Vehicle driving track determination method and device, processor, electronic equipment and vehicle

Abstract

The application discloses a driving track determination method and device of a vehicle, a processor, an electronic device and the vehicle. The method comprises the following steps: acquiring space-time perception information of a rear vehicle relative to the vehicle; detecting the rear vehicle based on the space-time perception information to obtain a detection result; in response to the detection result that the rear vehicle is in a risk area, adjusting an initial detection threshold of a radar detection device in the vehicle based on the space-time perception information to obtain a target detection threshold, wherein the risk area is used to represent an area in which echo power of the radar detection device is attenuated, and the target detection threshold is lower than the initial detection threshold; and determining a driving track of the rear vehicle in response to the space-time perception information meeting the target detection threshold. The application solves the technical problem of low determination accuracy of the driving track of the rear vehicle.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a method, apparatus, processor, electronic device, and vehicle for determining the driving trajectory of a vehicle. Background Technology

[0002] Currently, in areas directly behind a vehicle or within a very small angle of less than 15° relative to the vehicle's direction of travel, radar signals are prone to multiple propagation paths due to road surface reflections and reflections from the vehicle body (e.g., the rear bumper). This results in severe constructive and destructive interference of the receiver signal, causing the target echo signal-to-noise ratio of the radar signal to fluctuate periodically and drastically depending on the relative position of the vehicle and the vehicles behind it, as well as the road conditions (e.g., dry or wet).

[0003] In related technologies, fixed or distance-dependent detection thresholds are often used. These thresholds cannot effectively handle the instantaneous drop in signal-to-noise ratio (SNR) within a specific area. This causes real targets (e.g., vehicles behind) to be misjudged as noise because the instantaneous SNR falls below the threshold, resulting in a cycle of target loss and recapture, and causing discontinuous driving trajectories. Therefore, the technical problem of low accuracy in determining the driving trajectory of vehicles behind remains.

[0004] There is currently no effective solution to the aforementioned technical problems. Summary of the Invention

[0005] This application provides a method, apparatus, processor, electronic device, and vehicle for determining the driving trajectory of a vehicle, so as to at least solve the technical problem of low accuracy in determining the driving trajectory of following vehicles.

[0006] According to one aspect of the embodiments of this application, a method for determining the driving trajectory of a vehicle is provided. The method may include: acquiring spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle; detecting the vehicles behind based on the spatiotemporal perception information to obtain a detection result; in response to the detection result indicating that the vehicles behind are in a risk area, adjusting the initial detection threshold of a radar detection device in the vehicle based on the spatiotemporal perception information to obtain a target detection threshold, wherein the risk area represents an area where the echo power of the radar detection device attenuates, and the target detection threshold is lower than the initial detection threshold; and determining the driving trajectory of the vehicles behind based on the spatiotemporal perception information satisfying the target detection threshold.

[0007] Optionally, the spatiotemporal perception information includes at least one of the following: the azimuth angle, distance, and radial velocity of the vehicle behind relative to the vehicle. In response to the detection result indicating that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on the spatiotemporal perception information to obtain a target detection threshold. This includes: in response to the vehicle behind being in a risk area, based on the azimuth angle, distance, and radial velocity, calling a threshold adjustment function model to determine the downward adjustment amount of the initial detection threshold. The threshold adjustment function model is obtained by fitting azimuth angle samples, distance samples, and radial velocity samples. The azimuth angle samples represent the azimuth angle of the vehicle sample behind the vehicle sample relative to the vehicle sample; the distance samples represent the distance of the vehicle sample behind the vehicle sample relative to the vehicle sample; and the radial velocity samples represent the radial velocity of the vehicle sample behind the vehicle sample relative to the vehicle sample. The initial detection threshold is lowered using the downward adjustment amount to obtain the target detection threshold.

[0008] Optionally, in response to a vehicle behind being in a risk zone, a threshold adjustment function model is invoked based on the azimuth, distance, and radial velocity to determine the downward adjustment amount of the initial detection threshold. This includes: in response to a vehicle behind being in a risk zone, invoking the threshold adjustment function model to determine the azimuth component of the vehicle behind, where the azimuth component represents the degree of deviation of the vehicle behind relative to the vehicle; in response to a distance greater than a distance threshold, invoking the threshold adjustment function model to determine the distance component of the vehicle behind, where the distance component represents the component used to maintain stable detection of the vehicle behind within a target distance; in response to a radial velocity less than a first radial velocity threshold and an absolute value of the radial velocity greater than a second radial velocity threshold, invoking the threshold adjustment function model to determine the velocity component of the vehicle behind, where the first radial velocity threshold represents the radial approach of the vehicle behind and the second radial velocity threshold represents the critical velocity value at which a collision occurs between the vehicle behind and the vehicle; and determining the downward adjustment amount based on the azimuth component, distance component, and velocity component.

[0009] Optionally, the spatiotemporal perception information includes an initial signal-to-noise ratio (SNR), which represents the relative magnitude of the intensity of the echo signal received by the radar detection device relative to the noise. The method further includes: determining the spatiotemporal perception information as valid data that satisfies the target detection threshold in response to the initial SNR being greater than or equal to the target detection threshold.

[0010] Optionally, the method further includes: storing spatiotemporal perception information into a valid data list; and determining the driving trajectory of the following vehicle in response to the spatiotemporal perception information meeting the target detection threshold, including: performing correlation processing on the spatiotemporal perception information in the valid data list to obtain the initial driving trajectory of the vehicle; and filtering the initial driving trajectory to obtain the driving trajectory.

[0011] Optionally, the method further includes: in response to the following vehicle not being in the risk area, determining the initial detection threshold as the target detection threshold.

[0012] According to another aspect of the embodiments of this application, a vehicle trajectory determination device is also provided. The device may include: an acquisition unit for acquiring spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle; a detection unit for detecting the vehicles behind based on the spatiotemporal perception information and obtaining a detection result; an adjustment unit for adjusting the initial detection threshold of a radar detection device in the vehicle based on the spatiotemporal perception information in response to the detection result indicating that the vehicles behind are in a risk area, thereby obtaining a target detection threshold, wherein the risk area represents an area where the echo power of the radar detection device attenuates, and the target detection threshold is lower than the initial detection threshold; and a determination unit for determining the trajectory of the vehicles behind in response to the spatiotemporal perception information satisfying the target detection threshold.

[0013] According to another aspect of the embodiments of this application, a processor is also provided. The processor is used to run a program, wherein the program is executed by the processor to perform the methods described in the embodiments of this application.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0019] According to another aspect of the embodiments of this application, a vehicle is also provided. The vehicle includes a memory and a processor. The memory stores an executable program; the processor is used to run the program, which, when running, implements the methods described in the embodiments of this application.

[0020] In this embodiment, to determine the trajectory of a vehicle behind, spatiotemporal perception information of the vehicle behind relative to the vehicle can be obtained. Based on this spatiotemporal perception information, the vehicle behind can be detected to obtain a detection result. In response to the detection result indicating that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle can be adjusted based on the spatiotemporal perception information to obtain a target detection threshold. Here, the risk area represents the region where the echo power of the radar detection equipment attenuates, and the target detection threshold is lower than the initial detection threshold. In response to the spatiotemporal perception information meeting the target detection threshold, the trajectory of the vehicle behind can be determined. In other words, this embodiment intelligently identifies whether a vehicle behind is in a risk area by acquiring spatiotemporal perception information of the vehicle behind relative to the vehicle in real time. If the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on spatiotemporal perception information within that risk area. This lowers the initial detection threshold and ultimately achieves accurate output of the vehicle's trajectory. This overcomes the limitation of discontinuous vehicle trajectories caused by the method of lowering the detection threshold across the entire area in related technologies. Thus, it solves the technical problem of low accuracy in determining the trajectory of the vehicle behind and achieves the technical effect of improving the accuracy of determining the trajectory of the vehicle behind. Attached Figure Description

[0021] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0022] Figure 1 This is a flowchart of a method for determining the driving trajectory of a vehicle according to an embodiment of this application;

[0023] Figure 2 This is a schematic diagram of an adaptive detection threshold adjustment system for a vehicle rear corner radar according to an embodiment of this application;

[0024] Figure 3 This is a flowchart of an adaptive detection threshold adjustment method for a vehicle rear corner radar according to an embodiment of this application;

[0025] Figure 4 This is a flowchart of an adaptive threshold decision module implementation method according to an embodiment of this application;

[0026] Figure 5This is a schematic diagram of a vehicle trajectory determination device according to an embodiment of this application;

[0027] Figure 6 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, functional component, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, functional components, or devices.

