Road constraint determination method and device

Abstract

The present application discloses a method and device for determining road constraints, which are applied to the field of intelligent driving, and particularly relate to sensors in advanced driver assistance systems (ADAS) or autonomous driving systems, such as radars and / or cameras. In this method, the motion state of the target is determined based on the detection information of the target, and at least one road geometry of the road where the target is located is determined based on the detection information of the target; the road constraint of the target is determined based on the at least one road geometry and the motion state of the target, and the road constraint includes at least one of a road direction constraint and a road width constraint. The solution of the present application can improve the accuracy of determining road constraints and further improve the accuracy of target tracking. It further enhances the capabilities of autonomous driving or ADAS, and can be applied to vehicle networks, such as vehicle-to-vehicle (V2X), long-term evolution (LTE‑V) technology for inter-vehicle communication, and vehicle-to-vehicle (V2V).

Description

Technical Field

[0001] The present application relates to the field of target tracking technology, and in particular to a method and device for determining road constraints. Background Art

[0002] For targets on the ground, their motion is often somewhat predictable due to constraints imposed by the road or geographical environment. Therefore, target tracking technology is often used to predict the target's motion state. This is often done in advanced driving assistance systems (ADAS) or autonomous driving systems. In this case, the target can be one or more moving or stationary objects, such as bicycles, motor vehicles, people, or animals.

[0003] In ADAS or unmanned driving systems, tracking equipment is typically used to achieve target tracking. This equipment captures detection information transmitted by devices such as radar or imaging devices and uses target tracking technology to track the target. The detection information transmitted by radar typically includes information such as the target's range, azimuth, and velocity, while the detection information transmitted by imaging devices typically includes an image of the target. The tracking equipment then tracks the target based on this detection information and a pre-set algorithm (such as a Kalman filter).

[0004] Furthermore, considering that targets are subject to environmental constraints during their movement, for example, when moving on a road, they are constrained by road edges or lane markings, which restricts their movement to a specific area, another target tracking method is used. In addition to acquiring detection information, the tracking device can also utilize road constraints to further improve target tracking accuracy.

[0005] ADAS also typically needs to predict the target's trajectory over the next period of time, such as a 2-second prediction. To ensure accurate predictions, ADAS can also utilize road constraints to further improve the accuracy of its predictions of the target's future trajectory.

[0006] In this method, the road constraints generally include road direction constraints and road width constraints. Figure 1 As shown in the schematic diagram, in the existing method for determining road constraints, the road where the target is located is divided into at least one road segment, wherein each road segment is usually represented by two endpoints at the beginning and end of the road segment and the line between the two endpoints. For a curved road, it is divided into multiple connected road segments, for example, Figure 1In the road diagram shown, the road is divided into five segments connected end to end. In this case, the direction of the line connecting the end points of the segment where the vehicle is located is the road direction constraint, and the width of each segment is the road width constraint.

[0007] However, when determining road constraints using existing methods, the road scenes considered are relatively simple, while actual roads often have various situations such as discontinuity, intersections, and mergers, and the road conditions are relatively complex. Therefore, the road constraints obtained using existing technologies have large errors, resulting in low accuracy of road constraints, which further affects the accuracy of target tracking. Summary of the Invention

[0008] In order to solve the problem in the prior art of low accuracy of road constraints obtained when tracking a target, the embodiments of the present application disclose a road constraint determination method and apparatus.

[0009] In a first aspect, an embodiment of the present application discloses a method for determining a road constraint, comprising:

[0010] determining a motion state of the target based on the detection information of the target;

[0011] determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, each road geometry in the at least one road geometry being represented by at least one item of information;

[0012] A road constraint of the target is determined based on the at least one road geometry and a motion state of the target, the road constraint comprising at least one of a road direction constraint and a road width constraint.

[0013] In the above steps, the target's road constraints are determined based on the road geometry and the target's motion state. Road geometry reflects the geometric shape of the road on which the target is located. Therefore, compared to the prior art, the solution of the present embodiment can improve the accuracy of determining road constraints and further improve the accuracy of target tracking.

[0014] In an optional design, the method further includes:

[0015] determining at least one target road geometry among the at least one road geometry;

[0016] The determining of the road constraint of the target based on the at least one road geometry and the motion state of the target includes:

[0017] A road constraint of the target is determined based on the at least one target road geometry and a motion state of the target.

[0018] Through the above steps, the target road geometry in the road geometry can be determined. The target road geometry is the road geometry used for subsequent determination of road constraints. Determining the road constraints through the target road geometry can further improve the accuracy of determining the road constraints.

[0019] In an optional design, determining at least one target road geometry among the at least one road geometry includes:

[0020] For each road geometry of the at least one road geometry:

[0021] Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position;

[0022] acquiring a tangential direction angle of the target at the target position according to a lateral velocity and a radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range;

[0023] If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, the road geometry is determined to be the target road geometry.

[0024] Through the above steps, the target road geometry in the road geometry can be determined according to the tangent direction angle of the target at the target position and the tangent direction angle of the road geometry at the first position.

[0025] In an optional design, determining at least one target road geometry among the at least one road geometry includes:

[0026] For each road geometry of the at least one road geometry:

[0027] If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

[0028] Through the above steps, the target road geometry in the road geometry can be determined according to the distance between the target and the road geometry.

[0029] In an optional design, determining at least one target road geometry among the at least one road geometry includes:

[0030] For each road geometry of the at least one road geometry:

[0031] Obtaining a distance between the target and the road geometry;

[0032] According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

[0033] Through the above steps, the target road geometry in the road geometry can be determined according to the distance between the target and the road geometry, and the number of road geometries.

[0034] In an optional design, determining the road direction constraint of the target based on the at least one road geometry and the motion state of the target includes:

[0035] Determining at least one second position located in the at least one target road geometry, wherein the at least one second position is a position closest to the target in at least one first target road geometry, and the first target road geometry is the target road geometry where the second position is located;

[0036] A road direction constraint of the target is determined based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second position.

[0037] Through the above steps, the target road direction constraint can be determined according to the second position in the target road geometry and the confidence level of the target road geometry.

[0038] In an optional design, the confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry;

[0039] or,

[0040] The confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

[0041] In an optional design, when the confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry, the method further includes:

[0042] determining a confidence level of a road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, wherein the road parameter is at least one item of information used to characterize the target road geometry;

[0043] Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the method further includes:

[0044] The confidence level of the tangent direction angle of the target road geometry at the second position is determined according to the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

[0045] In an optional design, determining the road direction constraint of the target based on the confidence level of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0046] determining, according to the confidence level of the target road geometry, a weight value of a tangent direction angle of the at least one target road geometry at the at least one second position during fusion;

[0047] According to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

[0048] In an optional design, determining the weight value of the tangent direction angle of each target road geometry during fusion based on the confidence level of each target road geometry includes:

[0049] Determining a weight value of a tangent direction angle of each target road geometry at the second position based on a correspondence between the confidence level and the weight value of the road geometry and the confidence level of each target road geometry;

[0050] Alternatively, the weight value is determined by any of the following formulas:

[0051]

[0052]

[0053] Among them, w i represents the weight value of the tangent direction angle of the i-th target road geometry during fusion; φ(i) is the tangent direction angle of the i-th target road geometry at the second position; δ(φ(i)) is the confidence of the i-th target road geometry; n is the number of target road geometries; h(d i ) is the shortest distance between the target and the i-th target road geometry.

[0054] Through the above steps, the weight value of the tangent direction angle of the at least one target road geometry at the at least one second position during fusion can be determined.

[0055] In an optional design, determining the road direction constraint of the target based on the confidence level of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0056] Determine the tangent angle direction at a second position as the road direction constraint of the target, wherein the second position is a position closest to the target in a second target road geometry, and the second target road geometry is a target road geometry with the highest confidence among the at least one target road geometry.

[0057] In an optional design, determining the road direction constraint of the target based on the confidence level of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0058] Determine the tangent angle direction at a third position as the road direction constraint of the target, wherein the third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

[0059] In an optional design, determining a road width constraint for the target based on the geometry of the at least one target road and the motion state of the target includes:

[0060] Obtaining a straight line passing through the target position of the target and perpendicular to fourth target road geometries, wherein the fourth target road geometries are two target road geometries closest to the target and located on both sides of the target;

[0061] A distance between two intersection points is determined as a road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

[0062] In an optional design, determining a road width constraint for the target based on the geometry of the at least one target road and the motion state of the target includes:

[0063] determining at least one distance between the target and the at least one target road geometry, the at least one distance being a road width constraint for the target;

[0064] or,

[0065] determining at least one distance between the target and the at least one target road geometry, and determining a maximum or minimum value of the at least one distance as a road width constraint for the target;

[0066] or,

[0067] A distance between the target and the at least one target road geometry is determined, and an average of the at least one distance is determined as a road width constraint for the target.

[0068] In an optional design, the method further includes:

[0069] determining a measurement matrix including the road direction constraint;

[0070] The confidence level of the road direction constraint in the measurement matrix is determined by the road width constraint.

[0071] Through the above steps, a measurement matrix can be determined according to the road direction constraint and the road width constraint. When target tracking is performed using the measurement matrix, the accuracy of target tracking can be improved.

[0072] In an optional design, determining the confidence of the road direction constraint in the measurement matrix using the road width constraint includes:

[0073] determining the measurement noise corresponding to the target according to a mapping relationship between a road width constraint and measurement noise, and the road width constraint;

[0074] The confidence of the road direction constraint in the measurement matrix is determined according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target.

[0075] In an optional design, after determining the measurement noise corresponding to the target based on the mapping relationship between the road width constraint and the measurement noise and the road width constraint, the method further includes:

[0076] Determining a motion state change parameter of the target according to the motion state of the target;

[0077] When a comparison result of the motion state change parameter of the target and a corresponding threshold value indicates that it is necessary to determine a degree of change in the curvature or a degree of change in the curvature change rate of the target road geometry at a fourth position, determining the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position, the fourth position being a position located on the target road geometry and within a third distance range from the target;

[0078] When the degree of change of the curvature or the degree of change of the curvature change rate is greater than a third threshold, increasing the measurement noise corresponding to the target;

[0079] The determining the confidence of the road direction constraint in the measurement matrix according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes:

[0080] The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

[0081] Through the above steps, measurement noise can be adjusted based on the target's motion state change parameters. Furthermore, the confidence level of the road direction constraint in the measurement matrix can be determined based on the adjusted measurement noise. When the degree of change in curvature or the degree of change in the curvature change rate exceeds a third threshold, indicating a lane change, the measurement noise is increased. Therefore, through the above steps, the measurement matrix corresponding to a lane change can be considered, further improving target tracking accuracy.

[0082] In an optional design, the motion state change parameter includes:

[0083] an average value of a normalized innovation square (NIS) parameter corresponding to the at least one target road geometry;

[0084] or,

[0085] The curvature of the historical motion trajectory of the target or the degree of change of the curvature.

[0086] In an optional design, when the target motion state change parameter is an average value of the NIS parameter, and the target motion state change parameter is greater than a corresponding threshold, the comparison result indicates that it is necessary to determine the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at the target position;

[0087] or,

[0088] When the motion state change parameter of the target is the curvature of the historical motion trajectory of the target or the degree of change of the curvature, and the motion state change parameter of the target is less than the product of the curvature or the curvature change rate of the target's motion trajectory at the current moment and a preset positive number, the comparison result indicates that it is necessary to determine the degree of change of the curvature of the target road geometry at the target position or the degree of change of the curvature change rate.

[0089] In a second aspect, an embodiment of the present application provides a road constraint determination device, comprising:

[0090] at least one processing module;

[0091] The at least one processing module is configured to determine a motion state of the target based on the detection information of the target;

[0092] The at least one processing module is also used to determine at least one road geometry of the road where the target is located based on the detection information of the target, each road geometry of the at least one road geometry is represented by at least one item of information; and to determine the road constraint of the target based on the at least one road geometry and the motion state of the target, the road constraint including at least one of a road direction constraint and a road width constraint.

