Training method for driving intention prediction model, vehicle driving method and device

Abstract

The present disclosure provides a training method, a vehicle driving method, and a device for a driving intention prediction model. The method includes: obtaining an initial driving intention prediction model to be trained, and obtaining sample intention data and sample driving data of a sample vehicle to obtain a first training sample for the initial driving intention prediction model; training the initial driving intention prediction model according to the first training sample until the training is completed to obtain a trained candidate driving intention prediction model; obtaining sample driving style data of the sample vehicle, and adjusting the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model. The accuracy and efficiency of driving intention recognition are improved, the performance of the vehicle driving assistance function is optimized, the driving safety of the vehicle is improved, and the driving experience of the user is optimized.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and in particular, to a method for training a driving intention prediction model, a vehicle driving method, and a device. Background Art

[0002] With the development of society, more and more people choose vehicles as means of transportation. In the scenario of a vehicle changing lanes, a driver needs to observe the road conditions to determine whether the current road conditions allow for a lane change. In related technologies, radars and cameras configured on the vehicle can assist the driver in detecting the blind spots of the vehicle to provide an auxiliary reminder for the driver to change lanes.

[0003] In this scenario, the camera and radar can identify the current lane change requirement through the driver's activation of the turn signal, thereby activating the blind spot detection function of the vehicle. In this scenario, when the driver does not activate the turn signal, the blind spot detection function may not be activated, which may affect the safety of vehicle driving to a certain extent. Summary of the Invention

[0004] The present disclosure aims to at least solve one of the technical problems in the related technologies to some extent.

[0005] To this end, a first aspect of the present disclosure provides a method for training a driving intention prediction model.

[0006] A second aspect of the present disclosure provides a vehicle driving method.

[0007] A third aspect of the present disclosure provides a device for training a driving intention prediction model.

[0008] A fourth aspect of the present disclosure provides a vehicle driving device.

[0009] A fifth aspect of the present disclosure provides a vehicle.

[0010] A sixth aspect of the present disclosure provides an electronic device.

[0011] A seventh aspect of the present disclosure provides a computer-readable storage medium.

[0012] A first aspect of the present disclosure provides a method for training a driving intention prediction model. The method includes: obtaining an initial driving intention prediction model to be trained, and obtaining sample intention data and sample driving data of a sample vehicle to obtain a first training sample for the initial driving intention prediction model; training the initial driving intention prediction model according to the first training sample until the training ends to obtain a trained candidate driving intention prediction model; obtaining sample driving style data of the sample vehicle, and adjusting the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model.

[0013] In addition, the training method of the driving intention prediction model proposed in the first aspect of the present disclosure may further have the following additional technical features:

[0014] According to an embodiment of the present disclosure, the obtaining of the sample intention data and sample driving data of the sample vehicle to obtain the first training sample of the initial driving intention prediction model includes: performing lane-changing intention recognition on the sample intention data to obtain sample lane-changing intention data and sample non-lane-changing intention data from the sample intention data; obtaining the sample lane-changing driving data corresponding to the sample lane-changing intention data and the sample non-lane-changing driving data corresponding to the sample non-lane-changing intention data from the sample driving data; obtaining a first candidate sample according to the sample lane-changing intention data and the corresponding sample lane-changing driving data, and obtaining a second candidate sample according to the sample non-lane-changing intention data and the corresponding sample non-lane-changing driving data; and obtaining the first training sample according to the first candidate sample and the second candidate sample.

[0015] According to an embodiment of the present disclosure, the training of the initial driving intention prediction model with the first training sample until the training is completed to obtain a trained candidate driving intention prediction model includes: inputting the first training sample into the initial driving intention prediction model to obtain a first training result output by the initial driving intention prediction model; obtaining a first sample label of the first training sample to obtain a first training loss of the first training result based on the first sample label; adjusting the parameters of the initial driving intention prediction model according to the first training loss, and returning to obtain the next first training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training is completed to obtain the trained candidate driving intention prediction model.

[0016] According to an embodiment of the present disclosure, the obtaining of the sample driving style data of the sample vehicle and the adjustment of the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model includes: obtaining a plurality of preset sample driving styles and the initial driving style data of the sample vehicle, and performing style division on the initial driving style data according to the plurality of sample driving styles to obtain sample driving style data of each of the plurality of sample driving styles; for any one of the sample driving styles, obtaining the candidate driving intention prediction model corresponding to the sample driving style, and generating a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style; and training the candidate driving intention prediction model with the second training sample until the training is completed to obtain the trained target driving intention prediction model.

[0017] According to an embodiment of the present disclosure, for any sample driving style, obtaining a candidate driving intention prediction model corresponding to the sample driving style, and generating a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style includes: for any sample driving style, obtaining sample driving style intention data and sample driving style driving data under the sample driving style from the sample intention data and the sample driving data; and obtaining the second training sample of the candidate driving intention prediction model according to the sample driving style intention data and the sample driving style driving data.

[0018] According to an embodiment of the present disclosure, training the candidate driving intention prediction model with the second training sample until the training is completed to obtain the trained target driving intention prediction model includes: inputting the second training sample into the candidate driving intention prediction model to obtain a second training result output by the candidate driving intention prediction model; obtaining a second sample label of the second training sample to obtain a second training loss of the second training result based on the second sample label; adjusting parameters of the candidate driving intention prediction model according to the second training loss, and returning to obtain the next second training sample to continue model training on the candidate driving intention prediction model with adjusted parameters until the training is completed to obtain the trained target driving intention prediction model.

[0019] According to an embodiment of the present disclosure, after training the candidate driving intention prediction model with the second training sample until the training is completed to obtain the trained target driving intention prediction model, it includes: for any sample driving style, obtaining the target driving style of the target vehicle; in response to the target driving style matching the sample driving style, sending the target driving intention prediction model corresponding to the sample driving style to the target vehicle.

[0020] A second aspect of the present disclosure provides a vehicle driving method, including: obtaining a target driving intention prediction model configured on a target vehicle, where the target driving intention prediction model is obtained based on the training method of the driving intention prediction model proposed in the first aspect above; obtaining target driving style data of the target vehicle, as well as target intention data and target driving data of the vehicle within a target time range; inputting the target intention data, the target driving data, and the target driving style data into the target driving intention prediction model, and outputting a target predicted driving intention of the target vehicle through the target driving intention prediction model; and performing driving control on the target vehicle based on the target predicted driving intention.

[0021] In addition, the vehicle driving method proposed in the second aspect of the present disclosure may further have the following additional technical features:

[0022] According to an embodiment of the present disclosure, the method further includes: identifying whether there is corresponding update data for the target driving intention prediction model; in response to identifying that there is the update data, updating and adjusting the target driving intention prediction model according to the update data to obtain a new target driving intention prediction model.

[0023] According to an embodiment of the present disclosure, the driving control of the vehicle based on the target driving intention includes: in response to the target predicted driving intention being a lane change driving intention, controlling the lane change of the target vehicle through the turn signal control module and the lane change assistance module of the vehicle.

[0024] According to an embodiment of the present disclosure, after the driving control of the vehicle based on the target driving intention, it includes: in response to identifying that the lane change driving of the target vehicle ends, ending the lane change driving control of the target vehicle and uploading the target predicted driving intention output by the target driving intention prediction model to the cloud.

[0025] The present disclosure proposes a training device for a driving intention prediction model in a third aspect. The device includes: a first acquisition module for acquiring an initial driving intention prediction model to be trained, as well as acquiring sample intention data and sample driving data of a sample vehicle to obtain a first training sample of the initial driving intention prediction model; a first training module for training the initial driving intention prediction model according to the first training sample until the training ends to obtain a trained candidate driving intention prediction model; a second training module for acquiring sample driving style data of the sample vehicle and adjusting the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model.

