Tracking information determination method and device for automatic driving, equipment and storage medium
By acquiring and processing target-level data of lidar and millimeter wave radar, and using preset tracking lists and algorithms to match and correct targets, the problem of low accuracy and continuity of multiple observation source data in autonomous driving technology is solved, and the data of various observation source data is achieved. More efficient target tracking information processing.
Patent Information
- Application Number
- CN202411153884.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-26
AI Technical Summary
In the existing autonomous driving technology, the target tracking information processing method fails to fully utilize the observation characteristics of multiple observation sources, resulting in low data accuracy and continuity.
By obtaining target-level data and observation source information of lidar and millimeter wave radar, performing actual measurement data processing and matching, using a preset tracking list for target matching and correction, combining weighting and filtering algorithms to process data errors of different observation sources, improve the Accuracy and continuity of target tracking.
It improves the accuracy and continuity of target tracking information, makes full use of observation data from different sensors, enhances the comprehensive consideration of multiple measured information, and improves the accuracy and stability of target tracking.
Smart Images

Figure CN119024322A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of autonomous driving technology, and in particular to a method, device, equipment and storage medium for determining tracking information for autonomous driving. Background Art
[0002] Target tracking in autonomous driving is one of the key technologies to achieve autonomous driving. In practical applications, in order to achieve efficient and accurate target tracking, multiple algorithms and technologies are usually combined. Therefore, the target tracking information needs to be processed in multiple ways. According to the tracking information, the target is accurately matched, and the corresponding driving strategy is never formulated based on the target's position, speed and predicted behavior.
[0003] The existing methods for optimizing target tracking information include estimating the mean or variance data in the direction based on the detection point data of multiple frames, thereby realizing the screening of the detection point data; however, although such methods realize the processing of tracking data, they do not analyze the problem from the perspective of the observation source. The existing mainstream sensors are mostly lidar and millimeter-wave radar, which may lead to errors in the data and the problem that the continuous frame information is not effectively utilized.
[0004] How to effectively utilize continuous frame information processing and improve the accuracy of target tracking information data while taking into account the observation characteristics of multiple observation sources is a problem that the present invention urgently needs to solve. Summary of the invention
[0005] The present application provides a tracking information determination method, device, equipment and storage medium for autonomous driving, which are used to solve the problem of effectively utilizing continuous frame information processing and improving the accuracy of target tracking information data while considering the observation characteristics of multiple observation sources.
[0006] In a first aspect, the present application provides a method for determining tracking information of an autonomous driving system, the method comprising:
[0007] Acquire first target-level data and first observation source information of the tracked target, wherein the first observation source information is used to indicate that the first target-level data is monitored by a laser radar and / or a millimeter-wave radar;
[0008] Obtaining first measured data of the tracking target according to the first target level data and the first observation source information;
[0009] Through the preset tracking list, target matching is performed on the tracking target according to the first measured data to obtain a matching result; wherein the tracking list includes the pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times;
[0010] The first measured data is corrected according to the second measured data to obtain tracking information of the first measured data.
[0011] In a possible design, the first target-level data includes: a first transverse and longitudinal position, a first transverse and longitudinal speed, and target category information monitored by a laser radar, and / or a second transverse and longitudinal position, a second transverse and longitudinal speed, and a first transverse and longitudinal acceleration monitored by a millimeter-wave radar; the first measured data includes: first state information and first attribute information;
[0012] According to the first target level data and the first observation source information, first measured data of the tracking target is obtained, including:
[0013] When the first observation source information indicates that the first target level data is monitored by a laser radar or a millimeter wave radar alone, first state information is obtained according to the first transverse and longitudinal position and the first transverse and longitudinal velocity, or according to the second transverse and longitudinal position, the second transverse and longitudinal velocity, and the first transverse and longitudinal acceleration, and first attribute information is obtained according to the target category information and the first observation source information, or according to the first observation source information;
[0014] When the first observation source information indicates that the first target level data is jointly monitored by the laser radar and the millimeter wave radar, the first state information is obtained according to the first transverse and longitudinal position, the second transverse and longitudinal position, the first transverse and longitudinal velocity, the second transverse and longitudinal velocity and the first transverse and longitudinal acceleration, and the first attribute information is obtained according to the target category information and the first observation source information.
[0015] In a possible design, the first state information is obtained according to the first transverse and longitudinal position, the second transverse and longitudinal position, the first transverse and longitudinal speed, the second transverse and longitudinal speed, and the first transverse and longitudinal acceleration, including:
[0016] According to the first target level data, the position error and velocity error of the laser radar and the millimeter wave radar are calculated by the target matching algorithm;
[0017] A third transverse and longitudinal position is calculated by a weighted algorithm according to the first transverse and longitudinal position, the second transverse and longitudinal position and the position error, and a third transverse and longitudinal speed is calculated by a weighted algorithm according to the first transverse and longitudinal speed, the second transverse and longitudinal speed and the speed error;
[0018] The first state information is obtained according to the third transverse and longitudinal position, the third transverse and longitudinal speed and the first transverse and longitudinal acceleration.
[0019] In a possible design, the first measured data is measured data of the tracking target at a first moment, and the second measured data is measured data at a second moment, where the second moment is a moment before the first moment;
[0020] By using the preset tracking list, target matching is performed on the tracking target according to the first measured data to obtain a matching result, including:
[0021] Predicting first predicted data corresponding to the second measured data at the first moment;
[0022] Performing target matching on the tracking target according to the first predicted data and the first measured data to obtain a matching result;
[0023] When the matching result is successful, it is determined that the second measured data is measured data of the tracking target at the second moment.
[0024] In one possible design, the method further includes:
[0025] When the matching result is a matching failure, and the first observation source information indicates that the first target-level data is monitored by the laser radar alone, or by the laser radar and the millimeter-wave radar together, it is determined that the second measured data is not the measured data of the tracking target at the second moment;
[0026] The first measured data is written into the tracking list.
[0027] In a possible design, correcting the first measured data according to the second measured data to obtain tracking information of the first measured data includes:
[0028] Predicting second predicted data corresponding to the second measured data at the same time as the first measured data;
[0029] Correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data;
[0030] After correcting the first measured data according to the second measured data to obtain tracking information of the first measured data, the method further includes:
[0031] The first measured data and the tracking information are written into the tracking list.
