and the rear search area group size are referred to as the front group size and the rear group size respectively. When the main target movement direction is negative, the above situation is opposite; thirdly, when the main target movement direction is positive, extract the two measurements corresponding to the maximum values of the azimuth and pitch angles of the front group size in this radar cycle, and extract the two measurements corresponding to the minimum values of the azimuth and pitch angles of all members in the rear group size. The four measurements of the front group size and the rear group size are obtained through filter prediction. In the respective predicted values of the next radar cycle, when the main target motion direction is negative, the above situation is opposite; fourth, when the main target motion direction is positive, the two measurement values corresponding to the maximum value of the front group scale azimuth and pitch angle prediction, and the front boundary and upper boundary corresponding to the main target prediction tracking area are combined to form the area of the front prediction search area; the rear boundary and lower boundary corresponding to the main target prediction tracking area, and the two measurement values corresponding to the minimum value of the rear group scale azimuth and pitch angle prediction are combined to form the area of the rear prediction search area, wherein the front prediction search area refers to the prediction area of the front group scale, and the rear prediction search area refers to the prediction area of the rear group scale. When the main target motion direction is negative, the above situation is opposite;Fifth, tracking the main target's direction of motion will generate four search area situations: positive azimuth + positive pitch, positive azimuth + negative pitch, negative azimuth + positive pitch, and negative azimuth + negative pitch. This search area formation method is computationally simple, has clear physical and mathematical meanings, and features a search area whose size is synchronized with the swarm size in time and whose shape matches the swarm size in terms of boundaries. The main target group tracking of the group target radar refers to the tracking of the main target by using the group tracking method under the constraint of the group target formation criterion, wherein the group tracking of the main target is divided into two situations. One is that when the main target has been identified, the group tracking is implemented on the main target and all the targets around it that meet the group target formation criterion in the tracking beam; the other is that when only the area where the main target is located can be identified but it is impossible to confirm which one is the main target in the area, the group tracking is implemented on all the targets that meet the group target formation criterion in the area. These two situations are collectively referred to as the main target group tracking of the group target radar, referred to as the group tracking of the main target, wherein a single target is a special case of a group target, and the group tracking method refers to forming a group target according to the group target formation criterion, with the measurements of all members of the group target as elements. Forming equivalent measurements, a group target overall tracking method is performed using the group target equivalent measurements as the measurement, wherein the equivalent measurement is a virtual measurement obtained by weighted and normalized calculation of the measurements of all members of the group target; the main target group tracking has the characteristics of implementing overall group tracking for dense multi-targets that are difficult to distinguish, and also has the characteristics of realizing the integration of target tracking and situation awareness for sparse multi-targets that are distinguishable. Compared with the main target tracking method in the tracking and search working mode of the traditional phased array radar, the main target group tracking method of the group target radar has the characteristics of resisting interference from other target echoes entering the same distance wave gate as the main target, interference from other targets blocking the main target, interference from echoes generated by the multipath effect of the target on the main target, and interference caused by the main target itself being split into multiple targets; The group tracking plus group search working mode of the group target radar refers to the group target radar taking the main target as the benchmark and the group target formation criterion as the constraint, dividing the area of interest outside the main target tracking area into two front and rear search areas with real-time size adjustment and autonomous shape adjustment according to the movement direction of the main target, and adopting the alternating and continuous working mode of front search area group search, main target group tracking, and rear search area group search in sequence, which is called the group tracking plus group search working mode of the group target radar, wherein alternation means that the search and tracking are carried out in time sharing, and continuous means that the search and tracking are continuously cycled in the order of search, tracking, and search; the group tracking plus group search working mode of the group target radar includes two working modes, one is the group tracking plus group search working mode of the main target located in the tracking beam when the main target has been identified, and the other is the group tracking plus group search working mode of the main target located in the tracking area when only the area where the main target is located can be identified but which one in the area is the main target cannot be confirmed, wherein group tracking is adopted for the main target, and the traditional search method is adopted for the front and rear search areas of the main target, and the traditional tracking is adopted for the main target. The group tracking plus group search mode is a special case of the group target radar's group tracking plus group search mode. The main target within the tracking beam refers to the target identified as being tracked within the tracking beam, while the main target within the tracking area refers to the target identified as being within the tracking area, but the main target has not been confirmed. The traditional tracking mode is a closed-loop tracking method in which the radar beam points to and follows a single target. The group tracking plus group search mode of the group target radar has the following significant features: first, the group tracking method for the main target has strong anti-interference capability, is not easy to lose the target, and has a wide range of applicability; second, the front and rear search areas are constrained by the position of the main target, so that the search area does not deviate from the main target; third, the size and shape of the front and rear search areas match the group size, achieving accurate, effective, dynamic, and real-time full coverage of the search area. The group tracking plus group search mode of the group target radar is backward compatible with single-target tracking and multi-target tracking, and is applicable to both radar systems and optical systems, single radars and multi-radar networks, narrowband radars and wideband radars, and point targets and extended targets.

Citation Information

Patent Citations

  • Implementation method of phased array radar track and search working mode

    CN109164420A

  • Operating mode of phased array radar group target tracking

    CN113109804A