A confidence determination method, device, and equipment for a traffic moving target and a medium

Abstract

The application discloses a confidence determination method, device and equipment for a traffic moving target and a medium. The method comprises the following steps: acquiring a transverse and longitudinal relative distance and a transverse and longitudinal relative speed; determining a distance jump result and a distance jump confidence according to a difference value relationship between a difference of the transverse and longitudinal relative distance and a difference of the transverse and longitudinal relative speed; determining a corresponding difference relative speed according to a difference of the transverse and longitudinal relative distance in multiple periods, determining a speed jump result according to a difference value relationship between the difference relative speed and the transverse and longitudinal relative speed, and determining a speed jump confidence according to a relationship between the speed jump result and a speed jump threshold; and determining a confidence of a target vehicle according to a relationship between a collision time between a host vehicle and the target vehicle and a time threshold and a relationship between the distance jump confidence and the speed jump confidence. The application solves the technical problem of false triggering of a collision warning or automatic emergency braking of a vehicle for a transverse moving target.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and specifically to a method, apparatus, electronic device, and medium for determining the confidence level of a moving traffic target. Background Technology

[0002] The autonomous driving measurement systems are equipped with pure driver assistance cameras, driver assistance cameras plus radar, or radar-equipped vehicles. These systems detect vehicles (motorized and non-motorized) ahead in real time, determining the distance, position, and relative speed between the vehicle and the vehicle in front. When a potential collision hazard exists and the distance to the vehicle in front is less than a safe range, the system provides prompts to guide the driver to take timely action. The ADAS vision subsystem identifies pedestrians around the vehicle, detects the distance between pedestrians and the vehicle, and issues warnings accordingly. Once a warning is triggered, the system provides voice prompts to the driver to take timely measures, thus assisting in safe driving.

[0003] In the face of the aforementioned dangerous situations, if the driver does not respond to the alarm, or if the situation becomes extremely urgent and the danger escalates, the system will automatically apply emergency braking to avoid an accident or reduce the damage. However, during actual debugging or road testing, it was found that systems implementing the above functions, especially pure driver assistance camera solutions, are prone to falsely triggering FCW (Forward Collision Warning) or AEB (Autonomous Emergency Braking) alarms when encountering targets crossing the road, particularly two-wheeled vehicles, even without triggering an emergency scenario. This can interfere with the driver's normal driving and, in severe cases, may lead to rear-end collisions. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a method, device, electronic device and medium for judging the confidence of moving traffic targets, thereby solving the technical problem in the prior art that lateral moving targets can easily trigger forward collision warnings or vehicle automatic emergency braking systems even when an emergency scenario has not been triggered.

[0005] To solve the above problems, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a method for determining the confidence level of a moving traffic target, comprising:

[0007] Obtain the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle;

[0008] Based on the difference between the difference in the horizontal and vertical relative distances and the difference in the horizontal and vertical relative velocities, the distance jump result is determined, and based on the relationship between the distance jump result and the distance jump threshold, the distance jump confidence level is determined.

[0009] The corresponding differential relative velocity is determined based on the difference between the horizontal and vertical relative distances of multiple cycles. The velocity jump result is determined based on the difference relationship between the differential relative velocity and the horizontal and vertical relative velocity. The velocity jump confidence is determined based on the relationship between the velocity jump result and the velocity jump threshold.

[0010] Based on the distance jump confidence and the speed jump confidence, the relationship between the collision time between the vehicle and the forward target vehicle and the time threshold is compared to determine the confidence of the target vehicle.

[0011] In some embodiments, determining the distance jump result based on the difference between the difference in the lateral and longitudinal relative distances and the difference in the lateral and longitudinal relative velocities includes:

[0012] Construct a correlation model between the horizontal and vertical relative distances and the horizontal and vertical relative velocities;

[0013] The association model is discretized to obtain a discrete association model;

[0014] Based on the discrete correlation model, the difference between the horizontal and vertical relative distances and the difference between the horizontal and vertical relative velocities are determined.

[0015] In some embodiments, determining the distance jump confidence level based on the relationship between the distance jump result and the distance jump threshold includes:

[0016] Determine the distance jump threshold based on the horizontal and vertical relative distances;

[0017] Determine whether the distance jump result is less than the distance jump threshold;

[0018] If it is less than, then the confidence level of the relative horizontal and vertical distances is determined to be reliable.

