Vehicle assistance system, method, vehicle, and computer-readable storage medium

Abstract

A vehicle assistance system, method, vehicle and computer-readable storage medium are provided. The system includes: a data acquisition unit configured to detect the relative position of a target vehicle relative to a current vehicle, and when it is detected that the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle, data related to the target vehicle is collected in a plurality of consecutive preset time slots until the data acquisition unit detects that the target vehicle appears at the side of the current vehicle at a second preset angle relative to the current vehicle; a data processing unit configured to generate a matrix set by using the data collected in each preset time slot, and output the matrix set at the end of each preset time slot; and a calculation unit configured to calculate the possibility of the target vehicle merging by inputting each output matrix set into an artificial intelligence model. The solution of the present invention can improve the driving safety of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of vehicles, and more specifically, to a vehicle assistance system for calculating the possibility of a target vehicle merging with a current vehicle, a vehicle comprising the vehicle assistance system, a corresponding vehicle assistance method, and a computer-readable storage medium. Background Art

[0002] In an urban high-speed traffic environment, the current vehicle needs to timely and accurately predict the lane-changing behavior of the target vehicle (i.e., the vehicle traveling behind the current vehicle) in order to detect potential risks in a timely manner, and then adjust the driving state of the current vehicle accordingly in advance, thereby improving the driving safety of the current vehicle. Currently, some vehicle lane-changing detection technologies are based on collision volume detection, but the volume of the vehicle is usually difficult to accurately detect through cameras or other sensors. In addition, Time To Collision (TTC) is an important factor for evaluating lane-changing scenarios, but the definition of TTC cannot meet the new demand for critical assessment of lane-changing.

[0003] Therefore, there is a need for a new vehicle assistance technology that can timely and accurately predict the lane-merging behavior of the target vehicle. Summary of the invention

[0004] To this end, the present invention proposes a technical solution that can timely and accurately predict the lane-changing behavior of a target vehicle, and then adjust the driving state of the current vehicle accordingly in advance, which aims to improve the driving safety of the current vehicle.

[0005] According to a first aspect of the present invention, a vehicle assistance system is provided for calculating the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model, the vehicle assistance system comprising:

[0006] a data collection unit configured to detect a relative position of the target vehicle relative to the current vehicle, and when detecting that the target vehicle appears at a first preset angle relative to the current vehicle at the rear side of the current vehicle, start collecting data related to the target vehicle in a plurality of consecutive preset time slots until the data collection unit detects that the target vehicle appears at a second preset angle relative to the current vehicle at the side of the current vehicle, wherein the collected data at least includes the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle;

[0007] a data processing unit configured to generate a matrix set by using the data collected in each preset time slot, and output the matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a corresponding type of data in the collected data; and

[0008] The calculation unit is configured to calculate the possibility of the target vehicle merging into a lane by inputting each output matrix set into an artificial intelligence model.

[0009] In one embodiment, the artificial intelligence model can be a convolutional neural network model, preferably a LeNet-5 architecture.

[0010] In one embodiment, the first preset angle can be selected between 45 degrees and 60 degrees when the target vehicle is located behind the current vehicle; and / or the second preset angle can be selected from a position greater than 0 degrees to less than or equal to 10 degrees when the target vehicle is located behind the current vehicle, 0 degrees when the target vehicle is parallel to the current vehicle, and a position greater than 0 degrees to less than or equal to 45 degrees when the target vehicle is located in front of the current vehicle.

[0011] In one embodiment, the calculation of the merging possibility of the target vehicle may include: using an artificial intelligence model to separately calculate the merging possibility of the target vehicle represented by each matrix set, and then calculating the overall merging possibility of the target vehicle by multiplying each merging possibility by a corresponding preset weighted value and summing the values.

[0012] According to a second aspect of the present invention, a vehicle is provided, comprising the vehicle assistance system described in any one of the above embodiments.

