Method for Simultaneously Estimating the Motion and Shape of Target Vehicles

Abstract

The present disclosure relates to a method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of trajectory segments. This method, when implemented in real time by setting a preliminary distribution model defined by a small number of representative trajectory occurrence points using a linear correspondence between trajectory segments defined in an orthogonal coordinate system and the preliminary distribution model of the trajectory segments and the center position of the target vehicle, and performing a many-to-one data association between trajectory segment outputs, achieves high reliability in estimation performance even when the target vehicle's heading changes. The method for simultaneously estimating the movement and shape of a target vehicle using the preliminary distribution model of the trajectory segments is performed by a signal processing device that processes signals emitted from a plurality of radars installed at a host vehicle and reflected from the target vehicle.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of Korean Patent Application No. 10-2020-0128990, filed on October 6, 2020, which is hereby incorporated by reference in its entirety. Technical Field

[0003] The present disclosure relates to a method for simultaneously estimating the movement and shape of a target vehicle, and more particularly, to a method for simultaneously estimating the movement and shape of a target vehicle using, for example, a preliminary distribution model of radar tracklet outputs that has high reliability in estimation performance even when the heading of the target vehicle changes. Background Art

[0004] With the continued development of autonomous vehicles, there is a fundamental need for information about the surrounding environment of a self-driving vehicle (hereinafter referred to as the host vehicle), its absolute and relative position, and the heading and shape of another vehicle (hereinafter referred to as the target vehicle) moving in front of or beside it. If the host vehicle cannot predict the shape or heading of a target vehicle traveling around it while driving on the road, it may be difficult to prevent an accident with a target vehicle traveling near it.

[0005] To solve this problem, multiple radars are installed on the vehicle, and information collected from the radars is processed to identify the position of the vehicle, sense surrounding objects including target vehicles, people and animals around the vehicle, or identify free space.

[0006] Extended measurements from high-resolution radars with high range resolution include not only information about the target vehicle's movement but also crucial information for estimating the target vehicle's shape, including its size and width. Extended measurements refer to a set of polar coordinate measurements of a target vehicle detected by the high-resolution radar, including relative range, Doppler velocity, and azimuth. However, measurements acquired for a single target vehicle often include multiple irregular noises, and the number of measurements often changes.

[0007] Conventional methods for tracking the movement or shape of a target vehicle require matching extended measurements with a target vehicle shape model (extent model). To account for the irregular variations in the number and distribution of extended measurements depending on the radar's incident angle, particle filters, probability hypothesis density (PHD) filters, and random finite set (RGS) techniques are commonly used.

[0008] Under traditional technology, it is difficult to achieve stable target tracking performance because a lot of calculations are required and when the extended measurement values ​​are matched with the target vehicle shape model in the heading direction of the target vehicle, that is, in the direction facing the front of the target vehicle, it is not easy to achieve in real time, which may cause blur.

[0009] In particular, since gating is performed on the entire target vehicle shape model to suppress clutter, the complexity of the data association algorithm for determining whether the extended measurement value is generated by the target vehicle increases, which is caused by the severe nonlinear correspondence between the position of the target vehicle in the orthogonal coordinate system and the extended measurement value in the polar coordinate system.

[0010] Here, clutter of radar signals reflected from the target vehicle's surroundings rather than from the target vehicle is not a true measurement but a false measurement. Gating is the process of distinguishing probabilistically meaningful radar measurements based on predicted values, and data association is the process of determining whether a gated radar measurement is caused by the target vehicle.

[0011] Figure 1 A conventional data association algorithm is shown that uses extended measurements to directly correspond to a target vehicle shape model in a polar coordinate system.

[0012] Reference Figure 1 Three trajectory segments located within the polar coordinate gate that includes the polar coordinate target vehicle shape model are selected as associated extended measurements, and then a data association algorithm is performed on the selected associated extended measurements. Specifically, because a conventional target vehicle shape model is composed of an unmeasured number of extended measurement occurrence points (all points where radar rays intersect the target vehicle shape model), the computational complexity during the many-to-many data association process between the extended measurements and the target vehicle shape model increases exponentially, making it impossible to implement conventional target tracking algorithms based on extended measurements in real time.

