Motion information prediction method and device, computer device and storage medium

Abstract

The application relates to a motion information prediction method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring first fusion data; determining a detected target based on the first fusion data; adjusting the credibility of the detected target according to whether the detected target can be detected by a sensor; determining a to-be-detected target in the detected target according to the adjusted credibility; performing matching processing based on the prior prediction behavior of a reference vehicle around the to-be-detected target and the posterior behavior of the reference vehicle within a set time window; determining the detection information of the to-be-detected target according to the matching result; and predicting the behavior of the to-be-detected target according to the detection information of the to-be-detected target to obtain the predicted motion information of the to-be-detected target. The application can improve the accuracy of predicting the motion information of the detected target.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to methods, devices, computer equipment, storage media, and computer program products for predicting motion information. Background Technology

[0002] With the development of autonomous driving technology, the demand for perception fusion fault-tolerant technology is increasing in high-level autonomous driving. Perception fusion technology needs to perceive targets around the vehicle and predict their motion information (such as behavior, trajectory, etc.) to provide the system with correct behavior planning.

[0003] However, most current fault-tolerant technologies for perception fusion employ hardware redundancy. Even these hardware-based fault-tolerant technologies face scenarios where sensors miss or falsely detect objects, leading to ineffective perception of environmental information and incorrect predictions of target behavior and trajectories. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for predicting motion information to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for predicting motion information. The method includes:

[0006] Acquire the first fused data; the first fused data includes the detection data from each sensor of the target vehicle;

[0007] The target to be detected is determined based on the first fused data. The confidence level of the target to be detected is adjusted according to the sensor detection status of the target. Based on the adjusted confidence level, the target to be detected is determined among the targets to be detected.

[0008] The detection information of the target is determined based on the prior predicted behavior of reference vehicles around the target and the posterior behavior of the reference vehicles within a set time window.

[0009] Based on the detection information of the target to be detected, the motion information of the target to be detected is predicted to obtain the predicted motion information of the target to be detected.

[0010] In one embodiment, the confidence level is adjusted based on the sensor detection data of the target being detected, and based on the adjusted confidence level, the target to be detected among the detected targets includes:

[0011] For each target to be detected, the initial confidence level of the target is determined when it is first detected.

[0012] The credibility of the target is adjusted based on the initial credibility, the sensor detection status of the target, and the credibility adjustment strategy.

[0013] If the adjusted confidence level is lower than the first threshold, the target to be detected is identified as the target to be detected.

[0014] In one embodiment, adjusting the credibility of the detected target based on the initial credibility, the sensor detection status of the detected target, and the credibility adjustment strategy includes:

[0015] If the detected target is continuously detected, the credibility of the detected target is increased according to the preset credibility increase strategy and the initial credibility.

[0016] If no detection data of the target is detected within a preset time period, the credibility of the target is reduced according to the preset credibility reduction strategy and the initial credibility.

[0017] In one embodiment, a matching process is performed based on the prior predicted behavior of reference vehicles surrounding the target to be detected and the posterior behavior of the reference vehicles within a set time window. The detection information of the target to be detected is determined based on the matching results, including:

[0018] For each reference vehicle around the target to be detected, the first motion information of the reference vehicle is determined if the target to be detected does not exist, and the second motion information of the reference vehicle is determined if the target to be detected exists; the first motion information and the second motion information constitute the prior prediction behavior.

[0019] Based on the second fusion data of the reference vehicle under the set time window, the posterior behavior of the surrounding vehicles of the target to be detected at the current time is obtained;

[0020] Matching is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined based on the matching results.

[0021] In one embodiment, matching processing is performed based on first motion information, second motion information, and posterior behavior, and the detection information of the target to be detected is determined based on the matching result, including:

[0022] Determine the first matching result between the first motion information and the second motion information, the second matching result between the posterior behavior and the first motion information, and the third matching result between the posterior behavior and the second motion information;

[0023] Based on the first matching result, the second matching result, and the third matching result, as well as the mapping relationship between the preset matching result set and the credibility adjustment rule, the credibility adjustment rule of the target to be detected is determined.

[0024] The credibility of the target to be detected is adjusted by the credibility adjustment rules of the target to be detected to obtain the credibility of the target to be detected.

