Vehicle driving control method and device and computer readable storage medium

Abstract

The invention provides a vehicle driving control method and device and a computer readable storage medium, in the vehicle driving process, the current terrain is sensed to obtain vehicle end sensing information, the cloud reference information of the current terrain is obtained from the cloud based on the current position of the vehicle, and the vehicle end sensing information and the cloud reference information are arbitrated to obtain the vehicle driving control information. And the vehicle end sensing accuracy is improved. After the arbitrated target information is determined, the vehicle driving parameters adapting to the current terrain are determined based on the accurate target information, a driver is helped to better adapt to the current road condition, and the driving comfort and safety are effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle driving performance, and in particular to a vehicle driving control method, device and computer-readable storage medium. Background Art

[0002] In the process of driving a car, the driver's adaptability to road conditions and driving comfort and safety are crucial. With the development of technology, terrain mode and suspension adjustment technology have emerged, which have played a positive role in assisting drivers to cope with different terrains and road conditions, but there are also some disadvantages that cannot be ignored.

[0003] Terrain mode technology provides drivers with a variety of preset modes, such as sports mode, snow mode, mud mode, sand mode, etc. These modes can adjust the vehicle's torque output and driving force distribution according to different road types to better adapt to current road conditions. However, the current terrain mode technology has obvious problems in automatically identifying terrain. It mainly relies on the perception of the ADAS system (Advanced Driving Assistance System) or the analysis of vehicle-side power parameters. This recognition method has a large error. For example, on roads with complex alternating terrain, the vehicle may fail to adjust to the appropriate mode in time due to recognition errors, thereby affecting driving comfort and safety, and causing trouble to the driver.

[0004] Suspension adjustment technology is also the key to improving the driving experience. It adapts to different driving conditions and improves the vehicle's driving performance by adjusting components such as shock absorbers, spring preload, stabilizer bars, or using adjustable suspension systems. However, there are many problems with suspension adjustment at present. On the one hand, the mainstream suspension parameter adjustment mostly relies on the driver's experience and judgment, which is a big challenge for the driver. The driver needs to judge the road conditions and adjust them manually during driving, which is not only inconvenient to operate, but also may affect driving comfort and safety if the judgment is wrong. On the other hand, the existing suspension adjustment algorithm is relatively rigid, and its adaptability to complex terrain is limited. It cannot fully meet the requirements for driving comfort when driving outdoors, which reduces the driver's driving experience under complex road conditions.

[0005] Therefore, a new solution is urgently needed to solve the problem of poor existing driving experience. Summary of the invention

[0006] To overcome the problems existing in the related art, this specification provides a vehicle driving control method, device and computer-readable storage medium.

[0007] According to a first aspect of an embodiment of this specification, a vehicle driving control method is provided, which is applied to a vehicle-side controller, and the method includes:

[0008] During the driving process of the vehicle, the current terrain is sensed to obtain vehicle-side perception information;

[0009] Based on the current position of the vehicle, obtaining cloud reference information reflecting the current terrain from historical terrain data pre-stored in the cloud server;

[0010] In the case where the vehicle-side perception information and the cloud-side reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud-side reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud-side reference information to determine the target information;

[0011] Based on the target information, a vehicle driving parameter adapted to the current terrain is determined for controlling the travel of the vehicle.

[0012] According to a vehicle driving control method provided by the present application, during the vehicle driving process, the current terrain is sensed to obtain vehicle-side sensing information, including:

[0013] During the driving process of the vehicle, the current terrain is perceived based on the terrain perception model to obtain vehicle-side perception information;

[0014] The terrain perception model is trained based on historical terrain data pre-stored in the cloud server.

[0015] According to a vehicle driving control method provided by the present application, the vehicle-side perception information includes first terrain mode information, and the cloud-side reference information includes second terrain mode information.

[0016] When the vehicle-side perception information and the cloud reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information to determine the target information includes:

[0017] arbitrating the first terrain pattern information and the second terrain pattern information based on the confidence of the first terrain pattern information and the confidence of the second terrain pattern information to determine target information;

[0018] The determining of the vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle, comprises:

[0019] Based on the target information, determining a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle;

[0020] Based on the vehicle driving parameters, the driving of the vehicle is controlled to adapt to the current terrain.