[0030] According to an embodiment of this application, an embodiment of a method for determining the driving trajectory of a vehicle is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] Figure 1 This is a flowchart of a method for determining the driving trajectory of a vehicle according to an embodiment of this application, as shown below. Figure 1 As shown, the method may include the following steps.

[0032] Step S102: Obtain the spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle.

[0033] In the technical solution provided in step S102 of this application, in order to extract and parse the key parameters of the vehicle behind the vehicle in terms of spatial position and motion state in real time from the original signal of the vehicle's on-board rear corner radar detection device, and to provide a basis for the initial detection threshold adjustment of the radar detection device (e.g., radar) in the subsequent vehicle, the spatiotemporal perception information of the vehicle behind the vehicle relative to the vehicle can be obtained.

[0034] In this embodiment, the aforementioned spatiotemporal perception information may include the azimuth angle, distance, and radial velocity of vehicles behind the vehicle relative to the vehicle, as well as the initial signal-to-noise ratio initially estimated by the radar detection equipment.

[0035] Optionally, the aforementioned azimuth angle refers to the horizontal deflection angle of a rear vehicle (e.g., a target) relative to the longitudinal axis (direction of travel) of the vehicle (e.g., the vehicle itself). It can be positive on the right and negative on the left, and is used to determine whether the rear vehicle is located directly behind or in a risk area with a small deflection angle. For example, the risk area can be a high-risk multipath sensitive area.

[0036] Optionally, the aforementioned distance refers to the straight-line slant distance between the rear vehicle and the radar detection equipment of the vehicle (e.g., this vehicle), reflecting the distance of the rear vehicle.

[0037] Optionally, the aforementioned radial velocity is the velocity component of the following vehicle relative to the vehicle (e.g., this vehicle) along the radar line of sight, used to identify whether it is a high-threat following vehicle approaching at high speed.

[0038] Optionally, the aforementioned initial signal-to-noise ratio can characterize the intensity ratio of the radar received echo signal to the noise, and is the original physical basis for assessing whether the vehicle behind can be reliably detected.

[0039] In this embodiment of the application, by accurately obtaining the spatiotemporal perception information of the vehicles behind the vehicle relative to the vehicle, an accurate information basis is provided for subsequent judgment on whether the vehicles behind are in a risk area.

[0040] Step S104: Based on spatiotemporal perception information, detect vehicles behind and obtain detection results.

[0041] In the technical solution provided in step S104 of this application, after obtaining the spatiotemporal perception information of the vehicle behind the vehicle relative to the vehicle, the vehicle behind can be detected based on the spatiotemporal perception information to determine that the vehicle behind is in a risk area.

[0042] In this embodiment, it can be determined whether the vehicle behind is in a risk area based on the azimuth angle. For example, if the absolute value of the azimuth angle (|θ|) is less than or equal to the azimuth angle threshold (θ_th), it can be determined that the vehicle behind is in a risk area, that is, the vehicle behind is a target in a risk area.

[0043] Optionally, if the absolute value of the azimuth angle (|θ|) is greater than the azimuth angle threshold (θ_th), it can be determined that the vehicle behind is not in the risk area, that is, the vehicle behind is a target in the normal area.

[0044] Optionally, the above azimuth threshold can be 15°, and no specific limitation is made here.

[0045] In this embodiment of the application, vehicles behind are detected based on spatiotemporal perception information, providing an accurate basis for determining whether to adjust the initial detection threshold of the radar detection equipment in the vehicle.

[0046] Step S106: In response to the detection result that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on the spatiotemporal perception information to obtain the target detection threshold.

[0047] In the technical solution provided in step S106 of this application, the risk region can be used to represent the area where the echo power of the radar detection equipment attenuates, and the target detection threshold is lower than the initial detection threshold. The risk region can also be called the risk angle region. The target detection threshold can also be called the adaptive detection threshold.

[0048] In this embodiment, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on spatiotemporal perception information. This can proactively and accurately lower the radar detection threshold for specific areas where the radar echo signal attenuates due to multipath effects, thereby ensuring that targets are not lost in critical scenarios.

[0049] Optionally, the aforementioned risk area can refer to a narrow fan-shaped region where the absolute value of the azimuth angle of the vehicle behind is less than or equal to a preset threshold azimuth angle (e.g., 15°). Because the target is located directly behind the vehicle or at a very small angle, the radar beam is reflected multiple times by the road surface, rear bumper, and other structures within this risk area, creating numerous propagation paths. This results in severe constructive or destructive interference of the signal at the receiving end, causing a drastic, random, and periodic attenuation of the echo power, i.e., the signal-to-noise ratio (SNR). Even if the vehicle behind is actually present, the echo energy may briefly drop below the traditional detection threshold, leading to a misjudgment that there is no vehicle behind the vehicle. Therefore, this risk area can be considered an inherent signal attenuation region determined by both physical geometry and electromagnetic propagation characteristics.

[0050] Optionally, based on spatiotemporal perception information, the initial detection threshold of the radar detection equipment in the vehicle is adjusted. This adjustment is not simply a uniform reduction of the initial detection threshold, but rather employs a multi-parameter coupled dynamic modeling mechanism. For example, when it is determined that a vehicle behind is located in a risk zone, the threshold adjustment logic is triggered. The reduction amount Δ of the initial detection threshold of the radar detection equipment in the vehicle can be calculated jointly from the azimuth angle θ, the distance R, and the radial velocity Vr.

[0051] Optionally, the azimuth angle θ is used as the primary control factor. The closer it is to 0° (directly behind), the more severe the multipath attenuation, and the larger the reduction amount Δ. For distance R, vehicles at medium to long distances (e.g., R > 20m) have weaker echo signals due to path loss superposition. Therefore, the further the distance, the greater the reduction amount should be to compensate for the additional attenuation at long distances. For radial velocity Vr, when a vehicle is approaching at high speed (e.g., |Vr| > 20m / s), it can be determined that the vehicle is a high-threat target, and the reduction amount can be further increased to buy reaction time for emergency avoidance.

[0052] Optionally, the aforementioned azimuth angle θ, distance R, and radial velocity Vr are weighted and superimposed to form a downward adjustment Δ, which is then limited by upper and lower limits (e.g., 1~6dB) to finally obtain the target detection threshold Th_final = Th_base – Δ, ensuring that the adjustment is both sufficient and safe. Here, Th_base can be used to represent the initial detection threshold.

[0053] In this embodiment, within the risk area, even if the signal-to-noise ratio of a following vehicle temporarily drops below 10dB due to multipath fading, it can still be stably detected by adaptively lowering the initial detection threshold of the radar detection equipment in the vehicle to 7dB, thus avoiding periodic interruptions to the driving trajectory. Simultaneously, since the adjustment is triggered only within a narrow ±15° angle, the original initial detection threshold is maintained in other areas, achieving the effect of local enhancement and global stability.