[0025] If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

[0026] In one embodiment, for each reference vehicle surrounding the target to be detected, the first motion information of the reference vehicle when the target to be detected does not exist, and the second motion information of the reference vehicle when the target to be detected exists, include:

[0027] For each reference vehicle surrounding the target to be detected, based on the information of each reference vehicle surrounding the target, the danger level of each reference vehicle surrounding the target, and the map lane line information, the first motion information of the reference vehicle is predicted if the target to be detected does not exist; and based on the information of each reference vehicle surrounding the target, the danger level of each reference vehicle surrounding the target, and the map lane line information, the second motion information of the reference vehicle is predicted if the target to be detected exists.

[0028] Secondly, this application also provides a motion information prediction device. The device includes:

[0029] The acquisition module is used to acquire the first fused data; the first fused data includes the detection data of each sensor of the target vehicle;

[0030] The determination module determines the target to be detected based on the first fused data, adjusts the confidence level of the target to be detected based on the sensor detection status of the target to be detected, and determines the target to be detected among the targets to be detected based on the adjusted confidence level.

[0031] The matching module is used to perform matching processing based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles within a set time window, and to determine the detection information of the target to be detected based on the matching results.

[0032] The prediction module is used to predict the motion information of the target to be detected based on the detection information of the target to be detected, and obtain the predicted motion information of the target to be detected.

[0033] In one embodiment, the acquisition module is specifically used for:

[0034] For each target to be detected, the initial confidence level of the target is determined when it is first detected.

[0035] The credibility of the target is adjusted based on the initial credibility, the sensor detection status of the target, and the credibility adjustment strategy.

[0036] If the adjusted confidence level is lower than the first threshold, the target to be detected is identified as the target to be detected.

[0037] In one embodiment, the acquisition module is specifically used for:

[0038] If the detected target is continuously detected, the credibility of the detected target is increased according to the preset credibility increase strategy and the initial credibility.

[0039] If no detection data of the target is detected within a preset time period, the credibility of the target is reduced according to the preset credibility reduction strategy and the initial credibility.

[0040] In one embodiment, the determining module is specifically used for:

[0041] For each reference vehicle around the target to be detected, the first motion information of the reference vehicle is determined if the target to be detected does not exist, and the second motion information of the reference vehicle is determined if the target to be detected exists; the first motion information and the second motion information constitute the prior prediction behavior.

[0042] Based on the second fusion data of the reference vehicle under the set time window, the posterior behavior of the surrounding vehicles of the target to be detected at the current time is obtained;

[0043] Matching is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined based on the matching results.

[0044] In one embodiment, the determining module is specifically used for:

[0045] Determine the first matching result between the first motion information and the second motion information, the second matching result between the posterior behavior and the first motion information, and the third matching result between the posterior behavior and the second motion information;

[0046] Based on the first matching result, the second matching result, and the third matching result, as well as the mapping relationship between the preset matching result set and the credibility adjustment rule, the credibility adjustment rule of the target to be detected is determined.

[0047] The credibility of the target to be detected is adjusted by the credibility adjustment rules of the target to be detected to obtain the credibility of the target to be detected.

[0048] If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

[0049] In one embodiment, the determining module is specifically used for:

[0050] For each reference vehicle surrounding the target to be detected, based on the information of each reference vehicle surrounding the target, the danger level of each reference vehicle surrounding the target, and the map lane line information, the first motion information of the reference vehicle is predicted if the target to be detected does not exist; and based on the information of each reference vehicle surrounding the target, the danger level of each reference vehicle surrounding the target, and the map lane line information, the second motion information of the reference vehicle is predicted if the target to be detected exists.

[0051] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method of the first aspect.

[0052] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the method of the first aspect.

[0053] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method of the first aspect.

[0054] The aforementioned motion information prediction method, apparatus, computer equipment, storage medium, and computer program product acquire first fused data; the first fused data includes detection data from each sensor of the target vehicle; based on the first fused data, a target to be detected is determined; based on the sensor detection status of the target to be detected, the confidence level of the target to be detected is adjusted; based on the adjusted confidence level, a target to be detected is determined among the targets to be detected; based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles within a set time window, matching processing is performed, and the detection information of the target to be detected is determined based on the matching result; based on the detection information of the target to be detected, the motion information of the target to be detected is predicted to obtain the predicted motion information of the target to be detected. In this scheme, the target to be detected is identified among the targets to be detected through the fused data of each sensor, and then the motion information of the target to be detected is predicted through the prior and posterior behaviors of the targets surrounding the target to be detected. This can improve the accuracy of predicting the motion information of the target to be detected in scenarios where the detector misses or falsely detects targets. Attached Figure Description

[0055] Figure 1 This is a diagram illustrating the application environment of a motion information prediction method in one embodiment.