[0019] In some embodiments, obtaining the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle includes:

[0020] Obtain the vehicle's speed, vehicle position, target vehicle's lateral and longitudinal speeds, and target vehicle's lateral and longitudinal distances;

[0021] Calculate the standard deviations of the vehicle speed, the target vehicle's lateral and longitudinal speeds, and the target vehicle's lateral and longitudinal distances, respectively. Based on the standard deviations of the vehicle speed, the target vehicle's lateral and longitudinal speeds, and the target vehicle's lateral and longitudinal distances, determine the initial vehicle speed, the initial target vehicle's lateral and longitudinal speeds, and the initial target vehicle's lateral and longitudinal distances.

[0022] Based on preset parameter scores, probability scores are performed on the initial vehicle speed, the initial target vehicle lateral and longitudinal speeds, and the initial target vehicle lateral and longitudinal distances to determine the cleaning vehicle speed, the cleaning target vehicle lateral and longitudinal speeds, and the cleaning target vehicle lateral and longitudinal distances.

[0023] The relative speeds between the cleaning vehicle and the forward target vehicle are determined based on the relative relationship between the cleaning vehicle's speed and the target vehicle's lateral and longitudinal speeds.

[0024] The relative distances between the vehicle and the target vehicle are determined based on the relative relationship between the vehicle's position and the target vehicle's lateral and longitudinal distances.

[0025] In some embodiments, determining the confidence level of the target vehicle by comparing the collision time between the vehicle and the forward target vehicle with a time threshold based on the distance jump confidence level and the speed jump confidence level includes:

[0026] If both the distance jump confidence and the speed jump confidence are reliable, then the collision time between the vehicle and the forward target vehicle is determined.

[0027] Compare the relationship between the collision time and the time threshold;

[0028] If the collision time is less than the time threshold, the confidence level of the target vehicle is determined to be unreliable, and an alarm function is triggered.

[0029] In some embodiments, obtaining the time threshold includes:

[0030] The time threshold is determined based on the degree of influence of the target vehicle's lateral and longitudinal distances on the collision time.

[0031] In some embodiments, it also includes:

[0032] Mark the type of the target vehicle in the forward direction;

[0033] If the target vehicle type is a two-wheeled vehicle, then the time threshold, distance threshold, and speed threshold for the two-wheeled vehicle are determined based on the degree of influence of the target vehicle type on the time threshold, distance threshold, and speed threshold.

[0034] Secondly, the present invention also provides a confidence determination device for a moving traffic target, comprising:

[0035] The acquisition module is used to acquire the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle.

[0036] The distance jump confidence determination module is used to determine the distance jump result based on the difference between the difference between the horizontal and vertical relative distances and the difference between the horizontal and vertical relative velocities, and to determine the distance jump confidence based on the relationship between the distance jump result and the distance jump threshold.

[0037] The speed jump confidence determination module is used to determine the corresponding differential relative speed based on the difference between the horizontal and vertical relative distances of multiple cycles, determine the speed jump result based on the difference relationship between the differential relative speed and the horizontal and vertical relative speeds, and determine the speed jump confidence based on the relationship between the speed jump result and the speed jump threshold.

[0038] The target vehicle confidence determination module is used to determine the confidence level of the target vehicle by comparing the relationship between the collision time between the vehicle and the forward target vehicle and a time threshold based on the distance jump confidence level and the speed jump confidence level.

[0039] Thirdly, the present invention also provides an electronic device, comprising: a processor and a memory;

[0040] The memory stores a computer-readable program that can be executed by the processor;

[0041] When the processor executes the computer-readable program, it implements the steps in the confidence determination method for traffic movement targets as described above.

[0042] Fourthly, the present invention also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps in the confidence determination method for traffic moving targets as described above.