[0032] In a possible design, the second measured data is obtained according to the second target level data and the second observation source information; the second measured data includes: second state information, and second attribute information including the second observation source information; the second observation source information is used to indicate that the second target level data is monitored by a laser radar and / or a millimeter wave radar;
[0033] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, before correcting the first measured data by a filtering algorithm according to the second prediction data to obtain tracking information of the first measured data, the method further includes:
[0034] Obtaining a horizontal and vertical position difference according to the first horizontal and vertical position and a fourth horizontal and vertical position predicted by the second prediction data;
[0035] A fifth transverse and longitudinal speed is calculated by a weighted algorithm according to the transverse and longitudinal position difference, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0036] According to the first transverse and longitudinal speed and the fifth transverse and longitudinal speed in the first measured data, a sixth transverse and longitudinal speed is calculated by a mean value algorithm;
[0037] Replacing the first transverse and longitudinal speed in the first measured data with the sixth transverse and longitudinal speed;
[0038] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data, the method further includes:
[0039] The seventh transverse and longitudinal speed is calculated by a weighted algorithm according to the speed error, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0040] The first transverse and longitudinal speed in the first measured data is replaced by the seventh transverse and longitudinal speed;
[0041] When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, before correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data, the method further includes:
[0042] An eighth transverse and longitudinal speed is calculated by a mean value algorithm according to the second transverse and longitudinal speed and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0043] The second transverse and longitudinal speed in the first measured data is replaced by the eighth transverse and longitudinal speed;
[0044] When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data, the method further includes:
[0045] According to the second transverse and longitudinal speed and the first transverse and longitudinal acceleration, and the fourth transverse and longitudinal speed and the second transverse and longitudinal acceleration predicted by the second prediction data, a ninth transverse and longitudinal speed and a third transverse and longitudinal acceleration are calculated by a mean value algorithm;
[0046] The second transverse and longitudinal velocity and the first transverse and longitudinal acceleration in the first measured data are replaced by the ninth transverse and longitudinal velocity and the third transverse and longitudinal acceleration.
[0047] In a second aspect, the present application provides a tracking information determination device for autonomous driving, comprising:
[0048] An acquisition module, used to acquire first target level data and first observation source information of a tracked target, wherein the first observation source information is used to indicate that the first target level data is monitored by a laser radar and / or a millimeter wave radar;
[0049] A measurement module, used for obtaining first measured data of a tracking target according to first target level data and first observation source information;
[0050] A matching module, used to match the tracking target according to the first measured data through a preset tracking list to obtain a matching result; wherein the tracking list includes the pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times;
[0051] The result module is used to correct the first measured data according to the second measured data to obtain tracking information of the first measured data.
[0052] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;
[0053] Memory stores computer-executable instructions;
[0054] The processor executes computer-executable instructions stored in the memory to implement a method for determining tracking information for autonomous driving according to the first aspect of the invention.
[0055] In a fourth aspect, the present application provides a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by a processor, they are used to implement a method for determining tracking information for autonomous driving according to the invention content of the first aspect.
[0056] In a fifth aspect, the present application provides a computer program product, including: a computer program, which, when executed by a processor, is used to implement a method for determining tracking information for autonomous driving according to the invention content of the first aspect.
[0057] The present application provides a method, device, equipment and storage medium for determining tracking information of an autonomous driving, which obtains first target-level data and first observation source information of a tracking target, wherein the first observation source information is used to indicate that the first target-level data is monitored by a laser radar and / or a millimeter-wave radar; obtains first measured data of the tracking target based on the first target-level data and the first observation source information; performs target matching on the tracking target based on the first measured data through a preset tracking list to obtain a matching result; wherein the tracking list includes pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times; corrects the first measured data based on the second measured data to obtain tracking information of the first measured data. The following technical effects are achieved: target-level data and observation source information of the tracked target are obtained, different data information and the observation sources corresponding to the information are analyzed, and the observation source characteristics of different observation sources are considered; first, the measured information is obtained, and then the measured information is processed to obtain a more accurate tracking data information, which fully considers the observation data of different sensors, and is processed according to the classification of different sensors, thereby improving the accuracy of target tracking; the processed measured information is corrected with the original measured information, and multiple measured information is comprehensively considered to improve the accuracy of target tracking, and continuous frame data is processed to improve the continuity of target tracking. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0059] Figure 1 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 1 ;
[0060] Figure 2 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 2 ;
[0061] Figure 3 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 3 ;
[0062] Figure 4 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 4 ;
[0063] Figure 5 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 5 ;
[0064] Figure 6 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 6 ;
[0065] Figure 7 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 7 ;
[0066] Figure 8 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 8 ;
[0067] Fig. 9 A schematic diagram of the structure of a device for determining tracking information for autonomous driving provided in an embodiment of the present application;
[0068] Fig.10 A schematic diagram of the structure of the electronic device hardware provided in an embodiment of the present application.
[0069] Reference numerals:
[0070] 100 - device for determining tracking information of autonomous driving; 110 - acquisition module; 120 - measurement module; 130 - matching module; 140 - result module;
[0071] 200 - electronic device; 210 - processor; 220 - memory; 230 - communication component; 240 - bus. DETAILED DESCRIPTION
[0072] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0073] In the embodiments of the present application, words such as "first" and "second" are used to distinguish the same or similar items with basically the same functions and effects. Those skilled in the art will understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit the difference. It should be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way. In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more.
[0074] It should be noted that the "at..." in the embodiment of the present application can be the instant when a certain situation occurs, or can be a period of time after a certain situation occurs, and the embodiment of the present application does not specifically limit this. In addition, the method for determining the tracking information of dynamic driving provided in the embodiment of the present application is only an example, and the method for determining the tracking information of dynamic driving can also include more or less content.
[0075] In order to clearly describe the technical solutions of the embodiments of the present application, some terms and technologies involved in the embodiments of the present application are briefly introduced below:
[0076] Object Tracking: Object tracking is a key technology in the field of computer vision and autonomous driving, which involves the process of identifying and tracking one or more objects in a continuous sequence of images. This technology is particularly important for autonomous driving systems because it helps vehicles understand their surroundings, predict the movement of other objects, and make appropriate decisions based on this.
[0077] LiDAR: It is a remote sensing technology that measures distance by emitting laser beams and receiving the reflected signals. This technology can be used to create high-precision three-dimensional maps and is widely used in many fields, including but not limited to self-driving cars, terrain mapping, archeology, forest management, and drone systems. LiDAR is an important part of the autonomous driving system, which can help vehicles perceive the surrounding environment, detect obstacles, and achieve precise navigation.
[0078] Millimeter-wave radar: A radar system that uses the millimeter-wave frequency band. Millimeter-wave radar plays an important role in autonomous vehicle technology because it can provide reliable sensor data in all weather conditions, helping vehicles achieve environmental perception and safe driving.
[0079] Observation source: In this application, it refers to a sensor, that is, a sensor that detects data information.
[0080] Observation source information: in this application, it refers to the data information detected by the observation source.
[0081] Target-level data: Target-level data refers to data that is specifically focused on describing or predicting a specific target (or dependent variable, label) in data analysis or data science.
[0082] Target category: In this application, it refers to the type of detected target, such as cars and traffic signs.