[0056] Figure 2 This is a flowchart illustrating a motion information prediction method in one embodiment;

[0057] Figure 3 This is a flowchart illustrating the fault identification steps in one embodiment;

[0058] Figure 4 This is a schematic diagram of a target determination method in one embodiment;

[0059] Figure 5 This is a flowchart illustrating the credibility adjustment process in one embodiment;

[0060] Figure 6 This is a flowchart illustrating a motion information determination method in one embodiment;

[0061] Figure 7 This is a schematic diagram of targets surrounding the target to be detected in one embodiment;

[0062] Figure 8 This is a flowchart illustrating a motion information detection method in one embodiment;

[0063] Figure 9 This is a schematic diagram of a method for storing road constraint information using a matrix in one embodiment;

[0064] Figure 10 This is a schematic diagram of a method for predicting the prior behavior of a target to be detected in one embodiment;

[0065] Figure 11 This is a flowchart illustrating an example of a motion information prediction method in one embodiment;

[0066] Figure 12 This is a structural block diagram of a motion information prediction device in one embodiment;

[0067] Figure 13 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0069] The motion information prediction method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and IoT devices; IoT devices can include smart vehicle devices, etc. Server 104 can be implemented using a standalone server or a server cluster composed of multiple servers. The data analysis and processing involved in this solution can be performed by the terminal sending data to the server, or it can be performed by the terminal itself; this solution does not impose any limitations.

[0070] In one embodiment, such as Figure 2 As shown, a method for predicting motion information is provided, which can be applied to... Figure 1 Taking the vehicle-mounted terminal as an example, the explanation includes the following steps:

[0071] Step 202: Obtain the first fused data; the first fused data includes the detection data of each sensor of the target vehicle.

[0072] In this embodiment, the target vehicle may be equipped with multiple sensors, such as millimeter-wave radar, ultrasonic radar, and infrared cameras. After each sensor of the target vehicle detects data, it can send the detection data to the terminal. After receiving the detection data from each sensor, the terminal can summarize the detection data from each sensor to obtain the first fused data. For detection data from different sensors, the terminal can determine whether the detection data from different sensors belong to the same detected target based on information such as speed and position detected by each sensor.

[0073] Step 204: Determine the target to be detected based on the first fused data; adjust the confidence level of the target to be detected according to the sensor detection status of the target; and determine the target to be detected among the targets to be detected based on the adjusted confidence level.

[0074] In this embodiment, when initializing the data (such as confidence level) of the target being detected, the terminal simultaneously initializes and saves parameters such as the target's number, position, speed, acceleration, and category, thereby obtaining the target's operational data. Then, based on the sensor's detection signal of the target, the terminal increases or decreases the initial confidence level of the target, and then determines whether the target is the target to be detected based on the adjusted confidence level.

[0075] Step 206: Perform matching processing based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles within a set time window, and determine the detection information of the target to be detected based on the matching results.

[0076] Here, prior prediction behavior refers to the predicted information about the behavior and trajectory of targets surrounding the target to be detected, while posterior prediction behavior refers to the actual information about the behavior and trajectory of targets surrounding the target to be detected. Since the presence of the target to be detected will affect the surrounding targets, prior prediction behavior can include prediction information under two different conditions: the presence of the target to be detected and the absence of the target.

[0077] In this embodiment, the terminal performs behavior prediction and trajectory prediction on targets surrounding the target to be detected, obtaining prediction information (i.e., prior predicted behavior). This prediction information is then matched with the actual information (i.e., posterior behavior) of the targets surrounding the target to be detected, and the verification information of the target is determined based on the matching result. The verification information is used to characterize whether the target to be detected actually exists.

[0078] Step 208: Based on the detection information of the target to be detected, predict the motion information of the target to be detected to obtain the predicted motion information of the target to be detected.

[0079] In this embodiment of the application, if the target to be detected actually exists, the terminal predicts the behavior and trajectory of the target to be detected based on the behavior and trajectory of the targets around the target to be detected, and obtains the predicted behavior and trajectory of the target to be detected. The terminal then uses the predicted behavior and trajectory to plan the behavior of the vehicle.

