A lane recognition method based on millimeter wave radar target tracking

Abstract

The application discloses a lane recognition method based on millimeter wave radar target tracking, which comprises the following steps: acquiring the motion trajectory information of a target vehicle through a millimeter wave radar; acquiring the motion information of the vehicle, and judging the motion state of the vehicle according to the motion information; screening and counting the target vehicle associated with the motion state of the vehicle according to the motion state of the vehicle; calculating the distance distribution of the target vehicle relative to the lateral position or longitudinal position of the vehicle, so as to determine the lane center line position of the lane where the target vehicle is located; translating the lane center line position to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located; and processing the obtained multiple inner and / or outer lane line information of the same lane to obtain the fused lane position distribution. The application realizes lane recognition of the current driving road of the vehicle, improves the lane recognition precision, and realizes the intelligent driving auxiliary function of the vehicle by dynamically tracking the motion trajectory of the moving target through the millimeter wave radar.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving assistance technology for vehicles, specifically to a lane recognition method based on millimeter-wave radar target tracking. Background Technology

[0002] With the rapid development of 5G communication and vehicle-to-everything (V2X) technologies, autonomous driving has become a research hotspot. Core technologies in the field of autonomous driving include intelligent environmental perception, automatic navigation and positioning, driving behavior decision-making, and intelligent path planning and control. During autonomous driving, it is necessary to determine the lane the vehicle is currently traveling in, thus laying the foundation for subsequent driving behavior decisions and intelligent path planning and control.

[0003] Previously, to determine the lane a vehicle was traveling in, the following technologies were disclosed: using a vehicle-mounted camera to identify lane boundary lines and applying them to lane departure warning systems or lane keeping assist systems. However, while these systems are effective for lane recognition when lane boundary lines can be accurately identified, they cannot perform lane recognition when weather conditions or lane line wear prevents lane line recognition.

[0004] When lane lines cannot be identified, LiDAR and surrounding environment devices can be used to detect the vehicle's surroundings to determine the lane the vehicle is in. However, when the road environment changes and does not match the road structures pre-stored inside the vehicle, the vehicle's lane will also be unable to be detected.

[0005] Therefore, in unstructured roads, congested roads, or rainy or snowy weather, lane recognition based on vision and lidar is ineffective or cannot be recognized, affecting the assisted driving function. Summary of the Invention

[0006] In view of the above problems, embodiments of the present invention provide a lane recognition method based on millimeter-wave radar target tracking, which is used to solve the problem that in the prior art, the lane recognition effect based on vision and lidar is poor or cannot be recognized in unstructured roads, congested roads or rainy and snowy weather, thus affecting the assisted driving function.

[0007] According to one aspect of the present invention, a lane recognition method based on millimeter-wave radar target tracking is provided, comprising the following steps:

[0008] Acquire the motion trajectory information of the target vehicle using millimeter-wave radar;

[0009] Obtain the vehicle's motion information and determine the vehicle's motion status based on the motion information;

[0010] Based on the vehicle's motion status, filter and statistically analyze target vehicles that are associated with the vehicle's motion status;

[0011] Calculate the distance distribution of the target vehicle relative to the vehicle's lateral or longitudinal position to determine the position of the lane centerline of the lane where the target vehicle is located;

[0012] The center line of the lane is shifted half a lane to both sides to obtain the inner and outer lane line information of the lane where the target vehicle is located;

[0013] The obtained lane line information of multiple inner and / or outer lanes in the same lane is processed to obtain the fused lane position distribution.

[0014] Furthermore, as a preferred technical solution, the acquired trajectory information of the target vehicle includes the position and velocity components of the target vehicle relative to its own coordinate system.

[0015] Furthermore, as a preferred technical solution, the vehicle's motion information includes the vehicle's speed, steering wheel angle, and angular velocity;

[0016] The vehicle's motion states include straight driving, stationary driving, and cornering.

[0017] Furthermore, as a preferred technical solution, the target vehicles associated with the motion state of the vehicle include target vehicles moving in the same direction as the vehicle, target vehicles moving in the opposite direction to the vehicle, target vehicles moving laterally to the vehicle, and target vehicles moving diagonally to the vehicle.

[0018] Furthermore, as a preferred technical solution, the target vehicles associated with the motion state of this vehicle are specifically screened and statistically analyzed based on the vehicle's motion state, including:

[0019] When the vehicle is traveling straight, target vehicles moving in the same direction, opposite direction, or laterally are selected based on the speed component of the target vehicles, and the number of target vehicles moving in the same direction, opposite direction, or laterally is counted.

[0020] When the vehicle is stationary, target vehicles moving diagonally to the vehicle are selected based on the speed component of the target vehicle, and the number of target vehicles moving diagonally to the vehicle is counted.

[0021] When the vehicle is traveling on a curve, target vehicles moving in the same or opposite direction as the vehicle are selected based on the speed component of the target vehicle, and the number of target vehicles moving in the same or opposite direction as the vehicle is counted.

[0022] Furthermore, as a preferred technical solution, the specific selection criteria for the target vehicle include:

[0023] When the vehicle is traveling straight, the selection criteria for target vehicles moving in the same direction as the vehicle are: the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is greater than the second preset threshold.

[0024] When the vehicle is traveling straight, the selection criteria for target vehicles moving in the opposite direction are: the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is less than the third preset threshold.

[0025] When the vehicle is traveling straight, the selection criteria for target vehicles moving laterally to the vehicle are: the lateral velocity component of the target vehicle is greater than the first preset threshold, and the longitudinal velocity component of the target vehicle is less than the second preset threshold.

[0026] When the vehicle is stationary, the selection criteria for target vehicles moving diagonally to the vehicle are: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the first preset angle and the second preset angle.

[0027] When the vehicle is in a curve, the selection criteria for target vehicles moving in the same direction as the vehicle are: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the third preset angle and the fourth preset angle.

[0028] When the vehicle is in a curve, the selection criteria for the target vehicle moving in the opposite direction is: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the third preset angle and the fourth preset angle.

[0029] Furthermore, as a preferred technical solution, the lateral or longitudinal position of the target vehicle relative to the vehicle specifically includes: the target vehicle being located in the left side area, right side area, first front side area, second front side area, second rear side area, or second rear side area of ​​the vehicle.

[0030] The first front region is close to the front of the vehicle, and the second front region is far from the front of the vehicle.

[0031] The first rear side region is close to the rear of the vehicle, and the second rear side region is far from the rear of the vehicle.

[0032] Furthermore, as a preferred technical solution, when the vehicle is traveling straight, determining the position of the lane centerline based on a target vehicle moving in the same direction as the vehicle specifically includes:

[0033] Calculate the first distance distribution of the target vehicle relative to the vehicle in the left-side region to obtain the first driving center position of the target vehicle in the left-side region, and set the first driving center position of the target vehicle as the first lane centerline position of the vehicle in the left-side lane.