[0013] Furthermore, the nonlinear correspondence between the extended measurements and the position of the target vehicle leads to ambiguity when constructing the measurement equations, so that the estimation performance may vary greatly depending on the target heading. Summary of the Invention

[0014] Accordingly, the present disclosure is directed to a method for simultaneously estimating the motion and shape of a target vehicle using a preliminary distribution model of trajectory segments that substantially obviates one or more problems due to limitations and disadvantages of the related art.

[0015] The present disclosure aims to provide a method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of radar track segment outputs, with high reliability in estimation performance even when the heading of the target vehicle changes when implemented in real time.

[0016] The purpose of the present disclosure designed to solve these problems is not limited to the above-mentioned purpose, and other unmentioned purposes will be clearly understood by those skilled in the art based on the following detailed description of the present disclosure.

[0017] To achieve these objects and other advantages and in accordance with the purposes of the present disclosure as embodied and broadly described herein, a method for simultaneously estimating the movement and shape of a target vehicle (e.g., using a preliminary distribution model of trajectory segments) is provided, the method being performed by a signal processing device configured to process signals emitted from a plurality of radars mounted at a host vehicle and reflected from the target vehicle, the method comprising: defining, in an orthogonal coordinate system, at least one trajectory segment occurrence point corresponding to a specific point of the approximately polygonal shape of the target vehicle, and setting a coordinate function of a center point of the target vehicle relative to each of the defined points to set a preliminary distribution model; acquiring trajectory segment outputs from the plurality of radars; if at least one trajectory segment among the acquired trajectory segment outputs is located within a gate set based on the at least one trajectory segment occurrence point, selecting at least one trajectory segment occurrence point located within the gate as an associated trajectory segment; creating a measurement equation based on the associated trajectory segment, the position of each of the at least one trajectory segment occurrence point, and the coordinate function; calculating an estimated value of the state variable based on the state variable of the target vehicle and the measurement equation to update the state variable; and creating a predicted value of the target vehicle at a next time point based on the calculated estimated value of the state variable and the target vehicle movement equation in the host vehicle coordinate system.

[0018] It is to be understood that both the foregoing general description and the following detailed description of the present disclosure are exemplary and explanatory and are intended to provide further explanation of the disclosure as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this application, illustrate embodiments of the present disclosure and together with the description serve to explain the principles of the present disclosure. In the drawings:

[0020] Figure 1 A conventional data association algorithm for directly correlating a target vehicle shape model in a polar coordinate system using extended measurements is shown;

[0021] Figure 2 A coordinate system representing the geometric relationship between a vehicle equipped with a high-resolution radar (host vehicle) and a target vehicle is shown;

[0022] Figure 3 An embodiment of a method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of trajectory segments according to the present disclosure is shown;

[0023] Figure 4 shows preliminary distribution characteristics of trajectory segments illustrating where representative trajectory segments appear within a target vehicle shape defined by a rectangle;

[0024] Figure 5 Gating and data association of a preliminary distribution model of trajectory segments are shown. DETAILED DESCRIPTION

[0025] It should be understood that the term "vehicle" or "vehicular" or other similar terms used herein include: motor vehicles, such as passenger cars, including mobile utility vehicles (SUVs), buses, trucks, various commercial vehicles; watercraft, including various ships and boats; aircraft, etc., and include hybrid vehicles, electric vehicles, plug-in hybrid vehicles, hydrogen-powered vehicles and other alternative fuel vehicles (e.g., fuels derived from resources other than petroleum). As referred to herein, a hybrid vehicle is a vehicle having two or more power sources, such as a gasoline-powered vehicle and an electric vehicle.

[0026] The terms used herein are only used to describe the purpose of specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "one", "an" and "the" are also intended to include plural forms, unless the context clearly indicates otherwise. It will be further understood that when used in this specification, the terms "comprising" and / or "including" specify the presence of the features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. As used herein, the terms "and / or" include any and all combinations of one or more related listed items. Throughout the specification, unless explicitly described to the contrary, the word "comprising" and variations such as "comprising" or "including" will be understood to imply the inclusion of the elements but do not exclude any other elements. In addition, the "unit", "part", "piece", "module" described in the specification refer to a unit that processes at least one function and operation, and can be implemented by hardware components or software components and combinations thereof.

