Lane change intention recognition method, device, storage medium and electronic device

Abstract

Disclosed are a lane change intention recognition method, device, storage medium, and electronic device, relating to the field of intelligent driving technology. The method includes determining driving data of a target object at multiple times; adjusting the yaw rate in the driving data at each time based on the driving data at the multiple times to obtain an adjusted yaw rate at each time; determining a yaw rate error at each time based on the yaw rate at each time; and determining a lane change intention recognition result for the target object based on the driving data at each time, the adjusted yaw rate at each time, and the yaw rate error at each time. By adjusting the yaw rate and the yaw rate error at each time to determine the intention recognition result, the lane change intention recognition accuracy can be improved, and the method can better adapt to different driving environments and vehicle behaviors, providing greater flexibility.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent driving technology, and in particular to a method, device, storage medium, and electronic device for recognizing lane change intention. Background Art

[0002] In an increasingly complex road traffic environment, vehicle safety and smooth and orderly traffic flow are crucial. With the continuous advancement of technology, intelligent vehicle control and driving technologies are becoming increasingly important means of ensuring driving safety. Accurately identifying a vehicle's lane change intention is a key technology. Summary of the Invention

[0003] Currently, most related technologies rely on static thresholds to identify a vehicle's lane change intention. Because driving behavior in real-world environments changes rapidly, this approach is not adaptable to varying driving conditions. Consequently, lane change intention recognition results are relatively inaccurate. Therefore, accurately identifying lane change intentions is an urgent problem that needs to be solved.

[0004] In order to solve the above technical problems, the present disclosure provides a lane change intention recognition method, device, computer-readable storage medium and electronic device to accurately identify the lane change intention of a target object.

[0005] An embodiment of a first aspect of the present disclosure provides a method for identifying lane change intentions, comprising: determining driving data of a target object at multiple time points; adjusting a yaw rate in the driving data at each time point based on the driving data at the multiple time points to obtain an adjusted yaw rate at each time point; determining a yaw rate error at each time point based on the yaw rate at each time point; and determining a lane change intention identification result for the target object based on the driving data at each time point, the adjusted yaw rate at each time point, and the yaw rate error at each time point.

[0006] According to a second aspect of the present disclosure, an embodiment provides a lane change intention recognition device, comprising: a first determination module for determining driving data of a target object at multiple time points; a second determination module for adjusting the yaw rate in the driving data at each time point based on the driving data at the multiple time points to obtain an adjusted yaw rate at each time point; a third determination module for determining a yaw rate error at each time point based on the yaw rate in the driving data at each time point; and a fourth determination module for determining a lane change intention recognition result of the target object based on the driving data at each time point, the adjusted yaw rate at each time point, and the yaw rate error at each time point.

[0007] An embodiment of the third aspect of the present disclosure provides a computer-readable storage medium, which stores a computer program for executing the lane change intention recognition method provided by the embodiment of the first aspect of the present disclosure.

[0008] An embodiment of the fourth aspect of the present disclosure provides an electronic device, which includes: a processor; a memory for storing instructions executable by the processor; and a processor for reading instructions from the memory and executing the lane change intention recognition method provided in the embodiment of the first aspect of the present disclosure.

[0009] The fifth aspect embodiment of the present disclosure provides a computer program product, which, when executed by an instruction processor in the computer program product, executes the lane change intention recognition method provided by the first aspect embodiment of the present disclosure.

[0010] This disclosure provides a lane change intention recognition method that obtains real-time driving data of a target object at multiple moments. Based on the yaw rate in the driving data, the method determines an adjusted yaw rate and yaw rate error. Finally, based on the driving data at each moment, the adjusted yaw rate and yaw rate error are used to predict the lane change intention recognition result for the target object. This method not only improves the accuracy of lane change intention recognition but also better adapts to different driving environments and vehicle behaviors, offering greater flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 Schematic diagram of the composition of a lane change intention recognition system provided by an exemplary embodiment of the present disclosure.

[0012] Figure 2 This is one of the flowcharts of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0013] Figure 3 This is the second flowchart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0014] Figure 4 This is the third flowchart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0015] Figure 5 This is the fourth flowchart of the lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0016] Figure 6 This is the fifth flowchart of the lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0017] Figure 7 This is the sixth flowchart of the lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0018] Figure 8 This is the seventh flowchart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0019] Figure 9This is the eighth flowchart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0020] Figure 10 This is the ninth flowchart of a method for identifying lane change intention provided by an exemplary embodiment of the present disclosure.

[0021] Figure 11 This is the tenth flowchart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure.

[0022] Figure 12 1 is a schematic diagram of the structure of a lane change intention recognition device provided by an exemplary embodiment of the present disclosure.

[0023] Figure 13 FIG. 4 is a structural diagram of an electronic device provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] To explain the present disclosure, example embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. It should be understood that the present disclosure is not limited to the example embodiments.

[0025] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0026] Application Overview

[0027] In recent years, intelligent driving has become a key trend in the automotive industry. Intelligent driving can be further categorized into assisted driving and automated driving. Assisted driving, considered the initial stage of intelligent driving, provides drivers with features such as adaptive cruise control and automated parking, helping to reduce driver fatigue and improve driving safety. Automated driving, considered the advanced stage of intelligent driving, enables the vehicle to autonomously perceive its surroundings, plan routes, and respond to various traffic conditions without a driver.

[0028] Lane change recognition technology is developing rapidly to improve the safety of intelligent driving. However, it still faces several challenges. For example, vehicle sensor data is frequently affected by noise and jitter, resulting in a high level of abnormal data and, consequently, low accuracy in lane change intention recognition based on this data. Furthermore, most related technologies rely on static thresholds to determine lane change intention. However, in real-world driving environments, vehicle behavior is constantly changing, and this static threshold approach struggles to adapt to these changing conditions. Therefore, accurately identifying a vehicle's lane change intention remains an urgent challenge.

[0029] Exemplary Systems

[0030] Figure 1 FIG. 1 is a schematic diagram of a lane change intention recognition system provided by an exemplary embodiment of the present disclosure. Figure 1 As shown, the lane change intention recognition system includes a collection device 101 , a lane change intention recognition device 102 and a decision device 103 .