[0021] According to a vehicle driving control method provided by the present application, the target information includes target terrain pattern information.

[0022] The determining, based on the target information, a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle comprises:

[0023] Acquiring third terrain pattern information obtained by identifying the current terrain according to vehicle power parameters of the vehicle; the vehicle power parameters are used to reflect vehicle movement;

[0024] Based on the confidence level of the third terrain pattern information and the confidence level of the target information, arbitrating the third terrain pattern information and the target information to determine the target terrain pattern information;

[0025] A target terrain mode and a vehicle driving parameter corresponding to the target terrain mode are determined from the all-terrain mode of the vehicle according to the target terrain mode information.

[0026] According to a vehicle driving control method provided by the present application, the vehicle-side perception information includes first terrain feature information, and the cloud-side reference information includes second terrain feature information.

[0027] When the vehicle-side perception information and the cloud reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information to determine the target information includes:

[0028] In the case where the first terrain feature information and the second terrain feature information are inconsistent, arbitrating the first terrain feature information and the second terrain feature information based on the confidence level of the first terrain feature information and the confidence level of the second terrain feature information to determine target information;

[0029] The determining of the vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle, comprises:

[0030] Acquire a road bump feature model from the cloud server;

[0031] Determining vehicle driving parameters required for the vehicle to pass through the current terrain based on the road bump feature model and the target information data;

[0032] A suspension system of the vehicle is adjusted based on the vehicle driving parameters to control the vehicle to travel on a current terrain.

[0033] According to a vehicle driving control method provided by the present application, the method for constructing the road bump feature model comprises the following steps:

[0034] Acquire vehicle-side data when a preceding vehicle passes through a bumpy road, the vehicle-side data including vehicle-side status data and terrain feature information, and vehicle suspension system parameters obtained by analyzing the vehicle-side status data and terrain feature information;

[0035] The acquired vehicle-side data is subjected to correlation analysis to create a road bump feature model.

[0036] According to a vehicle driving control method provided by the present application, after acquiring cloud reference information reflecting the current terrain from historical terrain data pre-stored in a cloud server based on the current position of the vehicle, the method further includes:

[0037] Acquire a layer for representing current terrain features from the cloud server, wherein the layer is determined by the cloud server from a plurality of pre-stored layers according to the terrain features reflected by the cloud reference information;

[0038] The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

[0039] According to a vehicle driving control method provided by the present application, when the vehicle-side perception information and the cloud reference information are inconsistent, the vehicle-side perception information and the cloud reference information are arbitrated based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information. After determining the target information, the method further includes:

[0040] Determining a layer for representing current terrain features from a plurality of pre-stored layers according to the terrain features reflected by the target information;

[0041] The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

[0042] According to a vehicle driving control method provided by the present application, when the vehicle-side perception information and the cloud reference information are inconsistent, the vehicle-side perception information and the cloud reference information are arbitrated based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information. After determining the target information, the method further includes:

[0043] Generate a layer for representing current terrain features according to the target information;

[0044] The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

[0045] According to a vehicle driving control method provided by the present application, the layer includes at least one of a terrain layer and a bumpy layer.

[0046] The present application also provides a vehicle driving control device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the vehicle driving control method described above is implemented.

[0047] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the vehicle driving control method as described in any one of the above is implemented.

[0048] Compared with the current poor driving experience, the vehicle driving control method, device and computer-readable storage medium in the embodiments of this specification sense the current terrain to obtain vehicle-side perception information during vehicle driving, and obtain cloud reference information of the current terrain from the cloud based on the current position of the vehicle, and arbitrate the vehicle-side perception information and cloud reference information to improve the accuracy of vehicle-side perception. After determining the target information after arbitration, the vehicle driving parameters adapted to the current terrain are determined based on the accurate target information, helping the driver to better adapt to the current road conditions and effectively improving driving comfort and safety.

[0049] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the specification and, together with the description, serve to explain the principles of the specification.

