Multi-sensor data processing method and system based on rear wheel steering vehicle

Abstract

The invention discloses a vehicle multi-sensor data processing method and system based on rear wheel steering, and relates to the technical field of electric digital data processing. The vehicle multi-sensor data processing method based on rear wheel steering comprises the following steps: acquiring a timestamp offset coefficient; obtaining an error compensation accuracy coefficient; and obtaining a real-time evaluation index. According to the invention, the timestamp offset coefficient is obtained through the data synchronization period of the obtained sensor data in the sensor acquisition process, and then timestamp alignment is carried out to obtain the error compensation accuracy coefficient; and finally, the data delay duration of the sensor data to be transmitted is obtained in the data transmission process, a real-time evaluation index is obtained according to the data transmission rate, whether data processing is completed or not is judged based on the obtained real-time evaluation index, and the effect of improving the timeliness of rear wheel steering vehicle sensor data processing is achieved. The problem of low timeliness of data processing of a rear wheel steering vehicle sensor in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a method and system for processing multi-sensor data based on a rear-wheel steering vehicle. Background Art

[0002] With the continuous development of integrated circuits, micro-electromechanical systems, navigation systems and other fields, as well as the increasing market demand for safe driving of automobiles, the requirements for the accuracy, safety, real-time and reliability of positioning and attitude determination of vehicle-mounted integrated navigation products are becoming increasingly higher. Rear-wheel steering technology achieves more flexible and precise steering control by adjusting the angle of the rear wheels of the vehicle, thereby improving the vehicle's handling and stability. However, the implementation of this technology is inseparable from the support and processing of sensor data. These sensors can collect key data such as the vehicle's driving status, surrounding environment and road information in real time. Therefore, how to extract useful information from massive and heterogeneous sensor data and process it efficiently and accurately has become a problem that needs to be solved urgently.

[0003] Existing technology realizes accurate processing of steering data by deploying sensors on rear-wheel steering vehicles and collecting steering data and formatting the data at the same time, while monitoring the operating status of the sensors in real time and performing fault diagnosis. The results of the fault diagnosis are analyzed to identify obstacle information and formulate obstacle avoidance strategies.

[0004] For example, the invention patent with announcement number: CN117113284B announces a rear-wheel steering control method and a rear-wheel steering control system, including: performing distortion correction on images generated by visual sensors; performing filtering processing on point cloud data generated by distance sensors; and performing denoising processing on data generated by posture sensors.

[0005] For example, the invention patent announcement with announcement number: CN117407662B announces a sensor data processing method and system, including: based on the abnormal data uploaded by the first sensor, using a decision tree classification algorithm to perform abnormal data type analysis and generate a data type confirmation report; based on the data type confirmation report, using a threshold analysis method to perform secondary detection on the area where the first sensor is located to generate a secondary detection result; based on the secondary detection result, using a time series analysis algorithm to record the frequency of abnormalities continuously detected by the second sensor and generate an abnormal number statistical report; based on the abnormal number statistical report, using adaptive filtering technology and an adaptive minimum mean square error algorithm to generate noise processing data; based on the noise processing data, using a graph neural network and a particle swarm optimization algorithm to adjust the sensor network structure and generate an optimized network structure; based on the optimized network structure, using a graph database technology and an association rule learning algorithm to generate a sensor data knowledge graph.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, in a complex network environment, sensors usually have time asynchrony problems during data formatting and processing, resulting in inconsistent data arrival times at the processor, making it impossible to complete data parsing and processing in a short period of time, which in turn leads to delays in data diagnosis and processing time, and the problem of low timeliness in rear-wheel steering vehicle sensor data processing. Summary of the invention

[0008] The embodiments of the present application solve the problem of low timeliness of sensor data processing of rear-wheel steering vehicles in the prior art by providing a method and system for processing multi-sensor data of rear-wheel steering vehicles, thereby improving the timeliness of sensor data processing of rear-wheel steering vehicles.

[0009] The embodiment of the present application provides a multi-sensor data processing method based on a rear-wheel steering vehicle, comprising the following steps: step 1, obtaining sensor data of the sensor within a preset time period, and obtaining a timestamp offset coefficient according to a data synchronization period of the sensor data during the sensor acquisition process, wherein the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and a reference timestamp; step 2, aligning the timestamps of the acquired sensor data according to the acquired timestamp offset coefficient, counting the sensor data to be transmitted and comparing it with the sensor data to obtain an error compensation accuracy coefficient, and judging whether to perform data transmission based on the acquired error compensation accuracy coefficient, wherein the error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm; step 3, if data transmission is performed, obtaining a data delay time between the sensor acquisition and the reception of the sensor data to be transmitted by the central processing center, and obtaining a real-time evaluation index according to the data transmission rate, judging whether data processing is completed based on the acquired real-time evaluation index, wherein the real-time evaluation index is used to reflect the real-time performance of data processing of the sensor data within the data synchronization period.

[0010] Furthermore, the timestamp offset coefficient is obtained by the following method: obtaining the timestamp offset of the sensor within a preset time period, the timestamp offset including a first timestamp offset, a second timestamp offset and a third timestamp offset, the first timestamp offset being the absolute value of the difference between the steering angle sensor timestamp and the reference steering angle sensor timestamp, the second timestamp offset being the absolute value of the difference between the vehicle speed sensor timestamp and the reference vehicle speed sensor timestamp, the third timestamp offset being the absolute value of the difference between the lidar sensor timestamp and the reference lidar sensor timestamp; processing the data synchronization cycle of the sensor within the preset time period to obtain the synchronization cycle proportion, and obtaining the time difference offset coefficient based on the obtained timestamp offset and the synchronization cycle proportion and in combination with the maximum allowable timestamp offset, the maximum allowable timestamp offset including the maximum allowable first timestamp offset, the maximum allowable second timestamp offset and the maximum allowable third timestamp offset.

[0011] Furthermore, the error compensation accuracy coefficient is obtained by the following method: A1, judging whether the acquired timestamp offset coefficient is within the timestamp offset allowable range, if so, executing A2, otherwise continuing to monitor the timestamp alignment process; A2, counting the first timestamp alignment adjustment amount, the second timestamp alignment adjustment amount and the third timestamp alignment adjustment amount in the timestamp alignment process within a preset time period and performing a sum operation, and obtaining a preset timestamp alignment adjustment amount weight factor from a preset database; A3, obtaining a first timestamp compensation score, a second timestamp compensation score and a third timestamp compensation score according to the result of the sum operation and in combination with the acquired first timestamp offset, second timestamp offset and third timestamp offset, thereby obtaining the error compensation accuracy coefficient; the time The timestamp alignment adjustment weight factors include a first timestamp alignment adjustment weight factor, a second timestamp alignment adjustment weight factor and a third timestamp alignment adjustment weight factor; the first timestamp compensation score is the ratio of the absolute value of the difference between the average value of the first timestamp alignment adjustment and the initial first timestamp alignment adjustment in the previous time period to the first timestamp alignment adjustment; the second timestamp compensation score is the ratio of the absolute value of the difference between the average value of the second timestamp alignment adjustment and the initial second timestamp alignment adjustment in the previous time period to the second timestamp alignment adjustment; the third timestamp compensation score is the ratio of the absolute value of the difference between the average value of the third timestamp alignment adjustment and the initial third timestamp alignment adjustment in the previous time period to the third timestamp alignment adjustment.