[0043] Compared with existing technologies, the confidence assessment method, apparatus, device, and medium for traffic moving targets provided by this invention first acquires the lateral and longitudinal relative distances and lateral and longitudinal relative velocities between the vehicle and the forward target vehicle; then, based on the difference relationship between the difference in the lateral and longitudinal relative distances and the difference in the lateral and longitudinal relative velocities, a distance jump result is determined, and based on the relationship between the distance jump result and a distance jump threshold, a distance jump confidence level is determined; furthermore, based on the difference in the lateral and longitudinal relative distances over multiple periods, a corresponding differential relative velocity is determined, and based on the difference relationship between the differential relative velocity and the lateral and longitudinal relative velocities, a speed jump result is determined, and based on the relationship between the speed jump result and a speed jump threshold, a speed jump confidence level is determined; finally, based on the distance jump confidence level and the speed jump confidence level, the relationship between the collision time between the vehicle and the forward target vehicle and a time threshold is compared to determine the confidence level of the target vehicle. This invention determines the risk level of the vehicle to the vehicle by predicting the confidence level of distance and speed jumps of the vehicle in front, and then analyzing the collision time based on the distance and speed jumps. This improves the accuracy of the vehicle's alarm triggering when the vehicle is moving in front, and also improves the accuracy of judging scenarios where the vehicle is crossing or turning in front, thereby avoiding the problem of false alarms being triggered. Attached Figure Description

[0044] Figure 1 This is a flowchart of an embodiment of the confidence determination method for traffic moving targets provided by the present invention;

[0045] Figure 2 This is a schematic diagram of an embodiment of step S102 in the confidence determination method for traffic moving targets provided by the present invention;

[0046] Figure 3 This is a schematic diagram of an embodiment of the relationship between the distance jump threshold and the target distance in the confidence determination method for traffic moving targets provided by the present invention;

[0047] Figure 4 This is a flowchart of an embodiment of step S101 in the confidence determination method for traffic moving targets provided by the present invention;

[0048] Figure 5 This is a schematic diagram of an embodiment of the collision time calculation dimension in the confidence determination method for traffic moving targets provided by the present invention;

[0049] Figure 6 This is a schematic diagram of an embodiment of the confidence determination device for traffic moving targets provided by the present invention;

[0050] Figure 7 This is a schematic diagram of the operating environment of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0052] This invention provides a method for determining the confidence level of a moving traffic target. Please refer to [link to relevant documentation]. Figure 1 ,include:

[0053] S101. Obtain the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle.

[0054] S102. Based on the difference between the difference between the horizontal and vertical relative distances and the difference between the horizontal and vertical relative velocities, determine the distance jump result, and based on the relationship between the distance jump result and the distance jump threshold, determine the distance jump confidence level.

[0055] S103. Determine the corresponding differential relative velocity based on the difference between the horizontal and vertical relative distances of multiple cycles, determine the velocity jump result based on the difference relationship between the differential relative velocity and the horizontal and vertical relative velocity, and determine the velocity jump confidence level based on the relationship between the velocity jump result and the velocity jump threshold.

[0056] S104. Based on the distance jump confidence and the speed jump confidence, compare the relationship between the collision time between the vehicle and the forward target vehicle and the time threshold to determine the confidence of the target vehicle.

[0057] In this embodiment, firstly, the lateral and longitudinal relative distances and lateral and longitudinal relative velocities between the vehicle and the forward target vehicle are acquired; then, based on the difference between the difference in the lateral and longitudinal relative distances and the difference in the lateral and longitudinal relative velocities, a distance jump result is determined, and based on the relationship between the distance jump result and a distance jump threshold, a distance jump confidence level is determined; furthermore, based on the difference in the lateral and longitudinal relative distances over multiple periods, a corresponding differential relative velocity is determined, and based on the difference between the differential relative velocity and the lateral and longitudinal relative velocities, a speed jump result is determined, and based on the relationship between the speed jump result and a speed jump threshold, a speed jump confidence level is determined; finally, based on the distance jump confidence level and the speed jump confidence level, the relationship between the collision time between the vehicle and the forward target vehicle and a time threshold is compared to determine the confidence level of the target vehicle. This invention determines the risk level of the vehicle to the vehicle by predicting the confidence level of distance and speed jumps of the vehicle in front, and then analyzing the collision time based on the distance and speed jumps. This improves the accuracy of the vehicle's alarm triggering when the vehicle is moving in front, and also improves the accuracy of judging scenarios where the vehicle is crossing or turning in front, thereby avoiding the problem of false alarms being triggered.

[0058] It should be noted that the target vehicle in front includes the vehicle in front that is driving in the lane, the vehicle in front of the vehicle in front, etc., and may also include various types of vehicles parked on the side of the road, as well as other movable targets such as pedestrians and animals on the side of the road.