[0027] Furthermore, the control logic of the present disclosure may be embodied as a non-transitory computer-readable medium on a computer-readable medium containing executable program instructions executed by a processor, controller, or the like. Examples of computer-readable media include, but are not limited to, ROM, RAM, compact disk (CD)-ROMs, magnetic tapes, floppy disks, flash drives, smart cards, and optical data storage devices. The computer-readable medium may also be distributed among network-connected computer systems such that the computer-readable medium is stored and executed in a distributed manner, for example, via a telematics server or a controller area network (CAN).

[0028] For a full understanding of the present disclosure, the advantages in its operation, and the objectives attained by carrying out the present disclosure, reference must be made to the drawings which illustrate exemplary embodiments of the present disclosure and the matters set forth therein.

[0029] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The same reference numerals in the various drawings represent the same components.

[0030] The process of tracking an object in an image is to detect the object in each frame that constitutes the image, connect the movement route of the object between short image frames to detect trajectory segments (trajectory segments are information indicating the short movement route of the object), and connect the trajectory segments detected from all frames of the image to create a tracking trajectory of the object.

[0031] As used herein, a track segment (an intermediate output of the radar signal processing logic) refers to the orthogonal coordinate motion information of a target vehicle, temporarily calculated as an extended measurement. While the number of track segments for a given target vehicle may vary, the irregularities are not significant compared to the extended measurement.

[0032] The trajectory segments used in this disclosure consider the fact that the trajectory segments have a high probability of existing near a specific point of the target vehicle shape model, unlike the general extended measurement value with high irregularity. The representative positions where the trajectories appear on average are set as the preliminary distribution model.

[0033] First, the inertial coordinate system and the host vehicle coordinate system to be used in the following description will be described.

[0034] Figure 2 A coordinate system representing the geometric relationship between a vehicle equipped with a high-resolution radar (host vehicle) and a target vehicle is shown.

[0035] exist Figure 2 In the equation, I represents the inertial coordinate system. An inertial coordinate system is a reference system in which an object without an applied force keeps moving uniformly. V represents the own vehicle coordinate system, which has an origin as the center of the own vehicle. and the axis X, which is the forward direction of the vehicle V . represents the position of the target vehicle, v h represents the speed of the vehicle, v t represents the speed of the target vehicle, ψ represents the heading of the vehicle, φ represents the heading of the target vehicle, and a t represents the centripetal force of the target vehicle. X, Y, or Z represents one of the three-dimensional axes. In particular, it can be seen that the Z axis in the inertial coordinate system is the same as the Z axis in the own vehicle coordinate system.

[0036] In particular, Figure 2The relative movement between a vehicle equipped with high-resolution radar (host vehicle) and a target vehicle is shown.

[0037] Define the relative motion equation of the target vehicle.

[0038] The relative movement of the target vehicle defined in the inertial coordinate system is described as a constant turning rate (CT) movement model. When the relative movement of the target vehicle is discretized with respect to the time T and then described in the own vehicle coordinate system, this can be expressed by Mathematical Expression 1.

[0039] [Mathematical expression 1]

[0040]

[0041] In mathematical expression 1, u k Denotes process noise for reflecting uncertainty in the movement of the target vehicle and can be assumed to be normally distributed with an average value of 0 (zero) and a dispersion of Q. The vector and matrix in Mathematical Expression 1 are defined as represented by Mathematical Expression 2 and Mathematical Expression 3.

[0042] [Mathematical expression 2]

[0043]

[0044]

[0045] [Mathematical expression 3]

[0046]

[0047]

[0048] In Mathematical Expression 2, the heading ψ of the host vehicle can be approximated at each time using the yaw rate (or rotational angular velocity) of the host vehicle: k+1 ≈ψ k +ω h，k T.

[0049] The heading φ of the target vehicle in the inertial coordinate system k The heading γ of the target vehicle in the vehicle coordinate system k The relationship between can be expressed by mathematical expression 4.

[0050] [Mathematical expression 4]

[0051]

[0052] Next, a system model of the simultaneous target vehicle movement / shape estimation filter will be described.

[0053] The state variable x of the target vehicle finally estimated by the method proposed in the present invention can be expressed by mathematical expression 5, which has the center position (x, y) of the target vehicle, the speed v of the target vehicle, and the target vehicle's t , the heading of the target vehicle γ, the angular velocity of the target vehicle ω t , the length l of the target vehicle and the width w of the target vehicle are used as variables.