[0080] In the above-mentioned motion information prediction method, the existence of the target to be detected is determined by the behavior and trajectory of the targets around the target, and the behavior and trajectory of the target to be detected are predicted. This enables the prediction of motion information when the sensor misses or falsely detects the target, and can improve the accuracy of predicting the motion information of the target in scenarios where the detector misses or falsely detects the target.

[0081] In one embodiment, such as Figure 3 As shown, the target to be detected is determined based on the first fused data. The confidence level is adjusted according to the sensor detection status of the target. Based on the adjusted confidence level, the target to be detected among the detected targets includes:

[0082] Step 302: For each target to be detected, when the target is first detected, determine the initial credibility of the target.

[0083] In this embodiment, the terminal initializes the confidence level of the detected target based on the sensor's detection signal. This initial confidence level can be the median of the maximum confidence level. For the same target, the terminal determines whether the detection information from different sensors belongs to the detected target based on the target's position, velocity, and other information, and uses this information as the fusion target in the calculation; the fusion target is the same target detected by all sensors.

[0084] Step 304: Adjust the credibility of the target being detected based on the initial credibility, the sensor detection status of the target, and the credibility adjustment strategy.

[0085] In this embodiment, the terminal continuously updates the initial credibility of the detected target according to the credibility adjustment strategy, thereby adjusting the initial credibility of the detected target.

[0086] Step 306: If the adjusted confidence level is lower than the first threshold, the target to be detected is identified as the target to be detected.

[0087] In this embodiment of the application, if the adjusted confidence level of the detected target is greater than a first threshold, the terminal determines that the target is a real target; if the adjusted confidence level of the detected target is less than or equal to the first threshold, the terminal detects the location information of the detected target. Figure 4 As shown, 1 represents the vehicle itself, 2 represents the vehicle gradually moving out of the detection range, 3 represents an obstacle, and 4 represents an invisible target behind the obstacle. Then, further judgment is made based on the location information of the detected target. Specifically, if the location information of the detected target is not within the sensor's detection boundary and is not in an obstructed area, then the detected target is determined as a target to be detected. If the location information of the detected target is within the sensor's detection boundary and is in an obstructed area, the terminal treats this detected target as a normal target and removes its data. In this embodiment, the confidence level of the target to be detected is adjusted according to a confidence level adjustment strategy, which can determine whether the sensor is faulty and achieve fault identification.

[0088] In one embodiment, such as Figure 5 As shown, for each target being detected, the initial confidence level of the target is determined when it is first detected, including:

[0089] Step 502: If the detected target is continuously detected, increase the confidence of the detected target according to the preset confidence increase strategy and the initial confidence.

[0090] The detected target is included in the analysis as a fusion target.

[0091] In this embodiment, when the terminal continuously receives signals indicating that the target is being detected by the sensor, it increases the confidence level of the target according to a preset confidence level increase strategy until it reaches a preset maximum value. For example, a first value corresponding to a unit of time can be set, and then, based on the initial confidence level, the first value is increased every unit of time until it reaches the preset maximum value.

[0092] Step 504: If no detection data of the target is detected within a preset time period, reduce the confidence of the target according to a preset confidence reduction strategy and an initial confidence level.

[0093] In this embodiment, after the terminal initializes the confidence level of the target being detected, if it does not receive a signal from the sensor indicating that the target is being detected, it determines that no target can be associated with the stored operational data. Then, according to a preset confidence level reduction strategy, the confidence level of the target being detected is reduced. For example, a second value corresponding to a unit of time can be set. Then, based on the initial confidence level, the second value is reduced every unit of time until the confidence level of the target being detected is lower than a first threshold.

[0094] In this embodiment, by adjusting the credibility of the detected target in real time, it is possible to monitor whether the sensor is malfunctioning and to determine whether the loss of the detection signal is a normal situation.

[0095] In one embodiment, such as Figure 6 As shown, matching is performed based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles within a set time window. The detection information of the target to be detected is determined based on the matching results, including:

[0096] Step 602: For each reference vehicle around the target to be detected, determine the first motion information of the reference vehicle when the target to be detected does not exist, and the second motion information of the reference vehicle when the target to be detected exists; the first motion information and the second motion information constitute the prior prediction behavior.

[0097] In this embodiment, the terminal iterates through each reference vehicle around the target to be detected. If the target to be detected does not exist, the terminal predicts the behavior and trajectory of all vehicles around the target to obtain first motion information. If the target to be detected exists, the terminal predicts the behavior and trajectory of all vehicles around the target to obtain second motion information.