[0034] Calculate the second distance distribution of the target vehicle relative to the vehicle on the right side of the region to obtain the second driving center position of the target vehicle in the right side of the region, and set the second driving center position of the target vehicle as the second lane center line position of the right lane of the vehicle.

[0035] Calculate the third distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the third driving center position of the target vehicle in the front or rear region, and set the third driving center position of the target vehicle as the center line position of the third lane in the same lane as the vehicle.

[0036] Furthermore, as a preferred technical solution, when the vehicle is traveling straight, determining the position of the lane centerline based on a target vehicle moving in the opposite direction specifically includes:

[0037] Calculate the fourth distance distribution of the target vehicle relative to the vehicle in the left-hand region to obtain the fourth driving center position of the target vehicle in the left-hand region, and set the fourth driving center position of the target vehicle as the fourth lane center line position of the vehicle in the left-hand lane.

[0038] Calculate the fifth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the fifth driving center position of the target vehicle in the front or rear region, and set the fifth driving center position of the target vehicle as the position of the fifth lane center line in the same lane as the vehicle.

[0039] Furthermore, as a preferred technical solution, when the vehicle is traveling straight, determining the position of the lane centerline relative to a target vehicle moving laterally relative to the vehicle specifically includes:

[0040] Calculate the sixth distance distribution of the target vehicle relative to the vehicle in the first front area to obtain the sixth driving center position of the target vehicle in the first front area, and set the sixth driving center position of the target vehicle as the sixth lane centerline position of the vehicle in the first front lane.

[0041] Furthermore, as a preferred technical solution, it also includes:

[0042] Calculate the seventh distance distribution of the target vehicle relative to the vehicle in the second front area to obtain the seventh driving center position of the target vehicle in the second front area, and set the seventh driving center position of the target vehicle as the seventh lane center line position of the vehicle in the second front lane.

[0043] Furthermore, as a preferred technical solution, when the vehicle is stationary, determining the position of the lane centerline relative to a target vehicle moving diagonally relative to the vehicle specifically includes:

[0044] Calculate the eighth distance distribution of the target vehicle relative to the vehicle in the first rear side region to obtain the eighth driving center position of the target vehicle in the first rear side region, and set the eighth driving center position of the target vehicle as the eighth lane center line position of the vehicle in the first rear side lane.

[0045] Furthermore, as a preferred technical solution, it also includes:

[0046] Calculate the ninth distance distribution of the target vehicle relative to the vehicle in the second rear region to obtain the ninth driving center position of the target vehicle in the second rear region, and set the ninth driving center position of the target vehicle as the ninth lane centerline position of the vehicle in the second rear lane.

[0047] Furthermore, as a preferred technical solution, when the vehicle is traveling in a curve, determining the position of the lane centerline based on a target vehicle moving in the same direction as the vehicle specifically includes:

[0048] Calculate the tenth distance distribution of the target vehicle relative to the vehicle in the left-hand region to obtain the tenth driving center position of the target vehicle in the left-hand region, and set the tenth driving center position of the target vehicle as the tenth lane centerline position of the vehicle's left-hand lane.

[0049] Calculate the eleventh distance distribution of the target vehicle relative to the vehicle on the right side of the region to obtain the eleventh driving center position of the target vehicle in the right side of the region, and set the eleventh driving center position of the target vehicle as the eleventh lane centerline position of the right lane of the vehicle.

[0050] Calculate the twelfth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the twelfth driving center position of the target vehicle in the front or rear region, and set the twelfth driving center position of the target vehicle as the twelfth lane centerline position in the same lane as the vehicle.

[0051] Furthermore, as a preferred technical solution, when the vehicle is traveling on a curve, determining the position of the lane centerline relative to a target vehicle moving in the opposite direction specifically includes:

[0052] Calculate the thirteenth distance distribution of the target vehicle relative to the current vehicle in the left-hand region to obtain the thirteenth driving center position of the target vehicle in the left-hand region, and set the thirteenth driving center position of the target vehicle as the thirteenth lane centerline position of the current vehicle's left-hand lane.

[0053] Calculate the fourteenth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the fourteenth driving center position of the target vehicle in the front or rear region, and set the fourteenth driving center position of the target vehicle as the fourteenth lane centerline position in the same lane as the vehicle.

[0054] Furthermore, as a preferred technical solution, when the vehicle is driving on a curve, the acquisition of the inner and outer lane line information of the target vehicle's lane specifically includes:

[0055] The lane centerline position is shifted half a lane to both sides of the road curvature with reference to the vehicle's historical trajectory line, providing information on the inner and outer lane lines of the target vehicle's lane.

[0056] Furthermore, as a preferred technical solution, the first preset threshold, the second preset threshold, the third preset threshold, the first preset angle, the second preset angle, the third preset angle, and the fourth preset angle are set according to the vehicle speed and driving environment and / or road conditions and / or the computing power of the millimeter-wave radar.

[0057] The first preset threshold is set within the range of 0.8-1.5 m / s;

[0058] The second preset threshold is set within the range of 0.8-1.5 m / s;

[0059] The setting range of the third preset threshold is -0.8 to 1.5 m / s;

[0060] The first preset angle is set within the range of 15°-30°;

[0061] The second preset angle is set within the range of 60°-80°;

[0062] The setting range of the third preset angle is 60°-80°;

[0063] The fourth preset angle is set within the range of 80°-90°.

[0064] Compared with the prior art, the present invention has the following advantages:

[0065] In this embodiment of the invention, the movement trajectory of a moving target is dynamically tracked by millimeter-wave radar, and the lane of the vehicle's current driving road is identified based on the movement trajectory of the moving target. The lane identification accuracy is improved by using data processing algorithms, providing lane information for intelligent driving assistance of the vehicle, or providing fusion information for lane identification and other functions of other sensors.

[0066] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0067] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0068] Figure 1 The diagram shows a schematic flowchart of Embodiment 1 of a lane recognition method based on millimeter-wave radar target tracking provided by the present invention;

[0069] Figure 2 This illustration shows a schematic diagram of the lane recognition method based on millimeter-wave radar target tracking provided by the present invention, in which the vehicle is in a straight-moving state and the lane is recognized based on a target vehicle moving in the same direction as the vehicle.

[0070] Figure 3 This illustration shows a schematic diagram of the lane recognition method based on millimeter-wave radar target tracking provided by the present invention, in embodiment 3, when the vehicle is in a straight-moving state, and the lane is recognized based on a target vehicle moving in the opposite direction to the vehicle.

[0071] Figure 4 The diagram illustrates a scheme for lane recognition based on millimeter-wave radar target tracking, as described in Embodiment 4 of the present invention. It shows the lane recognition based on a target vehicle moving laterally relative to the vehicle when the vehicle is traveling straight.