[0031] The acquisition device 101 is used to collect driving data of a target object at multiple times. The acquisition device 101 may include one or more types of sensors. For example, the acquisition device 101 may include image sensors (such as vehicle-mounted cameras), radar, and lidar sensors for detecting the driving data of the target object. The vehicle-mounted camera is used to collect images of the target object at multiple times and can be installed on the vehicle's front windshield, rear windshield, rearview mirror, or all around the vehicle. The radar is used to collect the target object's position, speed, and direction of travel and can be installed on the vehicle's front bumper, rear bumper, or side. The lidar is used to collect the target object's position, shape, and motion state and can generally be installed on the vehicle's roof, front bumper, rear bumper, side, or near the interior rearview mirror. Of course, multiple data sources, such as collected image data, radar data, and lidar data, can also be fused to obtain the target object's driving data. Furthermore, the sensor can be one or multiple. For example, the vehicle-mounted camera can be a single forward-looking camera, dual forward-looking cameras, or a combination of a forward-looking camera and a surround-view camera. This application does not limit the type or number of acquisition devices.

[0032] The lane change intention recognition device 102 is used to obtain the driving data of the target object collected by the collection device 101 at multiple time points; then adjust the yaw rate in the driving data at each time point to obtain an adjusted yaw rate at each time point; then determine the yaw rate error at each time point based on the yaw rate at each time point; finally, determine the lane change intention recognition result of the target object based on the driving data at each time point, the adjusted yaw rate at each time point, and the yaw rate error at each time point, and send the lane change intention recognition result to the decision device 103.

[0033] The decision-making device 103 is used to receive the lane-changing intention recognition result of the target object sent by the lane-changing intention recognition device 102, and output a driving strategy according to the lane-changing intention recognition result of the target object.

[0034] The disclosed embodiments provide a lane change intention recognition system. The lane change intention recognition device in the system can collect driving data of a target object at multiple times through a collection device to determine an adjusted yaw rate and yaw rate error. The system then predicts a lane change intention recognition result for the target object based on the driving data at each time, the adjusted yaw rate, and the yaw rate error. The recognition result is then sent to a decision-making device, which then outputs a driving strategy. The system can determine the lane change intention recognition result based on dynamically changing driving data at multiple times. This processing method can better adapt to different driving environments and vehicle behaviors, has higher adaptability, and also makes the output lane change intention recognition result more accurate.

[0035] Exemplary Methods

[0036] Figure 2 This is a flow chart of a lane change intention recognition method provided by an exemplary embodiment of the present disclosure. This embodiment can be applied to electronic devices, such as Figure 2 As shown, the lane change intention recognition method includes the following steps:

[0037] Step 201: Determine the driving data of the target object at multiple moments.

[0038] The target object can be a dynamic object moving relative to the vehicle, such as another vehicle, another driving tool (e.g., a bicycle, an electric vehicle), a pedestrian, etc. The disclosed embodiments do not limit the type of target object, and the following embodiments are illustrative using another vehicle as the target object.

[0039] Driving data can include driving angle difference Yaw rate (ω i ), vehicle speed (v i ), acceleration (a i ) and other driving parameters.

[0040] In some examples, the process of determining the target object's driving data at multiple times may include: the vehicle uses a collection device to collect raw data of the target object at multiple times, and performs perception processing on the raw data to obtain the target object's driving data.

[0041] The acquisition device may include image sensors (such as on-board cameras), radars, lidars, and other sensors used to detect driving data of target objects. The raw data may include image data, radar data, and lidar data. These data can be used to determine the location of the target object, the relative speed between the vehicle and the target object, posture data, and motion data.

[0042] After obtaining the raw data, it can be sensed and processed to obtain the target object's driving data. Specifically, the data is acquired using an acquisition device such as an on-board camera, radar, or lidar, and transmitted to the vehicle's processing unit. The processing unit can then integrate and calibrate the acquired data using a perception algorithm to output the target object's position, the relative speed between the vehicle and the target object, attitude data, and motion data. The vehicle's position and attitude data can be used to calculate the target object's driving angle difference and yaw rate. The target object's speed and acceleration can be calculated using the vehicle's speed, acceleration, the target object's position, and the relative speed and motion information between the vehicle and the target object.

[0043] For example, a vehicle-mounted camera continuously captures images of the vehicle's surroundings (for example, capturing road markings, roadside buildings, the shapes and features of other vehicles, etc.) to obtain image data. Radar obtains radar data by emitting electromagnetic waves and receiving reflected waves. When electromagnetic waves encounter an object, they are reflected back. The radar calculates the distance, speed, and direction of the target object based on the time and frequency changes of the reflected waves. By continuously measuring the frequency changes of the reflected waves, the relative speed between the vehicle and the target object vehicle is determined. LiDAR uses a laser beam to scan the surrounding environment, which can provide very accurate distance measurement and object contour information. By rapidly rotating and emitting a laser beam, LiDAR data of the surrounding environment (for example, a three-dimensional point cloud map) can be generated. By analyzing the LiDAR data, the position and posture of the target object and the relative position and speed of the vehicle and the target object can be obtained.

[0044] For example, after acquiring data (including image data, radar data, and lidar data) using the aforementioned acquisition device, the data can be transmitted to the vehicle's processing unit. The processing unit can first analyze the image data using an image processing algorithm to determine the vehicle's position on the road. A fusion algorithm can then be used to integrate and calibrate the image data, radar data, and lidar data to determine the location of the target object, the relative speed between the vehicle and the target object, posture data, and motion data.

[0045] After obtaining the target object's driving data at multiple time points, the driving data can be preprocessed to improve the data quality of the driving data. In some examples, preprocessing the driving data can include filtering and denoising the driving data. Exemplarily, the filtering can include median filtering and mean filtering.

[0046] Because median filtering effectively eliminates impulse noise in driving data and prevents individual outliers from disrupting the overall driving data trend, mean filtering further smooths the driving data, reducing the impact of random noise and making the driving data fluctuations more stable and regular. By combining median and mean filtering, the impact of noise on driving data can be minimized, providing an accurate data foundation for subsequent lane change intention recognition.

[0047] Step 202 : adjusting the yaw rate in the driving data at each moment based on the driving data at multiple moments to obtain an adjusted yaw rate at each moment.

[0048] Yaw rate is the angular velocity of a vehicle's rotation about its vertical axis. It reflects the rate at which the vehicle's direction changes during driving. During normal driving, the vehicle's yaw rate is typically close to zero or fluctuates within a small range. When attempting to change lanes, the driver manipulates the steering wheel, causing the vehicle to begin turning, which can cause a significant change in the yaw rate. Therefore, after collecting driving data of a target object, the present disclosure can further process the yaw rate in the driving data to accurately derive the target object's lane change intention based on the yaw rate.

[0049] Illustratively, the present disclosure may further process the yaw rate in the driving data by adjusting the yaw rate corresponding to each of the multiple moments to obtain an adjusted yaw rate at each moment, and then using the adjusted yaw rate to determine the lane change intention of the target object.

[0050] Step 203: Determine the yaw rate error at each moment based on the yaw rate at each moment.