[0093] In an optional design, the at least one processing module is further configured to determine at least one target road geometry among the at least one road geometry;

[0094] The at least one processing module is specifically configured to determine a road constraint of the target based on the at least one target road geometry and a motion state of the target.

[0095] In an optional design, the at least one processing module is specifically configured to, for each road geometry of the at least one road geometry:

[0096] Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position;

[0097] acquiring a tangential direction angle of the target at the target position according to a lateral velocity and a radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range;

[0098] If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, the road geometry is determined to be the target road geometry.

[0099] In an optional design, the at least one processing module is specifically configured to, for each road geometry of the at least one road geometry:

[0100] If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

[0101] In an optional design, the at least one processing module is specifically configured to, for each road geometry of the at least one road geometry:

[0102] Obtaining a distance between the target and the road geometry;

[0103] According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

[0104] In an optional design, the at least one processing module is specifically configured to determine at least one second position located in the at least one target road geometry, the at least one second position being a position closest to the target in at least one first target road geometry, the first target road geometry being the target road geometry where the second position is located;

[0105] A road direction constraint of the target is determined based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second position.

[0106] In an optional design, the confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry;

[0107] or,

[0108] The confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

[0109] In an optional design, when the confidence level of the target road geometry is the confidence level of a road parameter of the target road geometry, the at least one processing module is further configured to determine the confidence level of the road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, where the road parameter is at least one item of information used to characterize the target road geometry;

[0110] Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the at least one processing module is further used to determine the confidence level of the tangent direction angle of the target road geometry at the second position based on the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

[0111] In an optional design, the at least one processing module is specifically configured to determine, based on the confidence level of the target road geometry, a weight value of a tangent direction angle of the at least one target road geometry at the at least one second position during fusion;

[0112] According to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

[0113] In an optional design, the at least one processing module is specifically used to determine the tangent angle direction at a second position as the road direction constraint of the target, wherein the second position is the position closest to the target in the second target road geometry, and the second target road geometry is the target road geometry with the highest confidence among the at least one target road geometry.

[0114] In an optional design, the at least one processing module is specifically used to determine the tangent angle direction at a third position as the road direction constraint of the target, wherein the third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

[0115] In an optional design, the at least one processing module is specifically configured to obtain a straight line passing through the target position of the target and perpendicular to fourth target road geometries, wherein the fourth target road geometries are two target road geometries closest to the target and located on both sides of the target;

[0116] A distance between two intersection points is determined as a road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

[0117] In an optional design, the at least one processing module is specifically configured to determine at least one distance between the target and the at least one target road geometry, the at least one distance being a road width constraint of the target;

[0118] or,

[0119] The at least one processing module is specifically configured to determine at least one distance between the target and the at least one target road geometry, and determine a maximum value or a minimum value of the at least one distance as a road width constraint for the target;

[0120] or,

[0121] The at least one processing module is specifically configured to determine a distance between the target and the at least one target road geometry, and determine an average value of the at least one distance as a road width constraint for the target.

[0122] In an optional design, the at least one processing module is further configured to determine a measurement matrix including the road direction constraint;

[0123] The confidence level of the road direction constraint in the measurement matrix is determined by the road width constraint.

[0124] In an optional design, the at least one processing module is specifically configured to determine the measurement noise corresponding to the target based on a mapping relationship between a road width constraint and the measurement noise, and the road width constraint;

[0125] The confidence of the road direction constraint in the measurement matrix is determined according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target.

[0126] In an optional design, the at least one processing module is further configured to, after determining the measurement noise corresponding to the target based on a mapping relationship between a road width constraint and the measurement noise and the road width constraint, determine a motion state change parameter of the target based on the motion state of the target;

[0127] When a comparison result of the motion state change parameter of the target and a corresponding threshold value indicates that it is necessary to determine a degree of change in the curvature or a degree of change in the curvature change rate of the target road geometry at a fourth position, determining the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position, the fourth position being a position located on the target road geometry and within a third distance range from the target;

[0128] When the degree of change of the curvature or the degree of change of the curvature change rate is greater than a third threshold, increasing the measurement noise corresponding to the target;

[0129] The determining the confidence of the road direction constraint in the measurement matrix according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes:

[0130] The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

[0131] In a third aspect, an embodiment of the present application provides a road constraint determination device, comprising:

[0132] at least one processor and memory;

[0133] Wherein, the memory is used to store program instructions;

[0134] The at least one processor is configured to call and execute program instructions stored in the memory. When the processor executes the program instructions, the apparatus executes the method as described in the first aspect.

[0135] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that:

[0136] The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to execute the method according to the first aspect.

[0137] In a fifth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when run on an electronic device, enables the electronic device to execute the method described in the first aspect.

[0138] In an embodiment of the present application, the road constraints of the target are determined by the road geometry and the target's motion state. Among them, the road geometry can reflect the geometric shape of the road where the target is located. In this case, through the solution of the present application, the road constraints of the target can be determined based on the geometric shape of the road where the target is located and the target's motion state. When determining road constraints in the prior art, the road scene considered is relatively simple, and the road is only represented by a series of points and road segments connecting these points. Therefore, compared with the prior art, the solution of the embodiment of the present application can improve the accuracy of determining road constraints, and further improve the accuracy of target tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0139] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0140] Figure 1 A schematic diagram of determining road constraints disclosed in the prior art;

[0141] Figure 2 A schematic diagram of a workflow of a method for determining road constraints disclosed in an embodiment of the present application;

[0142] Figure 3 A schematic diagram of an application scenario of a method for determining road constraints disclosed in an embodiment of the present application;

[0143] Figure 4 A schematic diagram of a workflow for determining target road geometry in a method for determining road constraints disclosed in an embodiment of the present application;

[0144] Figure 5 A schematic diagram of a workflow for determining a road direction constraint in a method for determining a road constraint disclosed in an embodiment of the present application;

[0145] Figure 6 This is a schematic diagram of an application scenario of another method for determining road constraints disclosed in an embodiment of the present application;

[0146] Figure 7 A schematic diagram of a workflow of another method for determining road constraints disclosed in an embodiment of the present application;

[0147] Figure 8 This is a schematic structural diagram of a road constraint determination device disclosed in an embodiment of the present application;

[0148] Figure 9 This is a schematic structural diagram of another device for determining a road constraint disclosed in an embodiment of the present application;

[0149] Figure 10 A schematic structural diagram of a tracking device disclosed in an embodiment of the present application;

[0150] Figure 11 This is a schematic structural diagram of another tracking device disclosed in an embodiment of the present application;

[0151] Figure 12 This is a schematic structural diagram of another tracking device disclosed in an embodiment of the present application;

[0152] Figure 13 This is a schematic structural diagram of another tracking device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0153] In order to solve the problem in the prior art of low accuracy of road constraints obtained when tracking a target, the embodiments of the present application disclose a road constraint determination method and apparatus.

[0154] The road constraint determination method disclosed in the embodiment of the present application is generally applied to a tracking device, which is provided with a processor. The processor can determine the road constraints of the road where the target is located through the solution disclosed in the embodiment of the present application.

[0155] The processor, when determining the road constraints, needs to apply detection information. The detection information can be obtained by a sensor. The sensor typically includes a radar and / or an imaging device. The sensor can be connected to the processor in the tracking device and transmit the detection information to the processor, so that the processor can determine the road constraints based on the received detection information according to the solution disclosed in the embodiment of the present application. The sensor can be set in the tracking device. Alternatively, the sensor can also be a device independent of the tracking device.

[0156] The tracking device can be installed in a variety of locations. It can be a stationary device installed at a traffic intersection or on the side of a highway. Alternatively, it can be installed on a moving object, such as a moving vehicle. In this case, the tracking device installed in the vehicle can also obtain the target's road constraints while the vehicle is in motion.

[0157] In one example, the tracking device may be an on-board processor in a vehicle, and the sensors include an on-board radar on the vehicle and an imaging device in the vehicle. When the vehicle moves on a certain road, the on-board device may obtain detection information transmitted by the on-board radar and imaging device, and determine the road constraints of a target on the road according to the scheme disclosed in each embodiment of the present application.

[0158] Furthermore, when obtaining the road constraints of the road where the target is located, the target may be a stationary object or an object in motion, which is not limited in the embodiments of the present application.

[0159] The road constraint determination method disclosed in the embodiment of the present application is introduced below with reference to specific drawings and workflows.

[0160] See also Figure 2 As shown in the workflow diagram, the road constraint determination method disclosed in the embodiment of the present application includes the following steps:

[0161] Step S11: Determine the motion state of the target based on the detection information of the target.

[0162] The target detection information can be acquired via at least one sensor. The at least one sensor includes a radar and / or an imaging device. The radar can be at least one of various types of radars, such as a laser radar, a millimeter-wave radar, or an ultrasonic radar. The imaging device can be at least one of a camera, an infrared sensor, or a video camera, etc., although this embodiment of the present application does not limit this.

[0163] For example, millimeter-wave radar uses electromagnetic waves to detect targets. Specifically, it transmits electromagnetic waves toward a target and receives echoes from the target. Based on these echoes, it obtains various types of detection information, including the target's distance from the emission point, velocity, and azimuth. This detection information includes the target's distance, azimuth, and radial velocity relative to the millimeter-wave radar.

[0164] Accordingly, the detection information of the target may include various types of information, for example, information that can be detected by a radar, or image information captured by an imaging device.

[0165] In the embodiment of the present application, the motion state of the target includes at least the target position of the target, and further, the motion state of the target may also include the target position and speed of the target. The motion state of the target can be determined by the detection information of the target.

[0166] Step S12: Determine at least one road geometry of the road where the target is located based on the detection information of the target, where each road geometry in the at least one road geometry is represented by at least one item of information.

[0167] Among them, road geometry refers to the geometric shape of the road, and road geometry is characterized by at least one information such as the road's orientation, road curvature (curvature), curvature direction, and length. It should be noted that road geometry can also be characterized by other information, which is not specifically limited in this application.

[0168] See also Figure 3 The scene diagram shown in FIG. 1 shows that the target's movable area can be determined by the road edge, road guardrail or lane line of the road where the target is located. Figure 3 In the figure, the solid line represents the road edge or road guardrail, and the dotted line represents the lane line. In the embodiment of the present application, a road geometry corresponds to the geometric shape of any one of the road edge, road guardrail, and lane line. For example, a certain road geometry can be the road geometry of the road edge of the road where the target is located, and accordingly, the information used to characterize the road geometry is at least one piece of information such as the orientation, curvature, bending direction, and length of the road edge; another road geometry can be the road geometry of the lane line of the road where the target is located, and accordingly, the information used to characterize the road geometry is at least one piece of information such as the orientation, curvature, bending direction, and length of the lane line.

[0169] In the process of determining road geometry in this application, the millimeter-wave radar can transmit electromagnetic waves toward the road edge or road guardrail, receive the echo feedback after the electromagnetic waves contact the road edge or road guardrail, and then obtain corresponding detection information based on the echo. The detection information is then transmitted to the processor set in the tracking device, so that the processor set in the tracking device can determine the road geometry of the road edge or road guardrail based on the received detection information. In this case, the detection information detected by the millimeter-wave radar includes information such as the distance or azimuth of the road edge or road guardrail relative to the millimeter-wave radar.

[0170] In an embodiment of the present application, a road edge model and / or a road guardrail model may be set, and the road edge model and / or the road guardrail model include parameters of the road edge model and / or the road guardrail, which are information characterizing the road geometry; then, based on the detection information, the specific values of each parameter are determined, and the values are substituted into the corresponding road edge model and / or road guardrail model to obtain the road edge geometry and / or road guardrail geometry.