[0098] Step 604: Based on the second fusion data of the reference vehicle under the set time window, obtain the posterior behavior of the surrounding vehicles of the target at the current time.

[0099] Among them, real fusion data refers to the detection signals of reference vehicles around the target to be detected by the sensor within a set time window.

[0100] In this embodiment, the terminal analyzes the real fused data of reference vehicles around the target to be detected within a set time window to determine the behavior and trajectory of the reference vehicles around the target at the current moment, i.e., the posterior behavior of the vehicles around the target at the current moment. The method for determining the behavior and trajectory of the reference vehicles around the target at the current moment can be a Bayesian network method.

[0101] Step 606: Perform matching processing based on the first motion information, the second motion information, and the posterior behavior, and determine the detection information of the target to be detected based on the matching results.

[0102] In this embodiment, by matching the prior predicted behavior of targets around the target to be detected with the prior predicted behavior and the posterior behavior, it is determined whether the target to be detected actually exists, thus realizing the prediction of motion information and improving the availability of the sensor in preventing false detections and missed detections. Figure 7 As shown, 1 represents the target to be detected, and 2, 3, 4, 5, and 6 represent reference vehicles surrounding the target.

[0103] In one embodiment, such as Figure 8 As shown, matching processing is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined according to the matching result, including:

[0104] Step 802: Determine the first matching result between the first motion information and the second motion information, the second matching result between the posterior behavior and the first motion information, and the third matching result between the posterior behavior and the second motion information.

[0105] In this embodiment, the terminal determines a first matching result by judging whether the first motion information and the second motion information match, determines a second matching result based on the matching result of the first motion information and the posterior behavior, and determines a third matching result based on the matching result of the second motion information and the posterior behavior.

[0106] Step 804: Determine the credibility adjustment rule for the target to be detected based on the first matching result, the second matching result, the third matching result, and the preset mapping relationship between the matching result set and the credibility adjustment rule.

[0107] In this embodiment of the application, matching processing is performed based on first motion information, second motion, and posterior behavior, and the detection information of the target is determined according to the matching result, including:

[0108] If the first motion information matches the second motion information and the posterior behavior matches the prior predicted behavior, then it is determined that the target to be detected does not affect the behavior of the surrounding environment targets and is consistent with the current vehicle behavior. The probability of the target's existence is low, thus reducing the credibility of the target to be detected.

[0109] If the first motion information matches the second motion information and the posterior behavior does not match the prior predicted behavior, then it is determined whether the abnormal target to be detected has any behavior that does not affect the surrounding environment, but does not match the current vehicle behavior result, so it cannot be determined whether it exists. The probability of the target's existence is not adjusted, and the credibility of the target to be detected remains unchanged.

[0110] If the first motion information and the second motion information do not match, but the posterior behavior matches the first motion information in the prior predicted behavior, it indicates that the probability of the target being detected is low, thus reducing the credibility of the target being detected.

[0111] If the first motion information and the second motion information do not match, but the posterior behavior matches the second motion information in the prior predicted behavior, it indicates that the probability of the target being detected is high, thus increasing the credibility of the target being detected.

[0112] If the first motion information and the second motion information do not match, and the first and second motion information in the posterior behavior and the prior prediction behavior do not match, it is impossible to determine whether the target to be detected exists, and its existence probability is not adjusted, so the credibility of the target to be detected remains unchanged.

[0113] Step 806: Adjust the current credibility of the target to be detected according to the credibility adjustment rules of the target to be detected, and obtain the credibility of the target to be detected.

[0114] In this embodiment, the terminal adjusts the current credibility of the target to be detected according to the credibility adjustment rule to obtain the credibility of the target to be detected, and determines whether the target to be detected exists based on the relationship between this credibility and the second threshold.

[0115] Step 808: If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

[0116] In this embodiment, after the credibility of the target to be detected is adjusted by the credibility adjustment rule, if the credibility is higher than the second threshold, the terminal determines that the inspection information of the target to be detected exists; if the credibility is lower than the second threshold, the terminal determines that the inspection information of the target to be detected does not exist.

[0117] In this embodiment, by determining the confidence adjustment rules and adjusting the confidence of the target to be detected according to the confidence adjustment rules, the determination of whether the target to be detected actually exists is realized, which can improve the availability of sensor missed detection and false detection.

