Reversing passing collision early warning method and system based on millimeter wave radar

Abstract

The invention relates to a reversing passing collision early warning method and system based on a millimeter wave radar. The method comprises the following steps: acquiring target point cloud information of a current frame in real time through a millimeter-wave radar; establishing a vehicle coordinate system, and calculating a left blind area and a right blind area; according to the target point cloud information, whether a target is located in the left blind area and / or the right blind area in the current frame is judged; responding to the situation that the target is located in the left blind area and / or the right blind area, and obtaining position information and size information of the target in the vehicle coordinate system; constructing a polygonal region corresponding to the target; calculating the intersection-to-union ratio; in response to the intersection-to-union ratio being greater than zero, determining that the target has entered a blind area; when the target enters the blind area, collision risk judgment is carried out according to the moving tracks of the vehicle and the target. The target entering the left and right blind areas can be rapidly detected, and the reliability and the accuracy of the early warning system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a reversing through-the-road collision warning method and system based on millimeter-wave radar. Background Art

[0002] As the focus of intelligent driving perception technology research, the vehicle reversing warning system generally perceives the surrounding environment through millimeter-wave radar, ultrasonic wave, laser radar and vehicle-mounted cameras, and monitors the traffic conditions around the vehicle in real time, especially in blind spots. Once the system detects a possible collision risk, such as a vehicle approaching, RCTA (Rear Cross Traffic Alert) will immediately activate the warning mechanism and warn the driver through flashing lights and sound reminders to avoid accidental collisions that may occur during reversing. In addition, the function of the RCTA system is not limited to reversing protection, it can also intelligently identify the dynamic environment around the vehicle, including pedestrians, animals or other motor vehicles that suddenly break in. If the system determines that there is a risk of collision, it will intervene quickly to assist the vehicle to slow down or stop to ensure driving safety.

[0003] Although the existing reversing warning system can improve driving safety to a certain extent, in complex traffic environments, multi-sensor data may be interfered by false targets, environmental noise and occlusion factors, resulting in misjudgment or missed judgment. Some systems do not have a high enough detection resolution for targets in blind spots, especially when the targets are small or moving at a low speed, and are prone to not being able to identify collision risks in a timely and accurate manner. Summary of the invention

[0004] To this end, the present invention provides a reversing cross-traffic collision warning method and system based on millimeter-wave radar, which can quickly detect targets entering the left and right blind spots and comprehensively judge the collision risk. At the same time, by effectively shielding interference factors such as false targets, the reliability and accuracy of the warning system are significantly improved, thereby providing drivers with more timely and accurate safety warnings.

[0005] In order to solve the above technical problems, the present invention provides a reversing through-the-road collision warning method based on millimeter-wave radar, comprising:

[0006] Acquire the target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar;

[0007] Establish the vehicle coordinate system and calculate the left and right blind spots;

[0008] According to the target point cloud information, determining whether the target is located in the left blind area and / or the right blind area in the current frame;

[0009] In response to the target being located in the left blind spot and / or the right blind spot, obtaining position information and size information of the target in the vehicle coordinate system according to the target point cloud information;

[0010] Constructing a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system;

[0011] Calculate the intersection-and-joint ratio of the polygonal area corresponding to the target and the left blind area and / or the right blind area;

[0012] In response to the intersection-over-combination ratio being greater than zero, determining that the target has entered a blind spot;

[0013] When the target has entered the blind spot, a collision risk determination is performed based on the respective running trajectories of the vehicle and the target.

[0014] In one embodiment of the present invention, a vehicle coordinate system is established, and a left blind spot and a right blind spot are calculated, including:

[0015] The vehicle coordinate system is defined as follows: the center of the rear axle of the vehicle is taken as the origin, the longitudinal front of the vehicle is the positive direction of the X axis, the lateral left side of the vehicle is the positive direction of the Y axis, and the vertical direction upward is the positive direction of the Z axis;

[0016] Calculate the coordinates of the first vertex in the left blind spot, whose x coordinate is -DISTANCEREAR+EGOCARLENGTH and y coordinate is 50;

[0017] The coordinates of the second vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is 50;

[0018] The coordinates of the third vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0;

[0019] The coordinates of the fourth vertex in the left blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0;

[0020] The coordinates of the first vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0;

[0021] The coordinates of the second vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0;

[0022] The coordinates of the third vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is -50;

[0023] The coordinates of the fourth vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR+EGOCARLENGTH, and its y coordinate is -50;

[0024] Among them, EGOCARLENGTH represents the length of the vehicle, in meters, and DISTANCEREAR represents the distance from the rear axle to the rear edge, in meters.