[0083] Measured information: Measured information is the observation information of a sensor detected individually.
[0084] Kalman Filter: Kalman filtering is an efficient recursive filtering method for estimating the state of a dynamic system, providing accurate estimates even when the system is subject to noise. Kalman filtering is widely used in many fields, including target tracking in autonomous vehicles, navigation systems, control systems, and signal processing.
[0085] With the rapid development of the economy, living standards are gradually improving, and people are paying more and more attention to travel convenience. As an advanced technology for existing vehicles, autonomous driving has solved the problems of driver fatigue and comfort to a certain extent. Target tracking in autonomous driving is particularly important, and the corresponding driving behavior can be predicted based on the information of the detected target. Therefore, target information processing is an important issue.
[0086] The existing methods for processing target tracking information are:
[0087] First, the semantic segmentation results of the image data information are obtained; second, the target elements are matched based on the semantic segmentation results; finally, the target elements are updated based on the matching results for new tracking.
[0088] Alternatively, the direction is estimated based on the detection point data, and the mean or variance data information in the direction is estimated to achieve the screening of the detection point data.
[0089] However, there are technical problems with this approach:
[0090] First, target matching and tracking based on image data or detection point data does not take into account the different characteristics of different observation sources that may lead to inaccurate target data information, which may result in low target matching accuracy.
[0091] The second aspect: Regarding the processing of data information, continuous frame data should be processed, and the issue of effective use of continuous frame information should be fully considered. The accuracy of target matching and tracking performed by the aforementioned data processing method may be low.
[0092] How to effectively utilize continuous frame information processing and improve the accuracy of target tracking information data while taking into account the observation characteristics of multiple observation sources is a problem that the present invention urgently needs to solve.
[0093] Based on this, the embodiments of the present application provide a method, device, equipment and storage medium for determining tracking information of autonomous driving, which can be used in the field of autonomous driving technology, aiming to solve the above technical problems of the prior art.
[0094] Figure 1 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 1 .like Figure 1 As shown, the method includes:
[0095] S101. Acquire first target level data and first observation source information of a tracked target.
[0096] The first observation source information is used to indicate that the first target level data is monitored by a laser radar and / or a millimeter wave radar.
[0097] Specifically, the vehicle body surrounding environment information is obtained, and the collected environmental information is basically processed to output target-level data, and the observation source information is obtained. When the observation source is the No. 1 sensor laser radar, the output target-level data includes the lateral and longitudinal position, lateral and longitudinal speed, and target category information; when the observation source is the No. 2 sensor millimeter-wave radar, the output target-level data includes the lateral and longitudinal position, lateral and longitudinal speed, and lateral and longitudinal acceleration.
[0098] Among them, the target missed detection rate and false detection rate of sensor No. 1 are low, the position accuracy is high, and the target category can be input. The above performance is higher than that of sensor No. 2. It can create observation targets through its single observation, and create new tracking targets through its single observation. The horizontal and vertical speeds of sensor No. 2 are more accurate than those of sensor No. 1, and it has horizontal and vertical acceleration information. It can create observation targets through its single observation, but it cannot create new tracking targets through its single observation. The target-level information output by the sensor can be used after confirming the existence of the target. That is, when the sensor No. 1 laser radar detects a target, a tracking list can be created later, while the sensor No. 2 millimeter-wave radar cannot create a target even if it detects a target detected by the sensor No. 1 laser radar.
[0099] S102. Obtain first measured data of the tracking target according to the first target level data and the first observation source information.
[0100] Specifically, based on the first target level data and the first observation source information, the first measured data of the tracking target is obtained. The first measured data includes three situations: measured information obtained by detecting by the lidar sensor alone, measured information obtained by detecting by the millimeter-wave radar alone, and measured information obtained by detecting by both the lidar and millimeter-wave radar sensors.
[0101] S103 . Perform target matching on the tracking target according to the first measured data through a preset tracking list to obtain a matching result.
[0102] The tracking list includes pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times.
[0103] Specifically, the first measured data is matched with the tracking information predicted at the corresponding time in the tracking list to obtain a matching result.
[0104] S104: Correct the first measured data according to the second measured data to obtain tracking information of the first measured data.
[0105] Specifically, the second measured data is corrected with the first measured data to obtain tracking information of the first measured data.
[0106] The present application provides a method, device, equipment and storage medium for determining tracking information of an autonomous driving, which obtains first target-level data and first observation source information of a tracking target, wherein the first observation source information is used to indicate that the first target-level data is monitored by a laser radar and / or a millimeter-wave radar; obtains first measured data of the tracking target based on the first target-level data and the first observation source information; performs target matching on the tracking target based on the first measured data through a preset tracking list to obtain a matching result; wherein the tracking list includes pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times; corrects the first measured data based on the second measured data to obtain tracking information of the first measured data. The following technical effects are achieved: target-level data and observation source information of the tracked target are obtained, different data information and the observation sources corresponding to the information are analyzed, and the observation source characteristics of different observation sources are considered; first, the measured information is obtained, and then the measured information is processed to obtain a more accurate tracking data information, which fully considers the observation data of different sensors, and is processed according to the classification of different sensors, thereby improving the accuracy of target tracking; the processed measured information is corrected with the original measured information, and multiple measured information is comprehensively considered to improve the accuracy of target tracking, and continuous frame data is processed to improve the continuity of target tracking.
[0107] Figure 2Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 2 In this embodiment Figure 1 Based on the embodiment, the specific steps of the method for determining tracking information of autonomous driving are described in detail. It should be noted that the second moment of this embodiment is the previous moment of the first moment, and the second measured data is the measured data of the second moment. Figure 2 As shown, this embodiment provides a method for determining tracking information of an autonomous driving system, including:
[0108] The principles of S201 and S101 are the same and will not be described in detail in this embodiment.
[0109] S202, determining whether the first target-level data indicated by the first observation source information is monitored by both the laser radar and the millimeter-wave radar;
[0110] Specifically, if the first observation source information indicates that the first target level data is monitored by both the laser radar and the millimeter wave radar, execute S204; otherwise, execute S203.
[0111] S203. When the first observation source information indicates that the first target level data is monitored solely by the laser radar or the millimeter wave radar, the first state information is obtained according to the first transverse and longitudinal positions and the first transverse and longitudinal speeds, or according to the second transverse and longitudinal positions, the second transverse and longitudinal speeds and the first transverse and longitudinal accelerations, and the first attribute information is obtained according to the target category information and the first observation source information, or according to the first observation source information.
[0112] The first target-level data includes: the first transverse and longitudinal position, the first transverse and longitudinal speed and target category information monitored by the laser radar, and / or, the second transverse and longitudinal position, the second transverse and longitudinal speed and the first transverse and longitudinal acceleration monitored by the millimeter-wave radar; the first measured data includes: first state information and first attribute information.