[0059] In step S102, a distance jump refers to a sudden change in the lateral or longitudinal relative distance between the vehicle and a forward target vehicle. This jump may indicate that the target vehicle has laterally shifted or accelerated / decelerated. For example, if the target vehicle suddenly changes lanes in front of the vehicle, the lateral relative distance between the two vehicles will jump; similarly, if the forward target vehicle suddenly crosses the road, it can also cause a distance jump relative to the vehicle. Detecting distance jumps helps the system promptly identify such changes and make corresponding decisions, thereby improving the accuracy of the vehicle alarm system's alerts for forward vehicles crossing the road or suddenly changing direction.

[0060] Furthermore, the difference refers to the distance variation within adjacent time intervals. If the difference exceeds a set threshold, it is determined that a distance jump has occurred. Based on the relationship between the distance jump result and the distance jump threshold, the confidence level of the distance jump can be determined. The confidence level can be measured based on the magnitude of the jump difference or the jump frequency; larger differences or more frequent jumps will result in a higher confidence level.

[0061] In step S103, a speed jump refers to a sudden change in the lateral or longitudinal relative speed between the vehicle and the target vehicle. This jump may indicate that the target vehicle is accelerating or decelerating, or that the relative speed between the two vehicles has changed abruptly. For example, if the target vehicle suddenly decelerates in front of the vehicle, a jump in the longitudinal relative speed between them will occur. Detecting speed jumps helps the system predict and respond to the target vehicle's motion behavior, thereby improving vehicle safety and reliability.

[0062] It should be noted that the camera first determines whether there is a stable following target (CIPV) in front of the vehicle. If there is, according to real vehicle tests and statistics from long-mileage road tests, the main false triggering in this scenario is FCW. At the same time, CIPV is present, and there is a high probability that lateral vehicles will not intrude. Therefore, the FCW function can be suppressed in this scenario, but the AEB function is retained to avoid the occurrence of emergency situations.

[0063] In some embodiments, please refer to Figure 2 The step of determining the distance jump result based on the difference between the difference in the lateral and longitudinal relative distances and the difference in the lateral and longitudinal relative velocities includes:

[0064] S201. Construct a correlation model between the horizontal and vertical relative distances and the horizontal and vertical relative velocities;

[0065] S202. Discretize the association model to obtain a discrete association model;

[0066] S203. Based on the discrete correlation model, determine the difference between the horizontal and vertical relative distances and the difference between the horizontal and vertical relative velocities.

[0067] In this embodiment, the correlation model is a mathematical model used to describe the relationship between lateral and longitudinal relative distances and lateral and longitudinal relative velocities. This model can be constructed using statistical methods, machine learning algorithms, or other mathematical modeling techniques. It can capture the correlation between relative distance and relative velocity; for example, as the distance to a target vehicle increases, the relative velocity may decrease. To better analyze and process the correlation model, it can be discretized. This means dividing the continuous range of distance and velocity into discrete intervals. Through discretization, the correlation model can be transformed into a discrete correlation model, where each interval represents a set of distance and velocity values. By calculating the difference between the differences in lateral and longitudinal relative distances and the differences in lateral and longitudinal relative velocities, the result of distance jumps can be determined. Based on the design and analysis of the discrete correlation model, the degree of abrupt change in distance and the change in relative velocity can be judged based on the magnitude and direction of the difference. By performing confidence assessments on distance jumps and velocity jumps,

[0068] In this embodiment, the relationship between relative distance and relative velocity is as follows:

[0069] D e dt = v 相对

[0070] Discretized

[0071]

[0072] Where D e Let T be the target's longitudinal relative distance, T be the camera's processing or signal cycle, and V be the target's relative velocity. Therefore, even if there is a difference between the target's relative distance and relative velocity, the difference should not be too large. Thus, the difference between the target's longitudinal relative distance and relative velocity is calculated, and the absolute value of the difference is used to determine the target's confidence level. If the absolute value exceeds a certain threshold, the target is considered to have low confidence. This strategy only uses the target's longitudinal distance before and after one cycle. A typical scenario that this strategy can filter is a jump in the target's relative distance, but not in the relative velocity.