[0051] Figure 1 is a flow chart of a vehicle driving control method according to an exemplary embodiment of the present specification;

[0052] Figure 2 is an all-terrain mode switching chain diagram shown in this specification according to an exemplary embodiment;

[0053] Figure 3 is a suspension automatic adjustment link diagram shown in this specification according to an exemplary embodiment;

[0054] Figure 4 It is a navigation map rendering link diagram shown in this specification according to an exemplary embodiment;

[0055] Figure 5 is a schematic diagram of a vehicle driving control device according to an exemplary embodiment of the present specification;

[0056] Figure 6 It is a schematic block diagram of a vehicle driving control device according to an exemplary embodiment of the present specification. DETAILED DESCRIPTION

[0057] Here, the technical solutions in the embodiments (or "implementations") of the present application will be described clearly and completely in conjunction with the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0058] If there are terms involving directional indications or positional relationships in the embodiments of the present application (such as up, down, left, right, front, back, inside, outside, top, bottom, center, vertical, horizontal, longitudinal, transverse, length, width, counterclockwise, clockwise, axial, radial, circumferential, etc.), such terms are only used to explain the relative positional relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings); if the specific posture changes, the directional indication or positional relationship will also change accordingly. In addition, the terms "first", "second", etc. involved in the embodiments of the present application are only used for the purpose of convenience of description and cannot be understood as indicating or implying relative importance.

[0059] The present application provides a vehicle driving control method, device and computer readable storage medium. The present application is described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the features of the following embodiments and implementations can be combined with each other.

[0060] When driving a car, the road conditions that drivers face are complex and changeable, which has a critical impact on the driving experience. From urban roads for daily commuting, to ordinary highways connecting urban and rural areas, to unpaved roads in remote areas, and even extremely harsh special road conditions, each one brings unique challenges to driving.

[0061] When dealing with these complex road conditions, terrain mode switching and suspension adjustment become two key factors affecting the driving experience. Terrain mode is designed to adjust the vehicle's power system, transmission system, etc. according to different terrain characteristics. Suspension adjustment also plays an important role in the driving experience. It adapts to the ups and downs of the road by changing the stiffness, damping and other parameters of the suspension.

[0062] Although all-terrain mode and suspension adjustment technology can help drivers better adapt to current road conditions and improve driving comfort and safety to a certain extent, the current technology still has the following shortcomings:

[0063] (1) The terrain pattern information recognition of the automatic all-terrain function is mainly achieved through ADAS system perception or vehicle-side power parameter analysis, and the analysis results have certain errors;

[0064] (2) The current mainstream suspension parameter adjustment function relies on the driver's manual control, which tests the driver's experience and judgment;

[0065] (3) The mainstream suspension adjustment algorithm is rigid and has low adaptability to complex terrain, which cannot meet the requirements of outdoor driving for driving comfort.

[0066] In order to solve the above technical problems, this specification provides a vehicle driving control method.

[0067] It aims to use the swarm intelligence of the cloud and vehicle-side integration, rely on the results of vehicle-side perception and cloud-side recognition for arbitration, and finally determine accurate target information through multi-party arbitration. This target information is used to determine the vehicle driving parameters that adapt to the current terrain, making the vehicle run more smoothly and improving the driving experience, in order to solve the problem of poor driving experience caused by inaccurate recognition of existing all-terrain modes and inconvenient suspension adjustment.

[0068] It should be noted that the vehicle driving control method is applied to a vehicle-side controller (hereinafter referred to as the vehicle-side).

[0069] Figure 1 This is a flow chart of a vehicle driving control method provided by an embodiment of this specification, comprising the following steps:

[0070] Step 100, during the driving process of the vehicle, the current terrain is sensed to obtain vehicle-side sensing information.

[0071] The acquisition of vehicle-side perception information mainly relies on the vehicle-side terrain perception model and the real-time perception data of various sensors.

[0072] As an example, the sensor includes but is not limited to the camera, laser radar or millimeter wave radar on the vehicle. The laser radar or millimeter wave radar installed near the chassis of the vehicle can emit electromagnetic waves and receive reflected waves, and detect terrain information such as the ups and downs of the road surface, potholes, and the distance to surrounding obstacles by analyzing the characteristics of the reflected waves. The camera can capture images in front of, to the side of, or even behind the vehicle, and use image recognition technology to identify the texture, color, signs, and general outline of the terrain of the road, such as determining whether it is a smooth asphalt road, a cracked cement road, or a potholed dirt road.