[0113] Specifically, when the first observation source information indicates that the first target level data is monitored by a laser radar or a millimeter wave radar alone, the first state information is obtained based on the first transverse and longitudinal position and the first transverse and longitudinal velocity, or based on the second transverse and longitudinal position, the second transverse and longitudinal velocity and the first transverse and longitudinal acceleration.
[0114] When the first observation source information is detected by the laser radar alone, the first state information includes the lateral and longitudinal position information, the lateral and longitudinal speed information and the target category, and the attribute information is the sensor laser radar detected alone.
[0115] When the first observation source information is detected by the millimeter-wave radar alone, the first state information includes the transverse and longitudinal position information, the transverse and longitudinal velocity information, and the transverse and longitudinal acceleration, and the attribute information is that of the sensor millimeter-wave radar detected alone.
[0116] S204, when the first observation source information indicates that the first target-level data is monitored jointly by the laser radar and the millimeter-wave radar, calculating the position error and velocity error of the laser radar and the millimeter-wave radar respectively according to the first target-level data through a target matching algorithm;
[0117] Specifically, the first state information is obtained according to the first transverse and longitudinal position, the second transverse and longitudinal position, the first transverse and longitudinal velocity, the second transverse and longitudinal velocity and the first transverse and longitudinal acceleration. When the first observer information is detected by the laser radar and the millimeter wave radar together, the target matching is performed according to the first target level data through the target matching algorithm, and the position error and velocity error of the laser radar and the millimeter wave radar are calculated.
[0118] S205, calculating a third transverse and longitudinal position by a weighted algorithm according to the first transverse and longitudinal position, the second transverse and longitudinal position and the position error, and calculating a third transverse and longitudinal speed by a weighted algorithm according to the first transverse and longitudinal speed, the second transverse and longitudinal speed and the speed error;
[0119] Specifically, the first transverse and longitudinal speeds, the second transverse and longitudinal speeds, speed errors and weights in the first target level data are obtained by a weighted algorithm, and the third transverse and longitudinal speeds are calculated based on the weights and errors.
[0120] S206. Obtain first state information according to the third transverse and longitudinal positions, the third transverse and longitudinal speeds, and the first transverse and longitudinal accelerations; and obtain first attribute information according to the target category information and the first observation source information.
[0121] Specifically, the first state information and the first attribute information are obtained according to the third transverse and longitudinal positions, the third transverse and longitudinal speeds and the first transverse and longitudinal accelerations, that is, the first measured data corresponding to the first moment are obtained.
[0122] S207, predicting first predicted data corresponding to the second measured data at the first moment;
[0123] The first measured data is the measured data of the tracking target at the first moment, and the second measured data is the measured data at the second moment, where the second moment is a moment before the first moment;
[0124] Specifically, through the preset tracking list, target matching is performed on the tracking target according to the first measured data to obtain a matching result. Prediction is performed according to the second measured data at the second moment before the first moment to obtain the first predicted data.
[0125] S208, performing target matching on the tracking target according to the first predicted data and the first measured data to obtain a matching result;
[0126] Specifically, target matching is performed on the tracking target according to the first predicted data and the first measured data to obtain a target matching result.
[0127] S209: Determine whether the matching result is successful.
[0128] Specifically, if the match succeeds, S210 is executed; if the match fails, S211 is executed. A successful match indicates that the tracking target information exists in the tracking list, that is, it is not the first time to be observed. A failed match indicates that the tracking target information does not exist in the tracking list, that is, it is the first time to be observed.
[0129] S210: When the matching result is successful, determine that the second measured data is measured data of the tracking target at the second moment.
[0130] Specifically, when the matching result is successful, that is, the tracking target information exists in the tracking list, it is determined that the second measured data is the measured data of the tracking target at the second moment.
[0131] S211. When the matching result is a matching failure, and the first observation source information indicates that the first target-level data is monitored by the laser radar alone, or by the laser radar and the millimeter-wave radar together, determine that the second measured data is not the measured data of the tracking target at the second moment; write the first measured data into the tracking list.
[0132] Specifically, when the matching result fails, that is, when the target is observed for the first time, and the first observation source information indicates that the first target level data is monitored by the lidar alone, or by the lidar and millimeter wave radar together, it is determined that the second measured data is not the tracking target. When the second measured data is measured, the first measured data is filled into the created tracking list, that is, a tracking list for the target observed for the first time is created.
[0133] When the millimeter-wave radar detects a target alone and there is no tracking information of the target in the tracking list, a tracking list is not created for the target. A tracking list is created only when the laser radar detects the target.
[0134] S212: Correct the first measured data according to the second measured data to obtain tracking information of the first measured data.
[0135] Specifically, the first measured data is corrected according to the second measured data to obtain tracking information of the first measured data, and the tracking information is written into the tracking list.
[0136] The present application provides a method for determining tracking information for autonomous driving. The method processes measured data according to different observation source characteristics of different observation sources to facilitate the processing of subsequent target information. A tracking list is created or not created for each observation source of the target monitored for the first time, so that later data is more stable.
[0137] Figure 3 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 3 In this embodiment Figure 1 Based on the embodiment, the method for determining the tracking information of the automatic driving is described in detail. The moment information used in this embodiment is not limited to the second moment, but uses the data of at least one moment for prediction, and does not limit the second measured data to the data information corresponding to the second moment, that is, the second measured data is the measured data corresponding to a moment before the first moment. Figure 3 As shown, this embodiment provides a method for determining tracking information of an autonomous driving system, including:
[0138] The principles of S301 - S306 are the same as those of S201 - S206 and will not be repeated in the embodiments of this application.
[0139] S307, predicting second predicted data corresponding to the second measured data at the same time as the first measured data;
[0140] Specifically, second predicted data corresponding to the same time as the first measured data is predicted based on the second measured data to obtain the second predicted data.
[0141] S308. Correct the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data.
[0142] According to the second predicted data, the first measured data is corrected through a filtering algorithm to obtain tracking information of the first measured data.
[0143] S309, writing the first measured data and tracking information into a tracking list;
[0144] Specifically, the tracking information is written into the tracking list to facilitate the matching of subsequent targets.
[0145] The present application provides a method for determining tracking information for autonomous driving, in which the time is no longer limited to the previous moment, but different moments. Data processing can be performed through multiple frames of measured data to obtain subsequent target matching data, and the data can be processed according to different situations. Continuous frame data can also be used to ensure the continuity of the data. A data smoothing processing method is set up to effectively utilize continuous frame data.
[0146] Figure 4Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 4 ; Figure 5 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 5 ; Figure 6 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 6 ; Figure 7 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 7 .