[0026] In addition, the training device for the driving intention prediction model proposed in the third aspect of the present disclosure may further have the following additional technical features:

[0027] According to an embodiment of the present disclosure, the first acquisition module is further configured to: perform lane-changing intention recognition on the sample intention data to obtain sample lane-changing intention data and sample non-lane-changing intention data from the sample intention data; obtain sample lane-changing driving data corresponding to the sample lane-changing intention data and sample non-lane-changing driving data corresponding to the sample non-lane-changing intention data from the sample driving data; obtain a first candidate sample according to the sample lane-changing intention data and the corresponding sample lane-changing driving data, and obtain a second candidate sample according to the sample non-lane-changing intention data and the corresponding sample non-lane-changing driving data; obtain the first training sample according to the first candidate sample and the second candidate sample.

[0028] According to an embodiment of the present disclosure, the first training module is further configured to: input the first training sample into the initial driving intention prediction model to obtain a first training result output by the initial driving intention prediction model; obtain a first sample label of the first training sample to obtain a first training loss of the first training result based on the first sample label; adjust the parameters of the initial driving intention prediction model according to the first training loss, and return to obtain the next first training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training ends, and obtain the trained candidate driving intention prediction model.

[0029] According to an embodiment of the present disclosure, the second training module is further configured to: obtain a plurality of preset sample driving styles and initial driving style data of the sample vehicle, and perform style division on the initial driving style data according to the plurality of sample driving styles to obtain sample driving style data of each of the plurality of sample driving styles; for any sample driving style, obtain the candidate driving intention prediction model corresponding to the sample driving style, and generate a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style; train the candidate driving intention prediction model through the second training sample until the training ends to obtain the trained target driving intention prediction model.

[0030] According to an embodiment of the present disclosure, the second training module is further configured to: for any sample driving style, obtain sample driving style intention data and sample driving style driving data under the sample driving style from the sample intention data and the sample driving data; obtain the second training sample of the candidate driving intention prediction model according to the sample driving style intention data and the sample driving style driving data.

[0031] According to an embodiment of the present disclosure, the second training module is further configured to: input the second training sample into the candidate driving intention prediction model to obtain a second training result output by the candidate driving intention prediction model; obtain a second sample label of the second training sample to obtain a second training loss of the second training result based on the second sample label; adjust the parameters of the candidate driving intention prediction model according to the second training loss, and return to obtain the next second training sample to continue model training on the candidate driving intention prediction model with adjusted parameters until the training ends, so as to obtain the trained target driving intention prediction model.

[0032] According to an embodiment of the present disclosure, the system further includes a distribution module configured to: for any sample driving style, obtain the target driving style of the target vehicle; in response to the target driving style matching the sample driving style, distribute the target driving intention prediction model corresponding to the sample driving style to the target vehicle.

[0033] A fourth aspect of the present disclosure provides a vehicle driving device, including: a second acquisition module configured to acquire a target driving intention prediction model configured on the target vehicle, where the target driving intention prediction model is obtained based on the training device of the driving intention prediction model proposed in the third aspect; a third acquisition module configured to acquire target driving style data of the target vehicle, as well as target intention data and target driving data of the vehicle within a target time range; a prediction module configured to input the target intention data, the target driving data, and the target driving style data into the target driving intention prediction model, and output a target predicted driving intention of the target vehicle through the target driving intention prediction model; and a control module configured to perform driving control on the target vehicle based on the target predicted driving intention.

[0034] In addition, the vehicle driving device proposed in the fourth aspect of the present disclosure may further have the following additional technical features:

[0035] According to an embodiment of the present disclosure, the device further includes an update module configured to: identify whether there is corresponding update data for the target driving intention prediction model; in response to identifying the existence of the update data, update and adjust the target driving intention prediction model according to the update data to obtain a new target driving intention prediction model.

[0036] According to an embodiment of the present disclosure, the control module is further configured to: in response to the target predicted driving intention being a lane change driving intention, perform lane change driving control on the target vehicle through the turn signal control module and the lane change assist module of the vehicle.

[0037] According to an embodiment of the present disclosure, the control module is further configured to: in response to identifying that the lane change driving of the target vehicle ends, end the lane change driving control of the target vehicle, and upload the target predicted driving intention output by the target driving intention prediction model to the cloud.

[0038] A fifth aspect of the present disclosure provides a vehicle, which is configured to implement the training method of the driving intention prediction model proposed in the first aspect and / or the vehicle driving method proposed in the second aspect above.

[0039] A sixth aspect of the present disclosure provides an electronic device, including: a processor; a memory for storing executable instructions of the processor; wherein, the processor is configured to execute the instructions to implement the training method of the driving intention prediction model proposed in the first aspect and / or the vehicle driving method proposed in the second aspect above.

[0040] A seventh aspect of the present disclosure provides a computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by the processor of the electronic device, enabling the electronic device to execute the training method of the driving intention prediction model proposed in the first aspect and / or the vehicle driving method proposed in the second aspect above.

[0041] The training method, vehicle driving method and device of the driving intention prediction model proposed in the present disclosure obtain an initial driving intention prediction model to be trained, as well as a first training sample constructed by sample intention data and sample driving data, and perform model training on the initial driving intention prediction model through the first training sample to obtain a candidate driving intention prediction model. Optionally, sample driving style data of a sample vehicle is obtained and the candidate driving intention prediction model is adjusted according to the sample driving style data to obtain a target driving intention prediction model. In the present disclosure, through the training of the initial driving intention prediction model and the adjustment of the candidate driving intention prediction model, a target driving intention prediction model is obtained, so that the target driving intention prediction model can learn the relationship between driver behavior data and driving intention, thereby realizing the prediction of the driver's driving intention. Compared with the method of only identifying the driving intention by starting the turn signal of the vehicle in the related art, the accuracy and efficiency of driving intention recognition are improved, the possibility of potential safety hazards in the vehicle driving caused by failure to recognize the driving intention of the vehicle is reduced, the performance of the vehicle driving assistance function is optimized, the driving safety of the vehicle is improved, and the driving experience of the user is optimized.

[0042] It should be understood that the content described in the present disclosure is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The above and / or additional aspects and advantages of the present disclosure will become apparent and be readily understood from the following description of embodiments in conjunction with the accompanying drawings, in which:

[0044] Figure 1 It is a schematic flowchart of a method for training a driving intention prediction model according to an embodiment of the present disclosure;

[0045] Figure 2 It is a schematic flowchart of a method for training a driving intention prediction model according to another embodiment of the present disclosure;

[0046] Figure 3 It is a schematic flowchart of a vehicle driving method according to an embodiment of the present disclosure;

[0047] Figure 4 It is a schematic flowchart of a vehicle driving method according to an embodiment of the present disclosure;

[0048] Figure 5 It is a schematic flowchart of a vehicle driving method according to another embodiment of the present disclosure;

[0049] Figure 6 It is a schematic structural diagram of a training device for a driving intention prediction model according to an embodiment of the present disclosure;

[0050] Figure 7 It is a schematic structural diagram of a vehicle driving device according to an embodiment of the present disclosure;

[0051] Figure 8 It is a block diagram of an electronic device according to an embodiment of the present disclosure. Detailed Description of the Embodiment

[0052] Embodiments of the present disclosure will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure, and should not be construed as a limitation of the present disclosure.

[0053] A method for training a driving intention prediction model, a vehicle driving method, and a device according to an embodiment of the present disclosure will be described below with reference to the accompanying drawings.

[0054] Figure 1 It is a schematic flowchart of a method for training a driving intention prediction model according to an embodiment of the present disclosure. As Figure 1 shown, the method includes:

[0055] S101, obtaining an initial driving intention prediction model to be trained, and obtaining sample intention data and sample driving data of a sample vehicle to obtain a first training sample of the initial driving intention prediction model.