[0073] The real-time perception data collected by these sensors will be transmitted to the vehicle-side controller. After complex data fusion, algorithm processing and model analysis, it will eventually form vehicle-side perception information, which can more comprehensively describe the characteristics and conditions of the terrain where the vehicle is currently located.

[0074] In some embodiments, step 100 may specifically include the following steps:

[0075] Step 110, during the driving process of the vehicle, the current terrain is sensed based on the terrain perception model to obtain vehicle-side perception information;

[0076] The terrain perception model is trained based on historical terrain data pre-stored in the cloud server.

[0077] The terrain perception model is trained by the vehicle-side data uploaded by the previous vehicle. Specifically, the vehicle-side data of the previous vehicle is uploaded to the cloud server (hereinafter referred to as the cloud) as basic data support. The vehicle-side data includes vehicle status data, driving control data, vehicle location information, chassis image data and ADAS (Advanced Driving Assistance System) perception data. The cloud processes the vehicle-side data based on algorithms such as wavelet transform, Fourier transform, and clustering.

[0078] Based on the processed vehicle-side data, the cloud extracts terrain features based on the NN (Neural Network) algorithm, trains the terrain perception model, and updates the terrain perception model to the vehicle-side through OTA (Over-the-Air Technology).

[0079] The terrain perception model can be dynamically updated based on vehicle-side data. The algorithm is evolutionary, continuously improving accuracy and precision, and can adapt to complex outdoor road conditions.

[0080] Step 200 , based on the current position of the vehicle, obtain cloud reference information reflecting the current terrain from historical terrain data pre-stored in the cloud server.

[0081] The cloud has a huge amount of historical terrain data. These data are accumulated by collecting, organizing and analyzing data uploaded by a large number of previous vehicles during their driving at different times and in different regions. They are historical data and serve as a reference for vehicle-side perception and identification.

[0082] The cloud will filter out the most relevant and matching historical terrain data based on the vehicle's location information, and send this data back to the vehicle as cloud reference information. The advantage of cloud reference information is that it can provide a more macro and comprehensive terrain perspective, because it integrates the experience data of many vehicles passing through the area at different times, which helps the vehicle understand the terrain conditions it will face in advance and make better preparations.

[0083] Of course, in the early stages of creating a terrain perception model, the data used to train and create the model is also historical terrain data. In this scenario, the terrain perception model and the cloud reference information given by the cloud may be consistent.

[0084] However, terrain features may change due to natural or man-made reasons, such as the alternation of seasons, where a grassland area in spring may become a snowy area in winter; or construction projects may cause terrain changes. Therefore, it is still necessary to use the current vehicle's terrain perception model, vehicle-side perception results, and vehicle-side power parameter arbitration to obtain more accurate results, while also continuously updating and optimizing cloud data. This part will be explained in detail in subsequent chapters.

[0085] Step 300, when the vehicle-side perception information and the cloud-side reference information are inconsistent, arbitrate the vehicle-side perception information and the cloud-side reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud-side reference information to determine the target information.

[0086] When terrain features change but the terrain perception model and cloud data are not updated synchronously, there will be inconsistencies between the vehicle-side perception information and the cloud-side reference information. Therefore, it is necessary to arbitrate the vehicle-side perception information and the cloud-side reference information to determine more reliable target information.

[0087] In some embodiments, according to the confidence of the vehicle-side perception information and the confidence of the cloud reference information, the vehicle-side perception information and the cloud reference information are arbitrated to determine the target information. The target information includes information for feedback of the terrain mode and terrain features of the current terrain.

[0088] As an example, the confidence of vehicle-side perception information mainly depends on the matching degree between vehicle-side terrain feature recognition and pre-training. Specifically, the vehicle-side terrain perception model is trained and constructed by a large number of pictures with clear terrain features (such as sand, mud, grass, snow, etc.).

[0089] During actual vehicle driving, the camera on the vehicle side will collect images of the surrounding terrain. These image data will be transmitted to the terrain perception model, and the model will compare and match the collected image features with various pre-trained terrain features. The confidence level is determined by the degree of match between the terrain features contained in the image and a specific terrain type (such as grass). For example, suppose the terrain perception model analyzes the collected image and finds that 70% of the image content meets the characteristics of grass terrain and 30% meets the characteristics of mud terrain. Then, for the terrain recognition result of grass, its confidence level is determined to be 70%.