[0147] The second target level data of this embodiment is the target level data output by the sensor at a certain time, which is not at the same time as the first target level data. Figure 3 On the basis of S308, the origin of the tracking information of the first measured data before obtaining the tracking information of the first measured data by correcting the first measured data by a filtering algorithm according to the second predicted data is specifically described, which is divided into four cases respectively, and these four cases are all based on the premise that the tracking list has been created:
[0148] The first case, such as Figure 4 As shown, the method includes:
[0149] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone.
[0150] S401 : Obtain a horizontal and vertical position difference according to the first horizontal and vertical positions and a fourth horizontal and vertical position predicted by the second prediction data.
[0151] Specifically, when the observation source information indicates that the target-level data is detected by the laser radar alone, and the tracking target does not contain the data information monitored by the millimeter-wave radar in the tracking list, the speed optimization will be performed through the continuous frame data of the tracking list. The horizontal and vertical position difference is obtained based on the first horizontal and vertical position and the fourth horizontal and vertical position predicted by the second prediction data.
[0152] S402 , calculating a fifth transverse and longitudinal speed by a weighted algorithm according to the transverse and longitudinal position difference, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data.
[0153] Specifically, the fifth transverse and longitudinal speed is calculated by a weighted algorithm according to the transverse and longitudinal position difference, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data.
[0154] S403 . Calculate a sixth transverse and longitudinal velocity by using a mean value algorithm according to the first transverse and longitudinal velocity and the fifth transverse and longitudinal velocity in the first measured data.
[0155] Specifically, according to the first transverse and longitudinal speed and the fifth transverse and longitudinal speed in the first actual measurement, the sixth transverse and longitudinal speed is calculated by the average algorithm.
[0156] S404: Replace the first transverse and longitudinal speeds in the first measured data with the sixth transverse and longitudinal speeds.
[0157] Specifically, the first transverse and longitudinal speeds in the first measured data are replaced with the sixth transverse and longitudinal speeds.
[0158] For the convenience of explanation, a specific embodiment is introduced here:
[0159] When the tracking information of the tracking target exists in the tracking list and the tracking information of the tracking target does not contain the tracking information of the tracking target monitored by the millimeter wave radar, the speed will be optimized through the continuous frame data of the tracking list.
[0160] Step 1: Calculate the horizontal and vertical position difference between the measured information and the latest frame of the matched tracking information, and calculate the horizontal and vertical speeds of the measured information based on the horizontal and vertical position difference. The horizontal and vertical speeds are calculated according to the following two formulas:
[0161]
[0162] but
[0163] Measured longitudinal velocity = b × measured lateral velocity + c × theoretical lateral velocity
[0164] Similarly,
[0165]
[0166] but
[0167] Measured lateral velocity = b × measured lateral velocity + c × theoretical lateral velocity
[0168] Here, a is a coefficient, which can be set to different data to subdivide the speed mutation situation, such as a≤3, 2≤a<3, 1≤a<2, and b and c are the assigned weights.
[0169] In the first case, assuming a≤3, b and c are 0.8 and 0.2 respectively, and the corresponding measured transverse and longitudinal velocities are 0.8×measured transverse velocity+0.2×theoretical transverse velocity.
[0170] In the second case, assuming 2≤a<3, b and c are 0.7 and 0.3 respectively, and the corresponding measured transverse and longitudinal velocities are 0.7×measured transverse velocity+0.3×theoretical transverse velocity.
[0171] In the third case, assuming 1≤a<2, b and c are 0.6 and 0.4 respectively, and the corresponding measured transverse and longitudinal velocities are 0.6×measured transverse velocity+0.4×theoretical transverse velocity.
[0172] Step 2: Take the average of the calculated measured horizontal and vertical speeds and the speed information in the tracking list to get the result, and use this result as the speed information in the measured information.
[0173] The second case, such as Figure 5 As shown, the method includes:
[0174] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, that is, the measured target is monitored by the laser radar alone, and the tracking list of the measured target contains the information of the millimeter wave radar.
[0175] S501 , calculating a seventh transverse and longitudinal speed by a weighted algorithm according to the speed error, the first transverse and longitudinal speeds, and the fourth transverse and longitudinal speeds predicted by the second prediction data.
[0176] Specifically, if the tracking information contains the tracking information of the tracking target, and the tracking information of the tracking target contains the data monitored by the millimeter-wave radar, and the laser radar is used to monitor the tracking target at the next moment, the tracking information is predicted based on the speed error and the time error to obtain the data monitored by the millimeter-wave radar.
[0177] The predicted tracking information and the speed information monitored by the laser radar at the corresponding moment are calculated by a weighted algorithm to obtain the speed information, which is the seventh horizontal and vertical speed. Figure 4 Specific embodiments introduced.
[0178] S502: Replace the first transverse and longitudinal speed in the first measured data with the seventh transverse and longitudinal speed.
[0179] Specifically, the transverse and longitudinal speed information in the first measured data is replaced with the seventh transverse and longitudinal speed.
[0180] For the convenience of explanation, a specific embodiment is introduced here:
[0181] Assume that the observation time is t c , then the horizontal and vertical velocities of the frame are calculated according to their horizontal and vertical accelerations and t aThe time difference of the moment is predicted, and t a The horizontal and vertical speeds of the millimeter wave radar’s predicted information at each moment;
[0182] The predicted lateral and longitudinal velocities are processed with the lateral and longitudinal velocities of the No. 1 sensor according to the velocity weighting method, and updated to the lateral and longitudinal velocities of the measured information of the No. 1 sensor.
[0183] The third case, such as Figure 6 As shown, the method includes:
[0184] When the first observation source information indicates that the first target level data is detected by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, that is, the radar has detected the tracking target before this moment and created a tracking list, and at this moment only the millimeter wave radar detects this target at this moment, that is, the millimeter wave radar detects the tracking target for the first time.
[0185] S601 . Calculate an eighth transverse and longitudinal speed by using a mean algorithm according to the second transverse and longitudinal speeds and the fourth transverse and longitudinal speeds predicted by the second prediction data.
[0186] Specifically, based on the second horizontal and vertical speeds and the second predicted data, the tracking information at the corresponding moment, i.e., the fourth horizontal and vertical speed, is predicted, and the fourth horizontal and vertical speeds and the horizontal and vertical speeds in the measured information at the corresponding moment are subjected to the mean algorithm to obtain the average speed, i.e., the eighth horizontal and vertical speed.
[0187] S602: Replace the second transverse and longitudinal speeds in the first measured data with the eighth transverse and longitudinal speeds.
[0188] Specifically, the second transverse and longitudinal speed in the first measured data is replaced with the eighth transverse and longitudinal speed.
[0189] The fourth case, such as Figure 7 As shown, the method includes:
[0190] When the first observation source information indicates that the first target level data is detected by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone. This means that the laser radar has previously detected the target and created a tracking list. The target was detected by the millimeter wave radar at the moments before this moment, and it is also detected by the millimeter wave radar at this moment.