[0054] Step S108: In response to the spatiotemporal perception information satisfying the target detection threshold, determine the driving trajectory of the vehicle behind.

[0055] In the technical solution provided in step S108 of this application, in response to the detection result indicating that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on spatiotemporal perception information. After obtaining the target detection threshold, the driving trajectory of the vehicle behind can be determined if the spatiotemporal perception information meets the target detection threshold. By actively and appropriately lowering the initial detection threshold, the real target can still be effectively detected even when the echo signal fades due to multipath interference, avoiding the phenomenon of periodic target loss.

[0056] In this embodiment, the spatiotemporal perception information satisfying the target detection threshold means that each valid spatiotemporal perception information (e.g., target point) retained after adjustment of the initial detection threshold of the radar detection equipment has an initial signal-to-noise ratio (SNR_est) higher than the target detection threshold after adaptive adjustment of the risk area (i.e., SNR_est≥Th_final), indicating that the target point is a real target and not noise or clutter.

[0057] Optionally, determining the trajectory of a vehicle behind can be achieved through a fusion algorithm of multi-frame data association and state estimation. For example, data association is performed on each valid target point in the current frame (e.g., nearest neighbor algorithm or probabilistic data association), and the target point is matched with the existing trajectory (e.g., target trajectory) in the previous frame. If the match is successful, the state of the trajectory (including position, velocity, acceleration, etc.) is updated. If it is a new target, a new trajectory is initialized. If the match fails and the target point continues to appear, it is determined to be a newly appearing target. Subsequently, state estimation algorithms such as Kalman filtering can be used to predict and correct the target's motion state, and under noise interference and point jitter, the continuous trajectory parameters of the target, including position coordinates, velocity vector, and heading angle, are smoothly output. The entire process iterates in real time within each radar frame period to ensure low latency and high continuity of the trajectory output.

[0058] Optionally, in traditional systems, because targets behind are frequently lost in risk areas, the target trajectory often jumps, is interrupted, or even disappears, causing the vehicle-mounted blind spot detection (BSD) or lane change assist (LCA) system to misjudge that there are no vehicles and miss the warning opportunity. However, the embodiments of this application avoid this defect of intermittent tracking through the above steps.

[0059] In this embodiment, when the spatiotemporal perception information meets the target detection threshold, the driving trajectory of the vehicle behind is determined, and the driving trajectory is no longer interrupted by echo signal fluctuations. In addition, algorithms such as Kalman filtering effectively suppress radar measurement noise and instantaneous jitter, resulting in a smoother and more accurate output driving trajectory.

[0060] In steps S102 to S108 of this application, if it is necessary to determine the driving trajectory of a vehicle behind, the spatiotemporal perception information of the vehicle behind the vehicle relative to the vehicle can be obtained; the vehicle behind can be detected based on the spatiotemporal perception information to obtain a detection result; in response to the detection result indicating that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle can be adjusted based on the spatiotemporal perception information to obtain a target detection threshold, wherein the risk area is used to represent the area where the echo power of the radar detection equipment is attenuated, and the target detection threshold is lower than the initial detection threshold; the driving trajectory of the vehicle behind can be determined in response to the spatiotemporal perception information meeting the target detection threshold. In other words, the embodiments of this application intelligently identify whether the vehicle behind is in a risk area by acquiring the spatiotemporal perception information of the vehicle behind the vehicle relative to the vehicle in real time. If the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on spatiotemporal perception information within that risk area. This lowers the initial detection threshold and ultimately achieves accurate output of the vehicle's trajectory. This overcomes the limitation of discontinuous vehicle trajectories caused by the method of lowering the detection threshold across the entire area in related technologies. Thus, it solves the technical problem of low accuracy in determining the trajectory of the vehicle behind and achieves the technical effect of improving the accuracy of determining the trajectory of the vehicle behind.

[0061] The method described in this embodiment will be further described below.

[0062] As an optional embodiment, the spatiotemporal perception information includes at least one of the following: the azimuth angle, distance, and radial velocity of the vehicle behind relative to the vehicle. Step S106, in response to the detection result that the vehicle behind is in a risk area, the initial detection threshold of the radar detection device in the vehicle is adjusted based on the spatiotemporal perception information to obtain a target detection threshold. This includes: in response to the vehicle behind being in a risk area, based on the azimuth angle, distance, and radial velocity, calling a threshold adjustment function model to determine the amount of reduction in the initial detection threshold, wherein the threshold adjustment function model is obtained by fitting azimuth angle samples, distance samples, and radial velocity samples. The azimuth angle samples are used to represent the azimuth angle of the vehicle behind relative to the vehicle sample, the distance samples are used to represent the distance of the vehicle behind relative to the vehicle sample, and the radial velocity samples are used to represent the radial velocity of the vehicle behind relative to the vehicle sample; the initial detection threshold is lowered using the reduction amount to obtain the target detection threshold.

[0063] In this embodiment, the threshold adjustment function model is not a static formula set based on experience, but a dynamic response function fitted by statistical modeling and regression analysis based on a large number of azimuth angle samples, distance samples, and radial velocity samples collected in real vehicle test environments. These samples are not theoretical values, but rather multiple frames of measurement data actually captured by radar under different geometric positions and motion states of vehicles behind in real road scenarios.

[0064] Optionally, the azimuth angle sample reflects the angular distribution of the following vehicles relative to the vehicle's longitudinal axis in the horizontal plane, the distance sample reflects the near and far distribution of the following vehicles in the slant distance dimension, and the radial velocity sample characterizes the speed characteristics of the following vehicles approaching or moving away along the radar line of sight.

[0065] Optionally, the aforementioned azimuth angle samples, distance samples, and radial velocity samples can be systematically collected, classified, and analyzed. For example, under dry / slippery road conditions, different vehicle speeds, and different background clutter environments, the signal-to-noise ratio attenuation of following vehicles under different combinations of azimuth angle, distance, and radial velocity can be statistically analyzed to establish a mapping relationship between input parameters and down-adjustment amounts.

[0066] Optionally, once the azimuth angle indicates that a vehicle has entered the risk zone (|θ|≤15°), the threshold adjustment function model is invoked. Based on the azimuth angle, distance, and radial velocity, the threshold adjustment function model is invoked to determine the downward adjustment amount of the initial detection threshold. This downward adjustment amount is then used to lower the initial detection threshold, thus obtaining the target detection threshold.

[0067] In this embodiment, the above steps upgrade the simple detection threshold adjustment rule, which originally relied on manual experience, into a data-driven, physically interpretable, and scene-adaptive intelligent decision-making system. Because the threshold adjustment function model is trained based on real attenuation data, the output downsampling amount more closely matches the actual signal-to-noise ratio fluctuation characteristics, resulting in a more accurate target detection threshold.

[0068] As an optional implementation, in response to a vehicle behind being in a risk zone, a threshold adjustment function model is invoked based on azimuth, distance, and radial velocity to determine the downward adjustment amount of the initial detection threshold. This includes: in response to a vehicle behind being in a risk zone, invoking the threshold adjustment function model to determine the azimuth component of the vehicle behind, where the azimuth component represents the degree of deviation of the vehicle behind relative to the vehicle; in response to a distance greater than a distance threshold, invoking the threshold adjustment function model to determine the distance component of the vehicle behind, where the distance component represents the component that maintains stable detection of the vehicle behind within a target distance; in response to a radial velocity less than a first radial velocity threshold and an absolute value of the radial velocity greater than a second radial velocity threshold, invoking the threshold adjustment function model to determine the velocity component of the vehicle behind, where the first radial velocity threshold represents the radial approach of the vehicle behind and the second radial velocity threshold represents the velocity threshold at which a collision occurs between the two vehicles; and determining the downward adjustment amount based on the azimuth component, distance component, and velocity component.