[0073] Using only the longitudinal distance of the target within a single cycle fails to filter out scenarios where the relative distance remains constant but the relative velocity changes drastically. Therefore, in another embodiment, the difference between the longitudinal relative distances of the target over several cycles is used to represent the relative velocity calculated from the relative distance. This difference is then compared to the measured relative velocity. If the absolute value of the difference exceeds a certain threshold, the target is considered to have low confidence. This strategy uses the difference between the longitudinal relative distances of the target over several cycles to represent the relative velocity calculated from the relative distance. A typical scenario this strategy can filter is a change in the target's relative velocity without a change in the relative distance.

[0074] In some embodiments, determining the distance jump confidence level based on the relationship between the distance jump result and the distance jump threshold includes:

[0075] Determine the distance jump threshold based on the horizontal and vertical relative distances;

[0076] Determine whether the distance jump result is less than the distance jump threshold;

[0077] If it is less than, then the confidence level of the relative horizontal and vertical distances is determined to be reliable.

[0078] In this embodiment, considering the inherent distance fluctuations in distant target recognition, the threshold is relatively large. The greater the target distance, the larger the threshold. For a specific embodiment, please refer to Table 1 and... Figure 3 Table 1 shows the relationship between target distance and distance jump threshold:

[0079] Table 1

[0080] Target distance / m 20 60 100 150 Distance jump threshold / m 2 6 10 10

[0081] Furthermore, the threshold for matching target distance with relative speed, i.e. the speed jump threshold, is also related to the target distance; the greater the target distance, the larger the threshold.

[0082] In some embodiments, the acquisition of the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle is described in [reference needed]. Figure 4 ,include:

[0083] S401, Obtain the vehicle's speed, vehicle position, target vehicle's lateral and longitudinal speeds, and target vehicle's lateral and longitudinal distances;

[0084] S402. Calculate the standard deviations of the vehicle speed, the target vehicle's lateral and longitudinal speeds, and the target vehicle's lateral and longitudinal distances respectively. Based on the standard deviations of the vehicle speed, the target vehicle's lateral and longitudinal speeds, and the target vehicle's lateral and longitudinal distances, determine the initial vehicle speed, the initial target vehicle's lateral and longitudinal speeds, and the initial target vehicle's lateral and longitudinal distances.

[0085] S403. Based on the preset parameter scores, perform probability scoring on the initial vehicle speed, the initial target vehicle lateral and longitudinal speeds, and the initial target vehicle lateral and longitudinal distances to determine the cleaning vehicle speed, the cleaning target vehicle lateral and longitudinal speeds, and the cleaning target vehicle lateral and longitudinal distances.

[0086] S404. Determine the relative speeds of the cleaning vehicle and the forward target vehicle in the lateral and longitudinal directions based on the relative relationship between the cleaning vehicle speed and the lateral and longitudinal speeds of the cleaning target vehicle.

[0087] S405. Determine the relative distance between the vehicle and the forward target vehicle based on the relative relationship between the vehicle's position and the horizontal and vertical distances between the vehicle and the target vehicle.

[0088] It should be noted that the vehicle's speed can be obtained through the vehicle's speed sensor or GPS positioning system. The vehicle's position can be measured using the vehicle's positioning system or relative reference points. The target vehicle's lateral and longitudinal speeds can be obtained through inter-vehicle communication or sensor data, while the target vehicle's lateral and longitudinal distances can be obtained through radar, cameras, or inter-vehicle communication.

[0089] Furthermore, standard deviation is a statistical indicator that measures the degree of dispersion of data. Calculating the standard deviation of a vehicle's speed, the target vehicle's lateral and longitudinal speeds, and lateral and longitudinal distances can help understand the variation in the data. A larger standard deviation indicates a higher degree of data dispersion, while a smaller standard deviation indicates a lower degree of data dispersion.

[0090] In this embodiment, the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle can be obtained. These results are of great significance for driver assistance systems and autonomous driving systems, and can be used for vehicle collision warning, trajectory planning, and decision-making.

[0091] In step S402, the camera generates two signals for each measured target parameter: the measured value and the standard deviation (STD). The formula for standard deviation is as follows:

[0092]

[0093] Here, n represents the amount of data, and x represents the target parameter. Therefore, the standard deviation (STD) indicates the dispersion of the measured values ​​from their mean. The larger the STD, the more dispersed the measured values ​​are from their mean. Signals with large jumps will also have a larger STD. Therefore, STD filtering can be used to filter out targets with large STDs.