[0012] Furthermore, the specific steps for obtaining the real-time evaluation index are: obtaining the data collection time, data transmission time and data processing time of the sensor within a preset time period, and obtaining the data collection time score, data transmission time score and data processing time score respectively in combination with the corresponding reference values; obtaining the data transmission rate of the sensor within the preset time period and obtaining a first data transmission rate score and a second data transmission rate score in combination with the corresponding reference value; processing the obtained first data transmission rate score and second data transmission rate score and combining them with the obtained data collection time score, data transmission time score and data processing time score to obtain the real-time evaluation index.

[0013] Furthermore, the specific limiting expression of the real-time evaluation index is:

[0014]

[0015] Where i is the number of the sensor, i=1,2,3, i=1 represents the steering angle sensor, i=2 represents the vehicle speed sensor, i=3 represents the lidar sensor, q is the number of the preset time period, q=1,2,...,Q, Q is the total number of preset time periods, e is a natural constant, SHI i.q represents the real-time evaluation index of the i-th sensor in the q-th preset time period, F i.q represents the first data transmission rate score of the i-th sensor in the q-th preset time period, F1 i.q represents the sensor data transmission rate of the i-th sensor in the q-th preset time period, F10 represents the reference sensor data transmission rate, R i.q R1 represents the second data transmission rate score of the i-th sensor in the q-th preset time period, i.q represents the data transmission rate of the sensor to be transmitted of the i-th sensor in the q-th preset time period, R10 represents the reference data transmission rate of the sensor to be transmitted, V i.q Indicates the data collection duration fraction of the i-th sensor in the q-th preset time period, V1 i.q represents the data collection duration of the i-th sensor in the q-th preset time period, V10 represents the reference data collection duration, N i.q N1 represents the data transmission duration fraction of the i-th sensor in the q-th preset time period, i.q represents the data transmission duration of the i-th sensor in the q-th preset time period, N10 represents the reference data transmission duration, and U i.q Indicates the data processing time fraction of the i-th sensor in the q-th preset time period, U1 i.q represents the data processing time of the i-th sensor in the q-th preset time period, U10 represents the reference data processing time, and BU qrepresents the error compensation accuracy coefficient in the timestamp alignment process within the qth preset time period, ΔBU represents the error compensation accuracy threshold range, and P i.q represents the data synchronization period of the ith sensor in the qth preset time period, and P0 represents the starting time point of the data synchronization period.

[0016] Furthermore, the specific process of judging whether data processing is completed based on the acquired real-time evaluation index is as follows: judging whether the acquired real-time evaluation index is within the real-time evaluation threshold range: if the acquired real-time evaluation index is within the real-time evaluation threshold range, data processing is completed; if the acquired real-time evaluation index is not within the real-time evaluation threshold range, the preset personnel is prompted to tune the sensor signal of the sensor and re-process the data.

[0017] The embodiment of the present application provides a multi-sensor data processing system based on a rear-wheel steering vehicle, including: a timestamp offset coefficient acquisition module, an error compensation accuracy coefficient acquisition module and a real-time evaluation index acquisition module; wherein the timestamp offset coefficient acquisition module is used to acquire sensor data of a sensor within a preset time period, and at the same time acquire a timestamp offset coefficient according to a data synchronization period of the sensor data during the sensor acquisition process, and the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and a reference timestamp; the error compensation accuracy coefficient acquisition module is used to perform timestamp alignment on the acquired sensor data according to the acquired timestamp offset coefficient, count the sensor data to be transmitted and compare it with the sensor data to obtain the error compensation accuracy coefficient, and at the same time determine whether to perform data transmission based on the acquired error compensation accuracy coefficient, and the error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm; the real-time evaluation index acquisition module is used to obtain the data delay time between the sensor acquisition and the reception of the sensor data to be transmitted by the sensor and obtain the real-time evaluation index according to the data transmission rate if data transmission is performed, and determine whether data processing is completed based on the acquired real-time evaluation index, and the real-time evaluation index is used to reflect the real-time performance of data processing of the sensor data within the data synchronization period.

[0018] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0019] 1. The timestamp offset coefficient is obtained by the data synchronization period of the acquired sensor data in the sensor acquisition process, and then the timestamp alignment is performed to obtain the error compensation accuracy coefficient. Finally, based on the acquired real-time evaluation index, it is judged whether the data processing is completed, thereby achieving the improvement of the accuracy and reliability of the acquisition of real-time evaluation indicators, and then achieving the improvement of the timeliness of the rear-wheel steering vehicle sensor data processing, which effectively solves the problem of low timeliness of the rear-wheel steering vehicle sensor data processing in the prior art.

[0020] 2. By obtaining the timestamp offset of the sensor within the preset time period, the data synchronization period of the sensor within the preset time period is processed to obtain the synchronization period ratio. At the same time, the time difference offset coefficient is obtained according to the obtained timestamp offset and synchronization period ratio and combined with the maximum allowable timestamp offset, thereby achieving more accurate acquisition of the timestamp offset coefficient and further achieving a more accurate evaluation of the sensor synchronization performance.

[0021] 3. By obtaining the data collection time, data transmission time and data processing time of the sensor within a preset time period, the data collection time score, data transmission time score and data processing time score are obtained in combination with the corresponding reference values. At the same time, the data transmission rate of the sensor within the preset time period is obtained and combined with the corresponding reference value to obtain the first data transmission rate score and the second data transmission rate score, and the real-time evaluation index is obtained accordingly, thereby achieving more accurate acquisition of the real-time evaluation index, and further achieving more accurate evaluation of the real-time performance of the sensor during the data processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flowchart of a multi-sensor data processing method based on a rear-wheel steering vehicle provided in an embodiment of the present application;

[0023] Figure 2 A specific flow chart of data processing provided in the embodiment of the present application;

[0024] Figure 3 A schematic diagram of the structure of a multi-sensor data processing system for a rear-wheel steering vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application solve the problem of low timeliness of sensor data processing of rear-wheel steering vehicles in the prior art by providing a method and system for processing multi-sensor data of rear-wheel steering vehicles. The method obtains sensor data of the sensor within a preset time period and obtains a timestamp offset coefficient according to a data synchronization period of the sensor data during the sensor acquisition process. The timestamps of the acquired sensor data are then aligned according to the obtained timestamp offset coefficient. The sensor data to be transmitted are counted and compared with the sensor data to obtain an error compensation accuracy coefficient. At the same time, it is determined whether to transmit the data based on the obtained error compensation accuracy coefficient. Finally, during the data transmission process, the data delay time between the sensor acquisition and the reception by the central processing center of the sensor data to be transmitted is obtained and a real-time evaluation index is obtained based on the data transmission rate. Based on the obtained real-time evaluation index, it is determined whether the data processing is completed, thereby improving the timeliness of sensor data processing of rear-wheel steering vehicles.