[0025] In one embodiment of the present invention, obtaining the position information and size information of the target in the vehicle coordinate system according to the target point cloud information includes:

[0026] Post-processing the target point cloud information to obtain the longitudinal position coordinates and the lateral position coordinates of the center of the target in the vehicle coordinate system, as well as the longitudinal length and the lateral width of the target;

[0027] When the length or width of the object is less than or equal to zero, it is set to 1.

[0028] In one embodiment of the present invention, constructing a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system includes:

[0029] Based on the longitudinal position coordinates and the lateral position coordinates of the center of the target in the vehicle coordinate system, and the longitudinal length and lateral width of the target, the coordinates of the four vertices of the target are calculated to obtain a polygonal area corresponding to the target.

[0030] In one embodiment of the present invention, when the target has entered the blind spot, a collision risk determination is performed according to the respective running trajectories of the vehicle and the target, including:

[0031] In response to the self-vehicle and the target both being in a straight-ahead state, determining whether the target and the self-vehicle belong to the same lane;

[0032] In response to the target and the vehicle being in the same lane, determining whether a velocity component of the target along a longitudinal direction, i.e., an X-axis direction, is greater than 0;

[0033] In response to a velocity component of the target in the longitudinal direction being greater than 0, calculating a vehicle collision time;

[0034] In response to the vehicle collision time being less than a preset first time threshold, a warning signal is triggered.

[0035] In one embodiment of the present invention, it also includes:

[0036] In response to the self-vehicle not being in a straight-moving state, determining whether there is an intersection between the self-vehicle driving trajectory and the target driving trajectory;

[0037] In response to the existence of an intersection between the own vehicle driving trajectory and the target driving trajectory, determining whether there is a possibility of collision according to the position information and driving direction of the target in the vehicle coordinate system;

[0038] In response to the possibility of a collision, whether to trigger a warning signal is determined based on a time difference between the time when the vehicle arrives at the intersection and the time when the target arrives at the intersection and the time when the vehicle arrives at the intersection.

[0039] In one embodiment of the present invention, determining whether to trigger a warning signal according to the time difference between the time when the vehicle arrives at the intersection and the time when the target arrives at the intersection and the time when the vehicle arrives at the intersection includes:

[0040] In response to the difference between the time when the ego vehicle arrives at the intersection and the time when the target arrives at the intersection being less than or equal to a preset second time threshold, determining whether the time when the ego vehicle arrives at the intersection is less than a preset third time threshold;

[0041] In response to the time taken for the vehicle to reach the intersection being less than a preset third time threshold, a warning signal is triggered.

[0042] In one embodiment of the present invention, determining whether the vehicle's driving trajectory and the target driving trajectory have an intersection includes:

[0043] A simultaneous equation is established for the vehicle's driving trajectory and the target driving trajectory, and the discriminant δ is calculated; when δ≤0, it indicates that the two trajectories have no intersection; when δ>0, it indicates that the simultaneous equation has two different solutions, that is, there is an intersection.

[0044] In one embodiment of the present invention, determining whether there is a possibility of collision based on the position information and the driving direction of the target in the vehicle coordinate system includes:

[0045] When two intersection points (x1, y1) and (x2, y2) are detected, if x1 and x2 are both greater than 0, it is determined that there is no possibility of collision;

[0046] When x1 is greater than or equal to 0 and x2 is less than 0, if the lateral position coordinate of the target is greater than y2 and is moving to the right, or if the lateral position coordinate of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision;

[0047] When x1 and x2 are both less than 0:

[0048] When y1 is less than or equal to y2, if the lateral position coordinate of the target is greater than y1 and is moving to the right, or the lateral position coordinate of the target is less than or equal to y1 and is moving to the left, it is determined that there is a possibility of collision;

[0049] When y1 is greater than y2, if the lateral position coordinate of the target is greater than y2 and is moving to the right, or the lateral position coordinate of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision.