[0171] In a feasible implementation, the road edge model can be represented by any one of formulas (1) to (4). Moreover, since road guardrails are usually set along the road edge, the road guardrail model can also be represented by any one of formulas (1) to (4). Formulas (1) to (4) are respectively:

[0172]

[0173]

[0174]

[0175] y(x)=y0 R,i +φ0 Ri x Formula (4).

[0176] In the coordinate systems corresponding to formulas (1) to (4), x represents the radial distance between the tracking device and the road edge or road guardrail, and y represents the lateral distance between the tracking device and the road edge or road guardrail. Furthermore, the origin of the coordinate system can also be the coordinates of the radar's location, or the origin of the coordinate system can be another location whose relative position to the radar is fixed. Furthermore, when the radar moves, the origin of the coordinate system will also change accordingly. For example, when the radar is a vehicle-mounted radar, the origin of the coordinate system can be the location of the headlights or the center of the axle, and the origin of the coordinate system will change as the vehicle moves.

[0177] In addition, in formulas (1) to (4), R represents the parameters in the formula determined by the detection information fed back by the radar; i represents the number of the road edge or road guardrail; y(x) represents the relationship between the i-th road edge or road guardrail determined based on the detection information fed back by the radar; y0 R,i represents the lateral offset of the i-th road edge or road guardrail from the tracker when x=0. The lateral offset refers to the lateral displacement of the target relative to the road edge or road guardrail in the above coordinate system; φ0 Ri C0 represents the heading of the ith road edge or road guardrail when x=0. The heading refers to the angle between the road edge or road guardrail and the vertical axis of the coordinate system in the above coordinate system. Ri represents the average curvature of the ith road edge or road guardrail; C1 Ri The average value of the rate of change of the curvature of the i-th road edge or road guardrail is represented by the radar feedback detection information. The processor of the tracking device can determine y0 R,i 、φ0 Ri 、C0Ri and C1 Ri The specific value of , thus we can get formula (1) to formula (4). Among them, y0 R,i 、φ0 Ri 、C0 Ri and C1 Ri This is information used to characterize the geometry of road edges or road guardrails.

[0178] Among them, formula (3) is more suitable for describing scenes with large road curvature, such as semicircular roads, while formula (4) is more suitable for scenes with small road curvature, such as straight roads. Formulas (1) and (2) are between the two and are more suitable for describing roads with curvatures between straight roads and semicircular roads. The processor executing the embodiment of the present application can select the corresponding formula according to the road conditions of the target road. Alternatively, the road edge model or road guardrail model can be described by a combination of one or more of formulas (1) to (4), for example, different formulas are used to represent the sections at roads with different curvatures.

[0179] Of course, the road edge and road guardrail models can also be represented by other formulas, and this embodiment of the present application does not limit this.

[0180] In addition, the sensor generally further includes an imaging device, which may be a camera, an infrared sensor, or a video camera, etc., and the embodiments of the present application do not limit this.

[0181] The imaging device can acquire a road image, wherein the road image includes lane lines, and the lane lines are usually marked with a special color, for example, lane lines are usually marked with yellow or white. In this case, after receiving the road image transmitted by the imaging device, the tracking device can extract edge information in the road image, and then determine whether the road image contains lane lines by combining color features and edge information. Of course, the road image acquired by the imaging device may also include road edges and / or road guardrails, etc. By processing the road image, the road edges and / or road guardrails included in the road image can also be determined, and the road edge model and / or road guardrail model can be determined by any one of formulas (1) to (4).

[0182] When lane markings are determined to be present in the road image, the specific values of the parameters in the lane model can be determined based on the detection information. These values are then substituted into the corresponding lane model to obtain the lane geometry. In this case, the detection information transmitted by the sensor refers to the road image captured by the imaging device.

[0183] In one feasible implementation, the lane line model can be expressed by the following formula:

[0184]

[0185]

[0186]

[0187] y(x)=y0 V,s +φ0 Vs x Formula (8).

[0188] In the coordinate systems corresponding to formulas (5) to (8), x represents the radial distance between the position of the tracking device and the road edge or road guardrail, and y represents the lateral distance between the position of the tracking device and the road edge or road guardrail. In addition, the origin of the coordinate system can also be the coordinates of the position where the imaging position is located, or the origin of the coordinate system can also be other positions, and the relative position between the other position and the imaging position is fixed. Moreover, when the imaging position moves, the origin of the coordinate system will also change accordingly. For example, when the imaging position is an on-board imaging position, the origin of the coordinate system can be the position of the headlight or the center position of the axle, and the origin of the coordinate system will change as the vehicle moves.

[0189] In addition, in formulas (5) to (8), V represents the parameters in the formula determined by the detection information fed back by the imaging device; s represents the lane number; y(x) represents the relationship between the sth lane line determined based on the detection information fed back by the imaging device; y0 V,s represents the lateral offset of the sth lane line from the tracking device when x=0; φ0 Vs Indicates the heading of the sth lane line when x=0; C0 Vs represents the average curvature of the sth lane line; C1 Vs The average value of the curvature change rate of the sth lane line is represented by the detection information fed back by the imaging device, which can determine y0 V,s 、φ0 Vs 、C0 Vs and C1 Vs The specific value of , thus we can get formula (5) to formula (8). Among them, y0 V,s 、φ0 Vs 、C0 Vs and C1 Vs This is the information used to characterize the geometry of the lane lines.

[0190] Among them, formula (7) is more suitable for describing scenes with large road curvature, such as semicircular roads, while formula (8) is more suitable for scenes with small road curvature, such as straight roads. Formulas (5) and (6) are between the two and are more suitable for describing roads with curvatures between straight roads and semicircular roads. The processor executing the embodiment of the present application can select the corresponding formula according to the road conditions of the target road. Alternatively, the lane line model can be described by a combination of one or more of formulas (5) to (8), for example, different formulas are used to represent the segments on roads with different curvatures.

[0191] Through formula (1) to formula (8), a model of road geometry can be obtained. In addition, in order to further improve the accuracy of road geometry, prior information can be combined with detection information to determine road geometry.

[0192] The prior information may be a pre-acquired map, which may be obtained in advance through a global positioning system (GPS) or simultaneous localization and mapping (SLAM). When determining the road geometry in conjunction with the prior information, the detection information transmitted by the sensor is matched and compared with the pre-acquired map to determine whether the environmental information represented by the detection information transmitted by the sensor is consistent with the environmental information displayed on the map. If they are consistent, the road geometry model is determined according to formulas (5) to (8).

[0193] Step S13: Determine a road constraint of the target based on the at least one road geometry and the motion state of the target, where the road constraint includes at least one of a road direction constraint and a road width constraint.

[0194] See also Figure 3 , the target's road orientation constraint can be determined based on parameters such as the tangent angle of the road geometry at a specific location. The tangent angle at a specific location refers to the angle between the tangent and the radial at that location. Additionally, the target's road width constraint can be determined based on parameters such as the distance between the target and the road geometry.

[0195] In an embodiment of the present application, the road constraints of the target are determined by the road geometry and the target's motion state. Among them, the road geometry can reflect the geometric shape of the road where the target is located. In this case, through the solution of the present application, the road constraints of the target can be determined based on the geometric shape of the road where the target is located and the target's motion state. When determining road constraints in the prior art, the road scene considered is relatively simple, and the road is only represented by a series of points and road segments connecting these points. Therefore, compared with the prior art, the solution of the embodiment of the present application can improve the accuracy of determining road constraints, and further improve the accuracy of target tracking.

[0196] Furthermore, in an embodiment of the present application, the method further includes:

[0197] At least one target road geometry of the at least one road geometry is determined.

[0198] In this case, determining the road constraint of the target based on the at least one road geometry and the motion state of the target includes:

[0199] A road constraint of the target is determined based on the at least one target road geometry and a motion state of the target.

[0200] In the embodiments of the present application, multiple road geometries can often be obtained. However, some of the road geometries may deviate significantly from the target's motion trajectory. If the road constraints are determined using road geometries with significant deviations, the accuracy of the road constraints will be reduced.

[0201] For example, when the road geometry is determined by detection information transmitted by a radar, and the target is a vehicle on the road, the electromagnetic waves generated by the radar often produce echoes when they come into contact with buildings beside the road or other vehicles on the road. In this case, the part of the road geometry determined by the detection information fed back by the radar may have a large deviation problem. In addition, when the road geometry is determined by detection information transmitted by an imaging device, errors may occur during the image processing process, which may also cause the determined part of the road geometry to be inaccurate. Therefore, it is necessary to determine the target road geometry in the road geometry through the above operation. The target road geometry is the road geometry used for the subsequent determination of road constraints. In this case, determining the road constraints through the target road geometry can further improve the accuracy of determining the road constraints.

[0202] In the embodiment of the present application, the target road geometry can be determined in a variety of ways. In the first feasible implementation, when the target's motion state includes the target's position and the target's speed, see Figure 4As shown in the workflow diagram, for each road geometry in the at least one road geometry, at least one target road geometry in the at least one road geometry may be determined by the following steps:

[0203] Step S121: Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position.

[0204] When the coordinates of the first position are (x1, y1), the tangent direction angle can be determined by the following formula:

[0205]

[0206] in, is the tangent direction angle of the road geometry at the first position; x1 is the x-axis coordinate of the first position in the ground coordinate system; φ0 is the heading of the road geometry when x=0; C0 represents the average curvature of the road geometry; C1 represents the average value of the rate of change of the curvature of the road geometry, and φ1 is the heading of the road geometry at the first position.

[0207] After determining φ1 by formula (9), the tangent direction angle can be determined by the following formula:

[0208]

[0209] Step S122: Acquire a tangent direction angle of the target at the target position according to the lateral velocity and radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range.

[0210] When the coordinates of the target are (x2, y2), the tangent direction angle of the target at the target position (x2, y2) can be determined by the following formula:

[0211]

[0212] in, is the tangential direction angle at the target location; vy2 is the lateral velocity at the target location; vx2 is the radial velocity at the target location, and the target location is the target position.

[0213] According to the detection information transmitted by the radar, the lateral velocity and radial velocity of the target at the target position can be determined, that is, vy2 and vx2 can be determined, so that the tangential direction angle of the target at the target position can be determined based on formula (11).

[0214] The first distance range is a pre-set distance range. The position of the target in the coordinate system can be determined using detection information from a radar or imaging device. Then, based on the position of the target in the coordinate system and the position of the first position in the coordinate system, the distance between the target position and the first position can be calculated. When the distance between the target position and the first position is within the first distance range, it indicates that the target position and the first position are relatively close.

[0215] Step S123: If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, determine that the road geometry is the target road geometry.

[0216] That is, when the difference between the tangent direction angle at the first position and the tangent direction angle at the target position satisfies the following formula, the road geometry can be determined to be the target road geometry:

[0217]

[0218] In the above formula, thresh represents the first threshold.

[0219] When the absolute value of the difference is less than the first threshold, it indicates that when the road geometry is at the first position, the difference between the tangent direction angle of the road geometry at the first position and the tangent direction angle at the position of the target is small, so that it can be determined that the deviation between the road geometry and the motion trajectory of the target is small, that is, the road geometry basically conforms to the motion trajectory of the target, so that the road constraint of the target can be determined based on the road geometry, and accordingly, the road geometry can be determined as the target road geometry.

[0220] In addition, when the absolute value of the difference is not less than the first threshold, it indicates that the deviation between the road geometry and the motion trajectory of the target is large, and the road geometry does not conform to the motion trajectory of the target. In this case, in order to avoid affecting the accuracy of determining the road constraints, the road constraints are usually not determined through the road geometry, that is, it is determined that the road geometry is not the target road geometry.

[0221] In the above description and Figure 4 In the embodiment of the present invention, the tangent direction angle of the road geometry at the first position is first determined, and then the tangent direction angle of the target at the target position is determined. In actual operation, there is no strict time limit for determining the two tangent direction angles. The tangent direction angle of the target at the target position can be determined first, and then the tangent direction angle of the road geometry at the first position is determined. Alternatively, the tangent direction angle of the road geometry at the first position and the tangent direction angle of the target at the target position can be determined simultaneously. This embodiment of the present application does not limit this.