[0069] In this embodiment, by employing a component decoupling and collaborative superposition threshold adjustment function model f_adjust(θ, R, Vr), multiple physical factors affecting radar echo signal attenuation and rear vehicle detection stability are structurally analyzed and modeled as azimuth components, range components, and velocity components, respectively, and the final down-adjustment amount is collaboratively calculated within the risk area.

[0070] Optionally, when it is confirmed that the vehicle behind is located in the risk zone (|θ|≤15°), the threshold adjustment function model can be invoked to calculate the azimuth component Δ_θ. This azimuth component characterizes the degree of geometric deviation of the vehicle behind from being directly behind the vehicle. When directly behind (θ=0°), the radar wave experiences the strongest specular reflection from the road surface and rear bumper, resulting in the most severe multipath interference and signal-to-noise ratio attenuation. As the deviation angle increases, the reflection path gradually weakens, and the attenuation decreases. Therefore, the azimuth component is inversely proportional to the deviation angle: Δ_θ=K_angle (θ_th-|θ|), where θ_th can be used to represent the risk threshold (e.g., 15°), and K_θ can be used to represent the angle adjustment coefficient, such as 0.3dB / degree. This means that the closer the vehicle is to the rear, the larger this azimuth component becomes, and the more significantly the initial detection threshold is lowered, thus accurately compensating for signal loss in the risk area.

[0071] Optionally, it can be determined whether the distance is greater than a distance threshold (e.g., R_near = 20 meters). If the distance is greater than the distance threshold, the distance component Δ_R is calculated. This distance component is used to compensate for the additional signal attenuation caused by the extended propagation path of mid-to-long-range targets. Within the risk area, even with the same azimuth angle, the echo energy of vehicles at a greater distance behind is inherently lower. If the same initial detection threshold is used as for vehicles at a closer distance behind, it is highly likely to lead to misjudgment and loss due to superimposed attenuation. Therefore, when R > R_near, Δ_R increases linearly with distance: Δ_R = K_dist (R-R_near), where K_dist can be used to represent the distance adjustment coefficient, such as 0.05dB / m, to ensure continuous sensitivity enhancement for high-risk targets at long distances and avoid the imbalance problem of being able to see up close but having a broken link at a distance.

[0072] Optionally, a dual-condition judgment can be applied to the radial velocity. If the radial velocity is less than a first radial velocity threshold (e.g., 0 m / s, indicating the target is approaching the vehicle), and the absolute value of the radial velocity is greater than a second radial velocity threshold (e.g., 20 m / s, approximately 72 km / h, representing the collision risk threshold for high-speed approach), then the velocity component Δ_V is activated. This velocity component does not simply reflect the magnitude of the speed, but rather identifies this high-risk situation of high-speed approach, providing additional detection redundancy for emergency collision avoidance scenarios. When a vehicle is detected approaching from behind at a near-collision speed in a straight line, even if the signal-to-noise ratio is only slightly higher than the initial detection threshold, the detection conditions should be proactively relaxed to allow the upper-level vehicle safety system 0.1 to 0.3 seconds of reaction time. In this case, Δ_V can be a fixed bias value (e.g., 1 dB) as a safety margin to ensure detection in high-risk scenarios.

[0073] Optionally, the three components mentioned above—azimuth component (reflecting geometric attenuation), range component (reflecting propagation loss), and velocity component (reflecting threat level)—are algebraically superimposed to form the down adjustment Δ = Δ_θ + Δ_R + Δ_V.

[0074] Optionally, Δ can be limited to a reasonable range [Δ_min, Δ_max], taking values ​​[1dB, 6dB]. If (Δ < Δ_min), then Δ = Δ_min; if (Δ > Δ_max), then Δ = Δ_max. This limitation ensures that the adjustment range is within a safe and controllable range, preventing false alarms due to excessive downward adjustment or loss of stability due to insufficient adjustment.

[0075] In this embodiment, Δ_θ ensures that the threshold adjustment is precisely focused on the region directly behind where multipath is most severe; Δ_R compensates for the additional path loss of targets at medium to long ranges; and Δ_V provides additional sensitivity margin when a high-speed approaching imminent threat is detected. These three elements work together to achieve refined and intelligent threshold adjustment.

[0076] As an optional embodiment, the spatiotemporal sensing information also includes an initial signal-to-noise ratio (SNR), which is used to determine the relative magnitude of the intensity of the echo signal received by the radar detection device relative to the noise. The method further includes: determining the spatiotemporal sensing information as valid data that satisfies the target detection threshold in response to the initial SNR being greater than or equal to the target detection threshold.

[0077] In this embodiment, the initial signal-to-noise ratio (SNR_est) is further introduced into the spatiotemporal perception information as a key judgment criterion, and the initial signal-to-noise ratio is explicitly used as the direct physical basis for determining whether the vehicle behind meets the target detection threshold, thereby forming a complete, closed-loop, and physically verifiable detection and confirmation mechanism.

[0078] Optionally, the initial signal-to-noise ratio (SNR) is the ratio of the echo signal energy to the background noise energy estimated by the radar for each potential target point after signal preprocessing, expressed in decibels (dB). The initial SNR directly reflects the detectability of the radar for the vehicle behind it at the current moment.

[0079] Optionally, when the initial signal-to-noise ratio (SNR) of the target point is greater than or equal to the target detection threshold Th_final corresponding to that target point, the spatiotemporal perception information can be determined as valid data that meets the target detection threshold, i.e., confirming that the vehicle behind is a real target, rather than noise, clutter, or false echoes. Since this ensures that even when the initial detection threshold is actively lowered, the basic principle that the signal strength must be sufficient to support reliable detection is upheld, preventing false detections due to excessive relaxation of the initial detection threshold, this judgment process is independent and cannot be skipped.

[0080] As an optional embodiment, the method further includes: storing spatiotemporal perception information into a valid data list; and determining the driving trajectory of a vehicle behind it in response to the spatiotemporal perception information meeting a target detection threshold, including: performing correlation processing on the spatiotemporal perception information in the valid data list to obtain the initial driving trajectory of the vehicle; and filtering the initial driving trajectory to obtain the driving trajectory.

[0081] In this embodiment, the target detection and tracking module in the vehicle can perform a judgment on each point Pi. If SNR_esti>=Th_final_i, Pi is confirmed as a valid point and added to the valid data list (e.g., the valid point list) Valid_List. Otherwise, Pi is discarded. Afterwards, the target detection and tracking module in the vehicle can perform association processing on the points in Valid_List, such as data association (e.g., nearest neighbor association) and filtering processing, such as state estimation (Kalman filtering), to update the driving trajectory of the tracked vehicle behind.

[0082] Optionally, filtering can eliminate interference from radar measurement noise and instantaneous fluctuations on the driving trajectory. Because radar points have inherent measurement errors in each frame, even if the target actually exists, the estimated position and velocity will fluctuate within a certain range. Filtering the initial driving trajectory allows estimation of its state (position, velocity, acceleration), outputting a smooth, continuous trajectory curve that conforms to physical laws, thus obtaining the driving trajectory.

[0083] Optionally, after obtaining the driving trajectory, the stable driving trajectory can be sent to the upper-level vehicle safety system through the vehicle control interface module.

[0084] In this embodiment, a robust trajectory construction process with four levels of separation—detection, storage, association, and filtering—is employed. This avoids the tightly coupled detection-tracking model of traditional systems, preventing the trajectory from breaking or diverging due to false or missed detections in a single frame. Data pre-screening is achieved through the isolation of the effective data list; continuous maintenance of following vehicles is achieved through association processing; and physical consistency and smoothness of the trajectory are achieved through filtering processing. This results in a highly modular and debuggable trajectory generation process.