[0050] The present invention also provides a millimeter-wave radar-based reverse vehicle cross-traffic collision warning system, comprising:

[0051] A millimeter wave radar module, used to obtain target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar;

[0052] Blind spot area calculation module, used to establish the vehicle coordinate system and calculate the left blind spot and the right blind spot;

[0053] a target position information and size acquisition module, configured to determine whether the target is located in the left blind spot and / or the right blind spot in the current frame according to the target point cloud information; and in response to the target being located in the left blind spot and / or the right blind spot, obtain the position information and size information of the target in the vehicle coordinate system according to the target point cloud information;

[0054] A target polygon generation module, used to construct a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system;

[0055] A blind spot determination module, configured to calculate an intersection-and-union ratio between a polygonal area corresponding to the target and the left blind spot and / or the right blind spot; in response to the intersection-and-union ratio being greater than zero, determining that the target has entered the blind spot;

[0056] The collision risk determination module is used to determine the collision risk according to the respective running trajectories of the vehicle and the target when the target has entered the blind spot.

[0057] The above technical solution of the present invention has the following advantages compared with the prior art:

[0058] The millimeter-wave radar-based reversing collision warning method and system described in the present invention screens and judges the lane, speed and posture information of the target according to the point cloud clustering tracking results, judges whether the target enters the alarm areas on the left and right sides, judges whether the target belongs to the lane and whether there is a possibility of collision, calculates whether the collision time is less than a threshold and makes a corresponding warning signal, further improving the reliability and anti-interference ability of the results.

[0059] Since the sensor used is a millimeter-wave radar, it has significant external anti-interference capabilities compared to other types of sensors and can better adapt to traffic monitoring needs in complex road environments. In addition, the point cloud occupancy probability is quickly derived through the radar equation. This method has the advantages of low hardware requirements, short running time and high reliability of results. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings.

[0061] Figure 1 It is a flow chart of the reversing vehicle through-the-road collision warning method based on millimeter-wave radar of the present invention.

[0062] Figure 2 It is a schematic diagram of the vehicle coordinate system of the present invention.

[0063] Figure 3 It is a schematic diagram of the blind spot alarm area of ​​the reversing warning function of the present invention. DETAILED DESCRIPTION

[0064] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0065] Example 1

[0066] Reference Figure 2 As shown, it is a schematic diagram of the coordinate system in the present invention. The direction parallel to the front of the vehicle is the X direction, the front is the positive direction, and the back is the negative direction. The direction perpendicular to the front of the vehicle is the Y direction, and the left side is the positive direction and the right side is the negative direction. The positive direction of the Z axis is the direction perpendicular to the earth and upward. The millimeter wave radar device involved in the present invention is installed at the rear bumper position of the vehicle.

[0067] Existing reversing warning systems (such as RCTA) can usually use millimeter-wave radar, ultrasound, camera or lidar for blind spot detection. However, traditional reversing warnings still have shortcomings in detection efficiency, anti-interference ability, and tracking and collision judgment of moving targets. Especially when a vehicle is reversing in a complex traffic scene, there may be vehicles, pedestrians or non-motor vehicles traveling sideways in the surrounding area. If the fusion analysis of the target trajectory is inaccurate or there is missed detection / false detection, it will lead to collision risks. To address these problems, an efficient and more reliable reversing collision warning system and method based on millimeter-wave radar is proposed.

[0068] Reference Figure 1 As shown, a reversing through-the-road collision warning method based on millimeter-wave radar includes:

[0069] Step S1, acquiring target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar.

[0070] It can be understood that the radar works periodically and outputs target point cloud information. The duration of one completion is called a frame, which will be repeated continuously during the operation of the radar.

[0071] Step S2: Establish a vehicle coordinate system and calculate the left blind spot (leftBsdRoi) and the right blind spot (rightBsdRoi).

[0072] Specifically, the vehicle coordinate system is defined as follows: the center of the vehicle's rear axle is taken as the origin, the longitudinal front of the vehicle is the positive direction of the X-axis, the lateral left side of the vehicle is the positive direction of the Y-axis, and the positive direction of the Z-axis is perpendicular to the ground and upward.