[0013] According to a third aspect of the present invention, a vehicle assistance method is provided for calculating the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model. The vehicle assistance method comprises the following steps:

[0014] Data collection step: detecting the relative position of the target vehicle with respect to the current vehicle, and when it is detected that the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle, starting to collect data related to the target vehicle in a plurality of consecutive preset time slots until it is detected that the target vehicle appears at the side of the current vehicle at a second preset angle relative to the current vehicle, wherein the collected data at least includes the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle;

[0015] A data processing step: generating a matrix set by using the data collected in each preset time slot, and outputting the matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a corresponding type of data in the collected data; and

[0016] Calculation step: Calculate the possibility of the target vehicle merging into another lane by inputting each output matrix set into an artificial intelligence model.

[0017] In one embodiment, the artificial intelligence model can be a convolutional neural network model, preferably a LeNet-5 architecture.

[0018] In one embodiment, the first preset angle can be selected between 45 degrees and 60 degrees when the target vehicle is located behind the current vehicle; and / or the second preset angle can be selected from a position greater than 0 degrees to less than or equal to 10 degrees when the target vehicle is located behind the current vehicle, 0 degrees when the target vehicle is parallel to the current vehicle, and a position greater than 0 degrees to less than or equal to 45 degrees when the target vehicle is located in front of the current vehicle.

[0019] In one embodiment, the calculation of the merging possibility of the target vehicle may include: using an artificial intelligence model to separately calculate the merging possibility of the target vehicle represented by each matrix set, and then calculating the overall merging possibility of the target vehicle by multiplying each merging possibility by a corresponding preset weighted value and summing the values.

[0020] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the computer program implements the vehicle assistance method described in any one of the above embodiments.

[0021] By using the technical solution of the present invention, the possibility of the target vehicle merging with the current vehicle can be efficiently and accurately determined by collecting relevant driving data of the target vehicle around the current vehicle within a set time period. Specifically, starting from the detection of the target vehicle appearing at the rear side of the current vehicle at a first preset angle, until the detection of the target vehicle being parallel to the current vehicle or the target vehicle appearing at the front side of the current vehicle at a second preset angle, data related to the target vehicle are collected in a plurality of consecutive preset time slots, and then the collected data are processed into at least three matrix layers, and finally the possibility of the target vehicle merging with the current vehicle is calculated based on the at least three matrix layers using an artificial intelligence model. In addition, the technical solution of the present invention can not only timely and accurately predict the merging behavior of the target vehicle, but also adjust the driving state of the current vehicle in advance in response to the merging behavior, thereby improving the driving safety of the current vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Non-limiting and non-exhaustive embodiments of the present invention are described by way of example with reference to the following drawings, in which:

[0023] Figure 1 is a schematic diagram showing a vehicle assistance system according to an embodiment of the present invention;

[0024] Figure 2 is a flow chart showing a vehicle assistance method according to one embodiment of the present invention;

[0025] Figure 3A , Figure 3B as well as Figure 3C The following diagrams respectively show the situation when the target vehicle is at the rear side of the current vehicle, the situation when the target vehicle and the current vehicle are traveling parallel to each other, and the situation when the target vehicle appears at the front side of the current vehicle in some embodiments. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and implementation schemes. It should be understood that the specific implementation schemes described herein are only used to explain this application and are not used to limit this application.

[0027] Figure 1 1 is a schematic diagram showing a vehicle assistance system 100 according to an embodiment of the present invention. The vehicle assistance system 100 is used to calculate the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model. The artificial intelligence model may be a trained convolutional neural network model or other machine learning models known in the art, preferably a LeNet-5 architecture.

[0028] like Figure 1 As shown in , the vehicle assistance system 100 includes: a data acquisition unit 101, a data processing unit 102 and a computing unit 103. The data acquisition unit 101 communicates with the data processing unit 102, and the data processing unit 102 communicates with the computing unit 103. The configuration and function of each unit of the vehicle assistance system 100 will be described in detail below.