[0026] The technical solution in the embodiment of the present application is to solve the problem of low timeliness of sensor data processing of the above-mentioned rear-wheel steering vehicle. The overall idea is as follows:

[0027] The timestamp offset coefficient is obtained by the data synchronization period of the acquired sensor data during the sensor acquisition process, and then the timestamp alignment is performed to obtain the error compensation accuracy coefficient. Finally, based on the obtained real-time evaluation index, it is judged whether the data processing is completed, thereby achieving the effect of improving the timeliness of sensor data processing for rear-wheel steering vehicles.

[0028] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0029] like Figure 1As shown, it is a flow chart of a multi-sensor data processing method for a rear-wheel steering vehicle provided in an embodiment of the present application. The multi-sensor data processing method for a rear-wheel steering vehicle provided in an embodiment of the present application includes the following steps: step one, acquiring sensor data of a sensor within a preset time period, and acquiring a timestamp offset coefficient according to a data synchronization period of the sensor data in a sensor acquisition process, wherein the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and a reference timestamp; step two, aligning the timestamps of the acquired sensor data according to the acquired timestamp offset coefficient and in combination with an error processing algorithm, counting the sensor data to be transmitted and comparing it with the sensor data to obtain an error compensation accuracy coefficient, and at the same time based on the acquired error The compensation accuracy coefficient determines whether data transmission is performed. Timestamp alignment is used to ensure that the sensor data collected by the sensor is synchronized in time. The sensor data to be transmitted is the sensor data after timestamp alignment. The error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm. Step three, if data transmission is performed, the data delay time between the sensor collection and the central processing center reception of the sensor data to be transmitted is obtained and the real-time evaluation index is obtained according to the data transmission rate. Based on the obtained real-time evaluation index, it is determined whether the data processing is completed. The data delay time is the sum of the data collection time, the data transmission time and the data processing time. The real-time evaluation index is used to reflect the real-time performance of the sensor data processing within the data synchronization period.

[0030] Among them, the sensors include steering angle sensors, vehicle speed sensors and lidar sensors; the steering angle sensor is used to measure the rear wheel steering angle of the specified vehicle in real time; the vehicle speed sensor is used to measure the rear wheel steering speed and rear wheel driving acceleration of the specified vehicle in real time; the lidar sensor is used to measure the distance between the specified vehicle and the obstacle in real time.

[0031] The sensor data includes rear wheel steering angle data, rear wheel steering speed data, rear wheel driving acceleration data and obstacle distance data of the specified vehicle; the data synchronization cycle includes the steering angle sensor acquisition cycle, the vehicle speed sensor acquisition cycle and the lidar sensor acquisition cycle; the sensor data timestamp includes the steering angle sensor timestamp (indicating the specific time point when the steering angle data is recorded or collected), the vehicle speed sensor timestamp (indicating the specific time point when the rear wheel steering speed data and the rear wheel driving acceleration data are recorded or collected) and the lidar sensor timestamp (indicating the time point when the obstacle distance data is recorded or collected); the reference timestamp includes the reference steering angle sensor timestamp, the reference vehicle speed sensor timestamp and the reference lidar sensor timestamp in the database; the sensor data to be transmitted includes the rear wheel steering angle data to be transmitted, the rear wheel steering speed data to be transmitted, the rear wheel driving acceleration data to be transmitted and the obstacle distance data to be transmitted corresponding to the specified vehicle after the sensor data is aligned with the timestamps.

[0032] In this embodiment, the reference steering angle sensor timestamp, the reference vehicle speed sensor timestamp and the reference lidar sensor timestamp are respectively represented by the sum and average of the historical timestamps of the steering angle sensor, the vehicle speed sensor and the lidar sensor in the preset database; Figure 2 As shown, this is a specific flow chart of data processing provided in an embodiment of the present application. This example improves the timeliness of rear-wheel steering vehicle sensor data processing by acquiring sensor data and timestamp offset coefficients, performing timestamp alignment and error compensation, as well as data transmission and real-time evaluation.

[0033] Further, the timestamp offset coefficient is obtained by the following method: obtaining the timestamp offset of the sensor within a preset time period (ie, ΔY in the constraint expression of the timestamp offset coefficient) i.q ), the timestamp offset includes a first timestamp offset, a second timestamp offset and a third timestamp offset, the first timestamp offset is the absolute value of the difference between the timestamp of the steering angle sensor and the timestamp of the reference steering angle sensor, the second timestamp offset is the absolute value of the difference between the timestamp of the vehicle speed sensor and the timestamp of the reference vehicle speed sensor, and the third timestamp offset is the absolute value of the difference between the timestamp of the lidar sensor and the timestamp of the reference lidar sensor; the data synchronization period of the sensor within the preset time period is processed to obtain the synchronization period ratio (that is, Z in the restriction expression of the timestamp offset coefficient i.q ), and according to the obtained timestamp offset and synchronization cycle ratio and the maximum allowed timestamp offset (that is, ΔY in the limiting expression of the timestamp offset coefficient i.max ) obtains the time difference offset coefficient, and the maximum allowed timestamp offset includes the maximum allowed first timestamp offset, the maximum allowed second timestamp offset and the maximum allowed third timestamp offset.

[0034] In this embodiment, the maximum allowable first timestamp offset is the maximum value of the first time offset of the steering angle sensor in the preset database within the historical time period, the maximum allowable second timestamp offset is the maximum value of the second timestamp offset of the vehicle speed sensor in the preset database within the historical time period, and the maximum allowable third timestamp offset is the maximum value of the third timestamp offset of the lidar sensor in the preset database within the historical time period.