[0222] In a second feasible implementation manner, for each road geometry in the at least one road geometry, at least one target road geometry in the at least one road geometry may be obtained by the following steps:

[0223] If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

[0224] The second distance range is a preset distance range.

[0225] In the above method, the distance between the target and the road geometry is first determined. When the distance between the target and the road geometry is within a second distance range, the road geometry can be determined to be the target road geometry.

[0226] When the target is located within the road geometry, the distance between the target and the road geometry is zero. The position of the target in the coordinate system can be determined using detection information from a radar or imaging device. When the position of the target in the coordinate system conforms to a model of the road geometry (e.g., equations (1) to (8)), the target is determined to be located within the road geometry.

[0227] In addition, when the target is not located in the road geometry, the distance between the target and the road geometry is the minimum distance between the target and the road geometry. Specifically, when determining the distance between the target and the road geometry, a corresponding road geometry diagram can be drawn according to the formula corresponding to the road geometry (e.g., formula (1) to formula (8)). The road geometry diagram is usually a curve used to represent the road edge, road guardrail or lane line. Then, the connecting line segments between the target location and each point in the curve can be obtained, where the length of the shortest connecting line segment is the distance between the target and the road geometry.

[0228] When the distance between the target and the road geometry is within a second distance range, it indicates that the target is located in the road geometry, or the target is closer to the road geometry. Therefore, the road geometry can be determined to be the target road geometry.

[0229] In a third possible implementation, for each road geometry in the at least one road geometry, determining at least one target road geometry in the at least one road geometry includes:

[0230] Obtaining a distance between the target and the road geometry;

[0231] According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

[0232] When the number of the at least one road geometry is 1, Num is 1. In addition, when the number of road geometries is greater than 1, Num can be determined as a positive integer not less than 2. For example, Num can be set to 2. In this case, the two road geometries closest to the target among the road geometries are determined as the target road geometries.

[0233] Alternatively, in another feasible implementation, a fixed value Num1 can be set. When the number of road geometries is greater than 1, Num is the difference between the number of road geometries and Num1. In this case, the larger the number of road geometries, the larger the number of target road geometries. This allows road constraints to be determined using a larger number of target road geometries when a larger number of road geometries are acquired. If the number of target road geometries is small, the presence of target road geometries with large errors will result in larger errors in the determined road constraints. Therefore, when determining road constraints using a larger number of target road geometries, the impact of target road geometries with large errors can be reduced, thereby improving the accuracy of determining road constraints.

[0234] Through the above method, the target road geometry can be obtained from the various road geometries, so that the target's road constraints can be subsequently determined based on the target road geometry. Since the deviation between the target road geometry and the target's motion trajectory is within a certain range, determining the target's road constraints based on the target road geometry can improve the accuracy of the determined road constraints.

[0235] In the embodiment of the present application, the road direction constraint can be determined in a variety of ways. In one of the feasible implementations, see Figure 5 As shown in the workflow diagram, determining the road direction constraint of the target based on the at least one target road geometry and the motion state of the target includes:

[0236] Step S131, determine at least one second position located in the at least one target road geometry, where the at least one second position is the position closest to the target in at least one first target road geometry, and the first target road geometry is the target road geometry where the second position is located.

[0237] The detection information transmitted by the radar or imaging device can determine the position coordinates (x2, y2) of the target in the coordinate system. Then, based on the position coordinates (x2, y2) of the target in the coordinate system and the target road geometry, the second position (x3, y3) closest to the target in the target road geometry can be determined.

[0238] Step S132: Determine a road direction constraint of the target according to the confidence level of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position.

[0239] Specifically, the tangent direction angle of the target road geometry at the second position (x3, y3) can be determined according to formula (13) or formula (14):

[0240]

[0241]

[0242] In formula (13) and formula (14), φ3 is the heading of the target road geometry at the second position (x3, y3); is the tangent direction angle of the target road geometry at the second position (x3, y3); φ0 is the heading of the target road geometry when x=0; C0 represents the average curvature of the target road geometry; C1 represents the average value of the rate of change of the curvature of the target road geometry.

[0243] Wherein, based on the detection information transmitted by the radar or imaging device, the tracking device can determine φ0, C0 and C1, and further determine the tangent direction angle of the target road geometry at the second position (x3, y3) through formulas (13) and (14).

[0244] After determining φ3 by formula (13) or formula (14), the tangent direction angle of the target on the target road geometry can be determined by the following formula:

[0245]

[0246] In addition, the confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry; or, the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

[0247] When the confidence level of the target road geometry is the confidence level of a road parameter of the target road geometry, the embodiment of the present application further includes the following steps: determining the confidence level of the road parameter of the target road geometry based on the variance or standard deviation of the road parameter of the target road geometry. The road parameter is at least one item of information used to characterize the target road geometry, that is, the road parameter can be at least one parameter selected from the group consisting of the orientation, curvature, curvature change rate, and length of the road.

[0248] When determining the confidence level of the road parameters of the target road geometry, the variance or standard deviation of the road parameters of each target road geometry can be obtained, where the variance or standard deviation is inversely proportional to the confidence level of the road parameters of the target road geometry, that is, the larger the variance or standard deviation, the smaller the confidence level of the road parameters of the target road geometry. Based on this, the confidence level of the road parameters of each target road geometry can be determined. Specifically, in an embodiment of the present application, a mapping relationship between the variance or standard deviation of the road parameters and the confidence level of the road parameters can be pre-set. After determining the variance or standard deviation of the road parameters, the confidence level of the road parameters of the target road geometry can be determined by querying the mapping relationship.

[0249] Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the method further includes:

[0250] The confidence level of the tangent direction angle of the target road geometry at the second position is determined according to the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

[0251] That is to say, when determining the tangent direction angle of each target road geometry at position (x3, y3) When the confidence level is , the tangent direction angle of each target road geometry at the second position (x3, y3) can be obtained. The variance or standard deviation of the target road geometry at position (x3, y3) is determined based on the variance or standard deviation. The confidence level of . Among them, the tangent direction angle The variance or standard deviation of the tangent angle The confidence is inversely proportional to the tangent direction angle The larger the variance or standard deviation is, the more accurate the tangent direction angle corresponding to the target road geometry is. The smaller the confidence level, the more the tangent direction angle corresponding to each target road geometry can be determined. Specifically, in an embodiment of the present application, a mapping relationship between the variance or standard deviation of the tangent direction angle and the confidence of the tangent direction angle at the second position can be pre-set. After determining the variance or standard deviation of the tangent direction angle, the confidence of the tangent direction angle of the target road geometry at the second position can be determined by querying the mapping relationship.

[0252] Determining the target road direction constraint based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position described in step S132 can be achieved in various ways.

[0253] In one embodiment, determining the target road direction constraint based on the confidence level of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0254] First, according to the confidence level of the target road geometry, a weight value of the tangent direction angle of the at least one target road geometry at the at least one second position during fusion is determined;

[0255] Then, according to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

[0256] In this method, generally, the higher the confidence level of a target road geometry is, the higher the weight value of the target road geometry is.

[0257] In the embodiment of the present application, it is necessary to determine the weight of the tangent direction angle of each target road geometry during fusion based on the confidence level of each target road geometry, so that the tangent direction angles of each target road geometry can be fused according to the weight. Generally, the higher the confidence level of a target road geometry, the higher the weight value of the tangent direction angle of the target road geometry during fusion.

[0258] Among them, the weight of the tangent direction angle of each target road geometry during fusion can be determined in a variety of ways. In a feasible implementation method, the correspondence between the confidence level and the weight value can be pre-set, and the weight value can be determined based on the correspondence.

[0259] Alternatively, in another feasible implementation, the weight value may be determined by formula (16) or formula (17):

[0260]

[0261]

[0262] In the above formula, w iThe weight value representing the tangent direction angle of the i-th target road geometry at the second position (x3, y3); is the tangent direction angle of the i-th target road geometry at the second position (x3, y3); is the confidence of the i-th target road geometry; n is the number of target road geometries; h(d i ) is the distance between the target and the i-th target road geometry. The distance between the target and the target road geometry refers to the shortest distance between the target and the line segments between each point included in the target road geometry.

[0263] In addition, according to the confidence level of each target road geometry, the tangent direction angle corresponding to each target road geometry is calculated. When performing fusion, the following formula can be used for fusion:

[0264]

[0265] in, is the tangent direction angle corresponding to the geometry of each target road The fusion result obtained after fusion; is the tangent direction angle of the i-th target road geometry at position (x1, y1), w i is the weight value of the i-th target road geometry; n is the number of target road geometries.

[0266] In the above steps, the tangent direction angles corresponding to the geometries of each target road are fused according to the confidence level of each target road geometry, and the fused result is used as the road direction constraint of the target.

[0267] In another feasible implementation, determining the road direction constraint of the target based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0268] The tangent angle direction at the second position is determined as the road direction constraint of the target.

[0269] The second position is the position of the second target road geometry closest to the target, and the second target road geometry is the target road geometry with the highest confidence among the at least one target road geometry. In addition, the confidence of the target road geometry is the confidence of the road parameters of the target road geometry, or the confidence of the target road geometry is the confidence of the tangent direction angle of the target road geometry at the second position.

[0270] In the above implementation, the road direction constraint of the target is determined by the tangent angle direction of the target road geometry with the highest confidence at the second position.

[0271] Alternatively, in another feasible implementation, determining the road direction constraint of the target based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes:

[0272] The tangent angle direction at the third position is determined as the road direction constraint of the target.

[0273] The third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

[0274] In the above implementation, the road direction constraint of the target is determined by the geometry of the target road that is closest to the target.

[0275] Furthermore, another method can be used to determine the target's road direction constraint. In this method, the at least one target road geometry is first fused to obtain a fused road geometry; then, the position in the fused road geometry that is closest to the target is determined; and the tangent direction angle at the position closest to the target is determined, and the tangent direction angle is used as the target's road direction constraint.

[0276] In the above method, each target road geometry needs to be fused to obtain a fused road geometry. When fusing the at least one target road geometry, a model for each target road geometry can be determined. Then, based on the confidence level of each target road geometry, the parameters in the model are fused to obtain fused parameters. These fused parameters are then substituted into the model to obtain a new model, which is the fused road geometry.

[0277] For example, when the geometric model of each target road is formula (1), the parameter y0 included in the geometric model of each target road is R,i 、φ0 Ri 、C0 Ri and C1 Ri After fusion, the fused parameters are substituted into formula (1), and the obtained new model can represent the fused road geometry.

[0278] Through the above scheme, road direction constraints can be obtained. Compared with the existing technology, the embodiment of the present application obtains road direction constraints based on road geometry, fully considering characteristics such as the curvature and heading of the road where the target is located. Therefore, the accuracy of the obtained road direction constraints is higher. Furthermore, target tracking can be performed using the road direction constraints obtained by the embodiment of the present application, which can further improve the accuracy of target tracking.

[0279] In the embodiments of the present application, the road width constraint can be represented in a variety of ways. For example, the width between two target road geometries closest to the target and on both sides of the target can be used as the target road width constraint; or the distance between the target and the at least one target road geometries can be used as the target road width constraint; or the maximum or minimum value of the distance between the target and the at least one target road geometries can be used as the target road width constraint; or the average value of the distance between the target and the at least one target road geometries can be used as the target road width constraint.

[0280] In this case, in a feasible manner, determining the road width constraint of the target according to the geometry of the at least one target road and the motion state of the target includes the following steps:

[0281] First, a straight line passing through the target position of the target and perpendicular to fourth target road geometries is obtained. The fourth target road geometries are two target road geometries closest to the target and located on both sides of the target respectively.