[0118] In one embodiment, for each reference vehicle surrounding the target to be detected, determining first motion information of the reference vehicle when the target to be detected does not exist, and second motion information of the reference vehicle when the target to be detected exists, includes:

[0119] Based on the information of each reference vehicle around the target, the danger level of each reference vehicle around the target, and the lane line information on the map, the first motion information and the second motion information of each reference vehicle around the target are predicted to obtain the prior predicted behavior.

[0120] The hazard level of the reference vehicle can be discretized using a road occupancy assessment method based on time-of-collision (TTC) and headway. The formula for calculating the time of collision is:

[0121]

[0122] In the embodiments of this application, such as Figure 9 As shown, the terminal uses a matrix to store road constraint information, defined here as matrix M, which includes 5 elements [M1, M2, M3, M4, M5], where: M1 represents whether there is a straight road ahead: 0 does not exist, 1 exists; M2 represents whether there is a decrease or increase in road on the left: 0 unchanged, 1 increases, 2 decreases; M3 represents whether there is a decrease or increase in road on the right: 0 unchanged, 1 increases, 2 decreases; M4 represents the lane line shape on the left: 0 dashed line, 1 solid line; M5 represents the lane line shape on the right: 0 dashed line, 1 solid line. Figure 10 As shown, 1 represents vehicle V_s, and includes 8 targets from 2 to 9. For lateral targets 5 and 6, the target features extracted are presence and absence. For front and rear targets 2 and 8, and side-front and side-rear targets 3, 4, 7, and 9, more motion and position information is needed. This motion and position information can be obtained using the collision time (TTC) and headway road occupancy determination methods. Based on the relationship between the first fused data of the detected targets and the road constraint information, d... ref For relative distance, v ref v is the relative velocity. ego To determine the ground velocity of the target to be detected, TTC threshold and Headway threshold are set, and the area is divided into three categories: area with danger, no danger, and moderate danger. Behavioral analysis is performed on the target to determine the first motion information and second motion information of the target.

[0123] In this embodiment, the prior predicted behavior of the target can be obtained by using information on each reference vehicle around the target, the danger level of each reference vehicle around the target, and the lane line information on the map to predict the target.

[0124] like Figure 11 As shown in the illustration, this application also provides an example of a method for predicting motion information, specifically including the following steps:

[0125] Step 1101: Obtain the first fused data; the first fused data includes the detection data of each sensor of the target vehicle.

[0126] Step 1102: Identify the target to be detected based on the first fused data, and determine the initial credibility of the target to be detected.

[0127] Step 1103: Determine whether the detected target is continuously detected.

[0128] If the detected target is continuously detected, proceed to step 1004; otherwise, proceed to step 1005.

[0129] Step 1104: Increase the credibility of the detected target according to the preset credibility enhancement strategy.

[0130] Step 1105: Reduce the credibility of the detected target according to the preset credibility reduction strategy.

[0131] Step 1106: Determine whether the adjusted confidence level is lower than the first threshold value.

[0132] If so, the target being detected is identified as the target to be detected, and step 1007 is executed; otherwise, the target being detected is identified as a real target.

[0133] Step 1107: Based on the information of each reference vehicle around the target to be detected, the danger level of each reference vehicle around the target, and the lane line information on the map, predict the prior predicted behavior M of each reference vehicle around the target.

[0134] The prior prediction behavior M includes first motion information M1 and second motion information M2. The first motion information is the motion information of reference vehicles around the target when the target to be detected does not exist, and the second motion information is the motion information of reference vehicles around the target when the target to be detected exists.

[0135] Step 1108: Based on the second fusion data of the reference vehicles under the set time window, obtain N of the surrounding vehicles of the target at the current time.

[0136] Step 1109: If M1 and M2 match and N and M match, reduce the confidence of the target to be detected.

[0137] If M1 and M2 match, but N and M do not match, the confidence level of the target to be detected remains unchanged.

[0138] If M1 and M2 do not match, but N matches M2 in M, the confidence level of the target to be detected is increased.

[0139] If M1 and M2 do not match, but N matches M1 in M, the confidence of the target to be detected is reduced.

[0140] If M1 and M2 do not match, and N does not match either M1 or M2, the credibility of the target to be detected remains unchanged.

[0141] Step 1110: If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

[0142] Step 1111: Predict the target behavior based on the target detection information to obtain the predicted motion information of the target to be detected.

[0143] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0144] Based on the same inventive concept, this application also provides a motion information prediction device for implementing the motion information prediction method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more motion information prediction device embodiments provided below can be found in the limitations of the motion information prediction method described above, and will not be repeated here.