[0035] Specifically, the specific restriction expression of the timestamp offset coefficient is (wherein the independent variable as the denominator of the expression is not 0):

[0036]

[0037] ΔY i.q =|Y1 i.q-Yi0|;

[0038] Where i is the number of the sensor, i=1,2,3, i=1 represents the steering angle sensor, i=2 represents the vehicle speed sensor, i=3 represents the lidar sensor, q is the number of the preset time period, q=1,2,...,Q, Q is the total number of preset time periods, e is a natural constant, YI i.q It represents the timestamp offset coefficient of the i-th sensor in the q-th preset time period, Z i.q represents the synchronization cycle ratio of the i-th sensor in the q-th preset time period, P i.q represents the data synchronization period of the ith sensor in the qth preset time period, P 1.q represents the steering angle sensor acquisition cycle within the qth preset time period, P 2.q represents the vehicle speed sensor acquisition cycle in the qth preset time period, P 3.q represents the laser radar sensor acquisition cycle in the qth preset time period, ΔY i.q Indicates the timestamp offset of the i-th sensor in the q-th preset time period, ΔY i.max represents the maximum allowed timestamp offset of the ith sensor in the qth preset time period, Y1 i.q represents the sensor data timestamp of the ith sensor in the qth preset time period, and Yi0 represents the reference timestamp of the ith sensor.

[0039] It should be understood that the timestamp offset coefficient (the smaller the better) decreases with the increase of the synchronization cycle ratio and increases with the increase of the timestamp offset. It should be noted that the synchronization cycle ratio also indirectly affects the value of the timestamp offset. When the synchronization cycle ratio increases, this means that the synchronization of sensor data during transmission increases, that is, the timestamp difference between sensors will decrease (that is, the timestamp offset decreases). By considering the above-mentioned indirect influence mechanism and combining the timestamp offset to calculate the timestamp offset coefficient, the synchronization quality and accuracy of the sensor data can be more accurately evaluated, which helps to timely discover and correct asynchronous data packets during the sensor data processing of rear-wheel steering vehicles, thereby improving the accuracy of data processing, and then realizing the improvement of the timeliness of sensor data processing of rear-wheel steering vehicles, effectively solving the problem of low timeliness of sensor data processing of rear-wheel steering vehicles in the prior art.

[0040] Further, the error compensation accuracy coefficient is obtained by the following method: A1, judging whether the obtained timestamp offset coefficient is within the allowed range of the timestamp offset, if so, executing A2, otherwise continuing to monitor the timestamp alignment process; A2, counting the first timestamp alignment adjustment amount, the second timestamp alignment adjustment amount and the third timestamp alignment adjustment amount in the timestamp alignment process within a preset time period and summing them respectively, and obtaining a preset timestamp alignment adjustment amount weight factor from a preset database; A3, obtaining a first timestamp compensation score (i.e., G in the restriction expression of the error compensation accuracy coefficient) according to the result of the summation operation and in combination with the obtained first timestamp offset, second timestamp offset and third timestamp offset q ), the second timestamp compensation score (i.e., H in the constraint expression of the error compensation accuracy coefficient q ) and the third timestamp compensation score (i.e., K in the constraint expression of the error compensation accuracy coefficient q ), and the error compensation accuracy coefficient is obtained accordingly; the timestamp alignment adjustment amount weight factor includes the first timestamp alignment adjustment amount weight factor (i.e., δ1 in the restriction expression of the error compensation accuracy coefficient), the second timestamp alignment adjustment amount weight factor (i.e., δ2 in the restriction expression of the error compensation accuracy coefficient) and the third timestamp alignment adjustment amount weight factor (i.e., δ3 in the restriction expression of the error compensation accuracy coefficient); the first timestamp compensation score is the ratio of the absolute value of the difference between the average value of the first timestamp alignment adjustment amount and the initial first timestamp alignment adjustment amount in the previous time period to the first timestamp alignment adjustment amount; the second timestamp compensation score is the ratio of the absolute value of the difference between the average value of the second timestamp alignment adjustment amount and the initial second timestamp alignment adjustment amount in the previous time period to the second timestamp alignment adjustment amount; the third timestamp compensation score is the ratio of the absolute value of the difference between the average value of the third timestamp alignment adjustment amount and the initial third timestamp alignment adjustment amount in the previous time period to the third timestamp alignment adjustment amount.

[0041] Specifically, the specific limiting expression of the error compensation accuracy coefficient is (wherein the independent variable as the denominator of the expression is not 0):

[0042]

[0043] Where i is the number of the sensor, i=1,2,3, i=1 represents the steering angle sensor, i=2 represents the vehicle speed sensor, i=3 represents the lidar sensor, q is the number of the preset time period, q=1,2,...,Q, Q is the total number of preset time periods, e is a natural constant, BU q represents the error compensation accuracy coefficient in the timestamp alignment process within the qth preset time period, δ1 represents the weight factor of the first timestamp alignment adjustment amount, G qIndicates the first timestamp compensation score in the timestamp alignment process within the qth preset time period, G1 q Indicates the first timestamp alignment adjustment amount during the timestamp alignment process in the qth preset time period, G1 q-1 represents the initial first timestamp alignment adjustment amount in the timestamp alignment process in the q-1th preset time period, δ2 represents the weight factor of the second timestamp alignment adjustment amount, and H q Indicates the second timestamp compensation score in the timestamp alignment process within the qth preset time period, H1 q Indicates the second timestamp alignment adjustment amount during the timestamp alignment process in the qth preset time period, H1 q-1 represents the initial second timestamp alignment adjustment amount in the timestamp alignment process in the q-1th preset time period, δ3 represents the weight factor of the third timestamp alignment adjustment amount, K q Indicates the third timestamp compensation score during the timestamp alignment process within the qth preset time period, K1 q K1 represents the third timestamp alignment adjustment amount in the timestamp alignment process within the qth preset time period, q-1 represents the initial third timestamp alignment adjustment amount in the timestamp alignment process within the q-1th preset time period, ΔY 1.q represents the first timestamp offset of the steering angle sensor in the qth preset time period, ΔE 2.q Indicates the second timestamp offset of the vehicle speed sensor in the qth preset time period, ΔS 3.q Indicates the third timestamp offset of the lidar sensor in the qth preset time period, YI i.q represents the timestamp offset coefficient of the i-th sensor in the q-th preset time period, and ΔYI represents the allowable range of the timestamp offset.

[0044] In this embodiment, the weight factor of the first timestamp alignment adjustment is the weight factor corresponding to the error compensation accuracy coefficient acquisition process in the preset database, which indicates the numerical value corresponding to the influence of the first timestamp alignment adjustment on the error compensation accuracy coefficient. When used, the weight factor corresponding to the first timestamp alignment adjustment can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the first timestamp alignment adjustment corresponding to the error compensation accuracy coefficient acquisition process of the first timestamp alignment adjustment and the weight factor corresponding to the first timestamp alignment adjustment in the preset database form a mapping set, and the real-time first timestamp alignment adjustment is input into the mapping set to obtain the corresponding first timestamp alignment adjustment weight factor, wherein the mapping relationship can be one-to-one or many-to-one. In this example, the value range of the weight factor of the first timestamp alignment adjustment is [0,1].