[0282] Then, a distance between two intersection points is determined as the road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

[0283] When the width between two target road geometries closest to the target and on both sides of the target is used as the road width constraint of the target, see Figure 6 In the scenario diagram shown, the target is located at (x2, y2) in the coordinate system. The intersection points of the two target roads closest to the target and perpendicular to the straight lines passing through this position are point A and point B. The distance between point A and point B is the road width. The straight lines perpendicular to the two target roads closest to the target can be expressed as:

[0284]

[0285] In formula (19), in, is the tangent direction angle of the two target road geometries closest to the target at (x2, y2), φ 1are the headings of the two target road geometries closest to the target at (x2, y2).

[0286] Based on formula (19) and the geometric relationship between the two target roads closest to the target, the distance between point A and point B can be determined, which is the road width.

[0287] In addition, in the embodiment of the present application, the distance between the target and each target road geometry can also be used as the target road width constraint. In this case, the target road width constraint is determined based on the at least one target road geometry and the motion state of the target, including:

[0288] At least one distance between the target and the at least one target road geometry is determined, the at least one distance being a road width constraint for the target.

[0289] Alternatively, the distance between the target and the at least one target road geometry is used as the road width constraint of the target. Accordingly, in this case, determining the road width constraint of the target based on the at least one target road geometry and the motion state of the target includes:

[0290] At least one distance between the target and the at least one target road geometry is determined, and a maximum or minimum value of the at least one distance is determined as a road width constraint for the target.

[0291] Alternatively, determining the road width constraint of the target based on the at least one target road geometry and the motion state of the target includes:

[0292] Determine a distance between the target and the at least one target road geometry, and determine an average of the at least one distance as a road width constraint for the target. In other words, the average of the at least one distance between the target and the at least one target road geometry is used as the road width constraint for the target.

[0293] In the above steps, the distance between the target and the target road geometry refers to the minimum distance between the target and the target road geometry. According to the formula corresponding to the target road geometry (e.g., formula (1) to formula (8)), a schematic diagram of the corresponding target road geometry is drawn. The target road geometry schematic diagram is usually a curve. The connecting line segments between the target position and each point on the curve are obtained. The length of the shortest connecting line segment is the distance between the target and the target road geometry.

[0294] Through the above steps, the road width constraint can be obtained. Compared with the existing technology, the embodiment of the present application obtains the road width constraint based on the road geometry, and fully considers the characteristics such as the curvature and heading of the road where the target is located. Therefore, the obtained road width constraint is more accurate. Furthermore, by performing target tracking based on the road width constraint obtained by the embodiment of the present application, the accuracy of target tracking can be further improved.

[0295] Furthermore, in the embodiment of the present application, target tracking can also be performed based on road direction constraints and road width constraints. In this case, see Figure 7 As shown in the workflow diagram, the embodiment of the present application also includes the following steps:

[0296] Step S14: Determine a measurement matrix including the road direction constraint, and determine the confidence of the road direction constraint in the measurement matrix through the road width constraint.

[0297] After obtaining the road direction and width constraints, the prior art determines a measurement matrix based on the road direction and width constraints. This measurement matrix is then used to estimate the target's motion state, completing target tracking. For example, this measurement matrix can be substituted into the measurement equation of a Kalman filter algorithm, and the target's motion state can be estimated based on this measurement equation. Of course, this measurement equation can also be substituted into other tracking algorithms, and this is not limited to this in the present embodiment.

[0298] However, through the solution of the embodiment of the present application, after obtaining at least one of the road direction constraint and the road width constraint, a measurement equation can be obtained based on the obtained road direction constraint and / or road width constraint, and the measurement matrix determined by the embodiment of the present application can be used to replace the measurement matrix in the prior art, that is, the measurement matrix determined by the embodiment of the present application is substituted into the tracking algorithm to achieve target tracking.

[0299] In this case, since at least one of the road direction constraint and the road width constraint can be obtained by the embodiment of the present application, and the accuracy of the road direction constraint and / or road width constraint obtained by the solution of the embodiment of the present application is higher. Therefore, the accuracy of the measurement matrix obtained by the embodiment of the present application is higher. Accordingly, compared with the prior art, when the solution of the embodiment of the present application is used to track a target, the accuracy is higher.

[0300] When the tracking algorithm used is the Kalman filter algorithm, in a feasible implementation, the measurement matrix can be represented by the following matrix:

[0301]

[0302] In matrix (1), β=atan2(y,x); φ is the road direction constraint determined in step S13. (x, y) represents the target's position in the coordinate system; vx represents the target's x-axis velocity, and vy represents the target's y-axis velocity. When vx and vy are measured by radar, they refer to the target's velocity relative to the radar. For example, if the target's actual x-axis velocity is 3 m / s, the radar and target's motion directions coincide, and the radar's actual velocity is 1 m / s, then vx is 2 m / s.

[0303] In addition, after determining the measurement matrix, it is also necessary to determine the confidence of the road direction constraint in the measurement matrix through the road width constraint. The confidence of the road direction constraint in the measurement matrix is usually related to the measurement noise n v Correlation, measurement noise n v The larger the value, the smaller the confidence of the road direction constraint in the measurement matrix. v It is usually proportional to the road width constraint. The larger the road width constraint, the larger the measurement noise n v That is, the larger the road width constraint is, the smaller the confidence of the road direction constraint in the measurement matrix is.

[0304] In a feasible implementation, the road width constraint and the measurement noise n can be set. v The mapping relationship between them, and the measurement noise n v The mapping relationship between the road width constraint and the confidence level of the road direction constraint in the measurement matrix is established, thereby enabling the confidence level of the road direction constraint in the measurement matrix to be determined using the road width constraint and the two mapping relationships described above. That is, first, the measurement noise corresponding to the target is determined based on the mapping relationship between the road width constraint and the measurement noise, as well as the road width constraint. Then, the confidence level of the road direction constraint in the measurement matrix is determined based on the mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix, as well as the measurement noise corresponding to the target.

[0305] Alternatively, a mapping relationship between the confidence of the road width constraint and the road direction constraint in the measurement matrix may be directly set. In this case, the confidence of the road direction constraint in the measurement matrix may be determined by the road width constraint and the mapping relationship.

[0306] Furthermore, after determining the confidence level of the road direction constraint in the measurement matrix, the target can be tracked through the measurement matrix and the tracking algorithm.

[0307] In the solution disclosed in the embodiments of this application, the target can operate in two-dimensional space or three-dimensional space. When the target moves in two-dimensional space, the target's road direction constraint and road width constraint can be determined using the above formula and algorithm. Alternatively, when the target moves in three-dimensional space, the target's height can be ignored when determining the target's road direction constraint and road width constraint. The target's road direction constraint and road width constraint can still be determined using the above formula and algorithm. In this case, the various parameters used in the calculation process are parameters of the plane in which the target is located.

[0308] In the above embodiment, a method for determining road constraints based on the geometry of the road where the target is located is disclosed. In addition, in the actual motion scene of the target, the target's motion state is often changeable, for example, sometimes there will be a situation such as changing lanes. When the target changes lanes, there may be a large deviation between the actual motion direction of the target and the tangent direction angle of the road geometry. Therefore, when the target changes lanes, it is necessary to further adjust the measurement noise n v That is, in the embodiment of the present application, it is also necessary to determine whether the target changes lanes. When the target changes lanes, it is necessary to further adjust the measurement noise n v , and by adjusting the measurement noise n v , determine the confidence of the road direction constraint in the measurement matrix.

[0309] Accordingly, the present application discloses another embodiment. In this embodiment, after determining the measurement noise corresponding to the target based on the mapping relationship between the road width constraint and the measurement noise and the road width constraint, the following steps are further included:

[0310] First, according to the motion state of the target, a motion state change parameter of the target is determined;

[0311] Then, when the comparison result of the motion state change parameter of the target and the corresponding threshold indicates that it is necessary to determine the degree of change of the curvature or the degree of change of the curvature change rate of the target road geometry at a fourth position, the degree of change of the curvature or the degree of change of the curvature change rate of the target road geometry at the fourth position is determined, and the fourth position is a position located on the target road geometry and the distance from the target is within a third distance range, wherein, when the degree of change of the curvature or the degree of change of the curvature change rate is greater than the third threshold, the measurement noise corresponding to the target is increased.

[0312] When the comparison result of the motion state change parameter of the target and the corresponding threshold value indicates that it is necessary to determine the degree of change of the curvature or the degree of change of the curvature change rate of the target road geometry at the fourth position, wherein the fourth position is a position located in the target road geometry and the distance from the target is within a third distance range. The third distance range is a pre-set or pre-defined range, and the third distance range may be the same as the first distance range, or may be different from the first distance range. This embodiment of the present application does not limit this. Since the distance between the fourth position and the target is within the third distance range, it indicates that the fourth position is closer to the target.

[0313] In the embodiment of the present application, whether the target may change lanes is determined based on the comparison result of the target's motion state change parameter and the corresponding threshold value. When the comparison result indicates that the target may change lanes, the judgment is further made based on the degree of change of the curvature of the target road geometry at the fourth position or the degree of change of the curvature change rate. In this case, when the degree of change of the curvature or the degree of change of the curvature change rate is greater than the third threshold value, it indicates that the target has changed lanes. In this case, it is necessary to increase the measurement noise n v .

[0314] In addition, when the comparison result of the target's motion state change parameter and the corresponding threshold indicates that there is no need to determine the degree of change of the curvature or the degree of change of the curvature change rate of the target road geometry at the fourth position, it indicates that the target has not changed lanes.

[0315] Accordingly, in the embodiment of the present application, determining the confidence of the road direction constraint in the measurement matrix based on the mapping relationship between the measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes:

[0316] The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

[0317] In an embodiment of the present application, the target's motion state is determined based on its motion state, and then the need to increase the measurement noise is determined based on the target's motion state change parameters. In other words, it is possible to determine whether the target has changed lanes, and to increase the measurement noise if a lane change has occurred. Through the scheme of the embodiment of the present application, it is possible to consider whether a lane change will occur during the target's motion, thereby determining the confidence level of the road direction constraint in the measurement matrix based on the target's various motion states. Accordingly, the accuracy of determining the confidence level of the road direction constraint in the measurement matrix can be improved, further improving the accuracy of target tracking.

[0318] Among them, the motion state change parameter can be a parameter in various forms. For example, the motion state includes the average value of the normalized innovation squared (NIS) parameter corresponding to the geometry of at least one target road, or the motion state change parameter includes: the curvature of the historical motion trajectory of the target or the degree of change of the curvature.

[0319] In one possible implementation of the embodiment of the present application, the average value of the NIS parameters corresponding to at least one target road geometry may be used as a motion state change parameter. When the motion state change parameter of the target is the average value of the NIS parameters and the motion state change parameter of the target is greater than a corresponding threshold, it indicates that the target may change lanes. It is necessary to determine the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at a fourth position. In order to further determine whether the target has changed lanes based on the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position.

[0320] The NIS parameter represents the degree of match between the target's motion state, as derived from the Kalman filter algorithm, and its actual motion state. A larger NIS parameter indicates a smaller match between the target's motion state and its actual motion state. When a target changes lanes, its NIS parameter often changes dramatically. Therefore, the average NIS parameter corresponding to each target's road geometry can be used as the motion state change parameter.

[0321] Among them, the NIS parameter corresponding to a target road geometry can be calculated by the following formula:

[0322]

[0323] Where NIS(i) represents the NIS parameter value of the target corresponding to the i-th target road geometry; y represents the y-axis coordinate value of the i-th target road geometry at the fourth position; T represents the transposition operation, It represents the y-axis coordinate value of the i-th target road geometry at the fourth position obtained by estimating the motion state of the target through the Kalman filter algorithm; S represents the new information covariance matrix obtained by the Kalman filter algorithm.

[0324] Then, the average value of the NIS parameters of each target road geometry is obtained based on the following formula:

[0325]

[0326] Where NIS-average represents the average value of the NIS parameters of each target road geometry; n represents the number of target road geometries.