[0054] [Mathematical expression 5]

[0055] x=[xyv t γ ωt lw] T

[0056] Using the heading angle of the target vehicle and the relative position of the target vehicle defined in the present vehicle coordinate system, the target movement model of mathematical expression 1 can be expressed as the target movement equation of the present vehicle coordinate system, and the target movement equation of the present vehicle coordinate system is expressed as the differential equation of the state vector in mathematical expression 5, which is expressed by the following mathematical expression 6.

[0057] [Mathematical expression 6]

[0058]

[0059] Variables constituting Mathematical Expression 6 are shown in Mathematical Expression 7.

[0060] [Mathematical expression 7]

[0061]

[0062]

[0063]

[0064] In mathematical expression 6, Δψ represents the yaw angle increment of the host vehicle, and Indicates the relative position of the target vehicle and the host vehicle.

[0065] Figure 3 An embodiment of a method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of trajectory segments according to the present disclosure is shown.

[0066] Reference Figure 3According to the present disclosure, a method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of trajectory segments (hereinafter referred to as a method for simultaneously estimating the movement and shape of a target vehicle) includes a target vehicle state variable and movement equation creation step (310), a trajectory segment preliminary distribution model setting step (320), a trajectory segment acquisition step (330), a gating and data association step (340), a trajectory segment measurement equation construction step (350), a measurement value updating step (360) and a system propagation step (370).

[0067] In the target vehicle state variable and movement equation creation step (310), the target vehicle movement model (mathematical expression 1) and the state variable of the target vehicle (mathematical expression 5) in the inertial coordinate system are set, and the relative position of the target vehicle and the heading angle of the target vehicle in the current vehicle coordinate system are applied to the movement model of the target vehicle (mathematical expression 1) to create the target vehicle movement equation (mathematical expression 6) in the current vehicle coordinate system.

[0068] In the trajectory segment preliminary distribution model setting step (320), a preliminary distribution model of the trajectory segments is set. Although the positions of the trajectory segments relative to the target vehicle are somewhat irregular, the positions of the trajectory segments are evenly distributed at specific points on the contour of the target vehicle shape.

[0069] Figure 4 The preliminary distribution characteristics of the trajectory segments are shown, which illustrate the locations where representative trajectory segments appear in the target vehicle shape defined by the rectangle.

[0070] Reference Figure 4 On the left side of , it can be seen that the preliminary distribution model of the trajectory segment includes specific points ① to ⑧ of the target vehicle shape, which is approximately rectangular in the orthogonal coordinate system. Figure 4 On the right side, it can be seen that each of the eight points can be described as the center position in the vehicle coordinate system. Length of the target vehicle width and heading (These are all predicted) functions, that is, the coordinate function h of the center point of the target vehicle, such as Figure 2 As shown in Figure 2, depending on the incident angle of the high-resolution radar, the high-resolution radar may not be able to perceive certain points of the preliminary distribution model that are blocked by the target vehicle's body. Therefore, the number of track segments output provided by the radar may vary.

[0071] Reference Figure 4 On the right side, we can see that the coordinates of the second point is the sum of the coordinates of the second point and the third point The coordinates of the fourth, fifth, seventh, and ninth points are calculated using a similar method.

[0072] In the trajectory segment acquisition step (330), a trajectory segment is acquired. The trajectory segment output used in the present disclosure is the sum of the coordinate function h of the center point of the target vehicle and the unknown measurement noise.

[0073] In the gating and data association step (340), a gating and data association process is performed for the trajectory segment obtained in the trajectory segment obtaining step (330) and the coordinate function h of the center point of the target vehicle set in the trajectory segment preliminary distribution model setting step (320) to determine the point of the target vehicle that generated the trajectory segment obtained in the trajectory segment obtaining step (330) and select the associated trajectory segment.

[0074] Figure 5 Gating and data association of a preliminary distribution model of trajectory segments are shown.

[0075] Figure 5 Indicated consideration Figure 4 The concept of processing the trajectory segment output actually obtained by the high-resolution radar is based on the correspondence between the coordinate function h of the center point of the target vehicle defined in .

[0076] Reference Figure 5 , for the points of the preliminary distribution model The three points ⑤, ⑥, and ⑦ are gates. The trajectory segments within the gate are considered to be valid trajectory segment outputs generated by the corresponding point, and the trajectory segment outputs outside the gate are considered to be trajectory segments generated by clutter or another point of the preliminary distribution model.

[0077] It can be seen that although there is only one trajectory segment in the gate at point ⑦, there are three trajectory segments at point ⑤ and two trajectory segments at point ⑥.