[0045] The second timestamp alignment adjustment weight factor is the weight factor corresponding to the error compensation accuracy coefficient acquisition process in the preset database, which indicates the numerical value corresponding to the influence of the second timestamp alignment adjustment on the error compensation accuracy coefficient. When used, the weight factor corresponding to the second timestamp alignment adjustment can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the second timestamp alignment adjustment corresponding to the error compensation accuracy coefficient acquisition process of the second timestamp alignment adjustment and the weight factor corresponding to the second timestamp alignment adjustment in the preset database form a mapping set, and the real-time second timestamp alignment adjustment is input into the mapping set to obtain the corresponding second timestamp alignment adjustment weight factor, wherein the mapping relationship can be one-to-one or many-to-one. In this example, the value range of the second timestamp alignment adjustment weight factor is [0,1].

[0046] In this example, the sum of the first timestamp alignment adjustment weight factor, the second timestamp alignment adjustment weight factor, and the third timestamp alignment adjustment weight factor is 1.

[0047] Specifically, timestamp alignment includes first timestamp alignment, second timestamp alignment and third timestamp alignment; the first timestamp alignment is used to obtain the first timestamp alignment adjustment amount after the first timestamp alignment (indicating the difference between the first timestamp offset after the first timestamp alignment and the first timestamp offset before the first timestamp alignment); the second timestamp alignment is used to obtain the second timestamp alignment adjustment amount after the second timestamp alignment (indicating the difference between the second timestamp offset after the second timestamp alignment and the second timestamp offset before the second timestamp alignment); the third timestamp alignment is used to obtain the third timestamp alignment adjustment amount after the third timestamp alignment (indicating the difference between the third timestamp offset after the third timestamp alignment and the third timestamp offset before the third timestamp alignment).

[0048] It should be added that the specific process for determining whether the timestamp alignment is completed is: real-time monitoring of the changes in the first timestamp offset, the second timestamp offset and the third timestamp offset during the timestamp alignment process; in the first timestamp alignment process, when the first timestamp alignment adjustment amount is equal to 0, the first timestamp alignment is completed; in the second timestamp alignment process, when the second timestamp alignment adjustment amount is equal to 0, the second timestamp alignment is completed; in the third timestamp alignment process, when the third timestamp alignment adjustment amount is equal to 0, the third timestamp alignment is completed; when the first timestamp alignment adjustment amount is equal to 0, the second timestamp alignment adjustment amount is equal to 0, and the third timestamp alignment adjustment amount is equal to 0, the timestamp alignment is completed.

[0049] It should be understood that when the preset time period is 1, the initial first timestamp alignment adjustment amount is the first timestamp offset, the initial second timestamp alignment adjustment amount is the second timestamp offset, and the initial third timestamp alignment adjustment amount is the third timestamp offset.

[0050] The error compensation accuracy coefficient (the larger the better) decreases with the increase of the first timestamp compensation score, the second timestamp compensation score and the third timestamp compensation score, wherein the first timestamp compensation score also indirectly affects the values ​​of the second timestamp compensation score and the third timestamp compensation score. Assuming that within the preset time period q=3, the first timestamp offset of the steering angle sensor increases (for example, from 0.1 milliseconds to 0.5 milliseconds), this will cause the data acquisition delay of the vehicle speed sensor, that is, the second timestamp offset increases, and the second timestamp compensation score decreases (because the compensation score is inversely proportional to the timestamp offset). Subsequently, due to the correlation between the vehicle speed and the laser radar measurement results (for example, an increase in vehicle speed may cause the obstacle distance measured by the laser radar to change), the third timestamp offset of the laser radar sensor will also increase. By considering the above-mentioned indirect influence mechanism, it is helpful to improve the accuracy and reliability of timestamp alignment, realize more accurate acquisition of the error compensation accuracy coefficient, and then realize the improvement of the timeliness of rear-wheel steering vehicle sensor data processing, which effectively solves the problem of low timeliness of rear-wheel steering vehicle sensor data processing in the prior art.

[0051] Furthermore, the specific process of determining whether to perform data transmission based on the obtained error compensation accuracy coefficient is as follows: determining whether the obtained error compensation accuracy coefficient is within the error compensation accuracy threshold range: if the obtained error compensation accuracy coefficient is within the error compensation accuracy threshold range (including the maximum and minimum values ​​of the historical error compensation accuracy coefficient), then it indicates that the result of the timestamp alignment meets the quality requirements of data transmission and data transmission is performed; if the obtained error compensation accuracy coefficient is not within the error compensation accuracy threshold range, then it indicates that the result of the timestamp alignment does not meet the quality requirements of data transmission, and data transmission is not performed at this time, and the preset personnel are prompted to take processing measures; the processing measures are to adjust the error processing algorithm parameters; the error processing algorithm parameters include filtering parameters, smoothing parameters and number of iterations.

[0052] In this embodiment, the error compensation accuracy threshold range represents the range corresponding to the maximum and minimum values ​​of the historical error compensation accuracy coefficient of the sensor in the historical time period in the preset database. The error processing algorithm parameters are adjusted by preset personnel. When the acquired error compensation accuracy coefficient is not within the error compensation accuracy threshold range, it is necessary to increase the filtering parameters, smoothing parameters and iteration parameters in the sensor to improve the timestamp alignment accuracy of the sensor data, thereby achieving the improvement of the accuracy and real-time performance of the sensor data to be transmitted during the transmission process.

[0053] Furthermore, the specific steps for obtaining the real-time evaluation index are as follows: when the obtained error compensation accuracy coefficient is within the error compensation accuracy threshold range and the data synchronization period is equal to the starting time point of the data synchronization period: obtain the data collection time, data transmission time and data processing time of the sensor within the preset time period, and respectively combine the corresponding reference values ​​to obtain the data collection time score (i.e., V in the restriction expression of the real-time evaluation index). i.q ), data transmission duration score (i.e., N in the restriction expression of real-time evaluation index i.q ) and the data processing time score (i.e., U in the limiting expression of the real-time evaluation index) i.q ); obtain the data transmission rate of the sensor within a preset time period and obtain the first data transmission rate score (i.e., the F in the restriction expression of the real-time evaluation index) in combination with the corresponding reference value i.q ) and the second data transmission rate score (i.e., R in the limiting expression of the real-time evaluation index) i.q ); Process the obtained first data transmission rate score and the second data transmission rate score and combine them with the obtained data collection duration score, data transmission duration score and data processing duration score to obtain a real-time evaluation index.