[0051] Since the acquisition device is inevitably affected by noise and vibration when collecting driving data of the target object, in order to further remove abnormal data caused by noise and vibration, the present disclosure can also determine the yaw rate error at each moment based on the yaw rate at each moment, so as to filter and reduce abnormal data and improve the signal-to-noise ratio of the data.

[0052] Step 204 : Determine a lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment.

[0053] After the adjusted yaw rate at each moment is determined in step 202 and the yaw rate error at each moment is determined in step 203, a lane change intention recognition result of the target object can be obtained based on information such as the yaw rate, adjusted yaw rate, and yaw rate error in the driving data at each moment.

[0054] The disclosed embodiments provide a method for lane change intention recognition. This method first acquires real-time driving data (e.g., angle difference, yaw rate, vehicle speed, and acceleration) of a target object at multiple moments in time. Based on the yaw rate in the driving data, it then obtains an adjusted yaw rate and yaw rate error. Finally, based on the driving data at each moment, the adjusted yaw rate and yaw rate error are used to determine the target object's lane change intention. This method further processes the yaw rate to better adapt to varying driving environments and vehicle changes, improving recognition accuracy and flexibility.

[0055] like Figure 3 As shown, in Figure 2 Based on the illustrated embodiment, the above step 202 of “adjusting the yaw rate in the driving data at each moment based on the driving data at multiple moments to obtain the adjusted yaw rate at each moment” may include the following steps 2021 to 2022:

[0056] Step 2021: Determine a dynamic adaptive factor based on the yaw rate, speed, and acceleration in the driving data at multiple moments.

[0057] In some examples, the dynamic adaptation factor F(t) satisfies the following expression:

[0058]

[0059] Among them, α, β and γ are weight coefficients. ω i is the yaw rate corresponding to the i-th frame, a i is the acceleration corresponding to the i-th frame, v i is the vehicle speed corresponding to the i-th frame.

[0060] For example, the driving data at multiple moments may include 25 frames of driving data. Using the yaw rate, acceleration, and vehicle speed in each frame of driving data and the aforementioned dynamic adaptive factor expression, the dynamic adaptive factor corresponding to each frame of driving data can be derived. For the aforementioned 25 frames of driving data, the dynamic adaptive factor corresponding to each frame of driving data is the same.

[0061] The above expression for the dynamic adaptive factor F(t) shows that it is determined based on the target object's real-time driving data (i.e., yaw rate, acceleration, and speed). Using the dynamic adaptive factor to adjust the yaw rate is equivalent to adjusting the target object's yaw rate based on its driving state. This adjusted yaw rate better reflects the target object's current driving environment and behavior, thus facilitating more accurate lane change intention recognition using the adjusted yaw rate.

[0062] Step 2022: Adjust the yaw rate at each moment based on the dynamic adaptive factor to obtain an adjusted yaw rate at each moment.

[0063] After the dynamic adaptive factor at each moment is obtained, the adjusted yaw rate at each moment can be determined using the dynamic adaptive factor at each moment and the yaw rate at each moment.

[0064] Illustratively, the process of determining the adjusted yaw rate at each moment using the dynamic adaptive factor at each moment and the yaw rate at each moment may be to multiply the dynamic adaptive factor at each moment and the yaw rate at each moment to obtain the adjusted yaw rate at each moment.

[0065] Among them, the adjusted yaw rate ω at each moment adjusted(t) Satisfies the following expression:

[0066] ω adjusted(t) =F(t)×ω i

[0067] It is understandable that the present disclosure exemplarily adopts a dynamic adaptive factor and a yaw rate to determine the adjusted yaw rate. A machine learning algorithm may also be adopted to determine an adjustment coefficient, and the yaw rate is adjusted using the adjustment coefficient. The present disclosure does not impose any restrictions on this, as long as it meets actual needs.

[0068] Because the driving environment and vehicle behavior are dynamic, the disclosed embodiments introduce a dynamic adaptive factor that dynamically changes with the target object's driving data to adjust the yaw rate, thereby generating an adjusted yaw rate. This adjusted yaw rate is more consistent with real-world driving data, and subsequent lane change intention recognition results determined using the adjusted yaw rate are more accurate.

[0069] like Figure 4 As shown, in Figure 2 Based on the embodiment shown, the above step 203 of “determining the yaw rate error at each moment based on the yaw rate at each moment” may include the following steps 2031 to 2032:

[0070] Step 2031 : Fitting the yaw rates at multiple moments to determine the predicted yaw rate at each moment.

[0071] In some examples, fitting processing is performed on the yaw rates at multiple moments in time in order to filter and reduce data noise, thereby effectively identifying and eliminating jitter data and improving the data signal-to-noise ratio.

[0072] Step 2032: Determine the yaw rate error at each moment based on the yaw rate at each moment and the predicted yaw rate at each moment.

[0073] After fitting the yaw rate at each moment to obtain the predicted yaw rate at each moment, the yaw rate at each moment can be compared with the predicted yaw rate to obtain the yaw rate error at each moment.

[0074] The disclosed embodiment performs fitting processing on the yaw rate to eliminate jittery data and improve the data signal-to-noise ratio, so as to facilitate more accurate prediction of the target object's lane change intention and reduce misjudgments and missed judgments.

[0075] like Figure 5 As shown, in Figure 4 Based on the illustrated embodiment, the above step 2031 of “performing fitting processing on the yaw rates at multiple moments to determine the predicted yaw rate at each moment” may include the following steps 20311 to 20312:

[0076] Step 20311: Determine fitting parameters of the yaw rate fitting function based on the yaw rate at each moment.

[0077] In some examples, the yaw rate fitting function is a function of how the yaw rate changes over time. The yaw rate fitting function is used to determine a predicted yaw rate at each moment.

[0078] Step 20312: Determine the predicted yaw rate at each moment based on the yaw rate fitting function.

[0079] After obtaining the yaw rate at each moment, various methods can be used to determine the predicted yaw rate at each moment. For example, sliding window filtering and polynomial fitting can be used to determine the predicted yaw rate at each moment. Of course, other methods can also be used to determine the yaw rate error at each moment, and this disclosure is not limited thereto.

[0080] In some examples, a process for determining a predicted yaw rate at each moment using sliding window filtering and polynomial fitting may include: first, setting a yaw rate fitting function; then, determining fitting parameters of the yaw rate fitting function based on the sliding window; and finally, determining the predicted yaw rate at each moment based on the yaw rate fitting function and the fitting parameters in the yaw rate fitting function.

[0081] According to the current driving environment and vehicle behavior, the yaw rate fitting function may be set to a polynomial function, and of course may also be set to other types of functions, which is not limited in the present disclosure.