[0073] Calculate the coordinates of the first vertex in the left blind spot, whose x coordinate is -DISTANCEREAR+EGOCARLENGTH and y coordinate is 50;

[0074] The coordinates of the second vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is 50;

[0075] The coordinates of the third vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0;

[0076] The coordinates of the fourth vertex in the left blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0;

[0077] The coordinates of the first vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0;

[0078] The coordinates of the second vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0;

[0079] The coordinates of the third vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is -50;

[0080] The coordinates of the fourth vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR+EGOCARLENGTH, and its y coordinate is -50;

[0081] Among them, EGOCARLENGTH represents the length of the vehicle, in meters, and DISTANCEREAR represents the distance from the rear axle to the rear edge, in meters.

[0082] That is, the coordinates of the four vertices in the left blind spot are:

[0083] leftRctaRoi.points[0].x=-DISTANCEREAR+EGOCARLENGTH;

[0084] leftRctaRoi.points[0].y=50;

[0085] leftRctaRoi.points[1].x=-8-DISTANCEREAR-EGOCARLENGTH;

[0086] leftRctaRoi.points[1].y=50;

[0087] leftRctaRoi.points[2].x=-8-DISTANCEREAR;

[0088] leftRctaRoi.points[2].y=0;

[0089] leftRctaRoi.points[3].x=-DISTANCEREAR;

[0090] leftRctaRoi.points[3].y=0;

[0091] The coordinates of the four vertices in the right blind area are:

[0092] rightRctaRoi.points[0].x=-DISTANCEREAR;

[0093] rightRctaRoi.points[0].y=0;

[0094] rightRctaRoi.points[1].x=-8-DISTANCEREAR;

[0095] rightRctaRoi.points[1].y=0;

[0096] rightRctaRoi.points[2].x=-8-DISTANCEREAR-EGOCARLENGTH;

[0097] rightRctaRoi.points[2].y=-50;

[0098] rightRctaRoi.points[3].x=-DISTANCEREAR+EGOCARLENGTH;

[0099] rightRctaRoi.points[3].y=-50;

[0100] Step S3: judging whether the target is located in the left blind spot and / or the right blind spot in the current frame according to the target point cloud information.

[0101] Step S4: in response to the target being located in the left blind spot and / or the right blind spot, obtaining position information and size information of the target in the vehicle coordinate system according to the target point cloud information.

[0102] Specifically include:

[0103] Post-process the target point cloud information to obtain the longitudinal position coordinates (posX_w, indicating the coordinates of the center position of the target along the X-axis, i.e., the forward direction of the vehicle) and lateral position coordinates (posY_w) of the center of the target in the vehicle coordinate system, as well as the longitudinal length (objLength) and lateral width (objWidth) of the target;

[0104] When the length or width of the target is less than or equal to zero, it is set to 1 to prevent errors in subsequent calculations.

[0105] Step S5: construct a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system. Specifically comprising:

[0106] Based on the longitudinal position coordinates and the lateral position coordinates of the center of the target in the vehicle coordinate system, and the longitudinal length and lateral width of the target, the coordinates of the four vertices of the target are calculated to obtain a polygonal area (objPoly) corresponding to the target.

[0107] Step S6: Calculate the intersection-and-joint ratio of the polygonal area corresponding to the target and the left blind spot and / or the right blind spot.

[0108] Specifically, the polygon_iou function is used to calculate the intersection over union (IoU) of the target polygon objPoly with the left blind area leftBsdRoi and the right blind area rightBsdRoi.

[0109] Step S7: In response to the intersection over union (IoU) ratio being greater than zero, determining that the target has entered a blind spot.

[0110] Step S8: When the target has entered the blind spot, a collision risk determination is performed based on the respective running trajectories of the vehicle and the target. Specifically, the following steps are performed:

[0111] Step S9: In response to the vehicle and the target both being in a straight-ahead state, determining whether the target and the vehicle belong to the same lane.

[0112] It is understandable that if both the ego vehicle and the target vehicle are traveling in a straight-ahead manner, the most direct and common scenario for a collision is when the two vehicles are in the same lane and relatively close to each other. If the lanes are different, a collision usually does not occur in a straight-ahead state.