[0054] Among them, the data collection time score is the ratio of the data collection time to the reference data collection time; the data transmission time score is the ratio of the data transmission time to the reference data transmission time; the data processing time score is the ratio of the data processing time to the reference data processing time; the data transmission rate includes the sensor data transmission rate and the sensor data transmission rate to be transmitted; the first data transmission rate score is the ratio of the sensor data transmission rate to the reference sensor data transmission rate; the second data transmission rate score is the ratio of the sensor data rate to be transmitted to the reference sensor data transmission rate to be transmitted.

[0055] Specifically, the specific restriction expression of the real-time evaluation index is (wherein the independent variable as the denominator of the expression is not 0):

[0056]

[0057] Where i is the number of the sensor, i=1,2,3, i=1 represents the steering angle sensor, i=2 represents the vehicle speed sensor, i=3 represents the lidar sensor, q is the number of the preset time period, q=1,2,...,Q, Q is the total number of preset time periods, e is a natural constant, SHI i.q represents the real-time evaluation index of the i-th sensor in the q-th preset time period, F i.q represents the first data transmission rate score of the i-th sensor in the q-th preset time period, F1 i.qrepresents the sensor data transmission rate of the i-th sensor in the q-th preset time period, F10 represents the reference sensor data transmission rate, R i.q R1 represents the second data transmission rate score of the i-th sensor in the q-th preset time period, i.q represents the data transmission rate of the sensor to be transmitted of the i-th sensor in the q-th preset time period, R10 represents the reference data transmission rate of the sensor to be transmitted, V i.q Indicates the data collection duration fraction of the i-th sensor in the q-th preset time period, V1 i.q represents the data collection duration of the i-th sensor in the q-th preset time period, V10 represents the reference data collection duration, N i.q N1 represents the data transmission duration fraction of the i-th sensor in the q-th preset time period, i.q represents the data transmission duration of the i-th sensor in the q-th preset time period, N10 represents the reference data transmission duration, and U i.q Indicates the data processing time fraction of the i-th sensor in the q-th preset time period, U1 i.q represents the data processing time of the i-th sensor in the q-th preset time period, U10 represents the reference data processing time, and BU q represents the error compensation accuracy coefficient in the timestamp alignment process within the qth preset time period, ΔBU represents the error compensation accuracy threshold range, and P i.q represents the data synchronization period of the ith sensor in the qth preset time period, and P0 represents the starting time point of the data synchronization period.

[0058] In this embodiment, the reference data collection duration is represented by the sum and average of the historical data collection durations of the sensors in the preset database within the historical time period, the reference data transmission duration is represented by the sum and average of the historical data transmission durations of the sensors in the preset database within the historical time period, and the reference data processing duration is represented by the sum and average of the historical data processing durations of the sensors in the preset database within the historical time period; the reference sensor data transmission rate is represented by the sum and average of the historical data transmission rates of the sensor data in the preset database within the historical time period, and the reference to-be-transmitted sensor data transmission rate is represented by the sum and average of the historical data transmission rates of the to-be-transmitted sensor data in the preset database within the historical time period.

[0059] It should be understood that the data synchronization period is equal to the starting time point of the data synchronization period (set by the preset personnel), indicating that the sensor transmits the sensor data to be transmitted synchronously. This synchronous transmission helps to ensure the data consistency and real-time performance between sensors, thereby improving the reliability and efficiency of data transmission.

[0060] Assuming that the error compensation accuracy threshold range is between 0 and 1, the statistical table of changes in real-time evaluation indicators is shown in Table 1:

[0061] Table 1 Statistics of changes in real-time evaluation indicators

[0062]

[0063] It should be understood that, from Table 1, the real-time evaluation index (the larger the better) increases with the increase of the first data transmission rate score and the second data transmission rate score, and decreases with the increase of the data acquisition time score, the data transmission time score and the data processing time score, among which the first data transmission rate score increases with the increase of the sensor data transmission rate, the second data transmission rate score increases with the increase of the sensor data transmission rate to be transmitted, the data acquisition time score increases with the increase of the data acquisition time, the data transmission time score increases with the increase of the data transmission time, and the data processing time score increases with the increase of the data processing time.

[0064] It should be noted that the data collection duration score also indirectly affects the values ​​of the data transmission duration score and the data processing score. Assuming that the sensor system of a rear-wheel steering vehicle needs to process data from multiple sensors in real time to control the steering of the vehicle, when the data transmission duration of a certain sensor (such as a steering angle sensor) increases (for example, due to network delays or the low communication rate of the sensor itself), the data of this sensor will arrive at the processing unit later than the data of other sensors, thereby increasing the data transmission duration and data processing duration. By considering the above-mentioned indirect impact mechanism, it is helpful to reduce the delay of data transmission and data processing, improve the accuracy of data transmission and data processing, and thus achieve the improvement of the timeliness of sensor data processing of rear-wheel steering vehicles, effectively solving the problem of low timeliness of sensor data processing of rear-wheel steering vehicles in the prior art.

[0065] Furthermore, the specific process of judging whether data processing is completed based on the acquired real-time evaluation index is as follows: judging whether the acquired real-time evaluation index is within the real-time evaluation threshold range: if the acquired real-time evaluation index is within the real-time evaluation threshold range (including being equal to the maximum and minimum values ​​of the historical real-time evaluation index), it indicates that the data processing process of the sensor meets the expected real-time requirements, and the data processing is completed at this time; if the acquired real-time evaluation index is not within the real-time evaluation threshold range, it indicates that the data processing process of the sensor does not meet the expected real-time requirements, and the preset personnel is prompted to tune the sensor signal of the sensor and re-process the data.

[0066] In this embodiment, the real-time evaluation threshold range represents the range corresponding to the maximum and minimum values ​​of the historical real-time evaluation indicators of the sensors in the preset database within the historical time period. When the preset personnel receive the tuning signal, they will adjust the software code related to the sensor data processing (such as updating the driver). After each adjustment, it is necessary to re-acquire the real-time evaluation indicator until the re-acquired real-time evaluation indicator is within the real-time evaluation threshold range to ensure the synchronization and compatibility of the sensors during the data processing process, reduce the reduction in real-time data processing due to data transmission errors, and help to improve the real-time and accuracy of the data processing system.