[0082] For example, the yaw rate fitting function ω fit (t) satisfies the following expression:

[0083] ω fit (t) = c0 + c1t + c2t 2 +c i t i +…+cn t n

[0084] Among them, c0, c1, c2, c i and c n is the fitting parameter.

[0085] After the yaw rate fitting function is set, fitting parameters of the yaw rate fitting function may be determined based on driving data at multiple moments.

[0086] Illustratively, the process of determining fitting parameters of a yaw rate fitting function based on driving data at multiple moments may include: setting a sliding parameter of a sliding window; dividing the driving data at multiple moments into at least one sliding window according to the sliding parameter of the sliding window; and determining the fitting parameters of the yaw rate fitting function based on the driving data within each sliding window.

[0087] The sliding parameters of the sliding window may include the size of the sliding window and the step size of the sliding window. For example, the size of the sliding window is set to M, that is, each sliding window includes M data points, and the data points in a sliding window are: (t1, ω1), (t2, ω2), ..., (t M ,ω M ). For example, M = 5. The step size of the sliding window is N, for example, N = 1.

[0088] For example, in conjunction with step 2021, the driving data at multiple moments may include 25 frames of driving data. When the sliding window size M=5 and the sliding window step size N=1, the 25 frames of driving data may be divided into 21 sliding windows, each of which includes 5 frames of driving data.

[0089] Among them, the data points in the first sliding window are: (t1,ω1), (t2,ω2),…, (t5,ω5); the data points in the second sliding window are: (t2,ω2), (t3,ω3),…, (t6,ω6); …; the data points in the 20th sliding window are: (t 20 ,ω 20 ),(t 21 ,ω 21 ),…,(t 24 ,ω 24 ); the data points in the 21st sliding window are: (t 21 ,ω 21 ),(t 22 ,ω 22 ),…,(t 25 ,ω 25 ).

[0090] Since the step size N = 1 of the sliding window is smaller than the size of the sliding window, the same data point exists in different sliding windows. For example, (t2, ω2) appears in both the first and second sliding windows. (t3, ω3) appears in the first, second, and third sliding windows, and so on.

[0091] Determining fitting parameters of the yaw rate fitting function based on the driving data within each sliding window may include performing polynomial fitting on the driving data within each sliding window to obtain fitting parameters of the yaw rate fitting function. It will be appreciated that, because the data points within each sliding window are different, the fitting parameters of the yaw rate fitting function determined for each sliding window are also different.

[0092] For example, after determining the data points in each sliding window, the least squares method can be used to substitute the data points in each sliding window into the yaw rate fitting function to determine the fitting parameters (c i ). Then substitute the fitting parameters into the yaw rate fitting function to obtain the predicted yaw rate at each moment.

[0093] After obtaining the data points in the above 21 sliding windows, the yaw rate fitting function ω can be determined by using the data points in a sliding window and the least squares method. fit (t) corresponds to a set of fitting parameters. For the 21 sliding windows mentioned above, 21 sets of fitting parameters can be determined. The fitting parameters corresponding to the data points in each sliding window are consistent. For example, for the data points in the first sliding window, the fitting parameters corresponding to (t1, ω1), (t2, ω2), …, (t5, ω5) are consistent.

[0094] Since the same data point exists in different sliding windows, one data point may correspond to multiple sets of fitting parameters. For example, (t2, ω2) is included in both the first and second sliding windows. Based on the first sliding window, one set of fitting parameters corresponding to (t2, ω2) can be determined, and based on the second sliding window, another set of fitting parameters corresponding to (t2, ω2) can be determined.

[0095] In this case, the multiple sets of fitting parameters corresponding to the data point can be averaged to obtain the processed fitting parameters corresponding to the data point. That is, the set of fitting parameters corresponding to the data point (t2, ω2) in the first sliding window and the other set of fitting parameters corresponding to the data point (t2, ω2) in the second sliding window are averaged to obtain the processed fitting parameters corresponding to (t2, ω2). The predicted yaw rate corresponding to the data point can then be obtained based on the processed fitting parameters corresponding to (t2, ω2) and the yaw rate fitting function. The processed fitting parameters and predicted yaw rate for other data points can also be obtained by referring to this method, which will not be repeated here.

[0096] The disclosed embodiments filter and reduce data noise by fitting a polynomial to the data points within a sliding window, and then use the predicted yaw rate to calculate the fitting error. Compared to simple filtering or moving average methods, this processing method is more efficient and can more accurately capture data trends and patterns.

[0097] like Figure 6 As shown, in Figure 5 Based on the illustrated embodiment, the above step 2032 of “determining the yaw rate error at each moment based on the adjusted yaw rate at each moment and the predicted yaw rate at each moment” may include the following steps 20321 to 20322:

[0098] Step 20321: Determine the difference between the yaw rate at each moment and the predicted yaw rate corresponding to each moment.

[0099] After determining the yaw rate at each moment and the corresponding predicted yaw rate at each moment, the difference between the yaw rate at each moment and the corresponding predicted yaw rate can be calculated to obtain the difference between the yaw rate at each moment and the corresponding predicted yaw rate at each moment.

[0100] Step 20322: Determine the yaw rate error at each moment based on the difference corresponding to each moment.

[0101] After the difference corresponding to each moment is determined, the yaw rate error at each moment may be determined based on the difference.

[0102] For example, the process of determining the yaw rate error at each moment based on the difference can be to determine the square of the difference between the yaw rate corresponding to each data point in each sliding window and the predicted yaw rate based on the difference between the yaw rate and the predicted yaw rate for each data point in each sliding window; and to sum the square of the difference for each data point in each sliding window to obtain the yaw rate error for each sliding window. The yaw rate error corresponding to each data point in each sliding window is the same.

[0103] Yaw rate error E fit (t) satisfies the following expression:

[0104]

[0105] Where M is the size of the sliding window, ω i is the yaw rate of the i-th frame, ω fit (t) is the predicted yaw rate of the i-th frame.

[0106] Combined with the above expression of yaw rate error, it can be seen that the yaw rate error is calculated based on a sliding window. According to the aforementioned step 20312, the same data point exists in different sliding windows. Then, when a data point is in different sliding windows, it may correspond to multiple yaw rate errors. In this case, the multiple yaw rate errors corresponding to the data point can be averaged to obtain the processed yaw rate error corresponding to the data point. For example, a yaw rate error corresponding to the data point (t2, ω2) in the first sliding window and another yaw rate error corresponding to the data point (t2, ω2) in the second sliding window are averaged to obtain the processed yaw rate error corresponding to (t2, ω2). The processed yaw rate error of other data points can also be obtained by referring to this method, which will not be repeated here.