[0029] Some units or the whole of the vehicle assistance system 100 may be arranged on the current vehicle and / or on a remote server. In one embodiment, the data acquisition unit 101 may obtain the acceleration, direction of travel, position and distance of the target vehicle relative to the current vehicle through various types of sensors arranged on the current vehicle. For example, the sensor may include a camera, a laser radar, a millimeter wave radar, an ultrasonic sensor or any other suitable sensor, or any combination thereof. Alternatively, in another embodiment, the relative position and / or distance of the target vehicle relative to the current vehicle may also be obtained through vehicle-to-vehicle communication V2V (via a mobile network or Wi-Fi, etc.) between the current vehicle and the target vehicle. For example, the data acquisition unit 101 may communicate with the target vehicle to obtain the GNSS position or navigation map data of the target vehicle provided by the vehicle global navigation satellite system (GNSS, global navigation satellite system) installed on the target vehicle, and then the data acquisition unit 101 may use the GNSS position or navigation map data of the target vehicle and the GNSS position or navigation map data of the current vehicle to calculate the relative position and / or distance of the target vehicle relative to the current vehicle.

[0030] The data acquisition unit 101 can be configured to detect the relative position of the target vehicle with respect to the current vehicle. When it is detected that the target vehicle appears on the rear side of the current vehicle at a first preset angle relative to the current vehicle, data related to the target vehicle begins to be collected in a plurality of consecutive preset time slots until the data acquisition unit 101 detects that the target vehicle appears on the side of the current vehicle at a second preset angle relative to the current vehicle, wherein the collected data includes at least the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle.

[0031] The first preset angle may be selected when the target vehicle is located between 45 and 60 degrees behind the current vehicle (including the endpoints of 45 and 60 degrees behind the current vehicle). Preferably, in one embodiment, the first preset angle refers to the target vehicle being located 45 degrees behind the current vehicle.

[0032] The second preset angle may be selected from the following: the target vehicle is located at a greater than 0 degree to less than or equal to 10 degrees behind the current vehicle, 0 degrees when the target vehicle is parallel to the current vehicle, and the target vehicle is located at a greater than 0 degree to less than or equal to 45 degrees in front of the current vehicle. For example, in one embodiment, the second preset angle refers to 0 degrees when the target vehicle is parallel to the current vehicle.

[0033] Specifically, for specific exemplary examples of the first preset angle and the second preset angle, reference may be made to the attached appendix. Figures 3A-3C , as described in further detail below.

[0034] The data processing unit 102 can be configured to generate a matrix set by using the data collected in each preset time slot, and output a matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a type of data corresponding to the collected data. For example, when the collected data are only the above-mentioned three types of data (that is, the acceleration of the target vehicle, the direction of travel and the distance between the target vehicle and the current vehicle), the matrix set output at the end of each preset time slot can only include three matrix layers, and the three matrix layers can be respectively related to the acceleration of the target vehicle, the direction of travel of the target vehicle and the distance between the target vehicle and the current vehicle. Of course, those skilled in the art can also think of collecting more travel-related data about the target vehicle to provide a more accurate prediction for the possibility of merging lines.

[0035] The time length of the preset time slot can be any value selected by those skilled in the art between 1s and 3s, including a preset time slot of 1s, a preset time slot of 2s, or a preset time slot of 3s. In an embodiment in which the preset time slot is 1s, in each preset time slot, relevant data can be collected at M moments (e.g., 100 moments) at a fixed time interval of 1 / M seconds (e.g., 1 / 100 or other values) to generate a matrix set covering all data collected at M moments from the beginning of the time slot to the end of the time slot at the end of the corresponding preset time slot.

[0036] The calculation unit 103 may be configured to calculate the merging possibility of the target vehicle by inputting each matrix set output into an artificial intelligence model. In one embodiment, the calculation of the merging possibility of the target vehicle may include: using the artificial intelligence model to calculate the merging possibility of the target vehicle represented by each matrix set, and then multiplying each merging possibility by a corresponding preset weight value and summing them to calculate the overall merging possibility of the target vehicle.

[0037] In one embodiment, assuming that four matrix sets (P1-P4) are generated at the end of four consecutive preset time slots (T1-T4) during the entire data collection process, each matrix set includes at least three matrix layers, then the overall lane merging possibility of the target vehicle can be calculated according to the following formula:

[0038] P overall=P1×w1+P2×w2+P3×w3+P4×w4 (1)

[0039] Where P overall represents the calculated overall merging possibility of the target vehicle, P1-P4 respectively represent the merging possibility of the target vehicle represented by each matrix set within the corresponding preset time slot, and w1-w4 respectively represent the weighted value of the merging possibility of the target vehicle corresponding to each matrix set. Preferably, w1<w2<w3<w4. However, w1-w4 can also be set according to empirical values, the time length of the preset time slot or the safety requirements of the vehicle. Of course, as those skilled in the art will understand, in the case where more or fewer time slots are included in the detection time period that meets the conditions, the above formula for calculating the overall merging possibility of the target vehicle also needs to be adaptively modified.