[0067] like Figure 3 As shown, it is a structural schematic diagram of a multi-sensor data processing system for a rear-wheel steering vehicle provided in an embodiment of the present application. The multi-sensor data processing system for a rear-wheel steering vehicle provided in an embodiment of the present application includes: a timestamp offset coefficient acquisition module, an error compensation accuracy coefficient acquisition module and a real-time evaluation index acquisition module; wherein, the timestamp offset coefficient acquisition module is used to acquire sensor data of the sensor within a preset time period, and at the same time acquire the timestamp offset coefficient according to the data synchronization period of the sensor data during the sensor acquisition process, and the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and the reference timestamp; the error compensation accuracy coefficient acquisition module is used to perform timestamp alignment on the acquired sensor data according to the acquired timestamp offset coefficient and in combination with the error processing algorithm, and to count the sensor data to be transmitted and compare it with the sensor data A comparison is performed to obtain the error compensation accuracy coefficient, and at the same time, whether data transmission is performed is determined based on the obtained error compensation accuracy coefficient. Timestamp alignment is used to ensure that the sensor data collected by the sensor is synchronized in time. The sensor data to be transmitted is the sensor data after timestamp alignment. The error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm; the real-time evaluation index acquisition module is used to obtain the data delay time between the sensor collection and the central processing center reception of the sensor data to be transmitted if data transmission is performed, and obtain the real-time evaluation index according to the data transmission rate. Based on the obtained real-time evaluation index, it is determined whether data processing is completed. The data delay time is the sum of the data collection time, the data transmission time and the data processing time. The real-time evaluation index is used to reflect the real-time performance of sensor data processing within the data synchronization period.

[0068] In this embodiment, the timestamp offset coefficient acquisition module can accurately capture and calculate the timestamp offset coefficient of the sensor in the data synchronization period, which ensures that all sensor data can maintain a high degree of consistency with the reference timestamp and ensures the synchronous processing of data by the sensor. The error compensation accuracy coefficient acquisition module accurately evaluates the accuracy and effectiveness of the error processing algorithm through timestamp alignment technology, thereby further optimizing the accuracy and reliability of data transmission. The real-time evaluation index acquisition module can monitor the entire process of data from collection to transmission to processing in real time, and can quickly determine whether data processing is completed and whether it meets the real-time requirements, thereby realizing a comprehensive evaluation of the time synchronization, error compensation capability and real-time data processing capability of the data processing system.

[0069] It should be added that, since the fusion operation needs to maintain the synchronization performance between sensors, the data processing in this example is performed before the fusion operation to ensure the accuracy and consistency of the sensor data before the fusion operation. Specifically, the inertial measurement unit (IMU), global navigation satellite system (GNSS) and odometer are fused through the Kalman filter (a recursive filter based on linear algebra and probability theory. It can predict the future state by weighted processing of past and present measurement data in the presence of uncertainty) to perform a fusion operation on the sensor data (for denoising and filtering), and combined with the rear wheel deflection angle information of the specified vehicle to ensure that the sensor data collected by the sensor at the same time point matches, thereby realizing the extraction of the lateral speed of the rear-wheel steering vehicle and the reduction of the positioning error, improving the accuracy and reliability of the positioning of the specified vehicle, and thus providing strong technical support for autonomous driving.

[0070] To summarize, the embodiment of the present application obtains the timestamp offset coefficient through the data synchronization period of the acquired sensor data during the sensor acquisition process, then performs timestamp alignment to obtain the error compensation accuracy coefficient, and finally determines whether the data processing is completed based on the acquired real-time evaluation index, thereby achieving improved accuracy and reliability in obtaining the real-time evaluation index, and further achieving improved timeliness of sensor data processing for rear-wheel steering vehicles, effectively solving the problem of low timeliness of sensor data processing for rear-wheel steering vehicles in the prior art.

[0071] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0072] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0075] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0076] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-sensor data processing method for a rear-wheel steering vehicle, characterized in that: The following steps are involved: Step 1: acquiring sensor data of the sensor within a preset time period, and acquiring a timestamp offset coefficient according to a data synchronization period of the sensor data during the sensor acquisition process, wherein the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and the reference timestamp; Step 2: align the timestamps of the acquired sensor data according to the acquired timestamp offset coefficient, collect statistics on the sensor data to be transmitted and compare them with the sensor data to obtain the error compensation accuracy coefficient, and determine whether to transmit the data based on the acquired error compensation accuracy coefficient. The error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm. Step three, if data transmission is performed, obtain the data delay time between the sensor data to be transmitted being collected and received by the central processing center, and obtain a real-time evaluation index based on the data transmission rate. Based on the obtained real-time evaluation index, determine whether the data processing is completed. The real-time evaluation index is used to reflect the real-time performance of data processing of the sensor data within the data synchronization period.

2. The multi-sensor data processing method based on rear-wheel steering vehicle according to claim 1, characterized in that: The sensors include a steering angle sensor, a vehicle speed sensor and a laser radar sensor; The steering angle sensor is used to measure the rear wheel steering angle of a specified vehicle in real time; The vehicle speed sensor is used to measure the rear wheel steering speed and rear wheel driving acceleration of the specified vehicle in real time; The laser radar sensor is used to measure the distance between the specified vehicle and the obstacle in real time; The sensor data includes rear wheel steering angle data, rear wheel steering speed data, rear wheel driving acceleration data and obstacle distance data of a specified vehicle; The data synchronization cycle includes a steering angle sensor acquisition cycle, a vehicle speed sensor acquisition cycle and a laser radar sensor acquisition cycle; The sensor data timestamp includes a steering angle sensor timestamp, a vehicle speed sensor timestamp and a lidar sensor timestamp.

3. The multi-sensor data processing method based on rear-wheel steering vehicle as claimed in claim 2, characterized in that: The timestamp offset coefficient is obtained by the following method: Obtaining a timestamp offset of the sensor within a preset time period, wherein the timestamp offset includes a first timestamp offset, a second timestamp offset, and a third timestamp offset, wherein the first timestamp offset is the absolute value of the difference between the timestamp of the steering angle sensor and the timestamp of the reference steering angle sensor, the second timestamp offset is the absolute value of the difference between the timestamp of the vehicle speed sensor and the timestamp of the reference vehicle speed sensor, and the third timestamp offset is the absolute value of the difference between the timestamp of the lidar sensor and the timestamp of the reference lidar sensor; The data synchronization cycle of the sensor within a preset time period is processed to obtain the synchronization cycle proportion, and the time difference offset coefficient is obtained based on the obtained timestamp offset and the synchronization cycle proportion and combined with the maximum allowable timestamp offset. The maximum allowable timestamp offset includes the maximum allowable first timestamp offset, the maximum allowable second timestamp offset and the maximum allowable third timestamp offset.

4. The method for processing multi-sensor data based on a rear-wheel steering vehicle as claimed in claim 1, characterized in that: The timestamp alignment includes a first timestamp alignment, a second timestamp alignment and a third timestamp alignment; The first timestamp alignment is used to obtain a first timestamp alignment adjustment amount after the first timestamp alignment; The second timestamp alignment is used to obtain a second timestamp alignment adjustment amount after the second timestamp alignment; The third timestamp alignment is used to obtain a third timestamp alignment adjustment amount after the third timestamp alignment; The sensor data to be transmitted include rear wheel steering angle data to be transmitted, rear wheel steering speed data to be transmitted, rear wheel driving acceleration data to be transmitted and obstacle distance data to be transmitted corresponding to the designated vehicle after the sensor data are aligned with the timestamps.