[0107] It is understandable that after obtaining the yaw rate at each moment and the predicted yaw rate corresponding to each moment, the yaw rate error at each moment can also be determined based on the mean square error of the yaw rate at each moment and the predicted yaw rate corresponding to each moment. The present disclosure does not impose any limitation on this.

[0108] In the embodiment of the present disclosure, after obtaining the yaw rate and the predicted yaw rate at each moment, the yaw rate error at each moment can be further obtained. Based on the yaw rate error, abnormal data in the driving data can be eliminated, thereby improving the accuracy of recognition.

[0109] like Figure 7 As shown, in Figure 2 Based on the illustrated embodiment, the above step 204 of "determining a lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment" may include the following steps 2041 to 2042:

[0110] Step 2041 : Determine a first lane change state based on the driving angle difference, the adjusted yaw rate, and the yaw rate error in the driving data at each moment.

[0111] In the present disclosure, determining whether a target object has changed lanes can be done from two perspectives: first, based on the target object's current dynamic driving trajectory. Second, based on the target object's historical dynamic driving trajectory. If both the first and second perspectives determine that the target object has a lane-changing tendency, the target object is considered to have a lane-changing intention.

[0112] The first aspect can be determined by the first lane changing state, and the second aspect can be determined by the second lane changing state.

[0113] Since the current dynamic driving trajectory of the target object is related to the yaw rate and the driving angle difference at each moment, after determining the adjusted yaw rate and yaw rate error at each moment, the first lane change state can be determined based on the adjusted yaw rate and yaw rate error at each moment, combined with the target object's driving angle difference. The driving angle difference refers to the angle of the target object's driving direction relative to the coordinate axis direction of the ego vehicle coordinate system. The driving angle difference can be calculated by the difference between the angle of the target object's driving direction relative to the coordinate axis direction of the world coordinate system and the angle of the ego vehicle's driving direction relative to the coordinate axis direction of the world coordinate system. For example, the coordinate axis direction of the ego vehicle coordinate system is the X-axis direction, and the coordinate axis direction of the world coordinate system is also the X-axis direction.

[0114] The vehicle coordinate system is based on the vehicle's center point (or the center point of the vehicle's rear axle) as its origin. The X-axis is parallel to the vehicle's direction of travel, the Y-axis is perpendicular to the vehicle's direction of travel, and the Z-axis is perpendicular to the ground. The positive direction of the X-axis is toward the front of the vehicle, the positive direction of the Y-axis is toward the right, and the positive direction of the Z-axis is toward the top.

[0115] The world coordinate system is a global coordinate system fixed to the ground and used to describe the vehicle's position in the environment. The world coordinate system's X-axis is parallel to the direction of the road, the Y-axis is perpendicular to the direction of the road, and the Z-axis is perpendicular to the ground.

[0116] Step 2042: Determine a second lane change state based on the driving data at each moment and the adjusted yaw rate at each moment.

[0117] The historical dynamic driving trajectory of the target object is related to the yaw rate and the driving data at each moment. Therefore, after the adjusted yaw rate at each moment is obtained, the second lane change state can be determined based on the adjusted yaw rate at each moment and the driving data at each moment.

[0118] Step 2043: Determine a lane change intention recognition result of the target object based on the first lane change state and the second lane change state.

[0119] When the first lane-changing state is a lane-changing trend and the second lane-changing state is also a lane-changing trend, the lane-changing intention recognition result of the target object is that the target object has a lane-changing trend, and the driving strategy of the vehicle can be outputted subsequently based on the lane-changing intention recognition result.

[0120] The embodiment provided by the present disclosure determines whether the target object has the intention to change lanes through data from two aspects (i.e., the current dynamic behavior trajectory of the target object and the historical dynamic behavior trajectory). Compared with the related technology using static thresholds, this method will be more accurate and more in line with the driving habits of the target object.

[0121] like Figure 8 As shown, in Figure 7 Based on the illustrated embodiment, the above step 2041 “determining the first lane change state based on the driving angle difference, the adjusted yaw rate, and the yaw rate error in the driving data at each moment” may include the following steps 20411 to 20413:

[0122] Step 20411: Determine the dynamic lane change angle threshold and the crossing angle threshold at each moment based on the adjusted yaw rate and the yaw rate error at each moment.

[0123] After the adjusted yaw rate and yaw rate error at each moment are obtained, the dynamic lane change angle threshold and the crossing angle threshold at each moment may be determined based on the adjusted yaw rate and yaw rate error at each moment.

[0124] The lane change angle refers to the angular difference between the target vehicle's actual driving direction and the original lane direction during the lane change process. The crossing angle refers to the angular difference between the vehicle's driving direction and the target lane direction during the process of crossing from the original lane to the target lane.

[0125] The lane change angle directly affects the collision risk between the ego vehicle and the target object. Excessively large lane change angles increase the likelihood of a collision, preventing the ego vehicle from changing lanes. A suitable crossing angle helps maintain vehicle stability during lane changes. However, excessive crossing angles can cause a dramatic shift in the vehicle's center of gravity, compromising the control and stability of the target object.

[0126] Exemplary dynamic lane change angle threshold Satisfies the following expression:

[0127]

[0128] Crossing angle threshold Satisfies the following expression:

[0129]

[0130] Among them, k1, k2, k3, k4 are preset thresholds, ω adjusted (t) is the adjusted yaw rate, E fit (t) is the yaw rate error.

[0131] When the target object is in a state of continuous motion, the driving data collected at each moment is also constantly changing. Therefore, the set dynamic lane change angle threshold and crossing angle threshold also change dynamically over time.

[0132] Step 20412: Determine the number of valid moments among the multiple moments based on the driving angle difference at each moment, the dynamic lane change angle threshold and the crossing angle threshold at each moment.

[0133] After collecting driving data at multiple moments, the effective moment can be determined from the multiple moments based on the driving angle difference at each moment, the dynamic lane change angle threshold and the crossing angle threshold at each moment, so as to facilitate the subsequent further deriving of the first lane change state based on the data at the effective moment.

[0134] Generally, a small driving angle difference indicates that the vehicle's direction of travel has not changed significantly, making a lane change unlikely. A large driving angle difference indicates that the vehicle is making a significant turn, making a lane change more likely. Therefore, the present disclosure utilizes the driving angle difference, the dynamic lane change angle threshold, and the crossing angle threshold to accurately derive the target object's lane change intention.

[0135] Step 20413: In response to the number of valid moments being greater than a number threshold, determining that the first lane-changing state has a lane-changing tendency.

[0136] In some examples, when the number of valid moments (n valid ) after that, you can valid With quantity threshold A comparison is performed to determine the first lane change state.