[0056] In the embodiments of the present disclosure, during the driving process of a vehicle, the driving intention of the driver can be recognized through relevant information of the driver. In this scenario, after recognizing the driving intention of the driver, the vehicle can achieve assisted driving for the driver based on the auxiliary functions of the vehicle.

[0057] Optionally, a model configured on the vehicle with driving intention recognition and assisted driving functions can be marked as the driving intention prediction model on the vehicle. Among them, if the model requires training, the driving intention prediction model that needs to be trained can be marked as the initial driving intention prediction model.

[0058] In the embodiments of the present disclosure, the samples for training the initial driving intention prediction model can be marked as the first training samples, and the vehicle for obtaining the first training samples can be marked as the sample vehicle.

[0059] Optionally, the corresponding intention data and driving data can be obtained from the sample vehicle and marked as the sample intention data and sample driving data of the sample vehicle respectively. Then, based on the construction method of training samples in related technologies, the sample intention data and sample driving data are processed to obtain the first training samples constructed from the sample intention data and sample driving data.

[0060] S102. Train the initial driving intention prediction model according to the first training samples until the training ends to obtain a trained candidate driving intention prediction model.

[0061] In the embodiments of the present disclosure, according to the model training method in related technologies, the initial driving intention prediction model can be trained through the first training samples until the training ends, and the trained model is determined as the candidate driving intention prediction model.

[0062] Optionally, the training end condition of the initial driving intention prediction model can be set based on the training rounds. For the current round of model training, if the current training round matches the preset training end condition, the training of the initial driving intention prediction model can be ended, and the model obtained at the end of the last round of training is determined as the trained candidate driving intention prediction model.

[0063] Optionally, the corresponding training end condition can be set according to the result output by the model training. For the current round of model training, if the output result of the current round of the model matches the preset training end condition, the training of the initial driving intention prediction model can be ended, and the model obtained at the end of the last round of training is determined as the trained candidate driving intention prediction model.

[0064] S103. Obtain the sample driving style data of the sample vehicle, and adjust the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model.

[0065] In the embodiments of the present disclosure, different drivers may have different driving styles. Among them, the driving style of a driver is related to information such as the vehicle speed, the pedaling force of the brake, and the steering angle of the vehicle when driving the vehicle.

[0066] As an example, it is set that the vehicle driving styles of drivers may include a steady style, an aggressive style, and a comfortable style. It can be understood that in a scenario where the vehicle needs to change lanes, when a driver with a steady style changes lanes, the vehicle usually changes lanes with a small steering angle and a low vehicle speed.

[0067] Moreover, a driver with an aggressive style usually changes lanes with a large steering angle and a high vehicle speed when changing lanes, and a driver with a comfortable style pays attention to energy consumption when driving the vehicle, and the vehicle speed and steering angle adopted when changing lanes are between the steady style and the aggressive style.

[0068] Optionally, relevant driving data of a sample vehicle within a sample time range can be obtained and analyzed in terms of the driving style dimension, so as to obtain sample driving style data of the sample vehicle. Among them, the sample driving style data can be determined by speed data and steering angle data in the relevant driving data of the sample vehicle, etc.

[0069] In this scenario, the candidate driving intention prediction model can be adjusted according to the sample driving style data of the sample vehicle, so that the candidate driving intention prediction model can learn the relationship between driving intention and driving style.

[0070] Furthermore, the model obtained after being adjusted according to the sample driving style data is determined as the adjusted target driving intention prediction model.

[0071] The present disclosure provides a method for training a driving intention prediction model, which obtains an initial driving intention prediction model to be trained, as well as a first training sample constructed from sample intention data and sample driving data, and trains the initial driving intention prediction model through the first training sample to obtain a candidate driving intention prediction model. Optionally, sample driving style data of the sample vehicle is obtained and the candidate driving intention prediction model is adjusted according to the sample driving style data to obtain a target driving intention prediction model. In the present disclosure, through the training of the initial driving intention prediction model and the adjustment of the candidate driving intention prediction model, a target driving intention prediction model is obtained, enabling the target driving intention prediction model to learn the relationship between driver behavior data and driving intention, thereby realizing the prediction of the driver's driving intention. Compared with the method in the related art that only identifies the driving intention through the activation of the vehicle's turn signal, the accuracy and efficiency of driving intention recognition are improved, the possibility of potential safety hazards in the vehicle's driving due to the failure to recognize the vehicle's driving intention is reduced, the performance of the vehicle's driving assistance function is optimized, the driving safety of the vehicle is improved, and the driving experience of the user is optimized.

[0072] In the above embodiment, regarding the acquisition of the target driving intention prediction model, it can be combined with Figure 2 for further understanding. Figure 2 As shown in Figure 2 is a schematic flowchart of a method for training a driving intention prediction model according to another embodiment of the present disclosure. The method includes:

[0073] S201, obtaining sample intention data and sample driving data of the sample vehicle to obtain a first training sample for the initial driving intention prediction model.

[0074] Optionally, lane-changing intention recognition is performed on the sample intention data to obtain sample lane-changing intention data and sample non-lane-changing intention data from the sample intention data.

[0075] In the embodiment of the present disclosure, the sample intention data of the sample vehicle may include lane-changing intention and non-lane-changing intention. Among them, the lane-changing intention can be marked as the sample lane-changing intention data in the sample intention data, and the non-lane-changing intention can be marked as the sample non-lane-changing intention data in the sample intention data.

[0076] Among them, the sample intention data of the driver of the sample vehicle can be collected through a data collection device configured on the vehicle. For example, the eye movement direction data of the driver can be collected through an eye tracker configured in the vehicle cockpit to obtain the sample intention data of the driver, and the head torsion angle data of the driver can also be collected through a camera configured on the vehicle to obtain the sample intention data of the driver. No specific limitation is made here.

[0077] Optionally, from the sample driving data, obtain the sample lane-changing driving data corresponding to the sample lane-changing intention data and the sample non-lane-changing driving data corresponding to the sample non-lane-changing intention data.

[0078] In the embodiments of the present disclosure, the sample driving data of the sample vehicle may include the driving data when the sample vehicle performs a lane change, or may include the driving data when the sample vehicle does not perform a lane change. In this scenario, the sample driving data can be divided according to the sample lane-changing intention data and the sample non-lane-changing intention data identified in the sample intention data, so as to obtain the corresponding data of the sample lane-changing intention data in the sample driving data and mark it as the sample lane-changing driving data, and obtain the corresponding data of the sample non-lane-changing intention data in the sample driving data and mark it as the sample non-lane-changing driving data.

[0079] Among them, it is possible to obtain the acquisition time range of the sample lane-changing intention data collected on the sample vehicle, obtain the driving data collected within the same acquisition time range from the sample driving data of the sample vehicle, and determine this part of the data as the sample lane-changing driving data corresponding to the sample lane-changing intention data.

[0080] And, obtain the acquisition time range of the sample non-lane-changing intention data collected on the sample vehicle, obtain the driving data collected within the same acquisition time range from the sample driving data of the sample vehicle, and determine this part of the data as the sample non-lane-changing driving data corresponding to the sample non-lane-changing intention data.

[0081] Correspondingly, it is also possible to obtain the vehicle coordinates when the sample lane-changing intention data is collected on the sample vehicle, obtain the driving data collected at the same vehicle coordinates from the sample driving data of the sample vehicle, and determine this part of the data as the sample lane-changing driving data corresponding to the sample lane-changing intention data.

[0082] And, obtain the vehicle coordinates when the sample non-lane-changing intention data is collected on the sample vehicle, obtain the driving data collected at the same vehicle coordinates from the sample driving data of the sample vehicle, and determine this part of the data as the sample non-lane-changing driving data corresponding to the sample non-lane-changing intention data.