[0145] In one embodiment, such as Figure 12 As shown, a motion information prediction device is provided, comprising: an acquisition module 1201, a determination module 1202, a matching module 1203, and a prediction module 1204, wherein:

[0146] The acquisition module 1201 is used to acquire first fused data; the first fused data includes detection data from each sensor of the target vehicle.

[0147] The determination module 1202 is used to determine the target to be detected based on the first fused data, adjust the confidence level of the target to be detected according to the sensor detection status of the target to be detected, and determine the target to be detected among the targets to be detected according to the adjusted confidence level.

[0148] The matching module 1203 is used to perform matching processing based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles under a set time window, and to determine the detection information of the target to be detected based on the matching result.

[0149] The prediction module 1204 predicts the behavior of the target based on the detection information of the target to be detected, and obtains the predicted motion information of the target to be detected.

[0150] In one embodiment, the acquisition module is specifically used for:

[0151] For each target to be detected, when the target is first detected, the initial confidence level of the target is determined;

[0152] The credibility of the detected target is adjusted based on the initial credibility, the sensor detection status of the detected target, and the credibility adjustment strategy.

[0153] If the adjusted confidence level is lower than the first threshold, the target being detected is identified as the target to be detected.

[0154] In one embodiment, the acquisition module is specifically used for:

[0155] If the detected target is continuously detected, the confidence of the detected target is increased according to the preset confidence increase strategy and the initial confidence.

[0156] If no detection data of the target is detected within a preset time period, the confidence of the target is reduced according to a preset confidence reduction strategy and an initial confidence level.

[0157] In one embodiment, the determining module is specifically used for:

[0158] For each reference vehicle surrounding the target to be detected, a first motion information of the reference vehicle is determined when the target to be detected does not exist, and a second motion information of the reference vehicle is determined when the target to be detected exists; the first motion information and the second motion information constitute a priori prediction behavior;

[0159] Based on the second fusion data of the reference vehicle under the set time window, the posterior behavior of the surrounding vehicles of the target to be detected at the current time is obtained;

[0160] The matching process is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined according to the matching result.

[0161] In one embodiment, the determining module is specifically used for:

[0162] Determine a first matching result between the first motion information and the second motion information, a second matching result between the posterior behavior and the first motion information, and a third matching result between the posterior behavior and the second motion information;

[0163] Based on the first matching result, the second matching result, and the third matching result, as well as the mapping relationship between the preset matching result set and the credibility adjustment rule, the credibility adjustment rule of the target to be detected is determined.

[0164] The current credibility of the target to be detected is adjusted by the credibility adjustment rule of the target to be detected, and the credibility of the target to be detected is obtained.

[0165] If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

[0166] In one embodiment, the determining module is specifically used for:

[0167] For each reference vehicle surrounding the target to be detected, based on the information of each reference vehicle surrounding the target to be detected, the danger level of each reference vehicle surrounding the target, and the map lane line information, the first motion information of the reference vehicle when the target to be detected does not exist is predicted; and based on the information of each reference vehicle surrounding the target to be detected, the danger level of each reference vehicle surrounding the target, and the map lane line information, the second motion information of the reference vehicle when the target to be detected exists is predicted.

[0168] Each module in the aforementioned motion information prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0169] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 13 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores fused data. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a motion information prediction method.

[0170] Those skilled in the art will understand that Figure 13 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0171] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0172] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0173] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments. 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, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0174] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0175] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0176] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting motion information, characterized in that, The method includes: Acquire first fused data; the first fused data includes detection data from each sensor of the target vehicle; Based on the first fused data, the target to be detected is determined. According to the sensor detection status of the target to be detected, the confidence level of the target to be detected is adjusted. Based on the adjusted confidence level, the target to be detected is determined among the targets to be detected. The detection information of the target is determined based on the prior predicted behavior of reference vehicles around the target and the posterior behavior of the reference vehicles within a set time window. Based on the detection information of the target to be detected, the motion information of the target to be detected is predicted to obtain the predicted motion information of the target to be detected. The step of adjusting the confidence level based on the sensor detection data of the target being detected, and determining the target to be detected among the detected targets based on the adjusted confidence level, includes: For each target to be detected, when the target is first detected, the initial confidence level of the target is determined; The credibility of the detected target is adjusted based on the initial credibility, the sensor detection status of the detected target, and the credibility adjustment strategy. If the adjusted confidence level is lower than the first threshold, the target being detected is identified as the target to be detected. The step of adjusting the credibility of the detected target based on the initial credibility, the sensor detection status of the detected target, and the credibility adjustment strategy includes: If the detected target is continuously detected, the confidence of the detected target is increased according to the preset confidence increase strategy and the initial confidence. If no detection data of the target is detected within a preset time period, the confidence of the target is reduced according to a preset confidence reduction strategy and an initial confidence level.