[0191] S701 . Calculate a ninth lateral and longitudinal speed and a third lateral and longitudinal acceleration by using a mean algorithm according to the second lateral and longitudinal speed and the first lateral and longitudinal acceleration, and the fourth lateral and longitudinal speed and the second lateral and longitudinal acceleration predicted by the second prediction data.
[0192] S702: Replace the second transverse and longitudinal velocity and the first transverse and longitudinal acceleration in the first measured data with the ninth transverse and longitudinal velocity and the third transverse and longitudinal acceleration.
[0193] Specifically, the speed information of the predicted tracking information before this moment and the speed in the measured data, the acceleration information of the predicted tracking information before this moment and the acceleration in the measured data are averaged through the mean algorithm to obtain the average value, which is the ninth lateral and longitudinal speed and the third lateral and longitudinal acceleration, and the speed information and acceleration information in the first measured data are replaced.
[0194] In addition to the above four situations, when the observed target is detected by both the laser radar and the millimeter wave radar, Figure 2 The weighted algorithms used in this application can all be referenced to Figure 4 In the specific embodiment introduced, the target matching algorithm is implemented by a conventional method in the art, and the filtering algorithm includes a Kalman filtering algorithm and an extended Kalman filtering algorithm.
[0195] To facilitate the explanation of the overall solution of the present application, a specific embodiment is introduced here to illustrate the overall solution of the present application:
[0196] Figure 8 Schematic diagram of the process of determining tracking information for autonomous driving provided in an embodiment of the present application Figure 8 ,like Figure 8 As shown, the measured information shown in the figure is the measured data of the aforementioned embodiment.
[0197] S801, obtaining vehicle body surrounding environment information and outputting target-level data.
[0198] S802: Obtain measured information.
[0199] There are two cases: The first case: the laser radar is used alone or the millimeter wave radar is used alone. For specific steps, refer to S203. The second case: the laser radar and the millimeter wave radar are used together to monitor the speed information. Then, through the weighted algorithm, different weights are given to the speed information monitored by the laser radar and the millimeter wave radar to obtain the speed information, and the speed information is used as the speed information in the measured information. For specific steps, refer to S204-S208.
[0200] S803: Create a tracking list for the target being monitored for the first time.
[0201] The measured information is matched with the tracking list. If the match fails, it is determined that the target is monitored for the first time, and it is determined whether the target is monitored by the laser radar. If so, a tracking list is created. Otherwise, it is determined that the target is monitored by the millimeter-wave radar and no tracking list is created.
[0202] S804: Update the tracking list for prediction and obtain prediction information.
[0203] Update the information in the tracking list, that is, predict the data information at the next moment.
[0204] S805: Target match the measured information with the predicted information in the tracking list, and update the measured information.
[0205] The measured information obtained in S802 is matched with the predicted information in the tracking list, which is divided into four cases: In the first case, the match fails, and it is determined that the monitoring target is monitored by the laser radar, then the step S803 is executed, otherwise the list is not created, and the specific steps refer to S211. The second case: the match is successful, and the measured information needs to be judged. For specific steps, refer to S210. The measured information is monitored by the laser radar and the millimeter wave radar together, and the correction is performed according to the measured information obtained in S802 and the predicted information at the corresponding time in the tracking list to obtain the tracking information. The third case: the match is successful, and the measured information is monitored by the millimeter wave radar alone, then the predicted information in the tracking list and the measured information are calculated by the mean algorithm to obtain the tracking information, and the specific steps refer to S601-S602 and S701-S702. The fourth case: the match is successful, and the measured information is monitored by the laser radar alone, it depends on whether there is tracking information monitored by the millimeter wave radar in the tracking list, and the specific steps refer to S401-S404 and S501-S502.
[0206] S806: Write the updated measured information into the tracking list.
[0207] The updated measured information is written into the tracking list. Steps S801-S806 can be executed continuously to create a multi-frame tracking list for each target. The multi-frame tracking list stores the tracking information of the multi-frames closest to the current frame when it is monitored.
[0208] The present application provides a method for determining tracking information for autonomous driving. According to different observation sources, the measured information is classified and processed, and finally the processed result is replaced with the result in the measured information to facilitate the implementation of subsequent steps. Continuous frame data is used, and the observation characteristics of different observation sources and the effective use of continuous frame information are fully considered.
[0209] The embodiment of the present invention can divide the electronic device or the main control device into functional modules according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation.
[0210] Fig. 9 This is a schematic diagram of the structure of the tracking information determination device for autonomous driving provided in an embodiment of the present application. Fig. 9 As shown, the tracking information determination device 100 for autonomous driving includes an acquisition module 110 , a measurement module 120 , a matching module 130 and a result module 140 .
[0211] An acquisition module 110 is used to acquire first target level data and first observation source information of a tracked target, wherein the first observation source information is used to indicate that the first target level data is monitored by a laser radar and / or a millimeter wave radar;
[0212] The measurement module 120 is used to obtain first measured data of the tracking target according to the first target level data and the first observation source information;
[0213] The matching module 130 is used to perform target matching on the tracking target according to the first measured data through a preset tracking list to obtain a matching result; wherein the tracking list includes the pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times;
[0214] The result module 140 is used to correct the first measured data according to the second measured data to obtain tracking information of the first measured data.
[0215] In a possible design, the first target-level data includes: a first transverse and longitudinal position, a first transverse and longitudinal speed, and target category information monitored by a laser radar, and / or a second transverse and longitudinal position, a second transverse and longitudinal speed, and a first transverse and longitudinal acceleration monitored by a millimeter-wave radar; the first measured data includes: first state information and first attribute information;
[0216] The measurement module 120 includes:
[0217] A separate detection module is used to obtain first state information according to the first transverse and longitudinal position and the first transverse and longitudinal velocity, or according to the second transverse and longitudinal position, the second transverse and longitudinal velocity and the first transverse and longitudinal acceleration, and to obtain first attribute information according to the target category information and the first observation source information, or according to the first observation source information, when the first observation source information indicates that the first target level data is monitored separately by the laser radar or the millimeter wave radar;
[0218] The joint detection module is used to obtain first state information according to the first transverse and longitudinal position, the second transverse and longitudinal position, the first transverse and longitudinal speed, the second transverse and longitudinal speed and the first transverse and longitudinal acceleration when the first observation source information indicates that the first target level data is jointly monitored by the laser radar and the millimeter wave radar, and to obtain first attribute information according to the target category information and the first observation source information.