[0083] It should be noted that the sample driving data may include the driving habit data of the driver of the sample vehicle such as the steering wheel angle, steering wheel rotation speed, steering wheel grip force, brake pedal stroke, brake pedal opening change rate, accelerator pedal stroke, accelerator pedal opening change rate, and turn signal lever state data, as well as the body dynamic driving data of the sample vehicle such as vehicle speed, longitudinal acceleration, lateral acceleration, and yaw rate data. It may also include other data that can describe the vehicle driving state, which is not specifically limited here.

[0084] Optionally, a first candidate sample is obtained according to the sample lane-changing intention data and the corresponding sample lane-changing driving data, and a second candidate sample is obtained according to the sample non-lane-changing intention data and the corresponding sample non-lane-changing driving data.

[0085] In an embodiment of the present disclosure, the training positive samples of the initial driving intention prediction model can be constructed according to the sample lane-changing intention data and the corresponding sample lane-changing driving data, and marked as the first candidate samples.

[0086] In addition, the training negative samples of the initial driving intention prediction model are constructed according to the sample non-lane-changing intention data and the corresponding sample non-lane-changing driving data, and marked as the second candidate samples.

[0087] Among them, the sample lane-changing intention data and the corresponding sample lane-changing driving data can be processed according to the sample construction method in the related art to obtain the first candidate sample constructed by the two, and the sample non-lane-changing intention data and the corresponding sample non-lane-changing driving data can be processed according to the sample construction method in the related art to obtain the second candidate sample constructed by the two.

[0088] Optionally, a first training sample is obtained according to the first candidate sample and the second candidate sample.

[0089] In an embodiment of the present disclosure, there is a corresponding training sample set for the training of the initial driving intention prediction model, and the samples in the training sample set can be marked as the first training samples.

[0090] In this scenario, a sample set composed of the first candidate sample and the second candidate sample can be obtained, and the sample set can be marked as the training sample set of the initial driving intention prediction model, and then the first training sample in the training sample set can be obtained.

[0091] S202. Train the initial driving intention prediction model according to the first training sample until the training ends to obtain a trained candidate driving intention prediction model.

[0092] Optionally, the first training sample is input into the initial driving intention prediction model to obtain a first training result output by the initial driving intention prediction model.

[0093] In an embodiment of the present disclosure, the first training sample can be input into the initial driving intention prediction model, the initial driving intention prediction model extracts features from the first training sample, and intention prediction is performed according to the extracted features.

[0094] Furthermore, the output result obtained by the initial driving intention prediction model based on the first training sample is marked as the first training result.

[0095] Optionally, obtain the first sample label of the first training sample to obtain the first training loss of the first training result based on the first sample label.

[0096] In the embodiments of the present disclosure, the label information of the first training sample can be obtained and marked as the first sample label of the first training sample.

[0097] Among them, the first sample label can be used to indicate whether the driving data in the first training sample is driving data under the lane-changing intention.

[0098] In the scenario where the first training sample is a sample constructed by sample lane-changing intention data and sample lane-changing driving data, the label information obtained based on the sample lane-changing intention data can be used as the first sample label of the first training sample in this scenario.

[0099] Correspondingly, in the scenario where the first training sample is a sample constructed by sample non-lane-changing intention data and sample non-lane-changing driving data, the label information obtained based on the sample non-lane-changing intention data can be used as the first sample label of the first training sample in this scenario.

[0100] In the embodiments of the present disclosure, the first training result and the first sample label can be processed by an algorithm according to the loss value acquisition algorithm in the related art, and then the loss value of the first training result based on the first sample label can be obtained according to the result of the algorithm processing, and this loss value can be determined as the first training loss.

[0101] Optionally, adjust the parameters of the initial driving intention prediction model according to the first training loss, and return to obtain the next first training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training ends, so as to obtain a trained candidate driving intention prediction model.

[0102] In the embodiments of the present disclosure, the model parameters of the initial driving intention prediction model can be adjusted according to the first training loss, and return to obtain the next training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training ends.

[0103] Among them, the relevant content of the end of the training of the initial driving intention prediction model can be understood in combination with the relevant content in the above embodiments, and will not be elaborated here.

[0104] Further, the model obtained by the end of the training of the initial driving intention prediction model is determined as a trained candidate driving intention prediction model.

[0105] S203. Obtain the sample driving style data of the sample vehicle, and adjust the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model.

[0106] Optionally, obtain a plurality of preset sample driving styles and the initial driving style data of the sample vehicle, and divide the initial driving style data according to the plurality of sample driving styles to obtain the sample driving style data corresponding to each of the plurality of sample driving styles.

[0107] In the embodiments of the present disclosure, the driving styles of different drivers are different. Among them, the driving style analysis can be performed through an open-source data set, so as to obtain a plurality of possible driving styles of the driver, and mark them as a plurality of sample driving styles.

[0108] In this scenario, the driving style data of the sample vehicle can be collected, and the collected driving style data can be classified according to the plurality of sample driving styles. Among them, the driving style data collected on the sample vehicle can be marked as the initial driving style data of the sample vehicle.

[0109] Furthermore, cluster the initial driving style data according to the plurality of sample driving styles. For any one of the multiple clusters obtained by clustering, the driving style data in the cluster can be obtained, and the driving style data can be marked as the sample driving style data corresponding to the sample driving style corresponding to the cluster.

[0110] It should be noted that the sample vehicle can be a plurality of preset test vehicles. In this scenario, the initial driving style data uploaded by the plurality of test vehicles can be continuously received within a set time range, and after receiving this part of the data, this part of the data can be classified according to the preset plurality of sample driving styles, so as to obtain the sample driving style data corresponding to each of the plurality of sample driving styles.

[0111] Optionally, for any sample driving style, obtain a candidate driving intention prediction model corresponding to the sample driving style, and generate a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style.

[0112] In the embodiments of the present disclosure, after the initial driving intention prediction model is trained to obtain the candidate driving intention prediction model, the candidate driving intention prediction model can be divided according to the driving style. It can be understood that after obtaining a plurality of initial driving intention prediction models, the plurality of initial driving intention prediction models are respectively trained to obtain a plurality of candidate driving intention prediction models. Furthermore, the plurality of candidate driving intention prediction models are divided in the driving style dimension according to the plurality of sample driving styles, so as to obtain the candidate driving intention prediction models corresponding to each of the plurality of sample driving styles.

[0113] In this scenario, the training sample of the candidate driving intention prediction model can be constructed according to the sample driving style data and marked as the second training sample.

[0114] Optionally, for any sample driving style, sample driving style intention data and sample driving style driving data under the sample driving style are obtained from the sample intention data and the sample driving data.

[0115] In the embodiments of the present disclosure, the sample intention data and the sample driving data include intention data and driving data under multiple sample driving styles. In this scenario, for any sample driving style, data belonging to the same style as the sample driving style can be obtained from the sample intention data and the sample driving data that can be collected from the sample vehicle, and used as the sample driving style intention data and the sample driving style driving data under the sample driving style.

[0116] Optionally, according to the sample driving style intention data and the sample driving style driving data, a second training sample of the candidate driving intention prediction model is obtained.

[0117] In the embodiments of the present disclosure, the sample driving style intention data and the sample driving style driving data can be processed by an algorithm according to the sample construction algorithm in the related art. Then, according to the result of the algorithm processing, a sample constructed from the sample driving style intention data and the sample driving style driving data is obtained, and it is used as the second training sample of the candidate driving intention prediction model.

[0118] Optionally, the candidate driving intention prediction model is trained with the second training sample until the training ends, and a trained target driving intention prediction model is obtained.

[0119] Among them, the second training sample can be input into the candidate driving intention prediction model to obtain a second training result output by the candidate driving intention prediction model.

[0120] In the embodiments of the present disclosure, the feature information of the second training sample can be extracted by the candidate driving intention prediction model, intention prediction is performed according to the extracted features, and the predicted result is used as the output result of the candidate driving intention prediction model. Among them, this output result can be determined as the second training result output by the candidate driving intention prediction model.