[0040] In another embodiment, when the target vehicle is detected to appear at the rear side of the current vehicle at a first preset angle relative to the current vehicle (the current moment is represented as T0), it is also possible to immediately query at the current moment T0 whether the relevant data about the target vehicle has been recorded in the previous preset time slot-T1 through the corresponding sensor or other vehicle-to-vehicle communication method. If the relevant data about the target vehicle has been recorded in the previous preset time slot-T1, the relevant data about the target vehicle recorded in the previous time slot-T1 can be directly collected when the trigger condition for starting collection is met (i.e., T0). The matrix set generated using the data recorded in the previous time slot-T1 can be represented as P0, and the corresponding weight can be represented as w0.

[0041] Therefore, the above formula for calculating the overall lane-merging possibility of the target vehicle can be adaptively modified as follows:

[0042] P overall =P0×w0+P1×w1+P2×w2+P3×w3+P4×w4 (2)

[0043] In addition, a lane merging possibility threshold may be preset so that the vehicle assistance system 100 can compare the calculated overall lane merging possibility of the target vehicle with the preset lane merging possibility threshold. overall ≥ threshold, it indicates that the target vehicle is likely to merge, and if P overall< threshold, it is prompted that the target vehicle is less likely to merge. The vehicle assistance system 100 can prompt the possibility of the target vehicle merging through voice broadcast, holographic projection, augmented reality display, vehicle-mounted display terminal display, and / or portable smart device (smart phone, smart watch, smart bracelet) worn by the vehicle user, and control the current vehicle to take corresponding measures such as deceleration braking or emergency braking when the possibility of the target vehicle merging is high.

[0044] According to another aspect of the present invention, a vehicle is provided, wherein the vehicle may include the above vehicle assistance system. The vehicle may be any type of vehicle, including a vehicle driven by a person or an autonomous vehicle.

[0045] Figure 2 is a flow chart illustrating a vehicle assistance method 200 according to one embodiment of the present invention.

[0046] The vehicle assistance method 200 is used to calculate the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model, and the vehicle assistance method 200 includes the following steps:

[0047] Data collection step 201: detecting the relative position of the target vehicle with respect to the current vehicle, and when it is detected that the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle, starting to collect data related to the target vehicle in a plurality of consecutive preset time slots until it is detected that the target vehicle appears at the side of the current vehicle at a second preset angle relative to the current vehicle, wherein the collected data at least includes the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle;

[0048] Data processing step 202: generating a matrix set by using the data collected in each preset time slot, and outputting the matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a corresponding type of data in the collected data; and

[0049] Calculation step 203: Calculate the possibility of the target vehicle merging into another lane by inputting each output matrix set into an artificial intelligence model.

[0050] It should be understood that the specific features described above in relation to the vehicle assistance system can also be similarly applied to the vehicle assistance method for similar extension. For the sake of simplicity, these specific features are not described in detail again.

[0051] Figure 3A , Figure 3B as well as Figure 3CThe following diagrams respectively show a situation in which a target vehicle appears behind a current vehicle, a situation in which the target vehicle and the current vehicle travel parallel to each other, and a situation in which the target vehicle appears in front of the current vehicle in some embodiments.

[0052] Reference now Figure 3A , Figure 3A The figure shows a situation where the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle. Figure 3A , the vehicle's driving direction (as indicated by the arrow), the position reference point R0 on the current vehicle, the position detection point R1 on the target vehicle, the baseline L0 horizontally extending from the reference point R0 about the current vehicle, and the straight line L1 connecting the position reference point R0 on the current vehicle and the position detection point R1 on the target vehicle are shown. Figure 3A As shown in , the target vehicle is located behind the current vehicle and the angle formed between the target vehicle and the current vehicle (i.e., the angle α1 between the straight line L1 and the reference line L0) is 45 degrees. Assuming that the first preset angle is 45 degrees, when the target vehicle is detected Figure 3A When the situation occurs, the data collection unit 101 can be triggered to start collecting data related to the target vehicle.