[0072] Figure 5 The diagram illustrates a schematic representation of a lane recognition method based on millimeter-wave radar target tracking, provided by the present invention, in embodiment 5, where the vehicle is in a parked state, and the lane is recognized based on a target vehicle moving diagonally relative to the vehicle.

[0073] Figure 6 The diagram illustrates a scheme for lane recognition based on a target vehicle moving in the same direction as the vehicle when the vehicle is in a curved driving state, according to Embodiment 6 of the present invention. Detailed Implementation

[0074] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0075] Example 1

[0076] In order to address the problem that lane recognition based on vision and lidar is ineffective or fails to recognize lanes in unstructured roads, congested roads, or rainy / snowy weather, thus affecting assisted driving functions, this embodiment discloses a lane recognition method based on millimeter-wave radar target tracking.

[0077] Figure 1 A flowchart of a first embodiment of a lane recognition method based on millimeter-wave radar target tracking according to the present invention is shown. Figure 1 As shown, the method includes the following steps:

[0078] Step S10: Obtain the motion trajectory information of the target vehicle through millimeter-wave radar.

[0079] Millimeter-wave radar, mounted on the vehicle, is used to assist in driving functions. Specifically, millimeter-wave radar is installed around the vehicle—at the front, rear, left, and right—to collect, process, and identify information about the surrounding environment, thereby enabling driver assistance functions.

[0080] In this embodiment, the millimeter-wave radar installed around the vehicle is also used to acquire the motion trajectory information of moving targets around the vehicle, that is, the motion trajectory information of the target vehicle described in this application.

[0081] Specifically, the acquisition of the target vehicle's trajectory information involves using existing target detection and tracking algorithms to calculate the target vehicle's trajectory information from the target vehicle's motion data acquired by millimeter-wave radar.

[0082] In this embodiment, the trajectory information of the target vehicle includes the position component and velocity component of the target vehicle relative to the vehicle's own coordinate system.

[0083] The trajectory information is represented by the parameter trace = [x,y,vx,vy], where (x,y) represents the position component of the target vehicle relative to the vehicle's coordinate system, and (vx,vy) represents the velocity component of the target vehicle relative to the vehicle's coordinate system; the forward direction of the vehicle is set as the Y-axis, and the right-hand direction is set as the X-axis.

[0084] Step S20: Obtain the motion information of the vehicle and determine the motion state of the vehicle based on the motion information.

[0085] Among them, the vehicle's motion states include straight-going state, stationary state, and curve-driving state;

[0086] The vehicle's motion information is obtained through the vehicle's CAN data, which includes at least the vehicle's speed, steering wheel angle, and angular velocity. Therefore, the vehicle's motion state can be directly determined by the vehicle's speed, steering wheel angle, and angular velocity, thus indicating that the vehicle can be in any one of the three states: straight driving, stationary, or driving on a curve.

[0087] Step S30: Filter and count target vehicles that are associated with the motion state of this vehicle based on the motion state of this vehicle.

[0088] In this step, since the target vehicle is acquired by millimeter-wave radars set around the vehicle, the target vehicle can be distributed on the left, right, front and rear sides of the vehicle.

[0089] By combining the vehicle's motion status with setting filtering conditions, it is possible to filter out target vehicles that are related to this vehicle from all target vehicles.

[0090] Step S40: Calculate the distance distribution of the target vehicle relative to the lateral or longitudinal position of the vehicle, thereby determining the position of the lane centerline of the lane where the target vehicle is located.

[0091] In this step, the target vehicle's lateral or longitudinal position relative to the vehicle depends on the vehicle's motion state. Therefore, after determining the target vehicle's motion state, this step involves obtaining the filtered target vehicles associated with the vehicle and calculating the distance distribution of the target vehicle's lateral or longitudinal position relative to the vehicle. Based on this distance distribution, the position of the lane centerline of the lane where the target vehicle is located can be determined.

[0092] Step S50: Shift the position of the center line of the lane to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0093] This step specifically involves: after obtaining the position of the lane center line, shifting the position of the lane center line to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0094] Step S60: Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0095] For example, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0096] Therefore, this step specifically involves: if two or more inner and / or outer lane line information of the same lane are identified, then a weighted average processing is required to obtain the fused lane position distribution.

[0097] This invention discloses a lane recognition method based on millimeter-wave radar target tracking. The method dynamically tracks the trajectory of a moving target using millimeter-wave radar, identifies the lane of the vehicle's current driving path based on the trajectory of the moving target, and improves lane recognition accuracy using data processing algorithms. This provides lane information for intelligent driving assistance or fusion information for lane recognition functions of other sensors.

[0098] Example 2

[0099] This embodiment discloses a lane recognition method based on millimeter-wave radar target tracking. Based on embodiment 1, it specifically discloses a target vehicle selection scheme associated with the vehicle's motion state and how to perform lane recognition on the vehicle's current driving road based on the selected target vehicles.

[0100] In this embodiment, based on embodiment 1, the motion states of the vehicle include straight driving, parking, and cornering.

[0101] The target vehicles can be located on the left, right, front, and rear sides of the vehicle. Therefore, based on the vehicle's movement status, the selected target vehicles will be located in the left, right, front, or rear areas of the vehicle.

[0102] The target vehicles associated with the motion state of this vehicle can be further divided into: target vehicles moving in the same direction as this vehicle, target vehicles moving in the opposite direction to this vehicle, target vehicles moving laterally to this vehicle, and target vehicles moving diagonally to this vehicle.

[0103] Therefore, this application needs to identify the lane of the current driving road of the vehicle based on the motion trajectory information of the target vehicle associated with the current motion state of the vehicle.

[0104] This embodiment illustrates how, when the vehicle is traveling straight, lane identification is performed based on a target vehicle moving in the same direction as the vehicle to identify the current road.

[0105] Please refer to the following for the specific implementation process:

[0106] Figure 2 The diagram illustrates a scheme for lane recognition based on a target vehicle moving in the same direction as the vehicle when the vehicle is traveling straight, according to Embodiment 2 of the present invention, which describes a lane recognition method based on millimeter-wave radar target tracking.

[0107] like Figure 2 As shown, when the vehicle is traveling straight, lane recognition based on target vehicles moving in the same direction as the vehicle specifically includes:

[0108] S4011. Filter target vehicles moving in the same direction as this vehicle and count the number of target vehicles.

[0109] This step specifically involves: filtering target vehicles moving in the same direction as this vehicle based on the speed component of the target vehicle, and counting their number.

[0110] The selection criteria are: the absolute value of the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is greater than the second preset threshold.

[0111] The first preset threshold and the second preset threshold are set according to the vehicle speed, driving environment and / or road conditions and / or the computing power of the millimeter-wave radar.

[0112] In this embodiment, the first preset threshold is set in the range of 0.8-1.5 m / s, and the second preset threshold is set in the range of 0.8-1.5 m / s.

[0113] In a preferred embodiment, the first preset threshold can be set to 1 m / s, and the second preset threshold can also be set to 1 m / s.