[0085] As an optional embodiment, the method further includes: determining an initial detection threshold as a target detection threshold in response to the following vehicle not being in a risk area.

[0086] In this embodiment, when the absolute value of the azimuth angle of the vehicle behind is determined to be greater than the azimuth angle threshold (e.g., |θ|>15°), it is determined that the vehicle behind is located in the conventional detection area behind or to the side of the vehicle. At this time, the initial detection threshold adjustment mechanism is not triggered, and the initial detection threshold, such as the radar's original basic detection threshold (Th_base), is directly used as the final target detection threshold (Th_final).

[0087] Optionally, the initial detection threshold can be set based on static or semi-adaptive factors such as target distance and environmental clutter level. For example, it can be fixed at 12dB at distances of 20–50 meters, or obtained by looking up a table based on distance. In long-term engineering practice, the initial detection threshold has been proven to effectively balance detection probability and false alarm rate in most non-rear-facing scenarios, ensuring stable detection of vehicles on the side while effectively suppressing false echoes caused by road surface reflections, roadside obstacles, raindrops, or billboards. Therefore, in non-risk areas, there is no need to introduce additional dynamic threshold reduction logic, avoiding unnecessary computational overhead and eliminating the risk of overly relaxed thresholds due to misjudgment or parameter drift.

[0088] In this embodiment, to determine the trajectory of a vehicle behind, spatiotemporal perception information of the vehicle behind relative to the vehicle can be obtained. Based on this spatiotemporal perception information, the vehicle behind can be detected to obtain a detection result. In response to the detection result indicating that the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle can be adjusted based on the spatiotemporal perception information to obtain a target detection threshold. Here, the risk area represents the region where the echo power of the radar detection equipment attenuates, and the target detection threshold is lower than the initial detection threshold. In response to the spatiotemporal perception information meeting the target detection threshold, the trajectory of the vehicle behind can be determined. In other words, this embodiment intelligently identifies whether a vehicle behind is in a risk area by acquiring spatiotemporal perception information of the vehicle behind relative to the vehicle in real time. If the vehicle behind is in a risk area, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on spatiotemporal perception information within that risk area. This lowers the initial detection threshold and ultimately achieves accurate output of the vehicle's trajectory. This overcomes the limitation of discontinuous vehicle trajectories caused by the method of lowering the detection threshold across the entire area in related technologies. Thus, it solves the technical problem of low accuracy in determining the trajectory of the vehicle behind and achieves the technical effect of improving the accuracy of determining the trajectory of the vehicle behind.

[0089] The technical solutions of the embodiments of this application will be illustrated below with reference to preferred embodiments.

[0090] Currently, BSD and LCA systems are key technologies for improving road safety. Core perception components can include corner radars (rear corner radars) positioned on either side of the vehicle's rear bumper. These radars are responsible for detecting targets in the blind spot to the side and rear of the vehicle and issuing warnings to the driver when there is a collision risk. However, in specific scenarios involving a vehicle rapidly approaching from directly behind or at an angle of very small (less than 15°) to the vehicle's direction of travel, the radar's target detection may experience periodic loss and re-acquisition, resulting in discontinuous target points / trajectories and an inability to generate stable and reliable target tracking output. This can lead to delayed warnings, missed detections, or temporary malfunctions in the BSD / LCA system, posing a safety hazard. This problem primarily stems from complex multipath propagation effects.

[0091] When the target vehicle is located directly behind or at a very small angle to the target vehicle, the radar waves, in addition to the direct path, are easily reflected by the road surface and the vehicle body (e.g., the rear bumper), forming multiple propagation paths. The superposition of these signals at the receiving end will produce severe constructive or destructive interference, causing the received target echo power, such as the SNR, to fluctuate violently and rapidly with relative position and road conditions (e.g., dry or wet).

[0092] In summary, the relevant technologies lack a dedicated solution that can effectively identify specific high-risk angle areas behind vehicles and intelligently and dynamically optimize radar detection thresholds within these areas to significantly improve the stability of detecting targets approaching in a straight line while ensuring a low false alarm rate.

[0093] To address the aforementioned issues, this application proposes an adaptive detection threshold adjustment method and system for vehicle rear-angle radar, thereby resolving the problems of unstable detection signals and discontinuous target trajectories caused by multipath effects when the rear-angle radar detects vehicles approaching from directly behind or at a small angle in a straight line in related technologies.

[0094] Figure 2 This is a schematic diagram of an adaptive detection threshold adjustment system for a vehicle rear corner radar according to an embodiment of this application, as shown below. Figure 2 As shown, the system includes: a radar transceiver module 201, a signal preprocessing and data acquisition module 202, an adaptive threshold decision module 203, a target detection and tracking module 204, a vehicle control interface module 205, and an upper-level vehicle safety system 206.

[0095] The radar transceiver module 201 serves as the RF front-end for millimeter-wave radar, used to transmit Frequency-Modulated Continuous Wave (FMCW) signals and receive reflected echoes from targets. Specifically, it includes circuit components such as a Voltage-Controlled Oscillator (VCO), a Power Amplifier (PA), a Transmit Antenna (TX), a Receive Antenna (RX), a Low-Noise Amplifier (LNA), and a mixer. The millimeter-wave radar RF front-end's function is to transmit the RF signal, receive the echo, down-convert the signal, and output an Intermediate Frequency (IF) analog signal.

[0096] The signal preprocessing and data acquisition module 202 may include an analog-to-digital converter (ADC), a digital signal processor (DSP), or a dedicated processing chip. Vehicle state information such as target angle, distance, and relative speed can be input into the signal preprocessing and data acquisition module 202. After receiving the IF signal and performing analog-to-digital conversion, a Fast Fourier Transform (FFT) is performed to separate the signal in the distance and velocity dimensions. Subsequently, initial detection algorithms such as Constant False Alarm Rate (CFAR) are used to extract the point information of potential targets. For each point, the signal preprocessing and data acquisition module 202 calculates the distance R, radial velocity Vr, azimuth angle θ, and initial signal-to-noise ratio SNR_est. The azimuth angle θ can be obtained by angle measurement algorithms such as Digital Beamforming (DBF) or Multiple Signal Classification (MUSIC). The azimuth angle θ is the angle of deviation of the target point relative to the longitudinal axis of the vehicle (i.e. the direction of vehicle movement), with the right side being positive and the left side being negative.

[0097] The adaptive threshold decision module 203 can be implemented using software algorithms in a DSP or microcontroller unit (MCU). It receives information on each target point to be confirmed (distance R, radial velocity Vr, azimuth angle θ, and initial signal-to-noise ratio SNR_est) from the signal preprocessing and data acquisition module 202. This adaptive threshold decision module 203 comprises two key sub-modules: a risk area judgment sub-module 2031 and a dynamic threshold calculation sub-module 2032.

[0098] Optionally, the risk area determination submodule 2031 can have a built-in configurable threshold register to store the low signal-to-noise ratio risk angle threshold θ_th (e.g., 15°). Its function is to compare the absolute value of the input azimuth angle θ with θ_th. If |θ|≤θ_th, the target point is determined to be located in a high-risk multipath sensitive area, and a flag signal Flag_Risk=1 is triggered; otherwise, Flag_Risk=0.

[0099] Optionally, the dynamic threshold calculation submodule 2032 can internally store or calculate a basic detection threshold Th_base in real time. Th_base can be a fixed value or a simple adaptive value obtained by looking up a table based on the distance R. The core is a built-in threshold adjustment function model. When Flag_Risk=1 is received, this threshold adjustment function model is activated. Based on the input parameters such as azimuth angle θ, distance R, and radial velocity Vr, the threshold adjustment function model calculates a threshold adjustment amount Δ, and finally outputs an adaptive detection threshold Th_adaptive=Th_base-Δ.