[0327] When NIS-average is greater than the corresponding threshold, it indicates that the target's motion state has changed significantly and the target may change lanes.

[0328] In another feasible implementation, the curvature of the target's historical motion trajectory or the degree of change of the curvature is used as a motion state change parameter.

[0329] The curvature or curvature change rate of the target's historical motion trajectory can be obtained by the following formula:

[0330]

[0331]

[0332] In the above formula, ω is the curvature of the target's historical motion trajectory or the degree of change of the curvature change rate, c1(r) is the curvature of the target's motion trajectory at the current moment or the curvature change rate; is the average value of the curvature or curvature change rate within the sliding window; s is the number of curvatures or curvature change rates within the sliding window, and the value of s depends on the size of the sliding window.

[0333] When calculating the degree of change of the curvature or curvature change rate of the historical motion trajectory of the target, the calculation is performed using a sliding window, wherein the sliding window includes s curvatures or curvature change rates, and when the curvatures or curvature change rates included in the sliding window are sorted in chronological order, the time difference between two adjacent curvatures or curvature change rates is within a first time period, and the time difference between the acquisition time of the most recently acquired curvature or curvature change rate in the sliding window and the current moment is within a second time period.

[0334] When the motion state change parameter is the curvature of the target's historical motion trajectory or the degree of change of the curvature, the corresponding threshold is usually the product of the curvature or curvature change rate of the target's motion trajectory at the current moment and a positive number, and when the motion state change parameter is less than the threshold, it indicates that the target may change lanes, and it is necessary to further determine the degree of change of the curvature of the target road geometry at the target position or the degree of change of the curvature change rate.

[0335] That is, when the target's motion state change parameter is the curvature of the target's historical motion trajectory or the degree of change in curvature, if c1(r) > m*ω, it is determined that the target's motion state has changed significantly and the target is likely to change lanes. Here, m is a preset positive number, for example, m can be set to 3. Of course, m can also be set to other values, and this embodiment of the application is not limited to this.

[0336] When the average value of the NIS parameters corresponding to the geometries of each target road is greater than the corresponding threshold, or when the curvature of the target's historical motion trajectory or the degree of change in curvature is less than the product of the curvature or curvature change rate of the target's motion trajectory at the current moment and a preset positive number, it indicates that the target's motion state has changed significantly. In this case, it is necessary to determine whether the target has changed lanes by the degree of change in the curvature or curvature change rate of the target road geometry at the target position.

[0337] If the curvature or curvature change rate of the target road geometry at the target location does not change, or the change is minimal, then the target's motion state change is inconsistent with the target road geometry. In this case, the target can be determined to have changed lanes. Alternatively, if the curvature or curvature change rate of the target road geometry at the target location changes significantly, then the target's motion state change is consistent with the target road geometry. Therefore, the target can be determined to have not changed lanes.

[0338] In the embodiment of the present application, when it is determined that the target changes lanes, the measurement noise n is further increased. v The increased noise amount can be a preset fixed value. In this case, the increased measurement noise n v is the sum of the measured noise obtained by the road width constraint and the noise amount. Alternatively, the measured noise n corresponding to the lane change of the target can be preset. v , the measurement noise n v is the larger measurement noise n v Then, according to the measured noise n v Determines the confidence level of the road direction constraints in the measurement matrix.

[0339] The following are embodiments of the apparatus of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the apparatus embodiments of the present invention, please refer to the method embodiments of the present invention.

[0340] In another embodiment of the present application, a road constraint determination device is disclosed, wherein the road constraint determination device includes: at least one processing module.

[0341] The at least one processing module is configured to determine the motion state of the target based on the detection information of the target;

[0342] determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, each road geometry in the at least one road geometry being represented by at least one item of information;

[0343] A road constraint of the target is determined based on the at least one road geometry and a motion state of the target, the road constraint comprising at least one of a road direction constraint and a road width constraint.

[0344] The at least one processing module included in the road constraint determination apparatus disclosed in the embodiments of this application is capable of executing the road constraint determination methods disclosed in the aforementioned embodiments of this application. By executing the road constraint determination methods disclosed in the aforementioned embodiments of this application, the at least one processing module is capable of determining the road constraints of a target, and the road constraints determined by the at least one processing module are more accurate than those in the prior art.

[0345] Furthermore, since the accuracy of the road constraint determined by the at least one processing module is relatively high, when target tracking is performed using the road constraint determined by the at least one processing module, the accuracy of target tracking can also be improved.

[0346] In addition, the at least one processing module can be logically divided into at least one module from a functional perspective. In an example of division, see Figure 8 As shown in the structural diagram, the at least one processing module can be divided into a motion state determination module 110, a road geometry determination module 120, and a road constraint determination module 130. It should be noted that the logical division here is merely an exemplary description to illustrate at least one function that the at least one processing module is configured to perform.

[0347] In this case, the motion state determination module 110 is used to determine the motion state of the target based on the detection information of the target;

[0348] The road geometry determination module 120 is configured to determine at least one road geometry of the road where the target is located based on the detection information of the target, wherein each road geometry of the at least one road geometry is represented by at least one piece of information;

[0349] The road constraint determination module 130 is configured to determine a road constraint of the target based on the at least one road geometry and the motion state of the target, wherein the road constraint includes at least one of a road direction constraint and a road width constraint.

[0350] Of course, the at least one processing module may also be divided in other ways, which is not limited in the embodiment of the present application.

[0351] Furthermore, in the road constraint determination device disclosed in an embodiment of the present application, the at least one processing module is also used to determine at least one target road geometry among the at least one road geometry.

[0352] In this case, the at least one processing module is specifically configured to determine a road constraint of the target based on the at least one target road geometry and a motion state of the target.

[0353] Furthermore, in the road constraint determination device disclosed in the embodiment of the present application, the at least one processing module is specifically configured to, for each road geometry in the at least one road geometry:

[0354] Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position;

[0355] acquiring a tangential direction angle of the target at the target position according to a lateral velocity and a radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range;

[0356] If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, the road geometry is determined to be the target road geometry.

[0357] In the above solution, it is possible to determine whether a certain road geometry is a target road geometry based on the tangent direction angle of the road geometry at the first position.

[0358] Furthermore, in the road constraint determination device disclosed in the embodiment of the present application, the at least one processing module is specifically configured to, for each road geometry in the at least one road geometry:

[0359] If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

[0360] In the above solution, it is possible to determine whether a certain road geometry is the target road geometry based on the distance between the road geometry and the target.

[0361] Furthermore, in the road constraint determination device disclosed in the embodiment of the present application, the at least one processing module is specifically configured to, for each road geometry in the at least one road geometry:

[0362] Obtaining a distance between the target and the road geometry;

[0363] According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

[0364] In the above solution, the target road geometry in the road geometry can be determined according to the distance between the target and the road geometry and the number of at least one road geometry.

[0365] Furthermore, in the road constraint determination device disclosed in the embodiments of the present application, a road direction constraint of a target can be determined in a variety of ways. In one feasible manner, the at least one processing module is specifically configured to determine at least one second position located in the at least one target road geometry, the at least one second position being the position closest to the target in at least one first target road geometry, the first target road geometry being the target road geometry where the second position is located;

[0366] A road direction constraint of the target is determined based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second position.

[0367] The confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry;

[0368] Alternatively, the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

[0369] When the confidence level of the target road geometry is a confidence level of a road parameter of the target road geometry, the processor is further configured to determine the confidence level of the road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, where the road parameter is at least one item of information used to characterize the target road geometry;

[0370] Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the processor is further used to determine the confidence level of the tangent direction angle of the target road geometry at the second position based on the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

[0371] In another method of determining the road direction constraint of the target, the at least one processing module is specifically configured to determine, based on the confidence level of the target road geometry, a weight value of a tangent direction angle of the at least one target road geometry at the at least one second position during fusion;

[0372] According to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

[0373] In another method of determining the road direction constraint of the target, the at least one processing module is specifically used to determine the tangent angle direction at a second position as the road direction constraint of the target, wherein the second position is the position closest to the target in the second target road geometry, and the second target road geometry is the target road geometry with the highest confidence among the at least one target road geometry.

[0374] In another method of determining the road direction constraint of the target, the at least one processing module is specifically used to determine the tangent angle direction at a third position as the road direction constraint of the target, wherein the third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

[0375] Furthermore, in the road constraint determination device disclosed in the embodiments of the present application, a road direction constraint of a target can be determined in a variety of ways. In one feasible manner, the at least one processing module is specifically configured to obtain a straight line passing through the target position of the target and perpendicular to a fourth target road geometry, wherein the fourth target road geometry is two target road geometries closest to the target and located on either side of the target;

[0376] A distance between two intersection points is determined as a road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

[0377] In another method of determining the road width constraint of the target, the at least one processing module is specifically used to determine at least one distance between the target and the at least one target road geometry, and the at least one distance is the road width constraint of the target; or, the processor is specifically used to determine at least one distance between the target and the at least one target road geometry, and determine the maximum value or minimum value of the at least one distance as the road width constraint of the target; or, the processor is specifically used to determine the distance between the target and the at least one target road geometry, and determine the average value of the at least one distance as the road width constraint of the target.

[0378] Furthermore, in an embodiment of the present application, the at least one processing module is further configured to determine a measurement matrix including the road direction constraint;

[0379] The confidence level of the road direction constraint in the measurement matrix is determined by the road width constraint.

[0380] The at least one processing module is specifically configured to determine the measurement noise corresponding to the target according to a mapping relationship between a road width constraint and measurement noise, and the road width constraint;

[0381] The confidence of the road direction constraint in the measurement matrix is determined according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target.

[0382] In addition, the at least one processing module is further configured to, after determining the measurement noise corresponding to the target based on the mapping relationship between the road width constraint and the measurement noise and the road width constraint, determine a motion state change parameter of the target based on the motion state of the target;

[0383] When a comparison result of the motion state change parameter of the target and a corresponding threshold value indicates that it is necessary to determine a degree of change in the curvature or a degree of change in the curvature change rate of the target road geometry at a fourth position, determining the degree of change in the curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position, the fourth position being a position located on the target road geometry and within a third distance range from the target;

[0384] When the degree of change of the curvature or the degree of change of the curvature change rate is greater than a third threshold, increasing the measurement noise corresponding to the target;

[0385] The determining the confidence of the road direction constraint in the measurement matrix according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes:

[0386] The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

[0387] The device disclosed in the embodiments of the present application can improve the accuracy of determining road constraints and further improve the accuracy of target tracking.

[0388] Corresponding to the above-mentioned road constraint determination method, in another embodiment of the present application, a road constraint determination device is also disclosed. Figure 9 As shown in the structural diagram, the road constraint determination device includes:

[0389] at least one processor 1101 and memory,

[0390] Wherein, the at least one memory is used to store program instructions;

[0391] The processor is configured to call and execute program instructions stored in the memory so as to enable the road constraint determination device to perform Figure 2 、 Figures 4 and 5 as well as Figure 7 All or part of the steps in the corresponding embodiments.

[0392] Furthermore, the device may further include: a transceiver 1102 and a bus 1103 , and the memory includes a random access memory 1104 and a read-only memory 1105 .

[0393] The processor is coupled to the transceiver, random access memory, and read-only memory through a bus. When the mobile terminal control device needs to be run, it is started by the basic input and output system fixed in the read-only memory or the bootloader in the embedded system to guide the device into normal operation. After the device enters the normal operation state, the application program and the operating system are run in the random access memory, so that the mobile terminal control device can execute Figure 2 、 Figures 4 and 5 as well as Figure 7 All or part of the steps in the corresponding embodiments.

[0394] The device of the embodiment of the present invention may correspond to the above Figure 2 、 Figures 4 and 5 as well as Figure 7 The road constraint determination device in the corresponding embodiment, and the processor in the road constraint determination device can realize Figure 2 、 Figures 4 and 5 as well as Figure 7 For the sake of brevity, the functions and / or various steps and methods implemented by the road constraint determination device in the corresponding embodiment are not repeated here.