[0114] Therefore, the target vehicles that satisfy |vx|<1m / s and vy>1m / s are the selected target vehicles.

[0115] S4012. Calculate the distance distribution of the target vehicle relative to the lateral position of this vehicle.

[0116] Since the target vehicle is located in one or more areas of the vehicle's left, right, front, or rear sides, this step specifically includes:

[0117] Calculate the first distance distribution of the target vehicle relative to the vehicle in the left-hand region. This first distance distribution can be considered as the average lateral position of all target vehicles in the left-hand region relative to the vehicle, denoted as Xleft1.

[0118] Calculate the second distance distribution of the target vehicle relative to the vehicle on the right side of the region. This second distance distribution can be considered as the average lateral position of all target vehicles in the right side of the vehicle relative to the vehicle, denoted as Xright1.

[0119] Calculate the third distance distribution of the target vehicle relative to the vehicle in the front or rear region. This third distance distribution can be considered as the average lateral position of all target vehicles in the front or rear region relative to the vehicle, denoted as Xcur1.

[0120] S4013. Based on the distance distribution of the target vehicle relative to the vehicle's lateral position, obtain the position of the lane centerline of the lane where the target vehicle is located.

[0121] Specifically, this step involves obtaining the first driving center position Xleft1 of the target vehicle in the left-side region based on the first distance distribution Xleft1 of the target vehicle relative to the vehicle in the left-side region, and setting the first driving center position Xleft1 of the target vehicle as the first lane centerline position of the vehicle in the left-side lane.

[0122] Based on the second distance distribution Xright1 of the target vehicle relative to the vehicle on the right side of the area, the second driving center position Xright1 of the target vehicle in the right side of the area is obtained, and the second driving center position Xright1 of the target vehicle is set as the second lane center line position of the right lane of the vehicle.

[0123] Based on the third distance distribution Xcur1 of the target vehicle relative to the vehicle in the front or rear region, the third driving center position Xcur1 of the target vehicle in the front or rear region is obtained, and the third driving center position Xcur1 of the target vehicle is set as the position of the center line of the third lane in the same lane as the vehicle.

[0124] S4014. Shift the position of the lane centerline to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0125] Specifically, this step involves shifting the center line of the first lane to both sides by half a lane to obtain the inner and outer lane line information Lane1 for the left lane of this vehicle.

[0126] The center line of the second lane is shifted half a lane to both sides to obtain the inner and outer lane line information of the right lane of this vehicle (Lane2).

[0127] The center line of the third lane is shifted half a lane to both sides to obtain the inner and outer lane line information (Lane3) of this lane.

[0128] S4015. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0129] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0130] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information in the same lane are identified, then a weighted average processing is required. That is, the identified inner and outer lane line information Lane1 of the left lane of the vehicle, the inner and outer lane line information Lane2 of the right lane of the vehicle, and the inner and outer lane line information Lane3 of the current lane are weighted and averaged to finally obtain the fused lane position LaneFuse1.

[0131] Example 3

[0132] This embodiment discloses a lane recognition method based on millimeter-wave radar target tracking. Specifically, based on embodiment 2, it discloses that when the vehicle is traveling straight, lane recognition is performed on the current driving road of the vehicle based on the target vehicle moving in the opposite direction.

[0133] Please refer to the following for the specific implementation process:

[0134] Figure 3 The diagram illustrates a scheme for lane recognition based on a target vehicle moving in the opposite direction to the vehicle when the vehicle is traveling straight, as described in Embodiment 3 of the present invention, which is a lane recognition method based on millimeter-wave radar target tracking.

[0135] like Figure 3 As shown, when the vehicle is traveling straight, lane recognition based on a target vehicle moving in the opposite direction specifically includes:

[0136] S4021. Filter target vehicles moving in the opposite direction to this vehicle and count the number of target vehicles.

[0137] This step specifically involves: filtering target vehicles moving in the opposite direction to the target vehicle based on the target vehicle's speed component, and counting their number.

[0138] The selection criteria are: the absolute value of the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is less than the third preset threshold.

[0139] The first and third preset thresholds are set based on the vehicle speed, driving environment and / or road conditions and / or the computing power of the millimeter-wave radar.

[0140] In this embodiment, the first preset threshold is set in the range of 0.8-1.5 m / s, and the third preset threshold is set in the range of -0.8--1.5 m / s.

[0141] In a preferred embodiment, the first preset threshold can be set to 1 m / s, and the third preset threshold can also be set to -1 m / s.

[0142] Therefore, the target vehicles that satisfy |vx|<1m / s and vy<-1m / s are the selected target vehicles.

[0143] S4022. Calculate the distance distribution of the target vehicle relative to the lateral position of this vehicle.

[0144] Since the target vehicle is located in one or more areas of the vehicle's left, right, front, or rear sides, this step specifically includes:

[0145] Calculate the fourth distance distribution of the target vehicle relative to the vehicle in the left-hand region. This fourth distance distribution can be considered as the average lateral position of all target vehicles in the left-hand region relative to the vehicle, denoted as Xleft2.

[0146] Calculate the fifth distance distribution of the target vehicle relative to the vehicle in the front or rear region. This fifth distance distribution can be considered as the average lateral position of all target vehicles in the front or rear region relative to the vehicle, denoted as Xcur2.

[0147] S4023. Based on the distance distribution of the target vehicle relative to the vehicle's lateral position, obtain the position of the lane centerline of the lane where the target vehicle is located.

[0148] Specifically, this step involves obtaining the fourth driving center position Xleft2 of the target vehicle in the left-side region based on the fourth distance distribution Xleft2 of the target vehicle relative to the vehicle in the left-side region, and setting the fourth driving center position Xleft2 of the target vehicle as the center line position of the fourth lane of the vehicle's left-side lane.

[0149] Based on the fifth distance distribution Xcur2 of the target vehicle relative to the vehicle in the front or rear region, the fifth driving center position Xcur2 of the target vehicle in the front or rear region is obtained, and the fifth driving center position Xcur2 of the target vehicle is set as the position of the fifth lane center line in the same lane as the vehicle.

[0150] S4024. Shift the position of the lane centerline to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0151] This step specifically involves shifting the center line of the fourth lane to both sides by half a lane to obtain the inner and outer lane line information (Lane4) for the left lane of this vehicle.

[0152] The center line of the fifth lane is shifted half a lane to both sides to obtain the inner and outer lane line information for this lane (Lane5).

[0153] S4025. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0154] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0155] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information in the same lane are identified, then a weighted average processing is required. That is, the identified inner and outer lane line information Lane4 of the left lane of this vehicle and the inner and outer lane line information Lane5 of this lane are weighted averaged to finally obtain the fused lane position LaneFuse2.