[0094] In step S403, each target has an internal score that helps determine whether the object actually exists or is a false alarm. This parameter is a floating-point number between 0 and 1; a value above 0.95 indicates a high probability of target existence, while a value below 0.6 indicates a low probability of target existence. Therefore, the existence probability signal can be used to filter targets, eliminating those with a low probability of existence.

[0095] It should be noted that, in order to reduce the impact on normal targets, for targets whose IDs do not change, if the conditions are met within a certain period of detection, then no further confidence assessment will be performed, and it will be assumed that the confidence will remain high thereafter.

[0096] In some embodiments, determining the confidence level of the target vehicle by comparing the collision time between the vehicle and the forward target vehicle with a time threshold based on the distance jump confidence level and the speed jump confidence level includes:

[0097] If both the distance jump confidence and the speed jump confidence are reliable, then the collision time between the vehicle and the forward target vehicle is determined.

[0098] Compare the relationship between the collision time and the time threshold;

[0099] If the collision time is less than the time threshold, the confidence level of the target vehicle is determined to be unreliable, and an alarm function is triggered.

[0100] In this embodiment, the threshold for the existence of a probability signal mainly considers the target category, target distance, and target TTC. First, the target category is considered because recognition of two-wheeled vehicle targets may be unstable, so the threshold design is relatively lenient. Second, the target distance is considered because distant targets are less dangerous, so the threshold design is relatively strict. Finally, the target TTC is considered because targets with higher TTC have longer collision times, so the threshold design is relatively strict.

[0101] Furthermore, the confidence level of the target vehicle is determined by comparing the collision time between the self-vehicle and the oncoming target vehicle with a time threshold. The collision time is calculated based on the current vehicle state information and the prediction model, while the time threshold is a pre-set value used to judge the urgency of the collision. If both the distance jump confidence and the speed jump confidence are considered reliable, then we can continue to calculate the collision time between the self-vehicle and the target vehicle. The calculated collision time is compared with the time threshold. If the collision time is less than the time threshold, it indicates that the collision risk is imminent or has exceeded the urgency threshold. In this case, the confidence level of the target vehicle is determined to be unreliable, and the alarm function and other emergency measures are activated.

[0102] It is important to note that the confidence assessment of distance jump confidence and velocity jump confidence may involve consideration of multiple factors, such as the accuracy of sensor data and environmental conditions. Furthermore, the determination of the time threshold also needs to be tailored to different application scenarios and safety standards.

[0103] In one specific embodiment, due to the incorrect angle of the vehicle entering laterally, the TTC judgment in both the lateral and longitudinal directions differs from that of vehicles with a high normal following overlap rate. This is specifically explained here.

[0104] In this scenario, the target vehicle and the driver vehicle are at an angle, so the lateral collision TTC calculation is as follows. Figure 5 This is a schematic diagram of TTC calculation dimensions.

[0105]

[0106]

[0107] Where D1 = the longitudinal distance between this vehicle and the vehicle in front at time t1;

[0108] D2 = The longitudinal distance between this vehicle and the vehicle in front at time t2;

[0109] W1 = The width of the vehicle in front is identified at time t1;

[0110] W2 = At ​​time t2, the width of the vehicle in front is identified;

[0111] C = Width variation coefficient;

[0112] TTC = Lateral Collision Time, which is the time required for the vehicle in front to collide with the vehicle in front at the current speed.

[0113] In some embodiments, it also includes:

[0114] Mark the type of the target vehicle in the forward direction;

[0115] If the target vehicle type is a two-wheeled vehicle, then the time threshold, distance threshold, and speed threshold for the two-wheeled vehicle are determined based on the degree of influence of the target vehicle type on the time threshold, distance threshold, and speed threshold.

[0116] In this embodiment, during vehicle identification and classification, different sensor data (such as cameras, radar, lidar, etc.) and machine learning algorithms or deep learning models can be used to classify and label forward target vehicles. These sensors can capture information about vehicle appearance, size, and motion behavior, enabling the autonomous driving system to identify and understand the type of target vehicle ahead. For two-wheeled vehicles, due to their different motion characteristics and stability compared to other vehicle types, the values ​​of time thresholds, distance jump thresholds, and speed jump thresholds need to be adjusted to better accommodate the behavior of two-wheeled vehicles.