[0053] It should be understood by those skilled in the art that although Figure 3A It is shown that the reference point R0 is located at the middle position of the front side of the current vehicle's front, and the detection point R1 is located at the middle position of the front side of the target vehicle's front. However, the positions of the reference point R0 and the detection point R1 can be reasonably selected according to the installation position of the sensor or the needs of other detection means.

[0054] As the current vehicle and the target vehicle continue to travel, the relative position of the target vehicle relative to the current vehicle continues to change. Figure 3B and Figure 3C The situation shown in .

[0055] Figure 3B The situation when the target vehicle is parallel to the current vehicle is shown. Figure 3B The driving direction of the vehicle is also shown in FIG. 1 (as indicated by the arrow), wherein the straight line L1 coincides with the reference line L0, that is, the angle α formed by the straight line L1 and the reference line L0 is 21 =0°. Assuming that the second preset angle is 0 degrees when the target vehicle is parallel to the current vehicle, at this time, the data acquisition unit 101 can stop collecting data related to the target vehicle, and then can use the collected data to generate a corresponding matrix set to further use the artificial intelligence model to calculate the possibility of the target vehicle merging.

[0056] Figure 3CThe figure shows the situation when the target vehicle appears in front of the current vehicle. Figure 3C The driving direction of the vehicle is also shown in FIG. 1 (as indicated by the arrow), wherein the target vehicle drives to the front side of the current vehicle, and the angle α formed by the straight line L1 and the reference line L0 of the current vehicle is 22 =30°. Assuming that the second preset angle is that the target vehicle is located 30 degrees in front of the current vehicle, at this time, the data acquisition unit 101 can stop collecting data related to the target vehicle, and then can generate a corresponding matrix set from the collected data to further use the artificial intelligence model to calculate the possibility of the target vehicle merging.

[0057] Preferably, the artificial intelligence model mentioned in this article can be trained in the following manner: by using the above-mentioned data collection steps, data processing steps and calculation steps to continuously perform machine learning on a large number of different actual or virtual lane-changing scenarios (e.g., the complete process of the actual lane-changing action), and then optimize various calculation parameters in the artificial intelligence model. For example, during the training process, different labels or weights can be set for different matrix layers in the matrix set, and different weights can be set for matrix sets of different time slots, so as to obtain an artificial intelligence model with optimal parameter settings and optimal prediction accuracy in the continuous machine learning process.

[0058] It should be understood that each unit of the vehicle auxiliary system of the present invention can be implemented in whole or in part by software, hardware, firmware or a combination thereof. Each of the units can be embedded in the processor of the computer device in the form of hardware or firmware or independent of the processor, or can be stored in the memory of the computer device in the form of software for the processor to call to perform the operation of each unit. Each of the units can be implemented as an independent component or module, or two or more units can be implemented as a single component or module.

[0059] It should be understood by those skilled in the art that Figure 1 The schematic diagram of the vehicle assistance system shown in the figure is only an exemplary block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device, processor or computer program embodying the solution of the present invention. The specific computer device, processor or computer program may include more or fewer components or modules than shown in the figure, or combine or split certain components or modules, or have different arrangements of components or modules.

[0060] The present invention can be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of the present invention to be performed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. One or more method steps / operations can be performed by one or more computer devices or processors, and one or more other method steps / operations can be performed by one or more other computer devices or processors. One or more computer devices or processors can perform a single method step / operation, or perform two or more method steps / operations.

[0061] It will be appreciated by a person skilled in the art that all or part of the steps of the auxiliary method of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-transitory computer-readable storage medium, and when the computer program is executed, the steps of the auxiliary method of the present invention are performed. Depending on the circumstances, any reference to memory, storage, database or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magnetic data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0062] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0063] Although the present invention has been described in conjunction with the embodiments, it will be appreciated by those skilled in the art that the above description and the accompanying drawings are exemplary and non-restrictive only, and the present invention is not limited to the disclosed embodiments. Various modifications and variations are possible without departing from the spirit of the present invention.