[0156] Example 4

[0157] This embodiment discloses a lane recognition method based on millimeter-wave radar target tracking. Specifically, based on embodiment 2, it discloses that when the vehicle is traveling straight, lane recognition is performed on the current driving road of the vehicle based on the target vehicle moving laterally to the vehicle.

[0158] Please refer to the following for the specific implementation process:

[0159] Figure 4 The diagram illustrates a scheme for lane recognition based on a target vehicle moving laterally relative to the vehicle when the vehicle is traveling straight, according to Embodiment 4 of the present invention, which describes a lane recognition method based on millimeter-wave radar target tracking.

[0160] like Figure 4 As shown, when the vehicle is traveling straight, lane recognition based on a target vehicle moving laterally relative to the vehicle specifically includes:

[0161] S4031. Filter target vehicles that are moving laterally to this vehicle and count the number of target vehicles.

[0162] This step specifically involves: filtering target vehicles that are moving laterally with the target vehicle based on the speed component of the target vehicle, and counting their number.

[0163] The selection criteria are: the absolute value of the lateral velocity component of the target vehicle is greater than a first preset threshold, and the absolute value of the longitudinal velocity component of the target vehicle is less than a second preset threshold.

[0164] The first preset threshold and the second preset threshold are set according to the vehicle speed, driving environment and / or road conditions and / or the computing power of the millimeter-wave radar.

[0165] In this embodiment, the first preset threshold is set in the range of 0.8-1.5 m / s, and the second preset threshold is set in the range of 0.8-1.5 m / s.

[0166] In a preferred embodiment, the first preset threshold can be set to 1 m / s, and the second preset threshold can also be set to 1 m / s.

[0167] Therefore, the target vehicles that satisfy |vx|>1m / s and |vy|<1m / s are the selected target vehicles.

[0168] S4032. Calculate the distance distribution of the target vehicle relative to the longitudinal position of this vehicle.

[0169] Since the target vehicle is distributed in one or more areas of the vehicle’s left, right, front or rear sides, and given that the target vehicle is moving laterally relative to the vehicle, it can be assumed that the vehicle is at an intersection and the target vehicle is located in the area in front of the vehicle.

[0170] In this embodiment, the front area of ​​the vehicle is divided into a first front area and a second front area. The first front area is close to the front of the vehicle, and the second front area is far from the front of the vehicle.

[0171] Therefore, this step specifically includes:

[0172] Calculate the sixth distance distribution of the target vehicle relative to the vehicle in the first front region. This sixth distance distribution can be considered as the average longitudinal position of all target vehicles in the first front region relative to the vehicle, denoted as Yfront1.

[0173] Furthermore, the seventh distance distribution of the target vehicle relative to the vehicle in the second front region can be calculated. This seventh distance distribution can be considered as the average longitudinal position of all target vehicles in the second front region relative to the vehicle, denoted as Yfront2.

[0174] S4033. Based on the distance distribution of the target vehicle relative to the longitudinal position of the vehicle, obtain the position of the lane centerline of the lane where the target vehicle is located.

[0175] Specifically, this step involves: based on the sixth distance distribution Yfront1 of the target vehicle relative to the vehicle in the first front area, obtaining the sixth driving center position Yfront1 of the target vehicle in the first front area, and setting the sixth driving center position Yfront1 of the target vehicle as the center line position of the sixth lane of the vehicle's first front lane.

[0176] Based on the seventh distance distribution Yfront2 of the target vehicle relative to the vehicle in the second front area, the seventh driving center position Yfront2 of the target vehicle in the second front area is obtained, and the seventh driving center position Yfront2 of the target vehicle is set as the center line position of the sixth lane of the second front lane of the vehicle.

[0177] S4034. Shift the position of the lane centerline to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0178] This step specifically involves shifting the center line of the sixth lane to both sides by half a lane to obtain the inner and outer lane line information of the first front lane of this vehicle, Lane 6.

[0179] The center line of lane seven is shifted half a lane to both sides to obtain the inner and outer lane line information of the second front lane of this vehicle (Lane 7).

[0180] S4035. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0181] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0182] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information of the same lane are identified, then a weighted average processing is required. That is, the inner and outer lane line information Lane6 of the first front lane of the vehicle and the inner and outer lane line information Lane7 of the second front lane of the vehicle are weighted and averaged to finally obtain the fused lane position LaneFuse3.

[0183] Example 5

[0184] This embodiment discloses a lane recognition method based on millimeter-wave radar target tracking. Specifically, based on embodiment 2, it discloses that when the vehicle is parked, lane recognition is performed on the current driving road of the vehicle based on a target vehicle moving diagonally to the vehicle.

[0185] Please refer to the following for the specific implementation process:

[0186] Figure 5 The diagram illustrates a scheme for lane recognition based on a target vehicle moving diagonally relative to the vehicle when the vehicle is stationary, as described in Embodiment 5 of the present invention, which is a lane recognition method based on millimeter-wave radar target tracking.

[0187] like Figure 5 As shown, when the vehicle is stationary, lane recognition based on a target vehicle moving diagonally relative to the vehicle specifically includes:

[0188] S4041. Filter target vehicles that are moving diagonally to this vehicle and count the number of target vehicles.

[0189] This step specifically involves: filtering target vehicles that are moving diagonally toward this vehicle based on the speed component of the target vehicle, and counting their number.

[0190] The selection criterion is: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the first preset angle and the second preset angle.

[0191] The first preset angle and the second preset angle are set according to the vehicle speed, driving environment and / or road conditions and / or the calculation capability of the millimeter-wave radar.

[0192] In this embodiment, the first preset angle is set within the range of 15°-30°, and the second preset angle is set within the range of 60°-80°.

[0193] In a preferred embodiment, the first preset angle can be set to 20° and the second preset angle can be set to 70°.

[0194] Therefore, the target vehicle that satisfies 20° < atan(|vy| / |vx|) < 70° is the selected target vehicle.

[0195] S4042. Calculate the distance distribution of the target vehicle relative to the longitudinal position of the host vehicle.

[0196] Since the target vehicles are distributed in one or more of the left area, right area, front area or rear area of the host vehicle, and considering that the selected target vehicles move longitudinally relative to the host vehicle and the host vehicle is in a parked state, it can be considered that the target vehicles are located in the rear area of the host vehicle.

[0197] In this embodiment, the rear area of the host vehicle is divided into a first rear area and a second rear area. The first rear area is close to the directly rear of the host vehicle, and the second rear area is far from the directly rear of the host vehicle.

[0198] Therefore, this step specifically includes:

[0199] Calculate the eighth distance distribution of the target vehicle relative to the host vehicle in the first rear area. This eighth distance distribution can be considered as the average longitudinal position of all target vehicles relative to the host vehicle in the first rear area, denoted as Yrear1.

[0200] Meanwhile, the ninth distance distribution of the target vehicle relative to the host vehicle in the second rear area can be further calculated. This ninth distance distribution can be considered as the average longitudinal position of all target vehicles relative to the host vehicle in the second rear area, denoted as Yrear2.