5. The method for processing multi-sensor data based on a rear-wheel steering vehicle as claimed in claim 1, characterized in that: The error compensation accuracy coefficient is obtained by the following method: A1, determine whether the acquired timestamp offset coefficient is within the timestamp offset allowable range, if so, execute A2, otherwise continue to monitor the timestamp alignment process; A2, counting a first timestamp alignment adjustment amount, a second timestamp alignment adjustment amount, and a third timestamp alignment adjustment amount in a timestamp alignment process within a preset time period and performing a sum operation, and obtaining a preset timestamp alignment adjustment amount weight factor from a preset database; A3, obtaining a first timestamp compensation score, a second timestamp compensation score, and a third timestamp compensation score according to the result of the summation operation and in combination with the obtained first timestamp offset, second timestamp offset, and third timestamp offset, and obtaining an error compensation accuracy coefficient accordingly; The timestamp alignment adjustment weight factor includes a first timestamp alignment adjustment weight factor, a second timestamp alignment adjustment weight factor and a third timestamp alignment adjustment weight factor; The first timestamp compensation score is a ratio of an absolute value of a difference between an average value of the first timestamp alignment adjustment amount and an initial first timestamp alignment adjustment amount in a previous time period to the first timestamp alignment adjustment amount; The second timestamp compensation score is a ratio of an absolute value of a difference between an average value of the second timestamp alignment adjustment amount and an initial second timestamp alignment adjustment amount in a previous time period to the second timestamp alignment adjustment amount; The third timestamp compensation score is a ratio of an absolute value of a difference between an average value of the third timestamp alignment adjustment amount and an initial third timestamp alignment adjustment amount in a previous time period to the third timestamp alignment adjustment amount.

6. The method for processing multi-sensor data based on a rear-wheel steering vehicle as claimed in claim 1, characterized in that: The specific process of determining whether to perform data transmission based on the obtained error compensation accuracy coefficient is as follows: Determine whether the obtained error compensation accuracy coefficient is within the error compensation accuracy threshold range: If the error compensation accuracy coefficient obtained is within the error compensation accuracy threshold range, data transmission is performed; If the obtained error compensation accuracy coefficient is not within the error compensation accuracy threshold range, data transmission will not be performed and the preset personnel will be prompted to take processing measures.

7. The multi-sensor data processing method based on a rear-wheel steering vehicle as claimed in claim 1, characterized in that: The specific steps for obtaining the real-time evaluation index are as follows: Obtain the data collection duration, data transmission duration and data processing duration of the sensor within a preset time period, and respectively obtain the data collection duration score, data transmission duration score and data processing duration score in combination with the corresponding reference values; Obtaining a data transmission rate of the sensor within a preset time period and obtaining a first data transmission rate score and a second data transmission rate score in combination with a corresponding reference value; The obtained first data transmission rate score and the second data transmission rate score are processed and combined with the obtained data collection duration score, data transmission duration score and data processing duration score to obtain a real-time evaluation index.

8. The method for processing multi-sensor data based on a rear-wheel steering vehicle as claimed in claim 7, characterized in that: The specific limiting expression of the real-time evaluation index is: Where i is the number of the sensor, i=1,2,3, i=1 represents the steering angle sensor, i=2 represents the vehicle speed sensor, i=3 represents the lidar sensor, q is the number of the preset time period, q=1,2,...,Q, Q is the total number of preset time periods, e is a natural constant, SHI i.q represents the real-time evaluation index of the i-th sensor in the q-th preset time period, F i.q represents the first data transmission rate score of the i-th sensor in the q-th preset time period, F1 i.q represents the sensor data transmission rate of the i-th sensor in the q-th preset time period, F10 represents the reference sensor data transmission rate, R i.q R1 represents the second data transmission rate score of the i-th sensor in the q-th preset time period, i.q represents the data transmission rate of the sensor to be transmitted of the i-th sensor in the q-th preset time period, R10 represents the reference data transmission rate of the sensor to be transmitted, V i.q Indicates the data collection duration fraction of the i-th sensor in the q-th preset time period, V1 i.q represents the data collection duration of the i-th sensor in the q-th preset time period, V10 represents the reference data collection duration, N i.q N1 represents the data transmission duration fraction of the i-th sensor in the q-th preset time period, i.q represents the data transmission duration of the i-th sensor in the q-th preset time period, N10 represents the reference data transmission duration, and U i.q Indicates the data processing time fraction of the i-th sensor in the q-th preset time period, U1 i.q represents the data processing time of the i-th sensor in the q-th preset time period, U10 represents the reference data processing time, and BU q represents the error compensation accuracy coefficient in the timestamp alignment process within the qth preset time period, ΔBU represents the error compensation accuracy threshold range, and P i.q represents the data synchronization period of the ith sensor in the qth preset time period, and P0 represents the starting time point of the data synchronization period.

9. The method for processing multi-sensor data based on a rear-wheel steering vehicle as claimed in claim 1, characterized in that: The specific process of judging whether data processing is completed based on the obtained real-time evaluation index is as follows: Determine whether the obtained real-time evaluation index is within the real-time evaluation threshold range: If the obtained real-time evaluation index is within the real-time evaluation threshold range, data processing is completed; If the acquired real-time evaluation index is not within the real-time evaluation threshold range, the preset personnel are prompted to tune the sensor signal of the sensor and re-process the data.

10. A multi-sensor data processing system for a rear-wheel steering vehicle, characterized in that: include: A timestamp offset coefficient acquisition module, an error compensation accuracy coefficient acquisition module, and a real-time evaluation index acquisition module; The timestamp offset coefficient acquisition module is used to acquire sensor data of the sensor within a preset time period, and acquire the timestamp offset coefficient according to the data synchronization period of the sensor data during the sensor acquisition process, and the timestamp offset coefficient is used to reflect the degree of deviation between the timestamp of the sensor data and the reference timestamp; The error compensation accuracy coefficient acquisition module is used to perform timestamp alignment on the acquired sensor data according to the acquired timestamp offset coefficient, count the sensor data to be transmitted and compare it with the sensor data to obtain the error compensation accuracy coefficient, and determine whether to perform data transmission based on the acquired error compensation accuracy coefficient. The error compensation accuracy coefficient is used to evaluate the accuracy and effectiveness of the error processing algorithm; The real-time evaluation index acquisition module is used to obtain the data delay time between the sensor data to be transmitted being collected from the sensor and received by the central processing center, and obtain the real-time evaluation index according to the data transmission rate, and judge whether the data processing is completed based on the obtained real-time evaluation index. The real-time evaluation index is used to reflect the real-time performance of the sensor data processing within the data synchronization period.

Images