[0395] It should be noted that this embodiment can also be based on a general physical server combined with a network device implemented by network function virtualization (English: Network Function Virtualization, NFV) technology, and the network device is a virtual network device (such as a virtual host, a virtual router or a virtual switch). The virtual network device can be a virtual machine (English: Virtual Machine, VM) running a program for sending notification message functions, and the virtual machine is deployed on a hardware device (for example, a physical server). A virtual machine refers to a complete computer system with complete hardware system functions simulated by software and running in a completely isolated environment. Those skilled in the art can virtualize multiple network devices with the above functions on a general physical server by reading this application. I will not go into details here.

[0396] Furthermore, the road constraint determination device disclosed in the embodiment of the present application can be applied to a tracking device. The tracking device needs to apply detection information when determining road constraints. The detection information can be obtained through a sensor. The sensor generally includes a radar and / or an imaging device. The sensor can be connected to the road constraint determination device in the tracking device and transmit detection information to the road constraint determination device so that the road constraint determination device determines the road constraint based on the received detection information and the method disclosed in the above embodiment of the present application. In addition, the sensor can be set in the tracking device, or the sensor can also be a device independent of the tracking device.

[0397] The tracking device using the road constraint determination device can be implemented in various forms. In one form, see Figure 10 As shown in the structural diagram, in this form, the road constraint determination device 210 disclosed in the embodiment of the present application is integrated into a fusion module 220. The fusion module 220 can be a software functional module and is carried by a chip or integrated circuit. Alternatively, the road constraint determination device can be a chip or integrated circuit.

[0398] The fusion module 220 is capable of connecting to at least one sensor 230 and acquiring detection information transmitted by the at least one sensor 230. The fusion module 220 can implement various fusion functions, such as fusing the detection information transmitted by the at least one sensor 230, and transmitting the fused detection information to the road constraint determination device 210, so that the road constraint determination device 210 can determine road constraints based on the fused detection information.

[0399] Exemplarily, the fusion processing performed by the fusion module 220 on the probe information may include screening and fusing the probe information. Screening the probe information involves deleting probe information with significant errors and determining road constraints based on the remaining probe information. For example, if the probe information includes the curvature of a road section, and a few of the curvatures differ significantly from the others, these few curvatures may be considered to have significant errors, and the fusion module 220 will therefore delete these few curvatures. In this case, the road constraint determination device 210 determines the road constraints based on the remaining curvatures, thereby improving the accuracy of determining the road constraints.

[0400] In addition, the fusion of detection information can be to determine multiple detection information of the same type at the same location, and obtain the fusion result of the multiple detection information of the same type, so that the road constraint determination device 210 determines the road constraint through the fusion result to improve the accuracy of determining the road constraint. For example, the fusion module 220 can be connected to multiple sensors and obtain the heading angles at the same location detected by the multiple sensors, that is, multiple heading angles at the same location can be obtained. In this case, the fusion module 220 can fuse the multiple heading angles through a fusion algorithm (for example, calculating the average value of the multiple heading angles), and the fusion result is the heading angle at the location. In this case, the road constraint determination device 210 determines the road constraint through the fusion result of the detection information corresponding to multiple sensors, thereby improving the accuracy of determining the road constraint.

[0401] In this form, the chip or integrated circuit carrying the fusion module 220 can serve as the tracking device, while the at least one sensor 230 is independent of the tracking device and can transmit detection information to the fusion module 220 via wired or wireless means. Alternatively, the at least one sensor 230 and the fusion module 220 together constitute the tracking device.

[0402] Alternatively, in another embodiment, the road constraint determination device disclosed in the embodiments of the present application is connected to a fusion module, which is connected to at least one sensor. After receiving detection information transmitted by the at least one sensor, the fusion module performs fusion processing on the received detection information and transmits the fusion result to the road constraint determination device, which then determines the road constraint.

[0403] In this case, the road constraint determination device and the fusion module can be implemented on the same chip or integrated circuit, or on different chips or integrated circuits, and this embodiment of the application does not limit this. It can also be understood that the road constraint determination device and the fusion module can be integrated or independently implemented.

[0404] In addition, in this form, the road constraint determination device and the fusion module are each part of the tracking device.

[0405] In another form, see Figure 11As shown in the structural diagram, in this embodiment, the road constraint determination device 310 disclosed in this application embodiment is built into a sensor 320, and is carried by a chip or integrated circuit within the sensor 320. After receiving detection information, the sensor 320 transmits the detection information to the road constraint determination device 310, which then determines the road constraint based on the detection information. Alternatively, the road constraint determination device can be a chip or integrated circuit within the sensor.

[0406] For example, when the sensor 320 is an imaging device, the imaging device can transmit the captured image information to the road constraint determination device 310, or the imaging device can process the image information after completing the shooting, determine the lane line model and / or the motion state of the target corresponding to the image information, and then transmit the lane line model and / or the motion state of the target to the road constraint determination device 310, so that the road constraint determination device 310 can determine the road constraint according to the scheme disclosed in each embodiment of the present application.

[0407] In addition, in this form, when road constraints are determined based on detection information from multiple sensors, other sensors can be connected to the sensor 320 in which the road constraint determination device 310 is built, and the other sensors can transmit the acquired detection information to the road constraint determination device 310, so that the road constraint determination device 310 applies the detection information transmitted by the other sensors to determine the road constraints.

[0408] In this form, the sensor 320 built into the road constraint determination apparatus 310 may serve as a tracking device.

[0409] In another form, see Figure 12 As shown in the structural schematic diagram, in this form, the road constraint determination device disclosed in the embodiment of the present application includes a first road constraint determination device 410 and a second road constraint determination device 420, wherein the first road constraint determination device 410 can be set in the sensor 430, and the second road constraint determination device 420 can be set in the fusion module 440.

[0410] In this case, the first road constraint determination device 410 can execute some steps of the road constraint determination method disclosed in each embodiment of the present application based on the detection information of the sensor 430, and transmit the determined result information to the second road constraint determination device 420, and the second road constraint determination device 420 determines the road constraint based on the result information.

[0411] In this form, the sensor 430 with the first road constraint determination device 410 built in, and the fusion module 440 with the second road constraint determination device 420 built in, together constitute a tracking device.

[0412] In another form, see Figure 13 As shown in the structural schematic diagram, in this form, the road constraint determination device 510 disclosed in the embodiment of the present application is independent of at least one sensor 520, and the road constraint determination device 510 is carried by a chip or an integrated circuit.

[0413] In this case, at least one sensor 520 may transmit detection information to the road constraint determination device 510 , and the road constraint determination device 510 determines the road constraint according to the solutions disclosed in various embodiments of the present application.

[0414] In this form, the chip or integrated circuit carrying the road constraint determination device 510 can serve as the tracking device, while the at least one sensor 520 is independent of the tracking device and can transmit detection information to the road constraint determination device 510 via wired or wireless means. Alternatively, the at least one sensor 520 and the road constraint determination device 510 together constitute the tracking device.

[0415] Of course, the road constraint determination device can also be implemented in other forms, and the embodiments of the present application do not limit this.

[0416] Furthermore, the road constraint determination device disclosed in the embodiments of the present application can be applied to the field of intelligent driving, and in particular, can be applied to advanced driver assistance systems (ADAS) or autonomous driving systems. For example, the road constraint determination device can be installed in a vehicle that supports advanced driver assistance functions or autonomous driving functions, and determine detection information based on sensors in the vehicle (such as radar and / or cameras) to determine road constraints based on the detection information, thereby realizing advanced driver assistance functions or autonomous driving functions.

[0417] In this case, the solution of the embodiments of the present application can improve the capabilities of autonomous driving or ADAS, and can therefore be applied to the Internet of Vehicles, for example, in systems such as vehicle-to-everything (V2X), long-term evolution technology-vehicle communication (LTE-V), and vehicle-to-vehicle (V2V).

[0418] Furthermore, the road constraint determination device disclosed in the embodiments of this application can also be placed at a specific location to track targets within the detection neighborhood of that location. For example, the road constraint determination device can be placed at an intersection and, based on the solutions provided in the embodiments of this application, determine the road constraints corresponding to targets in the area surrounding the intersection, thereby tracking the targets and achieving intersection detection.

[0419] In a specific implementation, the embodiment of the present application further provides a computer-readable storage medium, which includes instructions. The computer-readable storage medium set in any device can implement the following when it is run on a computer: Figure 2 、 Figures 4 and 5 as well as Figure 7 The computer-readable storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0420] In addition, another embodiment of the present application further discloses a computer program product comprising instructions, which, when the computer program product is run on an electronic device, enables the electronic device to implement the following steps: Figure 2 、 Figures 4 and 5 as well as Figure 7 All or part of the steps in the corresponding embodiments.

[0421] Furthermore, an embodiment of the present application also discloses a vehicle, which includes the road constraint determination device disclosed in the aforementioned embodiment of the present application.

[0422] In the vehicle disclosed in the embodiments of this application, the road constraint determination device includes at least one processor and memory. In this case, the road constraint determination device is typically implemented by a chip and / or integrated circuit built into the vehicle. The at least one processor and memory may be implemented by different chips and / or integrated circuits, or may be implemented by a single chip or integrated circuit.

[0423] Alternatively, it is understood that the road restraint device may also be a chip and / or an integrated circuit, wherein the chip is one chip or a collection of multiple chips, and the integrated circuit is one integrated circuit or a collection of multiple integrated circuits. For example, in one example, the road restraint device includes multiple chips, one of which serves as a memory in the road restraint device, and each of the other chips serves as a processor in the road restraint device.

[0424] Furthermore, the vehicle may be equipped with at least one built-in sensor to acquire the detection information required for determining road constraints. The sensor may include an onboard camera and / or onboard radar. Alternatively, the vehicle may be wirelessly connected to a remote sensor to utilize the detection information required for determining road constraints.

[0425] Furthermore, the vehicle may also be provided with a fusion module, and the road constraint determination device may be provided within the fusion module, or the road constraint determination device may be connected to the fusion module. The fusion module is connected to the sensors and performs fusion processing on the detection information transmitted by the sensors. The fusion processing result is then transmitted to the road constraint determination device, and the road constraint determination device determines the road constraint based on the fusion processing result.

[0426] Since the road constraint determination device disclosed in the aforementioned embodiments of the present application can improve the accuracy of determining road constraints, accordingly, the vehicle disclosed in the embodiments of the present application can enhance autonomous driving or ADAS capabilities.

[0427] The present application also discloses a system capable of determining road constraints using the methods disclosed in the aforementioned embodiments of the present application. The system includes a road constraint determination device and at least one sensor. The at least one sensor includes a radar and / or an imaging device. The at least one sensor is configured to acquire detection information of a target and transmit the detection information to the road constraint determination device. The road constraint determination device determines road constraints based on the detection information.

[0428] Furthermore, the system may also include a fusion module. The road constraint determination device may be disposed within the fusion module, or the road constraint determination device may be connected to the fusion module. The fusion module is connected to the sensors and performs fusion processing on the detection information transmitted by the sensors. The fusion processing results are then transmitted to the road constraint determination device, which then determines the road constraint based on the fusion processing results.

[0429] The various illustrative logic units and circuits described in the embodiments of the present application can be implemented or operated by a design of a general-purpose processor, a digital information processor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor can be a microprocessor, alternatively, the general-purpose processor can also be any traditional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital information processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital information processor core, or any other similar configuration to implement.

[0430] The steps of the methods or algorithms described in the embodiments of the present application can be directly embedded in hardware, software units executed by a processor, or a combination of the two. The software units can be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. For example, the storage medium can be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Alternatively, the storage medium can be integrated into the processor. The processor and storage medium can be provided in an ASIC, which can be provided in a UE. Alternatively, the processor and storage medium can be provided in different components within the UE.