[0201] S4043. Obtain the lane centerline position of the lane where the target vehicle is located according to the distance distribution of the target vehicle relative to the longitudinal position of the host vehicle.

[0202] This step is specifically: According to the eighth distance distribution Yrear1 of the target vehicle relative to the host vehicle in the first rear area, obtain the eighth driving center position Yrear1 of the target vehicle in the first rear area, and set the eighth driving center position Yrear1 of the target vehicle as the eighth lane centerline position of the first rear lane of the host vehicle.

[0203] According to the ninth distance distribution Yrear2 of the target vehicle relative to the host vehicle in the second rear area, obtain the ninth driving center position Yrear2 of the target vehicle in the second rear area, and set the ninth driving center position Yrear2 of the target vehicle as the ninth lane centerline position of the second rear lane of the host vehicle.

[0204] S4044. Translate the lane centerline position half a lane to both sides to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0205] This step specifically involves shifting the center line of the eighth lane to both sides by half a lane to obtain the inner and outer lane line information of the first rear lane of this vehicle (Lane8).

[0206] By shifting the center line of lane nine half a lane to either side, we obtain the inner and outer lane line information for the second rear lane of this vehicle (Lane 9).

[0207] S4045. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0208] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0209] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information in the same lane are identified, then a weighted average processing is required. That is, the inner and outer lane line information Lane8 of the first rear lane of the vehicle and the inner and outer lane line information Lane9 of the second rear lane of the vehicle are weighted and averaged to finally obtain the fused lane position LaneFuse4.

[0210] In this embodiment, when the vehicle is parked, lane identification is performed based on target vehicles moving in the same direction or laterally as the vehicle. See Embodiments 2 and 4 for details, which will not be repeated here.

[0211] Example 6

[0212] This embodiment discloses a lane recognition method based on millimeter-wave radar target tracking. Specifically, based on embodiment 2, it discloses that when the vehicle is driving on a curve, lane recognition is performed on the current driving road of the vehicle based on a target vehicle moving in the same direction as the vehicle.

[0213] Please refer to the following for the specific implementation process:

[0214] Figure 6 The diagram illustrates a schematic representation of a lane recognition method based on millimeter-wave radar target tracking, as described in Embodiment 6 of the present invention. It shows the lane recognition based on a target vehicle moving in the same direction as the vehicle when the vehicle is traveling on a curve.

[0215] like Figure 6 As shown, when the vehicle is traveling in a curve, lane recognition based on target vehicles moving in the same direction as the vehicle specifically includes:

[0216] S4051. Filter target vehicles moving in the same direction as this vehicle and count the number of target vehicles.

[0217] This step is specifically as follows: Filter out the target vehicles moving in the same direction as the vehicle according to the speed components of the target vehicles, and count their numbers.

[0218] And the filtering condition is: The arctangent value of the ratio of the longitudinal speed component to the lateral speed component of the target vehicle is between the third preset angle and the fourth preset angle.

[0219] Wherein, the third preset angle and the fourth preset angle are set according to the vehicle speed of the vehicle, the driving environment and / or road conditions and / or the calculation ability of the millimeter-wave radar.

[0220] In this embodiment, the setting range of the third preset angle is 60° - 80°, and the setting range of the fourth preset angle is 80° - 90°.

[0221] As a preferred embodiment, the third preset angle can be set to 70°, and the fourth preset angle can be set to 90°.

[0222] Therefore, the target vehicles that satisfy 70° < atan(|vy| / |vx|) <= 90° are the filtered target vehicles.

[0223] S4052. Calculate the distance distribution of the target vehicles at the same longitudinal position relative to the historical trajectory of the vehicle itself.

[0224] Since the target vehicles are distributed in one or more of the left area, right area, front area or rear area of the vehicle itself, this step specifically includes:

[0225] Calculate the tenth distance distribution of the target vehicles relative to the vehicle itself in the left area. This tenth distance distribution can be considered as the average lateral position of all the target vehicles in the left area relative to the vehicle itself, denoted as Xleft3.

[0226] Calculate the eleventh distance distribution of the target vehicles relative to the vehicle itself in the right area. This eleventh distance distribution can be considered as the average lateral position of all the target vehicles in the right area relative to the vehicle itself, denoted as Xright2.

[0227] Calculate the twelfth distance distribution of the target vehicles relative to the vehicle itself in the front area or rear area. This twelfth distance distribution can be considered as the average lateral position of all the target vehicles in the front area or rear area relative to the vehicle itself, denoted as Xcur3.

[0228] S4053. Obtain the position of the center line of the lane where the target vehicle is located according to the distance distribution of the target vehicle at the same longitudinal position relative to the historical trajectory of the vehicle itself.

[0229] Specifically, this step involves obtaining the tenth driving center position Xleft3 of the target vehicle in the left-side region based on the tenth distance distribution Xleft3 of the target vehicle relative to the vehicle in the left-side region, and setting the tenth driving center position Xleft3 of the target vehicle as the tenth lane centerline position of the vehicle in the left-side lane.

[0230] Based on the eleventh distance distribution Xright2 of the target vehicle relative to the vehicle on the right side of the area, the eleventh driving center position Xright2 of the target vehicle in the right side of the area is obtained, and the eleventh driving center position Xright2 of the target vehicle is set as the eleventh lane center line position of the right lane of the vehicle.

[0231] Based on the twelfth distance distribution Xcur3 of the target vehicle relative to the vehicle in the front or rear area, the twelfth driving center position Xcur3 of the target vehicle in the front or rear area is obtained, and the twelfth driving center position Xcur3 of the target vehicle is set as the twelfth lane centerline position in the same lane as the vehicle.

[0232] S4054. Shift the position of the lane centerline to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0233] This step specifically involves shifting the center line of the tenth lane half a lane to both sides to obtain the inner and outer lane line information (Lane10) for the left lane of this vehicle.

[0234] The center line of lane eleven is shifted half a lane to both sides to obtain the inner and outer lane line information of the right lane of this vehicle (Lane11).

[0235] The center line of lane 12 is shifted half a lane to both sides to obtain the inner and outer lane line information for this lane, Lane 12.

[0236] S4055. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0237] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0238] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information of the same lane are identified, then a weighted average processing is required. That is, the identified inner and outer lane line information Lane10 of the left lane of this vehicle, the inner and outer lane line information Lane12 of the right lane of this vehicle, and the inner and outer lane line information Lane13 of this lane are weighted averaged to finally obtain the fused lane position LaneFuse5.

[0239] Example 7

[0240] This example discloses a lane recognition method based on millimeter-wave radar target tracking. Specifically, on the basis of Example 2, it discloses that when the vehicle is in a curved driving state, the current driving road of the vehicle is recognized according to the target vehicle moving in the opposite direction of the vehicle.