[0121] Optionally, a second sample label of the second training sample is obtained to obtain a second training loss of the second training result based on the second sample label. The parameters of the candidate driving intention prediction model are adjusted according to the second training loss, and the process returns to obtaining the next second training sample to continue model training on the candidate driving intention prediction model with adjusted parameters until the training ends, and a trained target driving intention prediction model is obtained.

[0122] In the embodiments of the present disclosure, the label information of the second training sample can be determined as the second sample label of the second training sample.

[0123] Further, perform loss value algorithm processing on the second sample label and the second training result according to the loss value acquisition algorithm in the related technology, and then obtain the loss value of the second training result based on the second sample label as the second training loss according to the result of the algorithm processing.

[0124] Optionally, the model parameters of the candidate driving intention prediction model can be adjusted according to the second training loss, and the next second training sample can be obtained to continue training the candidate driving intention prediction model with the adjusted parameters until the training ends.

[0125] Among them, the training end condition of the candidate driving intention prediction model can be set according to the training round, and the training end condition of the candidate driving intention prediction model can also be set according to the training output result, which is not specifically limited here.

[0126] Optionally, for the current training round, if the candidate driving intention prediction model of this round meets the preset training end condition, the model training of the candidate driving intention prediction model can be ended, and the model obtained after the end of the current round of training can be determined as the trained target driving intention prediction model.

[0127] It should be noted that after training the target driving intention prediction models for multiple sample driving styles, for the vehicles that need to configure the model, the model can be configured according to the driving style of the vehicle.

[0128] Optionally, for any sample driving style, obtain the target driving style of the target vehicle.

[0129] Among them, the vehicle that needs to configure the model can be marked as the target vehicle. In this scenario, the historical driving style data of the target vehicle can be obtained and analyzed, so as to obtain the driving style of the target vehicle according to the analysis result and mark it as the target driving style.

[0130] Optionally, in response to the target driving style matching the sample driving style, the target driving intention prediction model corresponding to the sample driving style is sent to the target vehicle.

[0131] In the embodiments of the present disclosure, for any sample driving style, when the target driving style of the target vehicle is the same as or similar to the sample driving style, it can be determined that the target driving style of the target vehicle matches the sample driving style.

[0132] In this scenario, the target driving intention prediction model under the sample driving style can be obtained, and the target driving intention prediction model is sent and configured to the target vehicle.

[0133] The present disclosure provides a training method for a driving intention prediction model, enabling the target driving intention prediction model to learn the relationship between driver behavior data and driving intentions, thereby realizing the prediction of a driver's driving intention. Compared with the related art method of only identifying driving intentions by starting the vehicle's turn signal, the accuracy and efficiency of driving intention recognition are improved, the possibility of potential safety hazards in the vehicle's driving due to unrecognized driving intentions is reduced, the performance of the vehicle's driving assistance function is optimized, the driving safety of the vehicle is improved, and the driving experience of users is optimized.

[0134] The embodiments of the present disclosure also propose a vehicle driving method, which can be combined with Figure 3 Understand, Figure 3 is a flowchart of the vehicle driving method according to an embodiment of the present disclosure. As Figure 3 shown, the method includes:

[0135] S301, obtain the target driving intention prediction model configured on the target vehicle.

[0136] Among them, the target driving intention prediction model is obtained based on the training method of the driving intention prediction model proposed in the above Figures 1 to 2 embodiment.

[0137] In the embodiments of the present disclosure, a vehicle configured with a target driving intention prediction model can be marked as the target vehicle. In this scenario, the target vehicle can predict the driving intention of the driver through its configured target driving intention prediction model.

[0138] S302, obtain the target driving style data of the target vehicle, as well as the target intention data and target driving data of the target vehicle within the target time range.

[0139] In the embodiments of the present disclosure, the historical driving data of the target vehicle can be obtained, the driving style of the target vehicle can be obtained through the analysis of the historical driving data, and this driving style can be determined as the target driving style of the target vehicle.

[0140] In this scenario, the relevant data of the target vehicle under this target driving style can be obtained, and this part of the data can be marked as the target driving style data of the target vehicle.

[0141] Optionally, the time range in which the target vehicle needs to perform driving intention recognition and prediction can be marked as the target time range. In this scenario, the intention data and driving data of the target vehicle within the target time range can be obtained and marked as the target intention data and target driving data within the target time range, respectively.

[0142] Among them, the target intention data may at least include the eye movement data, head torsion angle data, etc. of the driver of the target vehicle, and the target driving data may at least include the steering wheel angle, steering wheel rotation speed, steering wheel grip force, brake pedal stroke, brake pedal opening change rate, accelerator pedal stroke, accelerator pedal opening change rate, turn signal lever state, etc. of the target vehicle, as well as the vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, etc. of the target vehicle, which are not specifically limited here.

[0143] S303. Input the target intention data, target driving data, and target driving style into the target driving intention prediction model, and output the target predicted driving intention of the target vehicle through the target driving intention prediction model.

[0144] In the embodiments of the present disclosure, the target driving intention prediction model is a model configured on the target vehicle for predicting the driving intention of the driver. In this scenario, the obtained target intention data, target driving data, and target driving style data can be input into the target driving intention prediction model configured on the target vehicle.

[0145] Through the model capabilities of the target driving intention prediction model, based on the target intention data, target driving data, and target driving style data, predict the current driving intention of the driver of the target vehicle, and determine the output prediction result as the target predicted driving intention of the target vehicle.

[0146] Among them, the target predicted driving intention may be that the target vehicle has a lane-changing driving intention, or the target vehicle does not have a lane-changing driving intention.

[0147] S304. Perform driving control on the target vehicle based on the target predicted driving intention.

[0148] In the embodiments of the present disclosure, a lane-changing assistance driving module is provided on the target vehicle. In this scenario, when the target driving intention prediction model outputs the target predicted driving intention, the lane-changing assistance driving module can perform driving control on the target vehicle according to the output target predicted driving intention to achieve assisted driving for the driver.

[0149] Optionally, the hardware configuration of the target vehicle can be adjusted based on a pre-set driving control strategy to achieve driving control of the target vehicle, or the central control software system configured on the target vehicle can be adjusted, and the driving control of the target vehicle can be achieved through the central control software system of the target vehicle, which is not specifically limited here.

[0150] The vehicle driving method proposed by the present disclosure obtains the target driving intention prediction model configured on the target vehicle, and obtains the target driving style data of the target vehicle, as well as the target intention data and target driving data of the target vehicle within the target time range. Then, based on the target style data, target intention data, and target driving data, the target prediction driving intention of the target vehicle is obtained through the target driving intention prediction model. Furthermore, the driving of the target vehicle is controlled according to the target prediction driving intention. In the present disclosure, through Figures 1 to 2 The target driving intention prediction model obtained by the method of the embodiment is used to predict the driving intention of the target vehicle, which improves the accuracy and efficiency of the driving intention prediction of the target vehicle. In the scenario where the auxiliary driving function of the vehicle is started by recognizing the driving intention of the vehicle, the situation where the auxiliary driving function of the vehicle cannot be started due to the failure to recognize the driving intention of the driver is avoided, which improves the stability of the driving intention prediction function and the auxiliary driving function of the vehicle, and optimizes the driving experience of the user.

[0151] In the above embodiment, regarding the driving control of the target vehicle, it can be combined with Figure 4 For further understanding, Figure 4 is a schematic flowchart of the vehicle driving method according to another embodiment of the present disclosure. As Figure 4 shown, the method includes:

[0152] S401, in response to the target prediction driving intention being a lane change driving intention, the target vehicle is controlled to perform lane change driving through the turn signal control module and the lane change auxiliary driving module of the vehicle.

[0153] In the embodiment of the present disclosure, the target prediction driving intention of the target vehicle may be a lane change driving intention. In this scenario, when it is recognized that the target prediction driving intention of the target vehicle is a lane change driving intention, the driving of the target vehicle can be controlled by controlling relevant hardware devices such as the turn signal of the target vehicle, and the driving control of the target vehicle in this scenario is marked as lane change driving control.