[0219] In a possible design, the common detection module includes:
[0220] An error module is used to calculate the position error and speed error of the laser radar and the millimeter wave radar respectively according to the first target level data through a target matching algorithm;
[0221] A speed calculation module, used to calculate the third transverse and longitudinal position by a weighted algorithm according to the first transverse and longitudinal position, the second transverse and longitudinal position and the position error, and to calculate the third transverse and longitudinal speed by a weighted algorithm according to the first transverse and longitudinal speed, the second transverse and longitudinal speed and the speed error;
[0222] The first state information module is used to obtain first state information according to the third transverse and longitudinal positions, the third transverse and longitudinal speeds, and the first transverse and longitudinal accelerations.
[0223] In a possible design, the first measured data is measured data of the tracking target at a first moment, and the second measured data is measured data at a second moment, where the second moment is a moment before the first moment;
[0224] The matching module 130 includes:
[0225] A prediction module, used to predict first predicted data corresponding to the second measured data at the first moment;
[0226] A target matching module, used for matching the tracking target according to the first predicted data and the first measured data to obtain a matching result;
[0227] The matching success module is used to determine, when the matching result is a matching success, that the second measured data is the measured data of the tracking target at the second moment.
[0228] In one possible design, it also includes:
[0229] A matching failure module, for determining that the second measured data is not the measured data of the tracking target at the second moment when the matching result is a matching failure and the first observation source information indicates that the first target level data is monitored by the laser radar alone or by the laser radar and the millimeter wave radar together;
[0230] The writing module is used to write the first measured data into the tracking list.
[0231] In one possible design, the result module 140 includes:
[0232] A second prediction module, used to predict second predicted data corresponding to the second measured data at the same time as the first measured data;
[0233] A tracking information module, used to correct the first measured data through a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data;
[0234] After the result module 140, it also includes:
[0235] The write list module is used to write the first measured data and the tracking information into the tracking list.
[0236] In a possible design, the second measured data is obtained according to the second target level data and the second observation source information; the second measured data includes: second state information, and second attribute information including the second observation source information; the second observation source information is used to indicate that the second target level data is monitored by a laser radar and / or a millimeter wave radar;
[0237] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, before the tracking information module, it also includes:
[0238] A position difference module, used for obtaining a horizontal and vertical position difference according to the first horizontal and vertical position and a fourth horizontal and vertical position predicted by the second prediction data;
[0239] a fifth speed module, configured to calculate a fifth transverse and longitudinal speed by a weighted algorithm according to the transverse and longitudinal position difference, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0240] a sixth speed calculation module, configured to calculate a sixth transverse and longitudinal speed by using a mean value algorithm according to the first transverse and longitudinal speed and the fifth transverse and longitudinal speed in the first measured data;
[0241] A sixth replacement module, used to replace the first transverse and longitudinal speed in the first measured data with a sixth transverse and longitudinal speed;
[0242] When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before the tracking information module, it also includes:
[0243] a seventh speed module, configured to calculate a seventh transverse and longitudinal speed by a weighted algorithm according to the speed error, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0244] A seventh replacement module, used to replace the first transverse and longitudinal speed in the first measured data with a seventh transverse and longitudinal speed;
[0245] When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, before the tracking information module, it also includes:
[0246] an eighth speed module, configured to calculate an eighth transverse and longitudinal speed by using a mean value algorithm according to the second transverse and longitudinal speed and the fourth transverse and longitudinal speed predicted by the second prediction data;
[0247] An eighth replacement module, used to replace the second transverse and longitudinal speed in the first measured data with an eighth transverse and longitudinal speed;
[0248] When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before the tracking information module, it also includes:
[0249] A transverse and longitudinal speed module, used for calculating a ninth transverse and longitudinal speed and a third transverse and longitudinal acceleration by using a mean value algorithm according to the second transverse and longitudinal speed and the first transverse and longitudinal acceleration, and the fourth transverse and longitudinal speed and the second transverse and longitudinal acceleration predicted by the second prediction data;
[0250] The ninth replacement module is used to replace the second transverse and longitudinal speed and the first transverse and longitudinal acceleration in the first measured data with the ninth transverse and longitudinal speed and the third transverse and longitudinal acceleration.
[0251] This embodiment provides a device for determining tracking information for autonomous driving, which can execute the method for determining tracking information for autonomous driving of the above-mentioned embodiment. Its implementation principle and technical effects are similar, and will not be described in detail in this embodiment.
[0252] In the specific implementation of the aforementioned method for determining tracking information for autonomous driving, each module can be implemented as a processor, and the processor can execute computer execution instructions stored in the memory, so that the processor executes the aforementioned method for determining tracking information for autonomous driving.
[0253] Fig.10 This is a schematic diagram of the structure of the electronic device hardware provided in the embodiment of the present application. Fig.10 As shown, the electronic device 200 includes: at least one processor 210 and a memory 220. The electronic device 200 also includes a communication component 230. The processor 210, the memory 220 and the communication component 230 are connected via a bus 240.
[0254] In the specific implementation process, at least one processor 210 executes the computer execution instructions stored in the memory 220, so that at least one processor 210 executes a method for determining tracking information of autonomous driving as performed by the electronic device side above.
[0255] The specific implementation process of the processor 210 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.
[0256] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.
[0257] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.
[0258] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.
[0259] The above-mentioned functions implemented by the electronic device and the main control device introduce the scheme provided by the embodiment of the present invention. It can be understood that in order to implement the above-mentioned functions, the electronic device or the main control device includes a hardware structure and / or software module corresponding to the execution of each function. In combination with the units and algorithm steps of each example described in the embodiment disclosed in the embodiment of the present invention, the embodiment of the present invention can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present invention.
[0260] The present application also provides a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by a processor, they are used to implement the above-mentioned method for determining tracking information of autonomous driving.
[0261] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.
[0262] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in an electronic device or a main control device.
[0263] The present application also provides a computer program product, which includes: a computer program, which is stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium. When the computer program is executed by the processor, it is used to implement the above-mentioned method for determining tracking information for autonomous driving.
[0264] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by hardware related to program instructions. The above program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above method embodiments are executed; and the above storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.
[0265] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments, and the above embodiments are only used to illustrate the technical solution of the present application rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for determining tracking information for autonomous driving, It is characterized in that The method comprises: Acquire first target-level data and first observation source information of a tracked target, wherein the first observation source information is used to indicate that the first target-level data is monitored by a laser radar and / or a millimeter-wave radar; Obtaining first measured data of the tracked target according to the first target level data and the first observation source information; By using a preset tracking list, target matching is performed on the tracking target according to the first measured data to obtain a matching result; wherein the tracking list includes pre-stored second measured data, and the matching result is a successful match, which is used to indicate that the second measured data and the first measured data are measured data of the tracking target at different times; The first measured data is corrected according to the second measured data to obtain tracking information of the first measured data.