[0100] The target detection and tracking module 204 receives point information from the signal preprocessing and data acquisition module 202 and the final detection threshold (Th_final, which can be Th_adaptive or Th_base) from the adaptive threshold decision module 203. It compares the SNR_est of each point with Th_final. If SNR_est ≥ Th_final, it is confirmed as a valid target point. Otherwise, it is considered noise or clutter and filtered out. The confirmed points are then fed into a tracking algorithm such as a Kalman filter to form a continuous and smooth target trajectory (Track).

[0101] The vehicle control interface module 205 may include a Controller Area Network (CAN) and a transceiver. The vehicle control interface module 205 can encapsulate the stable target trajectory (Track) output by the target detection and tracking module 204 into standard messages conforming to vehicle network protocols, such as CAN with Flexible Data-Rate (CAN FD) or Ethernet, and send them to upper-level vehicle safety systems 206 such as Blind Spot Detection (BSD) and Lane Change Assist (LCA).

[0102] Figure 3 This is a flowchart of an adaptive detection threshold adjustment method for a vehicle rear corner radar according to an embodiment of this application, such as... Figure 3 As shown, it includes the following steps.

[0103] Step S301: Obtain the raw radar detection data.

[0104] In this embodiment, after acquiring the original radar detection data, the signal preprocessing and data acquisition module 202 processes a frame of radar data and outputs N potential target traces Pi (i=1 to N), each trace containing data: (Ri, Vri, θi, SNR_esti).

[0105] Step S302: Calculate the target's relative azimuth.

[0106] In this embodiment, the relative azimuth angle of the target can be calculated.

[0107] Step S303: Determine whether the target is located in the low signal-to-noise ratio risk angle area behind.

[0108] In this embodiment, for each point Pi, the judgment submodule in the adaptive threshold decision module 203 performs the following operation: It reads a preset θ_th (with a value of 15°). If (|θi|<=θ_th), it is marked as a target in a risk area, i.e., Flag_Risk_i=1. Otherwise, it is marked as a target in a normal area, i.e., Flag_Risk_i=0.

[0109] Step S304: Use the standard / basic detection threshold.

[0110] In this embodiment, the basic detection threshold (e.g., the basic threshold) Th_base_i can be looked up in a table based on the distance Ri. For example, when Ri < 20m, Th_base = 12dB; when 20m <= Ri < 50m, Th_base = 10dB; when Ri >= 50m, Th_base = 8dB.

[0111] Step S305: Input the target angle, distance, relative speed and other status information.

[0112] In this embodiment, target angle, distance, relative speed and other status information can be input.

[0113] Step S306: Dynamically calculate the optimization threshold value based on the preset adjustment strategy / function model.

[0114] In this embodiment, the decision on whether and how to calculate the reduction amount Δ_i is based on Flag_Risk_i. If (Flag_Risk_i == 1), the threshold adjustment function model Δ_i = f_adjust(θi, Ri, Vri) is called to calculate the reduction amount. Afterwards, the adaptive threshold can be calculated using Th_adaptive_i = Th_base_i - Δ_i, thereby setting the final threshold Th_final_i = Th_adaptive_i. For regular regions, no special reduction is performed, and Th_final_i = Th_base_i.

[0115] Alternatively, a specific implementable embodiment of the threshold adjustment function model f_adjust(θ, R, Vr) is as follows.

[0116] This threshold adjustment function model aims to make the downward adjustment amount Δ positively correlated with the severity of multipath risk. The design is as follows.

[0117] Angular component Δ_θ: The risk is inversely proportional to |θ|: Δ_θ = K_angle (θ_th - |θ|). Here, K_angle is the angular adjustment coefficient, such as 0.3 dB / degree. When θ = 0° (directly behind), the downward adjustment amount is the largest, which is K_angle θ_th. When θ = θ_th (the boundary), the downward adjustment amount is 0, achieving a smooth transition with the conventional area.

[0118] Distance component Δ_R: In the risk area, more attention is given to medium and long - distance targets. If (R > R_near), where R_near is the short - distance threshold, for example, 20 meters. Specifically, Δ_R = K_dist (R - R_near). Here, K_dist is the distance adjustment coefficient, such as 0.05 dB / m. If Δ_R = 0, it means that the signal of the short - distance target is strong and no additional distance compensation is required.

[0119] Velocity component Δ_V: For high - threat targets approaching at high speed, an appropriate safety margin is added. If (Vr < Vr_th) and (|Vr| > V_high), then Δ_V = K_vel. Here, Vr_th is the approaching velocity threshold (negative value), and V_high is the high - speed threshold. Here, K_vel is the velocity adjustment bias, such as 1 dB. If (Vr < Vr_th) or (|Vr| > V_high) is not satisfied, then Δ_V = 0.

[0120] The comprehensive downward adjustment amount Δ = Δ_θ + Δ_R + Δ_V.

[0121] At the same time, to ensure system stability, Δ needs to be limited within a reasonable range [Δ_min, Δ_max], with values [1 dB, 6 dB]. If (Δ < Δ_min), then Δ = Δ_min; if (Δ > Δ_max), then Δ = Δ_max.

[0122] Step S307, output the optimized threshold value.

[0123] In this embodiment, the target detection and tracking module 204 performs the following steps for each trace Pi. If (SNR_esti >= Th_final_i), then confirm Pi as a valid trace and add it to the valid trace list Valid_List; otherwise discard Pi.

[0124] Step S308, apply the threshold value.

[0125] In this embodiment, the threshold value can be applied.

[0126] Step S309, perform target tracking and trajectory management.

[0127] In this embodiment, the target detection and tracking module 204 can perform data association (e.g., nearest neighbor association) and state estimation (Kalman filtering) on ​​the points in the Valid_List, and update the trajectories of all tracked targets.

[0128] Step S310: Output a stable and continuous target trajectory to the vehicle safety system.

[0129] In this embodiment, the stable trajectory information is finally sent to the upper-level vehicle safety system 206 through the vehicle control interface module 205.

[0130] In the embodiments of this application, the following significant beneficial effects can be achieved through the above steps.

[0131] Significantly improves detection stability and trajectory continuity: By identifying the risk area in the azimuth angle, the problem domain of multipath effect is accurately located. Within this area, the detection threshold is actively and appropriately reduced by adjusting the dynamic threshold of multiple parameters, so that the real target can still be effectively detected even when the signal fades due to multipath interference, avoiding the phenomenon of periodic target loss.

[0132] Effective control of the overall false alarm rate of the system: Threshold lowering is conditional and localized. It is only implemented in the narrow rear sector (±15°) where a physical increase in sensitivity is necessary. For larger areas such as the sides, the original, stricter threshold is maintained, avoiding the situation where solving one problem (rear instability) causes another problem (a surge in side false alarms). The conditional triggering mechanism (Flag_Risk) is key to ensuring this effect.

[0133] Enhancing the system's responsiveness to high-threat scenarios: The relative speed factor (Δ_V) introduced in the threshold adjustment function model enables the system to identify extremely high-risk scenarios such as vehicles approaching from behind at high speed in a straight line, and temporarily provides additional detection sensitivity. This buys valuable reaction time for functions such as Rear Collision Warning (RCW), improving active safety.

[0134] The following description uses a specific implementation method as an example for further explanation.

[0135] First, perform system initialization and parameter preset.

[0136] When the radar system is started, complete the following parameter configuration (these parameters can be embedded in the code or set through calibration tools).