[0154] As an example, as Figure 5 shown, the vehicle-end target driving intention prediction model collects the target intention data, target driving data, and target driving style data of the target vehicle through Figure 5 the data processing module shown, and transmits the collected data to Figure 5 the vehicle-end target driving intention prediction model shown, and predicts the driving intention of the target vehicle through Figure 5 the vehicle-end target driving intention prediction model shown, so as to output the target prediction driving intention.

[0155] In this example, when the target prediction driving intention is a lane change driving intention, it can be through Figure 5The shown turn signal control module turns on the turn signal of the target vehicle and, via Figure 5 the shown lane change assist driving module, assists the target vehicle in performing a lane change.

[0156] Among them, the lane change assist driving module can detect the driving blind area of the target vehicle after identifying the intention of the target vehicle to perform a lane change, so as to improve the driving safety of the target vehicle during the lane change process.

[0157] S402. In response to identifying that the lane change of the target vehicle has ended, the lane change control of the target vehicle is ended, and the target predicted driving intention output by the target driving intention prediction model is uploaded to the cloud.

[0158] In the embodiments of the present disclosure, when the lane change of the target vehicle ends, the lane change control of the target vehicle can be ended through Figure 5 the shown turn signal control module and the lane change assist driving module.

[0159] Among them, the turn signal of the target vehicle that has been turned on can be turned off through Figure 5 the shown turn signal control module, and the auxiliary driving function of Figure 5 the shown lane change assist driving module can be stopped, thereby ending the lane change control of the target vehicle.

[0160] Optionally, after the target vehicle ends the lane change, the target predicted driving intention output by the in-vehicle target driving intention prediction model can be uploaded to the cloud through the data transceiver module configured on the target vehicle. Among them, the data transceiver module can be the edge computing module configured on the target vehicle or other functional modules configured on the target vehicle that can perform data transceiver with the cloud server, and no specific limitation is made here.

[0161] It should be noted that there is a need to update the target driving intention prediction model configured on the target vehicle. Among them, it is possible to identify whether there is corresponding update data for the target driving intention prediction model. In response to identifying the existence of update data, the target driving intention prediction model is updated and adjusted according to the update data to obtain a new target driving intention prediction model.

[0162] In the embodiments of the present disclosure, the data generated by the cloud for updating the target driving intention prediction model configured on the vehicle side can be determined as the update data corresponding to the target driving intention prediction model.

[0163] As an example, as Figure 5 shown, it is set that Figure 5For the update and optimization of the model performed by the cloud target driving intention prediction model shown, in this example, due to the update and optimization of the cloud target driving intention prediction model, the vehicle-end target driving intention prediction model needs to follow the cloud target driving intention prediction model for update and optimization.

[0164] In this example, the optimization data of the cloud target driving intention prediction model based on the vehicle-end target driving intention prediction model can be determined as the update data of the vehicle-end target driving intention prediction model, and the update data is sent to the vehicle-end target driving intention prediction model through the data transceiver link between the two.

[0165] Furthermore, the vehicle-end target driving intention prediction model updates and optimizes the model based on the received update data, and determines the updated and optimized model as the new vehicle-end target driving intention prediction model.

[0166] In the vehicle driving method proposed by the present disclosure, when the target predicted driving intention is recognized as a lane-changing driving intention, lane-changing driving control is performed on the target vehicle, which avoids the situation where the assisted driving function of the vehicle cannot be started due to the failure to recognize the driver's driving intention, improves the stability of the vehicle's driving intention prediction function and the assisted driving function, and optimizes the user's driving experience.

[0167] Corresponding to the training methods of the driving intention prediction model proposed in the above several embodiments, an embodiment of the present disclosure also proposes a training device for the driving intention prediction model. Since the training device for the driving intention prediction model proposed in the embodiment of the present disclosure corresponds to the training methods of the driving intention prediction model proposed in the above several embodiments, the implementation manners of the above training methods of the driving intention prediction model are also applicable to the training device for the driving intention prediction model proposed in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0168] Figure 6 It is a schematic structural diagram of a training device for a driving intention prediction model according to an embodiment of the present disclosure. As Figure 6 shown, the training device 600 for the driving intention prediction model includes a first acquisition module 61, a first training module 62, and a second training module 63, where:

[0169] The first acquisition module 61 is configured to acquire an initial driving intention prediction model to be trained, and acquire sample intention data and sample driving data of a sample vehicle to obtain a first training sample of the initial driving intention prediction model;

[0170] The first training module 62 is configured to train the initial driving intention prediction model according to the first training sample until the training is completed to obtain a trained candidate driving intention prediction model;

[0171] The second training module 63 is configured to obtain sample driving style data of a sample vehicle and adjust a candidate driving intention prediction model according to the sample driving style data, so as to obtain an adjusted target driving intention prediction model.

[0172] In an embodiment of the present disclosure, the first acquisition module 61 is further configured to: perform lane change intention recognition on the sample intention data to obtain sample lane change intention data and sample non-lane change intention data from the sample intention data; obtain sample lane change driving data corresponding to the sample lane change intention data and sample non-lane change driving data corresponding to the sample non-lane change intention data from the sample driving data; obtain a first candidate sample according to the sample lane change intention data and the corresponding sample lane change driving data, and obtain a second candidate sample according to the sample non-lane change intention data and the corresponding sample non-lane change driving data; and obtain a first training sample according to the first candidate sample and the second candidate sample.

[0173] In an embodiment of the present disclosure, the first training module 62 is further configured to: input the first training sample into an initial driving intention prediction model to obtain a first training result output by the initial driving intention prediction model; obtain a first sample label of the first training sample to obtain a first training loss of the first training result based on the first sample label; adjust parameters of the initial driving intention prediction model according to the first training loss, and return to obtain the next first training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training ends, so as to obtain a trained candidate driving intention prediction model.

[0174] In an embodiment of the present disclosure, the second training module 63 is further configured to: obtain a plurality of preset sample driving styles and initial driving style data of the sample vehicle, and perform style division on the initial driving style data according to the plurality of sample driving styles to obtain sample driving style data of each of the plurality of sample driving styles; for any one of the sample driving styles, obtain a candidate driving intention prediction model corresponding to the sample driving style, and generate a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style; and train the candidate driving intention prediction model through the second training sample until the training ends, so as to obtain a trained target driving intention prediction model.

[0175] In an embodiment of the present disclosure, the second training module 63 is further configured to: for any one of the sample driving styles, obtain sample driving style intention data and sample driving style driving data under the sample driving style from the sample intention data and the sample driving data; and obtain a second training sample of the candidate driving intention prediction model according to the sample driving style intention data and the sample driving style driving data.

[0176] In an embodiment of the present disclosure, the second training module 63 is further configured to: input the second training sample into the candidate driving intention prediction model to obtain a second training result output by the candidate driving intention prediction model; obtain a second sample label of the second training sample to obtain a second training loss of the second training result based on the second sample label; adjust the parameters of the candidate driving intention prediction model according to the second training loss, and return to obtain the next second training sample to continue model training on the candidate driving intention prediction model with adjusted parameters until the training ends, so as to obtain a trained target driving intention prediction model.

[0177] In an embodiment of the present disclosure, the system further includes a distribution module, configured to: for any sample driving style, obtain the target driving style of the target vehicle; in response to the target driving style matching the sample driving style, distribute the target driving intention prediction model corresponding to the sample driving style to the target vehicle.