[0113] Step S10: In response to the target and the vehicle being in the same lane, determining whether a velocity component of the target along the longitudinal direction, i.e., the X-axis direction, is greater than 0.

[0114] If the component of the target velocity on the X-axis is greater than 0, it indicates that it is moving in the direction of the ego vehicle (for example, catching up from behind, or vice versa, the ego vehicle is reversing to approach the target).

[0115] Step S11 : In response to the velocity component of the target in the longitudinal direction being greater than 0, calculating the vehicle collision time (collisionTime, the distance of the target in the x direction divided by the velocity).

[0116] Collision Time is one of the key indicators for predicting collision risk. If the target and the ego vehicle are traveling in the same lane and the relative speed is clear in a simple scenario, the collision time can be obtained by dividing the relative distance (X-axis distance) between the target and the ego vehicle by the relative speed.

[0117] Step S12: In response to the vehicle collision time being less than a preset first time threshold, triggering a warning signal.

[0118] Step S13: In response to the vehicle not being in a straight-moving state, determining whether there is an intersection between the vehicle's driving trajectory and the target driving trajectory.

[0119] Step S14: In response to the existence of an intersection between the own vehicle driving trajectory and the target driving trajectory, determining whether there is a possibility of collision based on the position information and driving direction of the target in the vehicle coordinate system.

[0120] Specifically, determining whether there is an intersection between the vehicle's driving trajectory and the target driving trajectory includes:

[0121] A simultaneous equation is established for the vehicle's driving trajectory and the target driving trajectory, and the discriminant δ is calculated; when δ≤0, it indicates that the two trajectories have no intersection and no alarm is required; when δ>0, it indicates that the simultaneous equation has two different solutions (i.e., the two trajectories have an intersection mathematically), that is, there is an intersection.

[0122] Specifically, determining whether there is a possibility of collision according to the position information and the driving direction of the target in the vehicle coordinate system includes:

[0123] When two intersection points (x1, y1) and (x2, y2) are detected, if x1 and x2 are both greater than 0, it is determined that there is no possibility of collision;

[0124] When x1 is greater than or equal to 0 and x2 is less than 0, if the lateral position coordinate (posY_w) of the target is greater than y2 and is moving to the right, or the lateral position coordinate (posY_w) of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision;

[0125] When x1 and x2 are both less than 0:

[0126] When y1 is less than or equal to y2, if the target's lateral position coordinate (posY_w) is greater than y1 and it is moving to the right, or the target's lateral position coordinate (posY_w) is less than or equal to y1 and it is moving to the left, it is determined that there is a possibility of collision;

[0127] When y1 is greater than y2, if the lateral position coordinate (posY_w) of the target is greater than y2 and is moving to the right, or the lateral position coordinate (posY_w) of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision.

[0128] Step S15: In response to the possibility of collision, determine whether to trigger a warning signal according to the time difference between the time when the vehicle arrives at the intersection and the time when the target arrives at the intersection and the time when the vehicle arrives at the intersection. Specifically, it includes:

[0129] In response to the difference between the time when the ego vehicle arrives at the intersection and the time when the target arrives at the intersection being less than or equal to a preset second time threshold (2 seconds in this embodiment), determining whether the time when the ego vehicle arrives at the intersection is less than a preset third time threshold (3 seconds in this embodiment);

[0130] In response to the time taken for the vehicle to reach the intersection being less than a preset third time threshold, a warning signal is triggered.

[0131] Step S16, determine whether all targets have been determined. If all targets have been determined, end the determination logic of the current frame; otherwise, start from step S4 and execute sequentially.

[0132] Step S17, output the vehicle reversing warning effect, and end.

[0133] This embodiment makes full use of the environmental perception advantage of millimeter-wave radar, and combines the target motion state and lane information to make collision judgments, so as to provide reliable real-time warnings for the left and right blind spots when the vehicle is reversing.

[0134] Example 2

[0135] Based on the same inventive concept, this embodiment provides a reversing through collision warning system based on millimeter-wave radar, and the principle of solving the problem is similar to the reversing through collision warning method based on millimeter-wave radar, and the repeated parts are not repeated here.