[0431] It should be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0432] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0433] The same or similar parts between the various embodiments of this specification can be referred to in conjunction with each other, and each embodiment focuses on the differences between the other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0434] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus a necessary general-purpose hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention or certain portions of the embodiments.

[0435] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the embodiment of the road constraint determination device disclosed in this application, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0436] The above-described embodiments of the present invention do not limit the protection scope of the present invention.

Claims

1. A method for determining road constraints, characterized in that: include: determining a motion state of the target based on detection information obtained from detecting the target; determining, based on the detection information of the target, at least one road geometry of a road on which the target is located, each road geometry in the at least one road geometry being represented by at least one item of information; determining at least one target road geometry among the at least one road geometry; determining a road constraint for the target based on the at least one target road geometry and a motion state of the target, the road constraint comprising at least one of a road direction constraint and a road width constraint; The method further comprises: Determine a measurement matrix including the road direction constraint; and determine the confidence of the road direction constraint in the measurement matrix through the road width constraint, wherein the measurement matrix is used to estimate the motion state of the target to complete target tracking; the road direction constraint in the measurement matrix is determined based on the at least one target road geometry, and the deviation between the at least one target road geometry and the motion trajectory of the target is within a predetermined range; the confidence of the road direction constraint in the measurement matrix is negatively correlated with the road width constraint.

2. The method according to claim 1, characterized in that The determining of at least one target road geometry among the at least one road geometry comprises: For each road geometry of the at least one road geometry: Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position; acquiring a tangential direction angle of the target at the target position according to a lateral velocity and a radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range; If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, the road geometry is determined to be the target road geometry.

3. The method according to claim 1, characterized in that The determining of at least one target road geometry among the at least one road geometry comprises: For each road geometry of the at least one road geometry: If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

4. The method according to claim 1, wherein The determining of at least one target road geometry among the at least one road geometry comprises: For each road geometry of the at least one road geometry: Obtaining a distance between the target and the road geometry; According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

5. The method according to claim 1, wherein The determining, based on the at least one target road geometry and the motion state of the target, a road direction constraint of the target, comprises: determining at least one second position respectively located in the at least one target road geometry, the at least one second position being a position closest to the target in at least one first target road geometry, the first target road geometry being the target road geometry where the second position is located; A road direction constraint of the target is determined based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second position.

6. The method according to claim 5, characterized in that The confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry; or, The confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

7. The method according to claim 6, characterized in that When the confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry, the method further includes: determining a confidence level of a road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, wherein the road parameter is at least one item of information used to characterize the target road geometry; Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the method further includes: The confidence level of the tangent direction angle of the target road geometry at the second position is determined according to the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

8. The method according to claim 5, characterized in that The determining of the road direction constraint of the target based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes: determining, according to the confidence level of the target road geometry, a weight value of a tangent direction angle of the at least one target road geometry at the at least one second position during fusion; According to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

9. The method according to claim 5, characterized in that The determining of the road direction constraint of the target based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes: Determine the tangent angle direction at a second position as the road direction constraint of the target, wherein the second position is a position closest to the target in a second target road geometry, and the second target road geometry is a target road geometry with the highest confidence among the at least one target road geometry.

10. The method according to claim 5, characterized in that The determining of the road direction constraint of the target based on the confidence of the at least one target road geometry and the tangent direction angle of the at least one target road geometry at the at least one second position includes: Determine the tangent angle direction at a third position as the road direction constraint of the target, wherein the third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

11. The method according to claim 1, wherein Determining a road width constraint of the target based on the at least one target road geometry and a motion state of the target includes: Obtaining a straight line passing through the target position of the target and perpendicular to fourth target road geometries, wherein the fourth target road geometries are two target road geometries closest to the target and located on both sides of the target; A distance between two intersection points is determined as a road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

12. The method according to claim 1, characterized in that Determining a road width constraint of the target based on the at least one target road geometry and a motion state of the target includes: determining at least one distance between the target and the at least one target road geometry, the at least one distance being a road width constraint for the target; or, determining at least one distance between the target and the at least one target road geometry, and determining a maximum or minimum value of the at least one distance as a road width constraint for the target; or, A distance between the target and the at least one target road geometry is determined, and an average of the at least one distance is determined as a road width constraint for the target.

13. The method according to any one of claims 1 to 12, characterized in that Determining the confidence of the road direction constraint in the measurement matrix by using the road width constraint includes: determining the measurement noise corresponding to the target according to a mapping relationship between a road width constraint and measurement noise, and the road width constraint; The confidence of the road direction constraint in the measurement matrix is determined according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target.

14. The method according to claim 13, characterized in that After determining the measurement noise corresponding to the target according to the mapping relationship between the road width constraint and the measurement noise and the road width constraint, the method further includes: Determining a motion state change parameter of the target according to the motion state of the target; When a comparison result of the motion state change parameter of the target and a corresponding threshold value indicates that it is necessary to determine a degree of change in curvature or a degree of change in the curvature change rate of a target road geometry among the at least one road geometry at a fourth position, determining the degree of change in curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position, the fourth position being a position located on the target road geometry and within a third distance range from the target; When the degree of change of the curvature or the degree of change of the curvature change rate is greater than a third threshold, increasing the measurement noise corresponding to the target; The determining the confidence of the road direction constraint in the measurement matrix according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes: The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

15. A road constraint determination device, characterized in that: include: at least one processing module; The at least one processing module is configured to determine the motion state of the target based on detection information obtained by detecting the target; The at least one processing module is further configured to determine, based on the detection information of the target, at least one road geometry of a road on which the target is located, each road geometry of the at least one road geometry being characterized by at least one item of information; determine at least one target road geometry among the at least one road geometry; and determine a road constraint of the target based on the at least one target road geometry and a motion state of the target, the road constraint comprising at least one of a road direction constraint and a road width constraint; in, The at least one processing module is further used to determine a measurement matrix including the road direction constraint; and determine the confidence of the road direction constraint in the measurement matrix through the road width constraint, wherein the measurement matrix is used to estimate the motion state of the target to complete target tracking; the road direction constraint in the measurement matrix is determined based on the at least one target road geometry, and the deviation between the at least one target road geometry and the motion trajectory of the target is within a predetermined range; the confidence of the road direction constraint in the measurement matrix is negatively correlated with the road width constraint.

16. The device according to claim 15, characterized in that The at least one processing module is specifically configured to, for each road geometry of the at least one road geometry: Determine a tangent direction angle of the road geometry at a first position, where the tangent direction angle is an angle between a tangent and a radial direction of the road geometry at the first position; acquiring a tangential direction angle of the target at the target position according to a lateral velocity and a radial velocity of the target at the target position, wherein the distance between the target position and the first position is within a first distance range; If the absolute value of the difference between the tangent direction angle at the first position and the tangent direction angle at the target position is less than a first threshold, the road geometry is determined to be the target road geometry.

17. The device according to claim 15, characterized in that The at least one processing module is specifically configured to, for each road geometry of the at least one road geometry: If the distance between the target and the road geometry is within a second distance range, the road geometry is determined to be the target road geometry.

18. The device according to claim 15, characterized in that The at least one processing module is specifically configured to, for each road geometry of the at least one road geometry: Obtaining a distance between the target and the road geometry; According to the number of the at least one road geometry, the Num road geometries with the smallest distance are determined as the at least one target road geometry, where Num is a positive integer not less than 1.

19. The device according to claim 15, characterized in that The at least one processing module is specifically configured to determine at least one second position located in the at least one target road geometry, the at least one second position being a position closest to the target in at least one first target road geometry, the first target road geometry being the target road geometry where the second position is located; A road direction constraint of the target is determined based on a confidence level of the at least one target road geometry and a tangent direction angle of the at least one target road geometry at the at least one second position.

20. The device according to claim 19, characterized in that The confidence level of the target road geometry is the confidence level of the road parameters of the target road geometry; or, The confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position.

21. The device according to claim 20, characterized in that When the confidence level of the target road geometry is a confidence level of a road parameter of the target road geometry, the at least one processing module is further configured to determine the confidence level of the road parameter of the target road geometry based on a variance or a standard deviation of the road parameter of the target road geometry, where the road parameter is at least one item of information used to characterize the target road geometry; Alternatively, when the confidence level of the target road geometry is the confidence level of the tangent direction angle of the target road geometry at the second position, the at least one processing module is further used to determine the confidence level of the tangent direction angle of the target road geometry at the second position based on the variance or standard deviation of the tangent direction angle of the target road geometry at the second position.

22. The device according to claim 20, characterized in that The at least one processing module is specifically configured to determine, according to the confidence level of the target road geometry, a weight value of the tangent direction angle of the at least one target road geometry at the at least one second position during fusion; According to the weight value, a fusion result of fusing the tangent direction angles is determined as the road direction constraint of the target.

23. The device according to claim 20, characterized in that The at least one processing module is specifically used to determine the tangent angle direction at a second position as the road direction constraint of the target, wherein the second position is the position closest to the target in the second target road geometry, and the second target road geometry is the target road geometry with the highest confidence among the at least one target road geometry.

24. The device according to claim 20, characterized in that The at least one processing module is specifically used to determine the tangent angle direction at a third position as the road direction constraint of the target, wherein the third position is the position closest to the target in the third target road geometry, and the third target road geometry is the target road geometry closest to the target in the at least one target road geometry.

25. The device according to claim 15, wherein The at least one processing module is specifically configured to obtain a straight line passing through the target position of the target and perpendicular to fourth target road geometries, wherein the fourth target road geometries are two target road geometries closest to the target and located on both sides of the target; A distance between two intersection points is determined as a road width constraint of the target, where the two intersection points are two intersection points of the straight line and the fourth target road geometry.

26. The device according to claim 15, characterized in that The at least one processing module is specifically configured to determine at least one distance between the target and the at least one target road geometry, the at least one distance being a road width constraint for the target; or, The at least one processing module is specifically configured to determine at least one distance between the target and the at least one target road geometry, and determine a maximum value or a minimum value of the at least one distance as a road width constraint for the target; or, The at least one processing module is specifically configured to determine a distance between the target and the at least one target road geometry, and determine an average value of the at least one distance as a road width constraint for the target.

27. The device according to any one of claims 15 to 26, characterized in that The at least one processing module is specifically configured to determine the measurement noise corresponding to the target according to a mapping relationship between a road width constraint and the measurement noise, and the road width constraint; The confidence of the road direction constraint in the measurement matrix is determined according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target.

28. The device according to claim 27, characterized in that The at least one processing module is further configured to determine a motion state change parameter of the target according to the motion state of the target after determining the measurement noise corresponding to the target based on a mapping relationship between a road width constraint and the measurement noise and the road width constraint; When a comparison result of the motion state change parameter of the target and a corresponding threshold value indicates that it is necessary to determine a degree of change in curvature or a degree of change in the curvature change rate of a target road geometry among the at least one road geometry at a fourth position, determining the degree of change in curvature or the degree of change in the curvature change rate of the target road geometry at the fourth position, the fourth position being a position located on the target road geometry and within a third distance range from the target; When the degree of change of the curvature or the degree of change of the curvature change rate is greater than a third threshold, increasing the measurement noise corresponding to the target; The determining the confidence of the road direction constraint in the measurement matrix according to a mapping relationship between measurement noise and the confidence of the road direction constraint in the measurement matrix, and the measurement noise corresponding to the target, includes: The confidence level of the road direction constraint in the measurement matrix is determined according to the increased measurement noise and a mapping relationship between the measurement noise and the confidence level of the road direction constraint in the measurement matrix.

29. A road constraint determination device, characterized in that: include: at least one processor and memory; Wherein, the memory is used to store program instructions; The at least one processor is configured to call and execute program instructions stored in the memory, so that the apparatus performs the method according to any one of claims 1 to 14.

30. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the method according to any one of claims 1 to 14.

31. A computer program product comprising instructions, characterized in that When the computer program product is run on an electronic device, the electronic device is enabled to perform the method according to any one of claims 1 to 14.

Images