[0137] Specifically, when the number of valid moments is greater than the number threshold, that is, The first lane-changing state is considered to have a lane-changing tendency. It is considered that the first lane-changing state has no lane-changing tendency.

[0138] The present disclosure does not limit the order of determining the first lane change state and the second lane change state. The first lane change state may be determined first, and then the second lane change state; the second lane change state may be determined first, and then the first lane change state, or the first lane change state and the second lane change state may be determined simultaneously. In combination with step 2043, it can be seen that only when the first lane change state is a lane change trend and the second lane change state is a lane change trend, the lane change intention recognition result of the target object is a lane change trend. Therefore, when the first lane change state or the second lane change state is not a lane change trend, there is no need to check the result of the other lane change state, and the lane change intention recognition result of the target object can be directly determined as not having a lane change trend. When the first lane change state or the second lane change state is a lane change trend, it is necessary to continue to check the result of the other lane change state to finally output the lane change intention recognition result of the target object.

[0139] As can be seen from the foregoing, the lane change angle and the crossing angle have a significant impact on whether a lane change can be safely made. Therefore, the embodiment of the present application sets dynamic lane change angle thresholds and crossing angle thresholds based on adjusting the yaw rate and yaw rate error. The set dynamic lane change angle thresholds and crossing angle thresholds are used to determine the effective moments. Finally, based on the number of effective moments, the lane change intention of the target object can be accurately derived.

[0140] like Figure 9 As shown, in Figure 8 Based on the illustrated embodiment, the above step 20412 "determining the number of valid moments in the multiple moments based on the driving angle difference at each moment, the dynamic lane change angle threshold at each moment, and the crossing angle threshold at each moment" may include the following steps 204121 to 204123:

[0141] Step 204121, determining a first size relationship between the driving angle difference at each moment and the dynamic lane change angle threshold, and a second size relationship between the driving angle difference at each moment and the crossing angle threshold.

[0142] After determining the dynamic lane change angle threshold and the crossing angle threshold at each time, the driving angle difference at each time can be compared with the dynamic lane change angle threshold to determine a first magnitude relationship between the driving angle difference at each time and the dynamic lane change angle threshold. Furthermore, the driving angle difference at each time can be compared with the crossing angle threshold to determine a second magnitude relationship between the driving angle difference at each time and the crossing angle threshold. The effective time can then be determined based on the first magnitude relationship and the second magnitude relationship.

[0143] In some examples, the first size relationship between the driving angle difference at each moment and the dynamic lane change angle threshold includes: the driving angle difference at each moment is greater than the dynamic lane change angle threshold, and the driving angle difference at each moment is less than or equal to the dynamic lane change angle threshold.

[0144] The second magnitude relationship between the driving angle difference at each moment and the crossing angle threshold includes: the driving angle difference at each moment is greater than or equal to the crossing angle threshold, and the driving angle difference at each moment is less than the crossing angle threshold.

[0145] Step 204122, in response to the first size relationship corresponding to each moment being that the driving angle difference is greater than the dynamic lane change angle threshold, and the second size relationship being that the driving angle difference is less than the crossing angle threshold, determining the moment as a valid moment.

[0146] In some examples, the effective moment is the moment that meets the preset conditions. For example, the preset conditions are that the driving angle difference is greater than the dynamic lane change angle threshold and the driving angle difference is less than the crossing angle threshold. Satisfies the following expression:

[0147]

[0148] In combination with step 204121, a first magnitude relationship between the driving angle difference at each moment and the dynamic lane change angle threshold, as well as a second magnitude relationship between the driving angle difference at each moment and the crossing angle threshold, can be determined. By combining the above-described preset conditions, the first magnitude relationship, and the second magnitude relationship, a valid moment among the multiple moments can be determined.

[0149] Step 204123, determine the number of valid moments in the multiple moments.

[0150] In combination with step 204121 and step 204122, valid moments can be determined from multiple moments, and then the valid moments are counted to determine the number of valid moments.

[0151] Combined with step 20411, it can be seen that the lane change angle and the crossing angle can reflect to a certain extent whether the lane change can be completed safely. Therefore, the embodiment of the present application uses the dynamic lane change angle threshold and the crossing angle threshold to further determine the effective moments with lane change trends, and then the first lane change state can be accurately output based on the number of effective moments.

[0152] like Figure 10 As shown, in Figure 7 Based on the illustrated embodiment, the above step 2042 “determining the second lane change state based on the driving data at each moment and the adjusted yaw rate at each moment” may include the following steps 20421 to 20422:

[0153] Step 20421: Determine an average yaw rate corresponding to a preset time window based on the adjusted yaw rate at each moment.

[0154] In some examples, after determining the adjusted yaw rate at each of the multiple moments, a preset time window may be set, and the lane change intention of the target object may be further determined based on the yaw rate data within the preset time window.

[0155] Specifically, the lane-changing intention of the target object can be determined by the average yaw rate within the preset time window. Of course, other yaw rate data within the preset time window can also be used for determination, and the present disclosure does not limit this.

[0156] For example, the average yaw rate ω mean (t) satisfies the following expression:

[0157]

[0158] For example, the length of the preset time window is T, T=15, that is, one preset time window includes 15 frames of driving data, and the average yaw rate can be obtained based on the adjusted yaw rate of each frame in the 15 frames of driving data.

[0159] Generally speaking, if the average yaw rate is low and relatively stable, it indicates that the target object has maintained a relatively straight driving state for a long time and is unlikely to change lanes. This is because when driving in a straight line, the vehicle's direction changes less, and the yaw rate will also remain low. When the average yaw rate begins to increase and exceeds a certain threshold, it indicates that the target object is turning and may change lanes. If the average yaw rate is continuously high, it indicates that the target object's driving direction has changed significantly and continuously. Therefore, using the average yaw rate can output a more accurate second lane change status.

[0160] Step 20422: Determine a second lane change state based on the average yaw rate and the driving angle difference in the driving data at the last moment within the preset time window.

[0161] After determining the average yaw rate, a second lane change state can be determined based on the average yaw rate and a driving angle difference in driving data at the last moment within a preset time window. The second lane change state includes whether the target object has a lane change tendency or does not have a lane change tendency.

[0162] This embodiment of the present application uses the average yaw rate within a preset time window and the driving angle difference at the last moment within the preset time window to determine the second lane change state. Because the average yaw rate is derived from the yaw rate over a period of time, this assessment method better reflects the target vehicle's real-time driving state. Compared to static threshold assessment methods, this method better reflects real-world driving behavior and is more accurate.