[0241] The specific implementation process is as follows:

[0242] When the vehicle is in a curved driving state, the lane recognition according to the target vehicle moving in the opposite direction of the vehicle specifically includes:

[0243] S4061. Screen the target vehicles moving in the opposite direction of the vehicle and count the number of target vehicles.

[0244] This step is specifically: Screen the target vehicles moving in the opposite direction of the vehicle according to the speed components of the target vehicles and count their numbers.

[0245] And the screening condition is: The arctangent value of the ratio of the longitudinal speed component to the lateral speed component of the target vehicle is between the third preset angle and the fourth preset angle.

[0246] Among them, the third preset angle and the fourth preset angle are set according to the vehicle speed and driving environment and / or road conditions and / or the calculation ability of the millimeter-wave radar of the vehicle.

[0247] In this example, the setting range of the third preset angle is 60° - 80°, and the setting range of the fourth preset angle is 80° - 90°.

[0248] As a preferred example, the third preset angle can be set to 70°, and the fourth preset angle can be set to 90°.

[0249] Therefore, the target vehicle that satisfies 70° < atan(|vy| / |vx|) <= 90° is the screened target vehicle.

[0250] S4062. Calculate the distance distribution of the target vehicle relative to the same longitudinal position of the historical trajectory of the vehicle.

[0251] Since the target vehicles are distributed in one or more of the left area, right area, front area or rear area of the vehicle, this step specifically includes:

[0252] Calculate the thirteenth distance distribution of the target vehicle relative to the vehicle in the left area. This thirteenth distance distribution can be considered as the average lateral position of all target vehicles in the left area relative to the vehicle, denoted as Xleft4.

[0253] Calculate the fourteenth distance distribution of the target vehicle relative to the vehicle in the front or rear region. This fourteenth distance distribution can be considered as the average lateral position of all target vehicles in the front or rear region relative to the vehicle, denoted as Xcur4.

[0254] S4063. Based on the distance distribution of the target vehicle relative to the same longitudinal position on the historical trajectory of this vehicle, obtain the position of the lane centerline of the lane where the target vehicle is located.

[0255] Specifically, this step involves: based on the thirteenth distance distribution Xleft4 of the target vehicle relative to the vehicle in the left-side region, obtaining the thirteenth driving center position Xleft4 of the target vehicle in the left-side region, and setting the thirteenth driving center position Xleft4 of the target vehicle as the thirteenth lane centerline position of the vehicle in the left-side lane.

[0256] Based on the fourteenth distance distribution Xcur4 of the target vehicle relative to the vehicle in the front or rear area, the fourteenth driving center position Xcur4 of the target vehicle in the front or rear area is obtained, and the fourteenth driving center position Xcur4 of the target vehicle is set as the fourteenth lane centerline position in the same lane as the vehicle.

[0257] S4064. Shift the position of the lane centerline to both sides by half a lane to obtain the inner and outer lane line information of the lane where the target vehicle is located.

[0258] Specifically, this step involves shifting the center line of lane 13 half a lane to either side to obtain the inner and outer lane line information for the left lane of this vehicle (Lane13).

[0259] The center line of lane 14 is shifted half a lane to both sides to obtain the inner and outer lane line information for this lane, Lane 14.

[0260] S4065. Process the obtained inner and / or outer lane line information of the same lane to obtain the fused lane position distribution.

[0261] In this step, if at least two adjacent lanes are identified, then multiple inner and / or outer lane line information for the same lane will be identified.

[0262] Therefore, this step is specifically as follows: if two or more inner and / or outer lane line information in the same lane are identified, then a weighted average processing is required. That is, the inner and outer lane line information Lane13 of the left lane of this vehicle and the inner and outer lane line information Lane14 of this lane are weighted and averaged to finally obtain the fused lane position LaneFuse6.

[0263] In summary, the embodiments of the present invention dynamically track the trajectory of a moving target using millimeter-wave radar, identify the lane of the vehicle's current driving road based on the trajectory of the moving target, improve the lane identification accuracy using data processing algorithms, provide lane information for intelligent driving assistance of the vehicle, or provide fusion information for lane identification and other functions of other sensors.

[0264] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0265] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0266] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A lane recognition method based on millimeter-wave radar target tracking, characterized in that, Includes the following steps: Acquire the motion trajectory information of the target vehicle using millimeter-wave radar; Obtain the vehicle's motion information and determine the vehicle's motion status based on the motion information; Based on the vehicle's motion status, filter and statistically analyze target vehicles that are associated with the vehicle's motion status; Calculate the distance distribution of the target vehicle relative to the vehicle's lateral or longitudinal position to determine the position of the lane centerline of the lane where the target vehicle is located; The center line of the lane is shifted half a lane to both sides to obtain the inner and outer lane line information of the lane where the target vehicle is located; The obtained lane line information of multiple inner and / or outer lanes in the same lane is processed to obtain the fused lane position distribution.

2. The lane recognition method based on millimeter-wave radar target tracking according to claim 1, characterized in that, The acquired trajectory information of the target vehicle includes the position and velocity components of the target vehicle relative to its own coordinate system.

3. The lane recognition method based on millimeter-wave radar target tracking according to claim 2, characterized in that, The vehicle's motion information includes its speed, steering wheel angle, and angular velocity; The vehicle's motion states include straight driving, stationary driving, and cornering.

4. The lane recognition method based on millimeter-wave radar target tracking according to claim 3, characterized in that, Target vehicles associated with the motion state of this vehicle include target vehicles moving in the same direction as this vehicle, target vehicles moving in the opposite direction to this vehicle, target vehicles moving laterally to this vehicle, and target vehicles moving diagonally to this vehicle.

5. The lane recognition method based on millimeter-wave radar target tracking according to claim 4, characterized in that, Based on the vehicle's motion status, target vehicles associated with the vehicle's motion status are specifically selected and statistically analyzed, including: When the vehicle is traveling straight, target vehicles moving in the same direction, opposite direction, or laterally are selected based on the speed component of the target vehicles, and the number of target vehicles moving in the same direction, opposite direction, or laterally is counted. When the vehicle is stationary, target vehicles moving diagonally to the vehicle are selected based on the speed component of the target vehicle, and the number of target vehicles moving diagonally to the vehicle is counted. When the vehicle is traveling on a curve, target vehicles moving in the same or opposite direction as the vehicle are selected based on the speed component of the target vehicle, and the number of target vehicles moving in the same or opposite direction as the vehicle is counted.