[0078] When there is a single trajectory segment in the gate, the trajectory segment is selected as the trajectory segment generated by the corresponding point (hereinafter referred to as the associated trajectory segment). However, when there are multiple trajectory segments in the gate, it is necessary to further determine which trajectory segment is the associated trajectory segment. For example, the trajectory segment closest to the location where the representative trajectory segment appears in the preliminary distribution model can be selected as the trajectory segment generated by the corresponding point, i.e., the associated trajectory segment.

[0079] In the track segment measurement equation construction step (350), track segment measurement equations for estimating the movement and shape of the target vehicle are constructed using the associated track segments selected in the gating and data association step (340).

[0080] The trajectory segment measurement equation z is expressed by the following mathematical expression 8: k The associated trajectory segments and the predicted shape of the target vehicle in the orthogonal coordinate system can be used The predicted value of each point with the preliminary distribution model The corresponding relationship between Figure 4 The function h) is shown.

[0081] [Mathematical expression 8]

[0082] z k =h(x k )+v k

[0083] In mathematical expression 8, v k represents the measurement noise contained in the track segment output of the high-resolution radar. The measurement noise is considered to have a mean of 0 and a dispersion of R k Normal distribution.

[0084] In the measurement value updating step (360), a Kalman filter measurement updating process is performed on the result of the trajectory segment measurement equation constructed in the trajectory segment measurement equation construction step (350) and the state variable of the target vehicle represented by mathematical expression 5, thereby updating the estimated values ​​of the center position in the orthogonal coordinate system and the shape (length and width) of the target vehicle.

[0085] The Kalman filter measurement update process is performed using the state variables of the target vehicle represented by Mathematical Expression 5 and the trajectory segment measurement equation represented by Mathematical Expression 8, as shown in the following Mathematical Expression 9, to calculate the estimated values ​​of the target vehicle state variables, namely the center position (x, y) of the target vehicle, the speed v of the target vehicle t , the heading γ of the target vehicle, and the shape of the target vehicle (i.e., the length l and width w of the target vehicle).

[0086] [Mathematical expression 9]

[0087]

[0088]

[0089] In the system propagation step (370), a Kalman filter system propagation process is performed on the movement and shape of the target vehicle updated in the measurement value update step (360) to create the shape and center position of the target vehicle at the next time point.

[0090] The Kalman filter system propagation process is carried out using the measurement equation (mathematical expression 8) and the target movement equation (mathematical expression 6) in the vehicle coordinate system, so that the position and shape of the target vehicle at the next time point can be predicted by mathematical expression 10.

[0091] [Mathematical expression 10]

[0092]

[0093]

[0094] Predicted target vehicle center position (x, y) and shape This is reflected in the target vehicle state variables and motion equation creation step (310) to reconstruct a preliminary distribution model of the trajectory segments, and on this basis the data association and target vehicle motion / shape parameter estimation process is repeated.

[0095] The preliminary trajectory distribution model setting step (320) can be performed by a signal processing device (not shown) mounted on the host vehicle to process signals emitted from the plurality of radars mounted on the host vehicle and reflected from the target vehicle. The other steps (310 and 330 to 360) can be performed by a signal processing device mounted on the host vehicle or any device having signal processing and computing functions, such as a main computing device (not shown).

[0096] In the case where each step is performed by a signal processing device, the signal processing device may include a preliminary distribution model setting unit, a trajectory segment output acquisition unit, a trajectory segment selection unit, a measurement equation creation unit, a state variable management unit, a target vehicle tracking unit and a movement equation creation unit.

[0097] The preliminary distribution model setting unit may define an appearance point of at least one trajectory corresponding to a specific point of the target vehicle shape approximating a polygonal form in an orthogonal coordinate system and may set a coordinate function of the target vehicle center point relative to each defined point to set the preliminary distribution model.

[0098] The track segment output acquisition unit may acquire track segment outputs from a plurality of radars.

[0099] In a case where a trajectory segment among the acquired trajectory segment outputs exists in a gate set based on at least one trajectory segment appearance point, the trajectory segment selection unit may select at least one trajectory segment appearance point located in the gate as an associated trajectory segment.

[0100] The measurement equation creation unit may create the measurement equation based on the associated trajectory segment, the position of each of the at least one trajectory segment appearance point, and the coordinate function.