[0136] This embodiment provides a millimeter-wave radar-based reversing cross-traffic collision warning system, including:

[0137] A millimeter wave radar module, used to obtain target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar;

[0138] Blind spot area calculation module, used to establish the vehicle coordinate system and calculate the left blind spot and the right blind spot;

[0139] a target position information and size acquisition module, configured to determine whether the target is located in the left blind spot and / or the right blind spot in the current frame according to the target point cloud information; and in response to the target being located in the left blind spot and / or the right blind spot, obtain the position information and size information of the target in the vehicle coordinate system according to the target point cloud information;

[0140] A target polygon generation module, used to construct a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system;

[0141] A blind spot determination module, configured to calculate an intersection-and-union ratio between a polygonal area corresponding to the target and the left blind spot and / or the right blind spot; in response to the intersection-and-union ratio being greater than zero, determining that the target has entered the blind spot;

[0142] The collision risk determination module is used to determine the collision risk according to the respective running trajectories of the vehicle and the target when the target has entered the blind spot.

[0143] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0144] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0145] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0147] Finally, it should be noted that the above specific implementation methods are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A reversing vehicle collision warning method based on millimeter wave radar, characterized in that: include: Acquire the target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar; Establish the vehicle coordinate system and calculate the left and right blind spots; According to the target point cloud information, determining whether the target is located in the left blind area and / or the right blind area in the current frame; In response to the target being located in the left blind spot and / or the right blind spot, obtaining position information and size information of the target in the vehicle coordinate system according to the target point cloud information; Constructing a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system; Calculate the intersection-and-joint ratio of the polygonal area corresponding to the target and the left blind area and / or the right blind area; In response to the intersection-over-combination ratio being greater than zero, determining that the target has entered a blind spot; When the target has entered the blind spot, a collision risk determination is performed based on the respective running trajectories of the vehicle and the target.

2. The method for warning a vehicle's reversing collision based on millimeter-wave radar according to claim 1, characterized in that: Establish the vehicle coordinate system and calculate the left and right blind spots, including: The vehicle coordinate system is defined as follows: the center of the rear axle of the vehicle is taken as the origin, the longitudinal front of the vehicle is the positive direction of the X axis, the lateral left side of the vehicle is the positive direction of the Y axis, and the vertical direction upward is the positive direction of the Z axis; Calculate the coordinates of the first vertex in the left blind spot, whose x coordinate is -DISTANCEREAR+EGOCARLENGTH and y coordinate is 50; The coordinates of the second vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is 50; The coordinates of the third vertex in the left blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0; The coordinates of the fourth vertex in the left blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0; The coordinates of the first vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR and its y coordinate is 0; The coordinates of the second vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR, and its y coordinate is 0; The coordinates of the third vertex in the right blind spot are calculated, and its x coordinate is -8-DISTANCEREAR-EGOCARLENGTH, and its y coordinate is -50; The coordinates of the fourth vertex in the right blind spot are calculated, and its x coordinate is -DISTANCEREAR+EGOCARLENGTH, and its y coordinate is -50; Among them, EGOCARLENGTH represents the length of the vehicle, in meters, and DISTANCEREAR represents the distance from the rear axle to the rear edge, in meters.

3. The millimeter-wave radar-based back-end collision warning method according to claim 1, characterized in that: According to the target point cloud information, the position information and size information of the target in the vehicle coordinate system are obtained, including: Post-processing the target point cloud information to obtain the longitudinal position coordinates and the lateral position coordinates of the center of the target in the vehicle coordinate system, as well as the longitudinal length and the lateral width of the target; When the length or width of the object is less than or equal to zero, it is set to 1.

4. The method for warning a reversing vehicle through-the-road collision based on millimeter-wave radar according to claim 1, characterized in that: Constructing a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system, including: Based on the longitudinal position coordinates and the lateral position coordinates of the center of the target in the vehicle coordinate system, and the longitudinal length and lateral width of the target, the coordinates of the four vertices of the target are calculated to obtain a polygonal area corresponding to the target.