[0163] like Figure 11 As shown, in Figure 10Based on the illustrated embodiment, the above step 20422 “determining the second lane change state based on the average yaw rate and the driving angle difference in the driving data at the last moment within the preset time window” may include the following steps 204221 to 204222:

[0164] Step 204221: Determine a third magnitude relationship between the average yaw rate and the yaw rate threshold, and a fourth magnitude relationship between the driving angle difference in the driving data at the last moment in the preset time window and the angle difference threshold.

[0165] In some examples, a yaw rate threshold may be pre-set. and the angle difference threshold

[0166] After determining the average yaw rate within the preset time window, the average yaw rate can be compared with a yaw rate threshold to determine a third magnitude relationship between the average yaw rate and the yaw rate threshold. Furthermore, the driving angle difference in the driving data at the last moment within the preset time window can be compared with the angle difference threshold to determine a fourth magnitude relationship between the driving angle difference at the last moment and the angle difference threshold. Subsequently, the second lane change state can be determined based on the third and fourth magnitude relationships.

[0167] In some examples, the third magnitude relationship between the average yaw rate and the yaw rate threshold includes: the average yaw rate is greater than the yaw rate threshold, i.e. And the average yaw rate is less than or equal to the yaw rate threshold, that is

[0168] The fourth magnitude relationship between the driving angle difference at the last moment and the angle difference threshold includes: the driving angle difference at the last moment is greater than the angle difference threshold, that is, And the driving angle difference at the last moment is less than or equal to the angle difference threshold, that is

[0169] Step 204222: In response to the third magnitude relationship being that the average yaw rate is greater than the yaw rate threshold, and the fourth magnitude relationship being that the driving angle difference in the driving data at the last moment in the preset time window is greater than the angle difference threshold, determining that the second lane change state has a lane change tendency.

[0170] In combination with step 204221, a third magnitude relationship between the average yaw rate and the yaw rate threshold, and a fourth magnitude relationship between the driving angle difference in the driving data at the last moment in the preset time window and the angle difference threshold can be determined. If the third magnitude relationship indicates that the average yaw rate is greater than the yaw rate threshold, and the fourth magnitude relationship indicates that the driving angle difference in the driving data at the last moment in the preset time window is greater than the angle difference threshold, then the second lane change state can be determined as having a lane change tendency.

[0171] As can be seen from step 20421, the average yaw rate can reflect the lane change intention of the target object to a certain extent. Therefore, after obtaining the average yaw rate, the embodiment of the present application compares the average yaw rate with the yaw rate threshold, and compares the driving angle difference with the angle difference threshold, thereby obtaining an accurate second lane change state.

[0172] Exemplary devices

[0173] Based on the foregoing embodiments, the embodiments of the present disclosure provide a lane change intention recognition device. The modules included in the device, and the units included in each module, can be implemented by a processor in a computer device; of course, they can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0174] Figure 12 FIG. 1 is a schematic diagram of the structure of a lane change intention recognition device provided by an exemplary embodiment of the present disclosure. Figure 12 As shown, the lane change intention recognition device 1200 includes:

[0175] The first determining module 1201 is configured to determine the driving data of the target object at multiple moments.

[0176] The second determining module 1202 is configured to adjust the yaw rate in the driving data at each moment based on the driving data at multiple moments to obtain an adjusted yaw rate at each moment.

[0177] The third determining module 1203 is configured to determine a yaw rate error at each moment based on the yaw rate at each moment.

[0178] The fourth determining module 1204 is configured to determine a lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment.

[0179] In some embodiments, the second determination module 1202 is further configured to determine a dynamic adaptive factor based on the yaw rate, speed, and acceleration in the driving data at multiple moments; and adjust the yaw rate at each moment based on the dynamic adaptive factor to obtain an adjusted yaw rate at each moment.

[0180] In some embodiments, the third determination module 1203 is further configured to perform fitting processing on the yaw rates at multiple moments to determine a predicted yaw rate at each moment; and determine a yaw rate error at each moment based on the yaw rate at each moment and the predicted yaw rate at each moment.

[0181] In some embodiments, the third determining module 1203 is further configured to determine fitting parameters of a yaw rate fitting function based on the yaw rate at each moment; and determine a predicted yaw rate at each moment based on the yaw rate fitting function.

[0182] In some embodiments, the third determining module 1203 is further configured to determine a difference between the yaw rate at each moment and the predicted yaw rate corresponding to each moment; and determine a yaw rate error at each moment based on the difference corresponding to each moment.

[0183] In some embodiments, the fourth determination module 1204 is further configured to determine a first lane change state based on the driving angle difference, the adjusted yaw rate, and the yaw rate error in the driving data at each moment; determine a second lane change state based on the driving data and the adjusted yaw rate at each moment; and determine a lane change intention recognition result of the target object based on the first lane change state and the second lane change state.

[0184] In some embodiments, the fourth determination module 1204 is further configured to determine a dynamic lane change angle threshold and a crossing angle threshold at each moment based on the adjusted yaw rate and the yaw rate error at each moment; determine the number of valid moments among the multiple moments based on the driving angle difference at each moment, the dynamic lane change angle threshold, and the crossing angle threshold at each moment; and determine, in response to the number of valid moments being greater than the number threshold, that the first lane change state has a lane change trend.

[0185] In some embodiments, the fourth determination module 1204 is further used to determine a first size relationship between the driving angle difference at each moment and the dynamic lane change angle threshold, and a second size relationship between the driving angle difference at each moment and the crossing angle threshold; in response to the first size relationship corresponding to each moment being that the driving angle difference is greater than the dynamic lane change angle threshold, and the second size relationship being that the driving angle difference is less than the crossing angle threshold, determining the moment as a valid moment; and determining the number of valid moments among multiple moments.

[0186] In some embodiments, the fourth determination module 1204 is further configured to determine an average yaw rate corresponding to a preset time window based on the adjusted yaw rate at each moment; and determine a second lane change state based on a difference in driving angle between the average yaw rate and the driving data at the last moment in the preset time window.

[0187] In some embodiments, the fourth determining module 1204 is further configured to determine a third magnitude relationship between the average yaw rate and the yaw rate threshold, and a fourth magnitude relationship between the driving angle difference in the driving data at the last moment in the preset time window and the angle difference threshold; and in response to the third magnitude relationship being that the average yaw rate is greater than the yaw rate threshold, and the fourth magnitude relationship being that the driving angle difference in the driving data at the last moment in the preset time window is greater than the angle difference threshold, determining that the second lane change state has a lane change trend.

[0188] It should be noted that the description of the above exemplary device embodiments is similar to the description of the above method embodiments and has the same beneficial effects as the corresponding exemplary method embodiments. For technical details and corresponding beneficial technical effects not disclosed in the exemplary device embodiments of the present disclosure, those skilled in the art should refer to the description of the exemplary method embodiments of the present disclosure for understanding, and will not be repeated here.