[0137] Risk angle threshold θ_th: set to 15°. This value is derived from statistics on radar installation height, beam tilt angle, and the region with the most severe multipath fading from a large amount of measured data.

[0138] Basic detection threshold table: Stores distance-based basic threshold values ​​Th_base(R). For example, distance R < 20 meters: Th_base = 14dB (short distance, mainly for preventing false alarms); 20 meters ≤ R < 50 meters: Th_base = 12dB; R ≥ 50 meters: Th_base = 10dB (long distance, mainly for ensuring detection).

[0139] Threshold adjustment function model parameters: Angle adjustment coefficient K_angle = 0.3dB / degree; Range adjustment starting point R_near = 20 meters; Range adjustment coefficient K_dist = 0.05dB / meter; High-speed approach speed threshold V_high = 20 meters / second (72 kilometers / hour); Speed ​​adjustment offset K_vel = 1dB; Downward adjustment limits: Δ_min = 1dB, Δ_max = 6dB. Afterwards, the adaptive threshold decision module can be called in each radar signal processing cycle. Figure 4 This is a flowchart of an adaptive threshold decision module implementation method according to an embodiment of this application, such as... Figure 4 As shown, it includes the following steps.

[0140] Step S401: Calculate the basic threshold.

[0141] In this embodiment, the adaptive threshold decision module takes as input a single point trace information (R, Vr, θ, SNR_est) and outputs as the final detection threshold Th_final applied to that point trace. The basic threshold can be calculated by looking up a table based on distance.

[0142] Step S402, risk area determination.

[0143] In this embodiment, if the absolute value of the azimuth angle abs(θ) <= θ_th, then step S403 is executed. If abs(θ) > θ_th, then step S404 is executed.

[0144] Step S403, located in the risk zone, calculate the dynamic downward adjustment amount Δ.

[0145] In this embodiment, if abs(θ) <= θ_th, it is located in the risk zone, and the azimuth component Δ_θ is calculated using the following formula:

[0146] Δ_θ=K_angle (θ_th-abs(θ))

[0147] The distance component is calculated using the following steps: if R > R_near, then Δ_dist = K_dist. (R-R_near). Otherwise, Δ_dist=0.

[0148] The velocity component (safety enhancement) is calculated using the following steps, Δ_vel = 0. To determine if the target is approaching at high speed: a negative radial velocity Vr indicates approach, and if the absolute value is greater than the threshold (i.e., Vr < 0 and abs(Vr) > V_high), then Δ_vel = K_vel.

[0149] The following formula is used to synthesize and limit the downward adjustment:

[0150] Δ_total = Δ_θ + Δ_dist + Δ_vel

[0151] Step S404, located in a non-risk area, uses the basic threshold.

[0152] In this embodiment, the non-risk area uses a base threshold, Th_final=Th_base.

[0153] Step S405: Calculate and output the adaptive threshold.

[0154] In this embodiment, the adaptive threshold is calculated and output by Th_final = Th_base - Δ_total.

[0155] In higher-order implementations, the above function can receive signals from the vehicle's CAN bus to achieve intelligent linkage. For example, a new input could be added: the right turn signal status `TurnSignal_Right` (TRUE / FALSE). This would then call the basic calculation `Th_final`.

[0156] Enhanced logic: If the driver activates the right turn signal and the target is in the right rear risk zone, an additional safety margin is added.

[0157] Based on the already calculated adaptive threshold, the sensitivity is temporarily reduced by 1dB, that is, Th_final = Th_final - 1.0, to further improve the sensitivity.

[0158] For example, suppose that in a certain frame of data, the radar detects a target point, and the parameters calculated by the preprocessing module are: distance R = 40 meters; radial velocity Vr = -25 meters / second (the negative sign indicates that it is approaching the vehicle); azimuth angle θ = 5° (located to the right rear of the vehicle); signal-to-noise ratio SNR_est = 13dB.

[0159] The threshold calculation process using the method of this application is as follows.

[0160] Basic threshold lookup: Based on R=40 meters, the table shows Th_base=12dB.

[0161] Risk area determination: abs(θ)=5°, which is less than θ_th=15°, is determined to be a risk area target and enters the adaptive calculation path.

[0162] Calculate the dynamic downward adjustment amount Δ.

[0163] Angular component: Δ_θ = 0.3 (15-5) = 3.0 dB; Distance component: R > R_near (20 meters), therefore Δ_dist = 0.05 (40-20)=1.0dB; Velocity component: Vr is negative and abs(Vr)=25>V_high(20), so Δ_vel=1.0dB; Overall downsizing: Δ_total=3.0+1.0+1.0=5.0dB; Limiting: Δ_total remains at 5.0dB within the range [1, 6].

[0164] Calculate the final threshold: Th_final = Th_base - Δ_total = 12 - 5.0 = 7.0 dB.

[0165] Target detection: The SNR_est(13dB) of this point is greater than Th_final(7.0dB), so it is reliably identified as a valid target.

[0166] In this embodiment, if a traditional fixed or distance-based adaptive threshold (i.e., Th_final=Th_base=12dB) is used, the SNR_est (13dB) of the target point will only be slightly higher than the threshold. In the next cycle, if the SNR_est fluctuates and drops to 11dB due to multipath fading, the target will be missed, causing the trajectory to be interrupted. This application, however, dynamically reduces the threshold to 7.0dB, reserving a 6dB tolerance for SNR fluctuations, thereby ensuring the continuity of detection.

[0167] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0168] According to an embodiment of this application, a vehicle trajectory determination device is also provided. It should be noted that this vehicle trajectory determination device can be used to execute the vehicle trajectory determination method described in the embodiments.

[0169] Figure 5 This is a schematic diagram of a vehicle trajectory determination device according to an embodiment of this application. Figure 5As shown, the vehicle trajectory determination device 500 may include: an acquisition unit 502, a detection unit 504, an adjustment unit 506, and a determination unit 508.

[0170] The acquisition unit 502 is used to acquire spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle.

[0171] The detection unit 504 is used to detect vehicles behind based on spatiotemporal perception information and obtain detection results.

[0172] The adjustment unit 506 is used to adjust the initial detection threshold of the radar detection equipment in the vehicle based on the spatiotemporal perception information in response to the detection result that the vehicle behind is in the risk area, so as to obtain the target detection threshold. The risk area is used to represent the area where the echo power of the radar detection equipment is attenuated, and the target detection threshold is lower than the initial detection threshold.

[0173] The determination unit 508 is used to determine the driving trajectory of the vehicle behind in response to the spatiotemporal perception information meeting the target detection threshold.

[0174] Optionally, the spatiotemporal perception information includes at least one of the following: the azimuth angle, distance, and radial velocity of the following vehicle relative to the vehicle. The adjustment unit 506 includes: a calling subunit, used to, in response to the following vehicle being in a risk area, call a threshold adjustment function model based on the azimuth angle, distance, and radial velocity to determine the downward adjustment amount of the initial detection threshold, wherein the threshold adjustment function model is obtained by fitting azimuth angle samples, distance samples, and radial velocity samples, the azimuth angle samples are used to represent the azimuth angle of the following vehicle sample relative to the vehicle sample, the distance samples are used to represent the distance of the following vehicle sample relative to the vehicle sample, and the radial velocity samples are used to represent the radial velocity of the following vehicle sample relative to the vehicle sample; and a lowering subunit, used to lower the initial detection threshold using the downward adjustment amount to obtain the target detection threshold.