[0101] The state variable management unit can calculate the estimated value of the state variable of the target vehicle based on the state variable and the measurement equation of the target vehicle to update the state variable, and the target vehicle tracking unit can create a predicted value of the target vehicle at the next time point based on the calculated state variable estimated value and the target vehicle movement equation in the vehicle coordinate system.

[0102] In addition, the movement equation creation unit can set the movement model and state variables of the target vehicle in the inertial coordinate system, and can apply the relative position of the target vehicle and the heading angle of the target vehicle in the vehicle coordinate system to the movement model to create the target vehicle movement equation.

[0103] The present disclosure described above can be implemented as a computer-readable program stored in a non-transitory computer-readable recording medium. A computer-readable medium can be any type of recording device that stores data in a computer-readable manner. Computer-readable media may include, for example, a hard disk drive (HDD), a solid-state drive (SSD), a silicon disk drive (SDD), a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device.

[0104] As is apparent from the above description, in the method for simultaneously estimating the movement and shape of a target vehicle using a preliminary distribution model of trajectory segments according to the present disclosure, a small number of representative trajectory segment occurrence points are provided, and many-to-one data association is performed between trajectory segment outputs and individual output points, and thus the method is very suitable for real-time implementation of a target tracking algorithm.

[0105] In addition, the position of the target vehicle and the trajectory segment have a linear correspondence in the orthogonal coordinate system, which can fundamentally prevent the ambiguity problem that occurs when constructing the measurement equation, and improve the reliability of the estimation performance even if the heading of the target vehicle changes.

[0106] It should be noted that the effects of the present disclosure are not limited to the above effects, and those skilled in the art will clearly understand other effects not mentioned through the above description.

[0107] Although the technical concept of the present disclosure has been described with reference to the accompanying drawings, this is only an exemplary description of the preferred embodiments of the present disclosure and does not limit the present disclosure. In addition, those skilled in the art will understand that various modifications and variations are possible without departing from the scope of the technical concept of the present disclosure.

Claims

1. A method for simultaneously estimating the movement and shape of a target vehicle, the method being performed by a signal processing device that processes signals emitted from a plurality of radars installed at a host vehicle and reflected from the target vehicle, the method comprising: defining, in an orthogonal coordinate system, an appearance point of at least one trajectory segment corresponding to a specific point of a target vehicle shape approximating a polygon, and setting a coordinate function of a center point of the target vehicle relative to each defined point to set a preliminary distribution model; obtaining track segment outputs from the plurality of radars; If at least one trajectory segment among the acquired trajectory segment outputs is located in a gate set based on at least one trajectory segment appearance point, selecting the at least one trajectory segment appearance point located in the gate as an associated trajectory segment; creating a measurement equation based on the associated trajectory segment, the position of each of the at least one trajectory segment occurrence point, and the coordinate function; calculating an estimated value of the state variable based on the state variable of the target vehicle and the measurement equation to update the state variable; as well as Based on the calculated estimates of the state variables and the target vehicle's movement equations in the own vehicle's coordinate system, a predicted value for the target vehicle at the next point in time is created.

2. The method according to claim 1, further comprising: The state variables and a movement model of the target vehicle in an inertial coordinate system are set, and the relative position of the target vehicle in the host vehicle coordinate system and the heading angle of the target vehicle are applied to the movement model to create the target vehicle movement equation.

3. The method according to claim 2, wherein: The movement model is a constant turning rate movement model, ie, a CT movement model, and the CT movement model is discretized with respect to sampling time and then defined in the own vehicle coordinate system.

4. The method according to claim 1, wherein The state variable includes at least one of the following: a central position of the target vehicle, a speed of the target vehicle, a heading of the target vehicle, a rotational angular velocity of the target vehicle, or a length and a width of the target vehicle.

5. The method according to claim 1, wherein The estimated value of the state variable includes at least one of the following: a central position of the target vehicle, a speed of the target vehicle, a heading of the target vehicle, a rotational angular velocity of the target vehicle, or a length and a width of the target vehicle.

6. The method according to claim 1, wherein The polygon includes a quadrilateral.

7. The method according to claim 6, wherein: The specific point is located at each of the four corners and four sides of the quadrilateral.