2. The method according to claim 1, It is characterized in that The first target-level data includes: a first transverse and longitudinal position, a first transverse and longitudinal speed, and target category information monitored by a laser radar, and / or a second transverse and longitudinal position, a second transverse and longitudinal speed, and a first transverse and longitudinal acceleration monitored by a millimeter-wave radar; the first measured data includes: first state information and first attribute information; The obtaining, according to the first target level data and the first observation source information, first measured data of the tracked target includes: When the first observation source information indicates that the first target-level data is monitored by the laser radar or the millimeter-wave radar alone, the first state information is obtained according to the first transverse and longitudinal positions and the first transverse and longitudinal speeds, or according to the second transverse and longitudinal positions, the second transverse and longitudinal speeds, and the first transverse and longitudinal accelerations, and the first attribute information is obtained according to the target category information and the first observation source information, or according to the first observation source information; When the first observation source information indicates that the first target level data is jointly monitored by the laser radar and the millimeter wave radar, the first state information is obtained according to the first transverse and longitudinal positions, the second transverse and longitudinal positions, the first transverse and longitudinal speeds, the second transverse and longitudinal speeds and the first transverse and longitudinal accelerations, and the first attribute information is obtained according to the target category information and the first observation source information.
3. The method according to claim 2, It is characterized in that The obtaining of the first state information according to the first transverse and longitudinal position, the second transverse and longitudinal position, the first transverse and longitudinal speed, the second transverse and longitudinal speed, and the first transverse and longitudinal acceleration includes: According to the first target level data, the position error and speed error of the laser radar and the millimeter wave radar are calculated by a target matching algorithm; A third transverse and longitudinal position is calculated by a weighted algorithm according to the first transverse and longitudinal position, the second transverse and longitudinal position and the position error, and a third transverse and longitudinal speed is calculated by a weighted algorithm according to the first transverse and longitudinal speed, the second transverse and longitudinal speed and the speed error; The first state information is obtained according to the third transverse and longitudinal positions, the third transverse and longitudinal speeds, and the first transverse and longitudinal accelerations.
4. The method according to claim 2 or 3, It is characterized in that The first measured data is measured data of the tracking target at a first moment, and the second measured data is measured data at a second moment, where the second moment is a moment before the first moment; The step of performing target matching on the tracking target according to the first measured data through the preset tracking list to obtain a matching result includes: Predicting first predicted data corresponding to the second measured data at the first moment; Performing target matching on the tracking target according to the first predicted data and the first measured data to obtain the matching result; When the matching result is successful, it is determined that the second measured data is the measured data of the tracking target at the second moment.
5. The method according to claim 4, It is characterized in that The method further comprises: When the matching result is a matching failure, and the first observation source information indicates that the first target-level data is monitored by the laser radar alone, or is monitored by the laser radar and the millimeter-wave radar together, determining that the second measured data is not measured data of the tracked target at the second moment; The first measured data is written into the tracking list.
6. The method according to claim 3, It is characterized in that The correcting the first measured data according to the second measured data to obtain tracking information of the first measured data includes: Predicting second predicted data corresponding to the second measured data at the same time as the first measured data; Correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data; After correcting the first measured data according to the second measured data to obtain tracking information of the first measured data, the method further includes: The first measured data and the tracking information are written into the tracking list.
7. The method according to claim 6, It is characterized in that The second measured data is obtained according to the second target level data and the second observation source information; the second measured data includes: second state information, and second attribute information including the second observation source information; the second observation source information is used to indicate that the second target level data is monitored by the laser radar and / or the millimeter wave radar; When the first observation source information indicates that the first target-level data is monitored by the laser radar alone, and the second observation source information indicates that the second target-level data is detected by the laser radar alone, before correcting the first measured data by a filtering algorithm based on the second predicted data to obtain tracking information of the first measured data, the method further includes: Obtaining a horizontal and vertical position difference according to the first horizontal and vertical positions and a fourth horizontal and vertical position predicted by the second prediction data; Calculating a fifth transverse and longitudinal speed by a weighted algorithm according to the transverse and longitudinal position difference, the first transverse and longitudinal speed, and a fourth transverse and longitudinal speed predicted by the second prediction data; Calculate a sixth transverse and longitudinal speed by using a mean value algorithm according to the first transverse and longitudinal speed and the fifth transverse and longitudinal speed in the first measured data; replacing the first transverse and longitudinal speed in the first measured data with the sixth transverse and longitudinal speed; When the first observation source information indicates that the first target level data is monitored by the laser radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data, the method further includes: Calculating a seventh transverse and longitudinal speed by a weighted algorithm according to the speed error, the first transverse and longitudinal speed, and the fourth transverse and longitudinal speed predicted by the second prediction data; replacing the first transverse and longitudinal speed in the first measured data with the seventh transverse and longitudinal speed; When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the laser radar alone, before correcting the first measured data by a filtering algorithm based on the second predicted data to obtain tracking information of the first measured data, the method further includes: According to the second transverse and longitudinal speeds and the fourth transverse and longitudinal speeds predicted by the second prediction data, an eighth transverse and longitudinal speed is calculated by a mean algorithm; replacing the second transverse and longitudinal speed in the first measured data with the eighth transverse and longitudinal speed; When the first observation source information indicates that the first target level data is monitored by the millimeter wave radar alone, and the second observation source information indicates that the second target level data is detected by the millimeter wave radar alone, before correcting the first measured data by a filtering algorithm according to the second predicted data to obtain tracking information of the first measured data, the method further includes: According to the second transverse and longitudinal speed and the first transverse and longitudinal acceleration, and the fourth transverse and longitudinal speed and the second transverse and longitudinal acceleration predicted by the second prediction data, a ninth transverse and longitudinal speed and a third transverse and longitudinal acceleration are calculated by a mean algorithm; The second transverse and longitudinal speed and the first transverse and longitudinal acceleration in the first measured data are replaced by the ninth transverse and longitudinal speed and the third transverse and longitudinal acceleration.
8. A device for determining tracking information for autonomous driving, It is characterized in that include: an acquisition module, used to acquire first target-level data and first observation source information of a tracked target, wherein the first observation source information is used to indicate that the first target-level data is monitored by a laser radar and / or a millimeter-wave radar; A measurement module, used for obtaining first measured data of the tracked target according to the first target level data and the first observation source information; a matching module, configured to perform target matching on the tracking target according to the first measured data through a preset tracking list to obtain a matching result; wherein the tracking list includes pre-stored second measured data, and the matching result is a successful match, which indicates that the second measured data and the first measured data are measured data of the tracking target at different times; A result module is used to correct the first measured data according to the second measured data to obtain tracking information of the first measured data.
9. An electronic device, It is characterized in that include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; When the processor executes the computer-executable instructions stored in the memory, it is used to implement a method for determining tracking information for autonomous driving as described in any one of claims 1 to 7.
10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement a method for determining tracking information for autonomous driving as described in any one of claims 1 to 7.