[0175] Optionally, the calling subunit includes: a first determining subunit, used to call a threshold adjustment function model to determine the azimuth component of the rear vehicle in response to the rear vehicle being in a risk area, wherein the azimuth component is used to represent the degree of deviation of the rear vehicle relative to the vehicle; a second determining subunit, used to call a threshold adjustment function model to determine the distance component of the rear vehicle in response to the distance being greater than a distance threshold, wherein the distance component is used to represent the component for maintaining stable detection of the rear vehicle within a target distance; a third determining subunit, used to call a threshold adjustment function model to determine the velocity component of the rear vehicle in response to the radial velocity being less than a first radial velocity threshold and the absolute value of the radial velocity being greater than a second radial velocity threshold, wherein the first radial velocity threshold is used to represent the radial approach of the rear vehicle to the vehicle, and the second radial velocity threshold is used to represent the velocity threshold value at which the rear vehicle and the vehicle collide; and a fourth determining subunit, used to determine a downward adjustment amount based on the azimuth component, the distance component, and the velocity component.

[0176] Optionally, the spatiotemporal perception information includes an initial signal-to-noise ratio (SNR), which represents the relative magnitude of the intensity of the echo signal received by the radar detection device relative to the noise. The vehicle trajectory determination device 500 further includes a fifth determination subunit, which determines the spatiotemporal perception information as valid data that meets the target detection threshold in response to the initial SNR being greater than or equal to the target detection threshold.

[0177] Optionally, the vehicle trajectory determination device 500 further includes: a storage subunit for storing spatiotemporal perception information into a valid data list; and a determination unit 508 including: an association subunit for performing association processing on the spatiotemporal perception information in the valid data list to obtain the initial trajectory of the vehicle; and a filtering subunit for filtering the initial trajectory to obtain the driving trajectory.

[0178] Optionally, the vehicle trajectory determination device 500 further includes a sixth determination subunit, used to determine the initial detection threshold as the target detection threshold in response to the following vehicle not being in the risk area.

[0179] In this embodiment, the acquisition unit 502 acquires the spatiotemporal perception information of the vehicles behind the vehicle relative to the vehicle; the detection unit 504 detects the vehicles behind based on the spatiotemporal perception information and obtains the detection result; the adjustment unit 506, in response to the detection result indicating that the vehicles behind are in a risk area, adjusts the initial detection threshold of the radar detection equipment in the vehicle based on the spatiotemporal perception information to obtain the target detection threshold, wherein the risk area is used to represent the area where the echo power of the radar detection equipment is attenuated, and the target detection threshold is lower than the initial detection threshold; the determination unit 508, in response to the spatiotemporal perception information satisfying the target detection threshold, determines the driving trajectory of the vehicles behind, thereby solving the technical problem of low accuracy in determining the driving trajectory of the vehicles behind and achieving the technical effect of improving the accuracy of determining the driving trajectory of the vehicles behind.

[0180] This application also provides an electronic device 60, please refer to... Figure 6 , Figure 6 This is a structural diagram of an electronic device provided in one embodiment of the present application, including a processor 610 and a memory 620, wherein the memory 620 is used to store computer programs; the processor 610 is used to execute the programs stored in the memory 620 to implement the methods described in any embodiment of the present application.

[0181] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0182] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0183] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0184] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0185] According to another aspect of the embodiments of this application, a vehicle is also provided. The vehicle includes a memory and a processor. The memory stores an executable program; the processor is used to run the program, which, when running, implements the methods described in the embodiments of this application.

[0186] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0187] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0188] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0189] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0190] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0191] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for determining the driving trajectory of a vehicle, characterized in that, include: Acquire spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle. Based on the spatiotemporal perception information, the vehicle behind is detected, and the detection result is obtained; In response to the detection result that the vehicle behind is in a risk zone, the initial detection threshold of the radar detection device in the vehicle is adjusted based on the spatiotemporal perception information to obtain a target detection threshold. The risk zone is used to represent the area where the echo power of the radar detection device is attenuated, and the target detection threshold is lower than the initial detection threshold. In response to the spatiotemporal perception information satisfying the target detection threshold, the driving trajectory of the vehicle behind is determined.

2. The method according to claim 1, characterized in that, The spatiotemporal perception information includes at least one of the following: the azimuth angle, distance, and radial velocity of the rear vehicle relative to the vehicle. In response to the detection result indicating that the rear vehicle is in a risk zone, the initial detection threshold of the radar detection equipment in the vehicle is adjusted based on the spatiotemporal perception information to obtain a target detection threshold, including: In response to the following vehicle being in the risk area, based on the azimuth angle, the distance, and the radial velocity, a threshold adjustment function model is invoked to determine the downward adjustment amount of the initial detection threshold. The threshold adjustment function model is obtained by fitting azimuth angle samples, distance samples, and radial velocity samples. The azimuth angle samples represent the azimuth angle of the following vehicle sample relative to the vehicle sample; the distance samples represent the distance of the following vehicle sample relative to the vehicle sample; and the radial velocity samples represent the radial velocity of the following vehicle sample relative to the vehicle sample. The target detection threshold is obtained by lowering the initial detection threshold using the aforementioned adjustment amount.

3. The method according to claim 2, characterized in that, In response to the following vehicle being in the risk zone, based on the azimuth angle, the distance, and the radial velocity, a threshold adjustment function model is invoked to determine the downward adjustment amount of the initial detection threshold, including: In response to the following vehicle being in the risk zone, the threshold adjustment function model is invoked to determine the azimuth component of the following vehicle, wherein the azimuth component is used to represent the degree of deviation of the following vehicle relative to the vehicle. In response to the distance being greater than a distance threshold, the threshold adjustment function model is invoked to determine the distance component of the vehicle behind, wherein the distance component is used to represent the component that maintains stable detection of the vehicle behind within the target distance; In response to the radial velocity being less than a first radial velocity threshold and the absolute value of the radial velocity being greater than a second radial velocity threshold, the threshold adjustment function model is invoked to determine the velocity component of the following vehicle, wherein the first radial velocity threshold is used to represent the lower limit of the radial velocity between the following vehicle and the vehicle, and the second radial velocity threshold is used to represent the critical velocity value at which the following vehicle and the vehicle collide. The downward adjustment amount is determined based on the azimuth component, the distance component, and the velocity component.

4. The method according to claim 1, characterized in that, The spatiotemporal sensing information includes an initial signal-to-noise ratio (SNR), which represents the relative magnitude of the echo signal intensity received by the radar detection device relative to the noise. The method further includes: In response to the initial signal-to-noise ratio being greater than or equal to the target detection threshold, the spatiotemporal perception information is determined to be valid data that satisfies the target detection threshold.

5. The method according to claim 1, characterized in that, The method further includes: Store the spatiotemporal perception information into a list of valid data; In response to the spatiotemporal perception information satisfying the target detection threshold, the driving trajectory of the vehicle behind is determined, including: The spatiotemporal perception information in the effective data list is correlated to obtain the initial driving trajectory of the vehicle; The initial driving trajectory is filtered to obtain the driving trajectory.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: In response to the fact that the vehicle behind is not in the risk area, the initial detection threshold is determined as the target detection threshold.

7. A device for determining the driving trajectory of a vehicle, characterized in that, include: The acquisition unit is used to acquire spatiotemporal perception information of vehicles behind the vehicle relative to the vehicle. The detection unit is used to detect the vehicle behind based on the spatiotemporal perception information and obtain the detection result; An adjustment unit is configured to, in response to the detection result indicating that the vehicle behind is in a risk zone, adjust the initial detection threshold of the radar detection device in the vehicle based on the spatiotemporal perception information to obtain a target detection threshold, wherein the risk zone is used to represent the area where the echo power of the radar detection device is attenuated, and the target detection threshold is lower than the initial detection threshold; A determining unit is used to determine the driving trajectory of the vehicle behind in response to the spatiotemporal perception information satisfying the target detection threshold.

8. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 6 when it runs.

9. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 6.

10. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 6.

Images