2. The method according to claim 1, characterized in that, The matching process based on the prior predicted behavior of reference vehicles surrounding the target to be detected and the posterior behavior of the reference vehicles within a set time window, and the determination of the detection information of the target to be detected based on the matching results, includes: For each reference vehicle surrounding the target to be detected, a first motion information of the reference vehicle is determined when the target to be detected does not exist, and a second motion information of the reference vehicle is determined when the target to be detected exists; the first motion information and the second motion information constitute a priori prediction behavior; Based on the second fusion data of the reference vehicle under the set time window, the posterior behavior of the surrounding vehicles of the target to be detected at the current time is obtained; The matching process is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined according to the matching result.

3. The method according to claim 2, characterized in that, The matching process based on the first motion information, the second motion information, and the posterior behavior, and the determination of the detection information of the target to be detected based on the matching result, includes: Determine a first matching result between the first motion information and the second motion information, a second matching result between the posterior behavior and the first motion information, and a third matching result between the posterior behavior and the second motion information; Based on the first matching result, the second matching result, and the third matching result, as well as the mapping relationship between the preset matching result set and the credibility adjustment rule, the credibility adjustment rule of the target to be detected is determined. The current credibility of the target to be detected is adjusted by the credibility adjustment rule of the target to be detected, and the credibility of the target to be detected is obtained. If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

4. The method according to claim 2, characterized in that, The determination of first motion information of each reference vehicle when the target to be detected does not exist, and second motion information of the reference vehicle when the target to be detected exists, for each reference vehicle surrounding the target to be detected, includes: For each reference vehicle surrounding the target to be detected, based on the information of each reference vehicle surrounding the target to be detected, the danger level of each reference vehicle surrounding the target, and the map lane line information, the first motion information of the reference vehicle when the target to be detected does not exist is predicted; and based on the information of each reference vehicle surrounding the target to be detected, the danger level of each reference vehicle surrounding the target, and the map lane line information, the second motion information of the reference vehicle when the target to be detected exists is predicted.

5. A motion information prediction device, characterized in that, The device includes: An acquisition module is used to acquire first fused data; the first fused data includes detection data from each sensor of the target vehicle; The determination module determines the target to be detected based on the first fused data, adjusts the confidence level of the target to be detected according to the sensor detection status of the target to be detected, and determines the target to be detected among the targets to be detected according to the adjusted confidence level. The matching module is used to perform matching processing based on the prior predicted behavior of reference vehicles around the target to be detected and the posterior behavior of the reference vehicles within a set time window, and to determine the detection information of the target to be detected based on the matching results. The prediction module is used to predict the motion information of the target to be detected based on the detection information of the target to be detected, so as to obtain the predicted motion information of the target to be detected.

6. The apparatus according to claim 5, characterized in that, The determining module is specifically used to determine, for each reference vehicle around the target to be detected, a first motion information of the reference vehicle when the target to be detected does not exist, and a second motion information of the reference vehicle when the target to be detected exists; the first motion information and the second motion information constitute a priori prediction behavior; Based on the second fusion data of the reference vehicle under the set time window, the posterior behavior of the surrounding vehicles of the target to be detected at the current time is obtained; The matching process is performed based on the first motion information, the second motion information, and the posterior behavior, and the detection information of the target to be detected is determined according to the matching result.

7. The apparatus according to claim 6, characterized in that, The determining module is specifically used to determine a first matching result between the first motion information and the second motion information, a second matching result between the posterior behavior and the first motion information, and a third matching result between the posterior behavior and the second motion information; Based on the first matching result, the second matching result, and the third matching result, as well as the mapping relationship between the preset matching result set and the credibility adjustment rule, the credibility adjustment rule of the target to be detected is determined. The current credibility of the target to be detected is adjusted by the credibility adjustment rule of the target to be detected, and the credibility of the target to be detected is obtained. If the credibility of the target to be detected is higher than the second threshold, the detection information is determined to exist; if the credibility of the target to be detected is lower than the second threshold, the detection information is determined to not exist.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images