8. The method according to claim 1, wherein The coordinate function of the center point of the target vehicle includes a function of the center position in the vehicle coordinate system, the length of the target vehicle, the width of the target vehicle and the heading of the target vehicle. The center position in the vehicle coordinate system, the length of the target vehicle, the width of the target vehicle and the heading of the target vehicle are all predicted.

9. The method according to claim 8, wherein The coordinate function of the center point of the target vehicle is additionally expressed using the coordinates of another specific point of the target vehicle.

10. The method according to claim 1, wherein The predicted values ​​of the target vehicle include the predicted center position, length, and width of the target vehicle.

11. The method according to claim 10, further comprising: The preliminary distribution model is reconstructed based on the predicted value of the target vehicle.

12. A non-transitory computer-readable recording medium containing program instructions executed by a processor, the computer-readable recording medium comprising: program instructions for defining, in an orthogonal coordinate system, an appearance point of at least one trajectory segment corresponding to a specific point of an approximately polygonal target vehicle shape and setting a coordinate function of a center point of the target vehicle relative to each defined point to set a preliminary distribution model; Program instructions for acquiring track segment output from multiple radars; If at least one trajectory segment among the acquired trajectory segment outputs is located in a gate set based on at least one trajectory segment occurrence point, program instructions are provided for selecting the at least one trajectory segment occurrence point located in the gate as an associated trajectory segment; program instructions for creating a measurement equation based on associating a trajectory segment, a location of each of the at least one trajectory segment occurrence point, and the coordinate function; program instructions for calculating an estimate of the state variable based on the state variable of the target vehicle and the measurement equation to update the state variable; as well as Program instructions for creating a predicted value of the target vehicle at the next time point based on the calculated estimated values ​​of the state variables and the target vehicle movement equation in the host vehicle coordinate system.

13. A vehicle comprising: multiple radars; as well as A signal processing device, wherein the signal processing device comprises: a preliminary distribution model setting unit that defines, in an orthogonal coordinate system, an appearance point of at least one trajectory segment corresponding to a specific point of a target vehicle shape approximating a polygon, and sets a coordinate function of a center point of the target vehicle relative to each defined point to set a preliminary distribution model; A track segment output acquisition unit, which acquires track segment outputs from multiple radars; a trajectory segment selection unit, configured to select, if at least one trajectory segment among the acquired trajectory segment outputs is located in a gate set based on at least one trajectory segment appearance point, the at least one trajectory segment appearance point located in the gate as an associated trajectory segment; a measurement equation creating unit that creates a measurement equation based on the associated trajectory segment, the position of each of the at least one trajectory segment occurrence point, and the coordinate function; a state variable management unit that calculates estimated values ​​of the state variables based on the state variables of the target vehicle and the measurement equation to update the state variables; and The target vehicle tracking unit creates a predicted value of the target vehicle at a next time point based on the calculated estimated value of the state variable and the target vehicle movement equation in the vehicle coordinate system.

14. The vehicle of claim 13, further comprising: A movement equation creation unit sets the state variables and a movement model of the target vehicle in an inertial coordinate system, and applies the relative position of the target vehicle in the own vehicle coordinate system and the heading angle of the target vehicle to the movement model to create the movement equation of the target vehicle.

15. The vehicle of claim 14, wherein: The movement model is a constant turning rate movement model, ie, a CT movement model, and the CT movement model is discretized with respect to sampling time and then defined in the own vehicle coordinate system.

16. The vehicle of claim 13, wherein: The state variable includes at least one of the following: a central position of the target vehicle, a speed of the target vehicle, a heading of the target vehicle, a rotational angular velocity of the target vehicle, or a length and a width of the target vehicle.

17. The vehicle of claim 13, wherein: The estimated value of the state variable includes at least one of the following: a central position of the target vehicle, a speed of the target vehicle, a heading of the target vehicle, a rotational angular velocity of the target vehicle, or a length and a width of the target vehicle.

18. The vehicle of claim 13, wherein: The polygon includes a quadrilateral.

19. The vehicle of claim 13, wherein: The coordinate function of the center point of the target vehicle includes a function of the center position in the vehicle coordinate system, the length of the target vehicle, the width of the target vehicle and the heading of the target vehicle. The center position in the vehicle coordinate system, the length of the target vehicle, the width of the target vehicle and the heading of the target vehicle are all predicted.

20. The vehicle of claim 19, wherein: The coordinate function of the center point of the target vehicle is additionally expressed using the coordinates of another specific point of the target vehicle.

Images