Claims

1. A vehicle assistance system (100), for calculating the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model, the vehicle assistance system (100) comprising: A data acquisition unit (101) is configured to detect the relative position of the target vehicle relative to the current vehicle, and when it is detected that the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle, start to collect data related to the target vehicle in a plurality of consecutive preset time slots until the data acquisition unit (101) detects that the target vehicle appears at the side of the current vehicle at a second preset angle relative to the current vehicle, wherein the collected data at least includes the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle; A data processing unit (102) is configured to generate a matrix set by using the data collected in each preset time slot, and output the matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a corresponding type of data in the collected data; as well as A calculation unit (103) is configured to calculate the possibility of the target vehicle merging into a lane by inputting each matrix set output into an artificial intelligence model, wherein the calculation of the possibility of the target vehicle merging into a lane comprises: The artificial intelligence model is used to calculate the lane-changing possibilities of the target vehicle represented by each matrix set respectively, and then the overall lane-changing possibilities of the target vehicle are calculated by multiplying each lane-changing possibility by the corresponding preset weighted value and summing the values.

2. The vehicle assistance system (100) according to claim 1, wherein the artificial intelligence model is a convolutional neural network model.

3. The vehicle assistance system (100) according to claim 2, wherein the convolutional neural network model is a LeNet-5 architecture.

4. The vehicle assistance system (100) according to claim 1 or 2, wherein the first preset angle is selected between 45 degrees and 60 degrees when the target vehicle is located behind the current vehicle; and / or the second preset angle is selected from when the target vehicle is located greater than 0 degrees to less than or equal to 10 degrees behind the current vehicle, 0 degrees when the target vehicle is parallel to the current vehicle, and when the target vehicle is located greater than 0 degrees to less than or equal to 45 degrees in front of the current vehicle.

5. A vehicle, comprising the vehicle assistance system (100) according to any one of claims 1-4.

6. A vehicle assistance method (200), for calculating the possibility of a target vehicle merging with a current vehicle based on an artificial intelligence model, the vehicle assistance method (200) comprising the following steps: Data collection step (201): detecting the relative position of the target vehicle relative to the current vehicle, and when it is detected that the target vehicle appears at the rear side of the current vehicle at a first preset angle relative to the current vehicle, starting to collect data related to the target vehicle in a plurality of consecutive preset time slots until it is detected that the target vehicle appears at the side of the current vehicle at a second preset angle relative to the current vehicle, wherein the collected data at least includes the acceleration of the target vehicle, the direction of travel of the target vehicle, and the distance between the target vehicle and the current vehicle; Data processing step (202): generating a matrix set by using the data collected in each preset time slot, and outputting the matrix set at the end of each preset time slot, wherein the matrix set includes at least three matrix layers, and the at least three matrix layers are respectively related to a corresponding type of data in the collected data; as well as Calculation step (203): Calculating the possibility of the target vehicle merging into a lane by inputting each matrix set output into an artificial intelligence model, wherein the calculation of the possibility of the target vehicle merging into a lane comprises: The artificial intelligence model is used to calculate the lane-changing possibilities of the target vehicle represented by each matrix set respectively, and then the overall lane-changing possibilities of the target vehicle are calculated by multiplying each lane-changing possibility by the corresponding preset weighted value and summing the values.

7. The vehicle assistance method (200) according to claim 6, wherein the artificial intelligence model is a convolutional neural network model.

8. The vehicle assistance method (200) according to claim 7, wherein the convolutional neural network model is a LeNet-5 architecture.

9. The vehicle assistance method (200) according to claim 6 or 7, wherein the first preset angle is selected between 45 degrees and 60 degrees when the target vehicle is located behind the current vehicle; and / or the second preset angle is selected from when the target vehicle is located greater than 0 degrees to less than or equal to 10 degrees behind the current vehicle, 0 degrees when the target vehicle is parallel to the current vehicle, and when the target vehicle is located greater than 0 degrees to less than or equal to 45 degrees in front of the current vehicle.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the computer program implements the vehicle assistance method (200) according to any one of claims 6 to 9.

Images