[0189] Exemplary electronic devices

[0190] Figure 13 A structural diagram of an electronic device provided by an exemplary embodiment of the present disclosure is shown in FIG. Figure 13 As shown, the electronic device 1300 includes at least one processor 1301 and a memory 1302 .

[0191] The processor 1301 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 1300 to perform desired functions.

[0192] The memory 1302 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1301 may execute one or more computer program instructions to implement the traffic sign posture detection method and / or other desired functions of the various embodiments of the present disclosure described above.

[0193] In one example, the electronic device 1300 may further include an input device 1303 and an output device 1304 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0194] The input device 1303 may also include, for example, a keyboard, a mouse, and the like.

[0195] The output device 1304 can output various information to the outside, and may include, for example, a display, a speaker, a printer, a communication network and its connected remote output devices, etc.

[0196] Of course, to simplify, Figure 13 Only some of the components related to the present disclosure in the electronic device 1300 are shown, and components such as buses, input / output interfaces, etc. are omitted. In addition, according to specific application scenarios, the electronic device 1300 may further include any other appropriate components.

[0197] Exemplary computer program products and computer-readable storage media

[0198] In addition to the above-mentioned methods and devices, embodiments of the present disclosure may also provide a computer program product, including computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method for detecting the posture of traffic signs in each embodiment of the present disclosure described in the above-mentioned "Exemplary Method" section.

[0199] The computer program product may be written in any combination of one or more programming languages ​​to implement the operations of the disclosed embodiments, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0200] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor executes the steps in the method for detecting the posture of traffic signs in each embodiment of the present disclosure described in the above-mentioned "Exemplary Method" section.

[0201] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium is, for example, but not limited to, a system, device or component comprising electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0202] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and not restrictive, and should not be considered as essential to each embodiment of the present disclosure. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present disclosure to necessarily being implemented using these specific details.

[0203] Those skilled in the art may make various changes and modifications to the present disclosure without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is intended to include these modifications and variations.

Claims

1. A method for identifying lane change intention, comprising: Determine the target object's driving data at multiple times; adjusting the yaw rate in the driving data at each moment based on the driving data at the multiple moments to obtain an adjusted yaw rate at each moment; determining a yaw rate error at each moment based on the yaw rate at each moment; determining a lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment; The adjusting the yaw rate in the driving data at each moment based on the driving data at the multiple moments to obtain the adjusted yaw rate at each moment includes: determining a dynamic adaptive factor based on the yaw rate, speed, and acceleration in the driving data at the plurality of moments; The yaw rate at each moment is adjusted based on the dynamic adaptive factor to obtain an adjusted yaw rate at each moment.

2. The method according to claim 1, wherein The determining the yaw rate error at each moment based on the yaw rate at each moment includes: performing fitting processing on the yaw rates at the multiple moments to determine a predicted yaw rate at each moment; A yaw rate error at each moment is determined based on the yaw rate at each moment and the predicted yaw rate at each moment.

3. The method according to claim 2, wherein: The fitting process of the yaw rates at the plurality of moments to determine the predicted yaw rate at each moment includes: determining fitting parameters of a yaw rate fitting function based on the yaw rates at each moment; The predicted yaw rate at each moment is determined based on the yaw rate fitting function.

4. The method according to claim 3, wherein: The determining the yaw rate error at each moment based on the adjusted yaw rate at each moment and the predicted yaw rate at each moment includes: Determining a difference between the yaw rate at each moment and a predicted yaw rate corresponding to each moment; The yaw rate error at each moment is determined based on the difference corresponding to each moment.

5. The method according to any one of claims 1 to 4, wherein The determining the lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment includes: determining a first lane change state based on the driving angle difference in the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment; determining a second lane change state based on the driving data at each time and the adjusted yaw rate at each time; A lane change intention recognition result of the target object is determined based on the first lane change state and the second lane change state.

6. The method according to claim 5, wherein: The determining the first lane change state based on the driving angle difference in the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment includes: determining a dynamic lane change angle threshold and a crossing angle threshold at each moment based on the adjusted yaw rate at each moment and the yaw rate error at each moment; determining the number of valid moments in the plurality of moments based on the driving angle difference at each moment, the dynamic lane change angle threshold and the crossing angle threshold at each moment; In response to the number of the valid moments being greater than a number threshold, it is determined that the first lane change state has a lane change tendency.

7. The method according to claim 6, wherein: The determining the number of valid moments in the multiple moments based on the driving angle difference at each moment, the dynamic lane change angle threshold and the crossing angle threshold at each moment includes: Determining a first magnitude relationship between the driving angle difference at each moment and the dynamic lane change angle threshold, and a second magnitude relationship between the driving angle difference at each moment and the crossing angle threshold; In response to the first magnitude relationship corresponding to each of the moments being that the driving angle difference is greater than the dynamic lane change angle threshold, and the second magnitude relationship being that the driving angle difference is less than the crossing angle threshold, determining the moment as a valid moment; The number of the valid moments in the plurality of moments is determined.

8. The method according to claim 5, wherein The determining the second lane change state based on the driving data at each moment and the adjusted yaw rate at each moment includes: Determining an average yaw rate corresponding to a preset time window based on the adjusted yaw rates at each moment; The second lane change state is determined based on the average yaw rate and a driving angle difference in the driving data at a last moment in the preset time window.

9. The method according to claim 8, wherein The determining the second lane change state based on the average yaw rate and the driving angle difference in the driving data at the last moment in the preset time window includes: Determining a third magnitude relationship between the average yaw rate and the yaw rate threshold, and a fourth magnitude relationship between the driving angle difference in the driving data at the last moment in the preset time window and the angle difference threshold; In response to the third magnitude relationship being that the average yaw rate is greater than the yaw rate threshold, and the fourth magnitude relationship being that the driving angle difference in the driving data at the last moment in the preset time window is greater than the angle difference threshold, it is determined that the second lane change state has a lane change trend.

10. A lane change intention recognition device, comprising: A first determining module is used to determine the driving data of the target object at multiple moments; a second determining module, configured to determine a dynamic adaptive factor based on the yaw rate, speed, and acceleration in the driving data at the plurality of moments; Adjusting the yaw rate at each moment based on the dynamic adaptive factor to obtain the adjusted yaw rate at each moment; a third determining module, configured to determine a yaw rate error at each moment based on the yaw rate in the driving data at each moment; The fourth determination module is configured to determine a lane change intention recognition result of the target object based on the driving data at each moment, the adjusted yaw rate at each moment, and the yaw rate error at each moment.

11. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the lane change intention recognition method according to any one of claims 1 to 9.

12. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the lane change intention recognition method described in any one of claims 1 to 9.

Images