[0117] The time threshold, which is the time limit for assessing collision risk, can be adjusted based on the dynamic characteristics of two-wheeled vehicles. Because two-wheeled vehicles typically have higher agility and acceleration, they can change their state of motion more quickly. Therefore, a relatively short time threshold may be necessary to provide earlier warnings and allow for action to avoid potential collisions.

[0118] Distance and speed jump thresholds also need to be adjusted based on the characteristics of two-wheeled vehicles. Two-wheeled vehicles have relatively small turning radii and high centers of gravity, so they may experience significant lateral displacement during turns. Therefore, sudden changes in distance and speed are more common in two-wheeled vehicles. To effectively control risk, the distance and speed jump thresholds need to be lowered to more sensitively detect potential risks.

[0119] It should be noted that the above thresholds should also take into account other factors, such as road conditions, traffic flow, and weather conditions, in order to obtain a more accurate and reliable assessment and judgment.

[0120] Based on the above-described method for determining the confidence level of moving traffic targets, this embodiment of the invention also provides a corresponding confidence determination device 600 for moving traffic targets. Please refer to [link to relevant documentation]. Figure 6The confidence determination device 600 for the traffic moving target includes an acquisition module 610, a distance jump confidence determination module 620, a speed jump confidence determination module 630, and a target vehicle confidence determination module 640.

[0121] The acquisition module 610 is used to acquire the lateral and longitudinal relative distances and lateral and longitudinal relative speeds between the vehicle and the forward target vehicle.

[0122] The distance jump confidence determination module 620 is used to determine the distance jump result based on the difference relationship between the difference between the lateral and longitudinal relative distances and the difference between the lateral and longitudinal relative velocities, and to determine the distance jump confidence based on the relationship between the distance jump result and the distance jump threshold.

[0123] The speed jump confidence determination module 630 is used to determine the corresponding differential relative speed based on the difference between the horizontal and vertical relative distances of multiple cycles, determine the speed jump result based on the difference relationship between the differential relative speed and the horizontal and vertical relative speeds, and determine the speed jump confidence based on the relationship between the speed jump result and the speed jump threshold.

[0124] The target vehicle confidence determination module 640 is used to determine the confidence of the target vehicle by comparing the relationship between the collision time between the vehicle and the forward target vehicle and a time threshold based on the distance jump confidence and the speed jump confidence.

[0125] like Figure 7 As shown, based on the aforementioned confidence assessment method for traffic moving targets, the present invention also provides an electronic device, which can be a mobile terminal, desktop computer, laptop, handheld computer, server, or other computing device. The electronic device includes a processor 710, a memory 720, and a display 730. Figure 7 Only some components of the electronic device are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0126] In some embodiments, memory 720 may be an internal storage unit of the electronic device, such as a hard disk or memory. In other embodiments, memory 720 may be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Furthermore, memory 720 may include both internal and external storage units. Memory 720 is used to store application software and various types of data installed on the electronic device, such as program code installed on the electronic device. Memory 720 may also be used to temporarily store data that has been output or will be output. In one embodiment, memory 720 stores a traffic moving target confidence assessment program 740, which can be executed by processor 710 to implement the traffic moving target confidence assessment method of the various embodiments of this application.

[0127] In some embodiments, processor 710 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 720 or process data, such as executing a confidence judgment method for traffic movement targets.

[0128] In some embodiments, display 730 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 730 is used to display information from the confidence assessment device for the moving target in the traffic and to display a user interface for visualization. Components 710-730 of the electronic device communicate with each other via a system bus.

[0129] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The storage medium can be a memory, magnetic disk, optical disk, etc.

[0130] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for determining the confidence of a moving object in traffic, characterized by, The method comprises: obtaining the lateral and longitudinal relative distance and the lateral and longitudinal relative speed between the ego vehicle and the front target vehicle; determining the distance jump result according to the difference between the difference of the lateral and longitudinal relative distance and the difference of the lateral and longitudinal relative speed, and determining the distance jump confidence according to the relationship between the distance jump result and the distance jump threshold value; determining the corresponding difference relative speed according to the difference of the lateral and longitudinal relative distance in multiple periods, determining the speed jump result according to the difference between the difference relative speed and the lateral and longitudinal relative speed, and determining the speed jump confidence according to the relationship between the speed jump result and the speed jump threshold value; determining the confidence of the target vehicle according to the relationship between the collision time between the ego vehicle and the front target vehicle and the time threshold value based on the distance jump confidence and the speed jump confidence.