[0178] The present disclosure provides a training device for a driving intention prediction model, which obtains an initial driving intention prediction model to be trained, as well as a first training sample constructed from sample intention data and sample driving data, and performs model training on the initial driving intention prediction model through the first training sample to obtain a candidate driving intention prediction model. Optionally, sample driving style data of a sample vehicle is obtained and the candidate driving intention prediction model is adjusted according to the sample driving style data to obtain a target driving intention prediction model. In the present disclosure, through the training of the initial driving intention prediction model and the adjustment of the candidate driving intention prediction model, a target driving intention prediction model is obtained, enabling the target driving intention prediction model to learn the relationship between driver behavior data and driving intentions, thereby realizing the prediction of the driver's driving intentions. Compared with the method in the related art that only identifies driving intentions by starting the vehicle's turn signal, the accuracy and efficiency of driving intention recognition are improved, the possibility of potential safety hazards in the vehicle's driving due to unrecognized driving intentions is reduced, the performance of the vehicle's driving assistance function is optimized, the driving safety of the vehicle is improved, and the driving experience of the user is optimized.

[0179] Corresponding to the vehicle driving methods proposed in the above several embodiments, an embodiment of the present disclosure also provides a vehicle driving device. Since the vehicle driving device proposed in the embodiment of the present disclosure corresponds to the vehicle driving methods proposed in the above several embodiments, the implementation manners of the above vehicle driving methods are also applicable to the vehicle driving device proposed in the embodiment of the present disclosure and will not be described in detail in the following embodiments.

[0180] Figure 7 is a schematic structural diagram of a vehicle driving device according to an embodiment of the present disclosure, as Figure 7 shown, the vehicle driving device 700 includes a second acquisition module 71, a third acquisition module 72, a prediction module 73, and a control module 74, where:

[0181] A second acquisition module 71, configured to acquire a target driving intention prediction model configured on a target vehicle, where the target driving intention prediction model is obtained based on the training device for the driving intention prediction model proposed in the foregoing Figure 6 embodiment;

[0182] A third acquisition module 72, configured to acquire target driving style data of the target vehicle, as well as target intention data and target driving data of the vehicle within a target time range;

[0183] A prediction module 73, configured to input the target intention data, the target driving data, and the target driving style data into the target driving intention prediction model, and output a target predicted driving intention of the target vehicle through the target driving intention prediction model;

[0184] A control module 74, configured to perform driving control on the target vehicle based on the target predicted driving intention.

[0185] In an embodiment of the present disclosure, the device further includes an update module, configured to: identify whether there is corresponding update data for the target driving intention prediction model; in response to identifying that there is update data, update and adjust the target driving intention prediction model according to the update data to obtain a new target driving intention prediction model.

[0186] In an embodiment of the present disclosure, the control module 74 is further configured to: in response to the target predicted driving intention being a lane change driving intention, perform lane change driving control on the target vehicle through a turn signal control module and a lane change assist module of the vehicle.

[0187] In an embodiment of the present disclosure, the control module 74 is further configured to: in response to identifying that the lane change driving of the target vehicle ends, end the lane change driving control of the target vehicle, and upload the target predicted driving intention output by the target driving intention prediction model to the cloud.

[0188] The vehicle driving device proposed in the present disclosure acquires a target driving intention prediction model configured on a target vehicle, as well as target driving style data of the target vehicle and target intention data and target driving data of the target vehicle within a target time range, and then obtains a target predicted driving intention of the target vehicle through the target driving intention prediction model according to the target style data, the target intention data, and the target driving data, and then performs driving control on the target vehicle according to the target predicted driving intention. In the present disclosure, through Figures 1 to 2The target driving intention prediction model obtained by the method of the embodiment is used to predict the driving intention of the target vehicle, which improves the accuracy and efficiency of the driving intention prediction of the target vehicle. In the scenario where the auxiliary driving function of the vehicle is started by identifying the driving intention of the vehicle, the situation where the auxiliary driving function of the vehicle cannot be started due to the failure to identify the driving intention of the driver is avoided, the stability of the driving intention prediction function and the assisted driving function of the vehicle is improved, and the driving experience of the user is optimized.

[0189] To achieve the above embodiment, the present disclosure also provides a vehicle, wherein the vehicle is used to implement the training method of the driving intention prediction model and / or the vehicle driving method proposed in the above embodiment.

[0190] To achieve the above embodiment, the present disclosure also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0191] Figure 8 The block diagram of the electronic device 800 according to an embodiment of the present disclosure is shown in Figure 8 As shown, the electronic device 800 includes a memory 801, a processor 802, and a computer program stored in the memory 801 and executable on the processor 802. When the processor 802 executes the program instructions, the training method of the driving intention prediction model and / or the vehicle driving method provided in the above embodiment are implemented.

[0192] For the training method of the driving intention prediction model proposed by the present disclosure, a first set of lane change evaluation parameters between a first vehicle on a first lane and a second vehicle on a second lane, and a second set of lane change evaluation parameters between the second vehicle and a third vehicle on the first lane are obtained, and based on the first set of lane change evaluation parameters and / or the second set of lane change evaluation parameters, it is identified whether the second vehicle meets the corresponding target lane change condition, and when it is identified that the second vehicle meets the target lane change condition, the second vehicle is controlled to change lanes from the second lane to the first lane. In the present disclosure, whether the second vehicle meets the target lane change condition is evaluated through the first set of lane change evaluation parameters and / or the second set of lane change evaluation parameters, and the evaluation is based on multiple parameters between the first vehicle and the second vehicle and / or multiple parameters between the second vehicle and the third vehicle, which increases the number of parameters used in the evaluation process, expands the evaluation parameter range, improves the accuracy of the evaluation result of whether the second vehicle meets the target lane change condition, and further improves the accuracy of the determination result of whether the second vehicle can change lanes, optimizes the safety of the vehicle during lane change driving, and further optimizes the driving experience of the user.

[0193] The vehicle driving method proposed by the present disclosure involves obtaining the target driving intention prediction model configured on the target vehicle, as well as obtaining the target driving style data of the target vehicle, the target intention data, and the target driving data of the target vehicle within the target time range. Then, based on the target style data, target intention data, and target driving data, the target predicted driving intention of the target vehicle is obtained through the target driving intention prediction model, and then the target vehicle is controlled for driving according to the target predicted driving intention. In the present disclosure, through Figures 1 to 2 The target driving intention prediction model obtained by the method of the embodiment is used to predict the driving intention of the target vehicle, improving the accuracy and efficiency of the driving intention prediction of the target vehicle. In the scenario where the auxiliary driving function of the vehicle is activated by recognizing the driving intention of the vehicle, it avoids the situation where the auxiliary driving function of the vehicle cannot be activated due to the failure to recognize the driving intention of the driver, improves the stability of the driving intention prediction function and the assisted driving function of the vehicle, and optimizes the driving experience of the user.

[0194] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0195] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0196] The program code for implementing the methods per se can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.

[0197] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0198] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0199] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), the Internet, and blockchain network.

[0200] A computer system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services (“Virtual Private Server”, or simply “VPS”). The server can also be a server of a distributed system, or a server combined with blockchain.

[0201] In the description of this specification, descriptions with reference to the terms “one embodiment”, “some embodiments”, “example”, “specific example”, or “some examples” etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, without conflict, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples.

[0202] In addition, the terms “first” and “second” are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with “first” and “second” can explicitly or implicitly include at least one of the features. In the description of the present disclosure, “a plurality of” means at least two, such as two, three, etc., unless otherwise specifically defined.

[0203] Any process or method description represented in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may be executed not in the order shown or discussed, including in a substantially simultaneous manner according to the functions involved or in a reverse order, which should be understood by those skilled in the art to which the embodiments of the present disclosure pertain.

[0204] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered a sequenced list of executable instructions for implementing a logical function and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable medium on which a program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpretation, or otherwise appropriate processing if necessary, and then stored in a computer memory.

[0205] It should be understood that various parts of the present disclosure can be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: discrete logic circuits having logic gates for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0206] Those of ordinary skill in the art can understand that all or part of the steps carried out in implementing the above-described embodiment methods can be completed by a program instructing relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiment.