5. The method for warning a reversing vehicle through-the-road collision based on millimeter-wave radar according to claim 1, characterized in that: When the target has entered the blind spot, a collision risk determination is performed based on the respective running trajectories of the vehicle and the target, including: In response to the self-vehicle and the target both being in a straight-ahead state, determining whether the target and the self-vehicle belong to the same lane; In response to the target and the vehicle being in the same lane, determining whether a velocity component of the target along a longitudinal direction, i.e., an X-axis direction, is greater than 0; In response to a velocity component of the target in the longitudinal direction being greater than 0, calculating a vehicle collision time; In response to the vehicle collision time being less than a preset first time threshold, a warning signal is triggered.

6. The millimeter-wave radar-based back-up vehicle collision warning method according to claim 5, characterized in that: Also includes: In response to the self-vehicle not being in a straight-moving state, determining whether there is an intersection between the self-vehicle driving trajectory and the target driving trajectory; In response to the existence of an intersection between the own vehicle driving trajectory and the target driving trajectory, determining whether there is a possibility of collision according to the position information and driving direction of the target in the vehicle coordinate system; In response to the possibility of a collision, whether to trigger a warning signal is determined based on a time difference between the time when the vehicle arrives at the intersection and the time when the target arrives at the intersection and the time when the vehicle arrives at the intersection.

7. The millimeter-wave radar-based back-end collision warning method according to claim 6, characterized in that: Determining whether to trigger a warning signal according to a time difference between the time when the vehicle arrives at the intersection and the time when the target arrives at the intersection and the time when the vehicle arrives at the intersection includes: In response to the difference between the time when the ego vehicle arrives at the intersection and the time when the target arrives at the intersection being less than or equal to a preset second time threshold, determining whether the time when the ego vehicle arrives at the intersection is less than a preset third time threshold; In response to the time taken for the vehicle to reach the intersection being less than a preset third time threshold, a warning signal is triggered.

8. The millimeter-wave radar-based back-up vehicle collision warning method according to claim 6, characterized in that: Determining whether the vehicle's driving trajectory and the target driving trajectory have an intersection includes: A simultaneous equation is established for the vehicle's driving trajectory and the target driving trajectory, and the discriminant δ is calculated; when δ≤0, it indicates that the two trajectories have no intersection; when δ>0, it indicates that the simultaneous equation has two different solutions, that is, there is an intersection.

9. The method for warning a vehicle crossing back collision based on millimeter wave radar according to claim 8, characterized in that: Determining whether there is a possibility of collision based on the position information and the driving direction of the target in the vehicle coordinate system includes: When two intersection points (x1, y1) and (x2, y2) are detected, if x1 and x2 are both greater than 0, it is determined that there is no possibility of collision; When x1 is greater than or equal to 0 and x2 is less than 0, if the lateral position coordinate of the target is greater than y2 and is moving to the right, or if the lateral position coordinate of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision; When x1 and x2 are both less than 0: When y1 is less than or equal to y2, if the lateral position coordinate of the target is greater than y1 and is moving to the right, or the lateral position coordinate of the target is less than or equal to y1 and is moving to the left, it is determined that there is a possibility of collision; When y1 is greater than y2, if the lateral position coordinate of the target is greater than y2 and is moving to the right, or the lateral position coordinate of the target is less than or equal to y2 and is moving to the left, it is determined that there is a possibility of collision.

10. A millimeter-wave radar-based reversing collision warning system, characterized in that: include: A millimeter wave radar module, used to obtain target point cloud information of the current frame in real time through the millimeter wave radar; wherein the target point cloud information includes the angle of the target point, the radial velocity and the distance relative to the millimeter wave radar; Blind spot area calculation module, used to establish the vehicle coordinate system and calculate the left blind spot and the right blind spot; a target position information and size acquisition module, configured to determine whether the target is located in the left blind spot and / or the right blind spot in the current frame according to the target point cloud information; and in response to the target being located in the left blind spot and / or the right blind spot, obtain the position information and size information of the target in the vehicle coordinate system according to the target point cloud information; A target polygon generation module, used to construct a polygonal area corresponding to the target according to the position information and size information of the target in the vehicle coordinate system; A blind spot determination module, configured to calculate an intersection-and-union ratio between a polygonal area corresponding to the target and the left blind spot and / or the right blind spot; in response to the intersection-and-union ratio being greater than zero, determining that the target has entered the blind spot; The collision risk determination module is used to determine the collision risk according to the respective running trajectories of the vehicle and the target when the target has entered the blind spot.

Images