2. The method according to claim 1, wherein The method comprises: constructing the correlation model between the lateral and longitudinal relative distance and the lateral and longitudinal relative speed; discretizing the correlation model to obtain the discrete correlation model; determining the difference of the lateral and longitudinal relative distance and the difference of the lateral and longitudinal relative speed based on the discrete correlation model.

3. The traffic moving object confidence determination method according to claim 2, characterized by, The method comprises: determining the distance jump threshold value according to the lateral and longitudinal relative distance; determining whether the distance jump result is less than the distance jump threshold value; if yes, determining that the confidence of the lateral and longitudinal relative distance is reliable.

4. The method according to claim 1, wherein The method comprises: obtaining the ego vehicle speed, the ego vehicle position, the target vehicle lateral and longitudinal speed, and the target vehicle lateral and longitudinal distance; calculating the standard deviation of the ego vehicle speed, the target vehicle lateral and longitudinal speed, and the target vehicle lateral and longitudinal distance respectively, and determining the initial ego vehicle speed, the initial target vehicle lateral and longitudinal speed, and the initial target vehicle lateral and longitudinal distance based on the corresponding standard deviation of the ego vehicle speed, the target vehicle lateral and longitudinal speed, and the target vehicle lateral and longitudinal distance; performing probability scoring on the initial ego vehicle speed, the initial target vehicle lateral and longitudinal speed, and the initial target vehicle lateral and longitudinal distance based on the preset parameter score to determine the ego vehicle speed, the target vehicle lateral and longitudinal speed, and the target vehicle lateral and longitudinal distance; determining the lateral and longitudinal relative speed between the ego vehicle and the front target vehicle based on the relative relationship between the ego vehicle speed and the target vehicle lateral and longitudinal speed; determining the lateral and longitudinal relative distance between the ego vehicle and the front target vehicle based on the relative relationship between the ego vehicle position and the target vehicle lateral and longitudinal distance.

5. The traffic moving object confidence determination method according to claim 4, characterized by, The method comprises: if the distance jump confidence and the speed jump confidence are both reliable, determining the collision time between the ego vehicle and the front target vehicle. comparing the collision time with a time threshold value; if the collision time is less than the time threshold value, determining that the confidence of the target vehicle is not reliable, and triggering an alarm function.

6. The traffic moving object confidence determination method according to claim 5, characterized by, obtaining the time threshold value, comprising: determining the time threshold value according to the influence degree of the lateral and longitudinal distance of the target vehicle on the collision time.

7. The method according to claim 1, wherein further comprising: labeling the type of the forward target vehicle; if the type of the forward target vehicle is a two-wheeled vehicle, determining the time threshold value, the distance jump threshold value and the speed jump threshold value of the two-wheeled vehicle according to the influence degree of the type of the forward target vehicle on the time threshold value, the distance jump threshold value and the speed jump threshold value.

8. A confidence determination device for a moving object in traffic, characterized in that comprising: an obtaining module, configured to obtain lateral and longitudinal relative distances and lateral and longitudinal relative speeds between a host vehicle and a forward target vehicle; a distance jump confidence determining module, configured to determine a distance jump result according to a difference relationship between a difference of the lateral and longitudinal relative distances and a difference of the lateral and longitudinal relative speeds, and determine a distance jump confidence according to a relationship between the distance jump result and a distance jump threshold value; a speed jump confidence determining module, configured to determine a difference relative speed corresponding to the difference of the lateral and longitudinal relative distances according to a plurality of periods of the difference of the lateral and longitudinal relative distances, determine a speed jump result according to a difference relationship between the difference relative speed and the lateral and longitudinal relative speeds, and determine a speed jump confidence according to a relationship between the speed jump result and a speed jump threshold value; a target vehicle confidence determining module, configured to compare a collision time between the host vehicle and the forward target vehicle with a time threshold value according to the distance jump confidence and the speed jump confidence, and determine a confidence of the target vehicle.

9. An electronic device, comprising: comprising: a processor and a memory; the memory stores a computer readable program which can be executed by the processor; the processor executes the computer readable program to realize steps in the confidence determination method of the moving target vehicle according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, the computer readable storage medium stores one or more programs which can be executed by one or more processors to realize steps in the confidence determination method of the moving target vehicle according to any one of claims 1-7.

Images