[0207] In addition, in each of the various embodiments of the present disclosure, the functional units can be integrated into one processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0208] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

[0209] It should be understood that various forms of the processes shown above can be used, reordering, adding, or deleting steps. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved. This is not limited herein.

[0210] The above specific embodiments do not constitute a limitation to the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present disclosure should be included within the protection scope of the present disclosure.

Claims

1. A training method for a driving intention prediction model, characterized in that, The method includes: Obtaining an initial driving intention prediction model to be trained, and obtaining sample intention data and sample driving data of a sample vehicle to obtain a first training sample of the initial driving intention prediction model; Training the initial driving intention prediction model according to the first training sample until the training ends to obtain a trained candidate driving intention prediction model; Obtaining sample driving style data of the sample vehicle, and adjusting the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model; The obtaining sample intention data and sample driving data of the sample vehicle to obtain a first training sample of the initial driving intention prediction model includes: Performing lane change intention recognition on the sample intention data to obtain sample lane change intention data and sample non-lane change intention data from the sample intention data; From the sample driving data, obtaining sample lane change driving data corresponding to the sample lane change intention data and sample non-lane change driving data corresponding to the sample non-lane change intention data; Constructing a training positive sample of the initial driving intention prediction model according to the sample lane change intention data and the corresponding sample lane change driving data, and marking it as a first candidate sample, and constructing a training negative sample of the initial driving intention prediction model according to the sample non-lane change intention data and the corresponding sample non-lane change driving data, and marking it as a second candidate sample; Obtaining the first training sample according to the first candidate sample and the second candidate sample.

2. The method according to claim 1, wherein The training the initial driving intention prediction model according to the first training sample until the training ends to obtain a trained candidate driving intention prediction model includes: Inputting the first training sample into the initial driving intention prediction model to obtain a first training result output by the initial driving intention prediction model; Obtaining a first sample label of the first training sample to obtain a first training loss of the first training result based on the first sample label; Adjusting parameters of the initial driving intention prediction model according to the first training loss, and returning to obtain the next first training sample to continue model training on the initial driving intention prediction model with adjusted parameters until the training ends to obtain the trained candidate driving intention prediction model.

3. The method according to claim 1, characterized in that, The obtaining sample driving style data of the sample vehicle, and adjusting the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model includes: Obtaining a preset plurality of sample driving styles and initial driving style data of the sample vehicle, and performing style division on the initial driving style data according to the plurality of sample driving styles to obtain sample driving style data of each of the plurality of sample driving styles; For any sample driving style, obtaining the candidate driving intention prediction model corresponding to the sample driving style, and generating a second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style; Train the candidate driving intention prediction model with the second training sample until the training is completed, and obtain the trained target driving intention prediction model.

4. The method according to claim 3, wherein For any sample driving style, obtaining the candidate driving intention prediction model corresponding to the sample driving style, and generating the second training sample of the candidate driving intention prediction model according to the sample driving style data under the sample driving style, includes: For any sample driving style, obtain the sample driving style intention data and sample driving style driving data under the sample driving style from the sample intention data and the sample driving data; According to the sample driving style intention data and the sample driving style driving data, obtain the second training sample of the candidate driving intention prediction model.

5. The method according to claim 3, wherein The training of the candidate driving intention prediction model with the second training sample until the training is completed to obtain the trained target driving intention prediction model includes: Input the second training sample into the candidate driving intention prediction model to obtain the second training result output by the candidate driving intention prediction model; Obtain the second sample label of the second training sample to obtain the second training loss of the second training result based on the second sample label; Adjust the parameters of the candidate driving intention prediction model according to the second training loss, and return to obtain the next second training sample to continue the model training for the candidate driving intention prediction model with adjusted parameters until the training is completed, and obtain the trained target driving intention prediction model.

6. The method according to any one of claims 3-5, characterized in that, After training the candidate driving intention prediction model according to the second training sample until the training is completed to obtain the trained target driving intention prediction model, includes: For any sample driving style, obtain the target driving style of the target vehicle; In response to the target driving style matching the sample driving style, send the target driving intention prediction model corresponding to the sample driving style to the target vehicle.

7. A vehicle driving method, characterized in that, The method includes: Obtain the target driving intention prediction model configured on the target vehicle, where the target driving intention prediction model is obtained based on the training method of the driving intention prediction model according to any one of the above claims 1-6; Obtain the target driving style data of the target vehicle, as well as the target intention data and target driving data of the vehicle within the target time range; Input the target intention data, the target driving data, and the target driving style data into the target driving intention prediction model, and output the target predicted driving intention of the target vehicle through the target driving intention prediction model; Based on the target predicted driving intention, perform driving control on the target vehicle.

8. The method according to claim 7, wherein The method further includes: Identify whether there is corresponding update data for the target driving intention prediction model; In response to identifying the existence of the update data, update and adjust the target driving intention prediction model according to the update data to obtain a new target driving intention prediction model.

9. The method according to claim 7, wherein The performing driving control on the vehicle based on the target predicted driving intention includes: In response to the target predicted driving intention being a lane change driving intention, the lane change driving control of the target vehicle is performed through the turn signal control module and the lane change assist module of the vehicle.

10. The method according to claim 9, wherein After performing the driving control on the vehicle based on the target driving intention, it includes: In response to recognizing the end of the lane change driving of the target vehicle, the lane change driving control of the target vehicle is ended, and the target predicted driving intention output by the target driving intention prediction model is uploaded to the cloud.

11. A training device for a driving intention prediction model, characterized in that, The device includes: A first acquisition module, configured to acquire an initial driving intention prediction model to be trained, and acquire sample intention data and sample driving data of a sample vehicle to obtain a first training sample of the initial driving intention prediction model; A first training module, configured to train the initial driving intention prediction model according to the first training sample until the training is completed to obtain a trained candidate driving intention prediction model; A second training module, configured to acquire sample driving style data of the sample vehicle, and adjust the candidate driving intention prediction model according to the sample driving style data to obtain an adjusted target driving intention prediction model; The acquisition of the sample intention data and sample driving data of the sample vehicle to obtain the first training sample of the initial driving intention prediction model includes: Performing a lane change intention recognition on the sample intention data to obtain sample lane change intention data and sample non-lane change intention data from the sample intention data; From the sample driving data, acquiring sample lane change driving data corresponding to the sample lane change intention data and sample non-lane change driving data corresponding to the sample non-lane change intention data; According to the sample lane change intention data and the corresponding sample lane change driving data, constructing a training positive sample of the initial driving intention prediction model and marking it as a first candidate sample, and according to the sample non-lane change intention data and the corresponding sample non-lane change driving data, constructing a training negative sample of the initial driving intention prediction model and marking it as a second candidate sample; According to the first candidate sample and the second candidate sample, obtaining the first training sample.

12. A vehicle driving device, characterized in that, The device includes: A second acquisition module, configured to acquire a target driving intention prediction model configured on a target vehicle, where the target driving intention prediction model is obtained based on the training device of the driving intention prediction model according to claim 11 above; A third acquisition module, configured to acquire target driving style data of the target vehicle, as well as target intention data and target driving data of the vehicle within a target time range; A prediction module, configured to input the target intention data, the target driving data, and the target driving style data into the target driving intention prediction model, and output the target predicted driving intention of the target vehicle through the target driving intention prediction model; A control module, configured to perform driving control on the target vehicle based on the target predicted driving intention.

13. A vehicle, characterized in that, The vehicle is used to implement the method according to any one of claims 1-6 and / or claims 7-10 above.

14. An electronic device, characterized in that, It includes: A processor; A memory for storing executable instructions of a processor; Wherein, the processor is configured to execute instructions to implement the method described in any one of claims 1-6 and / or claims 7-10.

15. A computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to execute the method described in any one of claims 1-6 and / or claims 7-10.

Images