6. The lane recognition method based on millimeter-wave radar target tracking according to claim 5, characterized in that, The specific criteria for selecting target vehicles include: When the vehicle is traveling straight, the selection criteria for target vehicles moving in the same direction as the vehicle are: the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is greater than the second preset threshold. When the vehicle is traveling straight, the selection criteria for target vehicles moving in the opposite direction are: the lateral velocity component of the target vehicle is less than the first preset threshold, and the longitudinal velocity component of the target vehicle is less than the third preset threshold. When the vehicle is traveling straight, the selection criteria for target vehicles moving laterally to the vehicle are: the lateral velocity component of the target vehicle is greater than the first preset threshold, and the longitudinal velocity component of the target vehicle is less than the second preset threshold. When the vehicle is stationary, the selection criteria for target vehicles moving diagonally to the vehicle are: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the first preset angle and the second preset angle. When the vehicle is in a curve, the selection criteria for target vehicles moving in the same direction as the vehicle are: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the third preset angle and the fourth preset angle. When the vehicle is in a curve, the selection criteria for the target vehicle moving in the opposite direction is: the arctangent of the ratio of the longitudinal velocity component to the lateral velocity component of the target vehicle is between the third preset angle and the fourth preset angle.

7. The lane recognition method based on millimeter-wave radar target tracking according to claim 5, characterized in that, The lateral or longitudinal position of the target vehicle relative to the vehicle specifically includes: the target vehicle being located in the left side area, right side area, first front side area, second front side area, first rear side area, or second rear side area of ​​the vehicle. The first front region is close to the front of the vehicle, and the second front region is far from the front of the vehicle. The first rear side region is close to the rear of the vehicle, and the second rear side region is far from the rear of the vehicle.

8. The lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When the vehicle is traveling straight, determining the position of the lane centerline based on a target vehicle moving in the same direction as the vehicle includes: Calculate the first distance distribution of the target vehicle relative to the vehicle in the left-side region to obtain the first driving center position of the target vehicle in the left-side region, and set the first driving center position of the target vehicle as the first lane centerline position of the vehicle in the left-side lane. Calculate the second distance distribution of the target vehicle relative to the vehicle on the right side of the region to obtain the second driving center position of the target vehicle in the right side of the region, and set the second driving center position of the target vehicle as the second lane center line position of the right lane of the vehicle. Calculate the third distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the third driving center position of the target vehicle in the front or rear region, and set the third driving center position of the target vehicle as the center line position of the third lane in the same lane as the vehicle.

9. A lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When the vehicle is traveling straight, determining the position of the lane centerline relative to a target vehicle moving in the opposite direction specifically includes: Calculate the fourth distance distribution of the target vehicle relative to the vehicle in the left-hand region to obtain the fourth driving center position of the target vehicle in the left-hand region, and set the fourth driving center position of the target vehicle as the fourth lane center line position of the vehicle in the left-hand lane. Calculate the fifth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the fifth driving center position of the target vehicle in the front or rear region, and set the fifth driving center position of the target vehicle as the position of the fifth lane center line in the same lane as the vehicle.

10. A lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When the vehicle is traveling straight, determining the position of the lane centerline relative to a target vehicle moving laterally relative to the vehicle specifically includes: Calculate the sixth distance distribution of the target vehicle relative to the vehicle in the first front area to obtain the sixth driving center position of the target vehicle in the first front area, and set the sixth driving center position of the target vehicle as the sixth lane centerline position of the vehicle in the first front lane.

11. A lane recognition method based on millimeter-wave radar target tracking according to claim 10, characterized in that, Also includes: Calculate the seventh distance distribution of the target vehicle relative to the vehicle in the second front area to obtain the seventh driving center position of the target vehicle in the second front area, and set the seventh driving center position of the target vehicle as the seventh lane center line position of the vehicle in the second front lane.

12. The lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When this vehicle is stationary, determining the position of the lane centerline relative to a target vehicle moving diagonally relative to this vehicle specifically includes: Calculate the eighth distance distribution of the target vehicle relative to the vehicle in the first rear side region to obtain the eighth driving center position of the target vehicle in the first rear side region, and set the eighth driving center position of the target vehicle as the eighth lane center line position of the vehicle in the first rear side lane.

13. A lane recognition method based on millimeter-wave radar target tracking according to claim 12, characterized in that, Also includes: Calculate the ninth distance distribution of the target vehicle relative to the vehicle in the second rear region to obtain the ninth driving center position of the target vehicle in the second rear region, and set the ninth driving center position of the target vehicle as the ninth lane centerline position of the vehicle in the second rear lane.

14. A lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When this vehicle is traveling on a curve, determining the position of the lane centerline based on a target vehicle moving in the same direction as this vehicle specifically includes: Calculate the tenth distance distribution of the target vehicle relative to the vehicle in the left-hand region to obtain the tenth driving center position of the target vehicle in the left-hand region, and set the tenth driving center position of the target vehicle as the tenth lane centerline position of the vehicle's left-hand lane. Calculate the eleventh distance distribution of the target vehicle relative to the vehicle on the right side of the region to obtain the eleventh driving center position of the target vehicle in the right side of the region, and set the eleventh driving center position of the target vehicle as the eleventh lane centerline position of the right lane of the vehicle. Calculate the twelfth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the twelfth driving center position of the target vehicle in the front or rear region, and set the twelfth driving center position of the target vehicle as the twelfth lane centerline position in the same lane as the vehicle.

15. A lane recognition method based on millimeter-wave radar target tracking according to claim 7, characterized in that, When this vehicle is traveling on a curve, determining the position of the lane centerline relative to a target vehicle moving in the opposite direction specifically includes: Calculate the thirteenth distance distribution of the target vehicle relative to the vehicle in the left-hand region to obtain the thirteenth driving center position of the target vehicle in the left-hand region, and set the thirteenth driving center position of the target vehicle as the thirteenth lane centerline position of the vehicle in the left-hand lane. Calculate the fourteenth distance distribution of the target vehicle relative to the vehicle in the front or rear region to obtain the fourteenth driving center position of the target vehicle in the front or rear region, and set the fourteenth driving center position of the target vehicle as the fourteenth lane centerline position in the same lane as the vehicle.

16. A lane recognition method based on millimeter-wave radar target tracking according to claim 3, characterized in that, When the vehicle is traveling on a curve, the acquisition of the inner and outer lane markings of the target vehicle's lane specifically includes: The position of the lane center line is shifted half a lane to both sides of the road curvature with reference to the vehicle's historical trajectory line, providing information on the inner and outer lane lines of the target vehicle's lane.

17. A lane recognition method based on millimeter-wave radar target tracking according to claim 6, characterized in that, The first preset threshold, the second preset threshold, the third preset threshold, the first preset angle, the second preset angle, the third preset angle, and the fourth preset angle are set according to the vehicle speed, driving environment and / or road conditions and / or the computing power of the millimeter-wave radar. The first preset threshold is set within the range of 0.8-1.5 m / s; The second preset threshold is set within the range of 0.8-1.5 m / s; The setting range of the third preset threshold is -0.8 to 1.5 m / s; The first preset angle is set within the range of 15°-30°; The second preset angle is set within the range of 60°-80°; The setting range of the third preset angle is 60°-80°; The fourth preset angle is set within the range of 80°-90°.

Images