[0090] The confidence of the terrain pattern information in the cloud depends on the breadth of the data source, the timeliness of the data, and the reliability of the data processing process. If the historical terrain data of a certain area comes from a large number of vehicles and covers a variety of driving conditions, then the confidence of the cloud reference information in this area is relatively high. However, if the data is outdated or the number of previous vehicles is small, such as if the area has recently undergone large-scale road construction or terrain transformation, but the cloud data has not been updated in time, then its confidence will be reduced.

[0091] During the arbitration process, when the confidence level of the vehicle-side perception information is higher than the set threshold (such as 70%), the vehicle-side perception result will be used as the target information for the final vehicle control. When the confidence level of the cloud reference information is higher than the set threshold (such as 70%), the cloud reference result will be used as the target information for the final vehicle control.

[0092] In other examples, the arbitration process will comprehensively consider the confidence weights of the vehicle-side perception information and the cloud-based reference information. For example, if the confidence of the vehicle-side perception information is higher, while the confidence of the cloud-based reference information is lower, then the vehicle-side perception information will be more inclined when determining the target information; conversely, if the confidence of the cloud-based reference information has a clear advantage, more reference will be made to the cloud-based data. Through this weighted average or other reasonable arbitration algorithms, the more reliable target information in the current situation is ultimately determined to ensure that the vehicle can make subsequent driving parameter adjustments based on relatively accurate terrain information. This part is explained in detail in the subsequent chapters.

[0093] In other embodiments, if the vehicle-side perception information is inconsistent with the cloud-side reference information, the current data (the currently perceived vehicle-side perception data and / or the cloud-side reference information) is considered abnormal data, and the data is transmitted back from the vehicle to the cloud to analyze the cause of the abnormality. In other examples, the cloud further trains the terrain perception model through the abnormal data to improve the model accuracy, and sends the updated terrain perception model to the vehicle through OTA (Over-the-Air Technology) to form a new cycle.

[0094] Step 400: determining vehicle driving parameters adapted to the current terrain based on the target information, so as to control the driving of the vehicle.

[0095] After determining the target information, the vehicle's control system will calculate and determine the appropriate driving parameters based on this information to solve the problem of poor driving experience caused by inaccurate all-terrain mode recognition and inconvenient suspension adjustment.

[0096] When the target information is determined to be terrain mode information, the vehicle control system will make a series of complex and precise adjustments to the vehicle driving parameters based on the terrain mode information. Each mode can adjust the vehicle's torque output and drive force distribution according to the road type to achieve effective control of the vehicle's driving and ensure that it adapts to the corresponding terrain. For example, if the target information shows that the current terrain is an uphill section with a large slope, the vehicle's power system control unit will increase the torque output of the engine or motor to ensure that the vehicle has enough power to climb the slope. At the same time, the transmission control module may adjust the gear position so that the engine or motor operates within the optimal power output speed range.

[0097] For the suspension system, if the target information indicates that the road surface is bumpy, the suspension control system will accordingly increase the damping coefficient of the suspension and reduce the compression stroke of the spring to enhance the stability and handling of the vehicle and prevent excessive shaking of the vehicle body.

[0098] In this embodiment, through the comprehensive adjustment and control of these vehicle driving parameters, the vehicle can better adapt to different terrain conditions and provide the driver with a safe and comfortable driving experience.

[0099] In terms of improving the driving experience, the two aspects involved are improvements in all-terrain mode recognition and suspension adjustment, which are specifically explained through the following two embodiments.

[0100] In some embodiments, a vehicle driving control method based on automatic all-terrain mode switching.

[0101] The perception information includes first terrain pattern information, and the cloud reference information includes second terrain pattern information.

[0102] Reference Figure 2 As shown, Figure 2 The vehicle driving control method comprises:

[0103] Step a1, during the driving process of the vehicle, sensing the current terrain and obtaining first terrain pattern information;

[0104] Step a2, based on the current position of the vehicle, obtaining second terrain pattern information reflecting the current terrain from historical terrain data pre-stored in the cloud server;

[0105] Step a3, when the first terrain pattern information and the second terrain pattern information are inconsistent, arbitrating the first terrain pattern information and the second terrain pattern information based on the confidence of the first terrain pattern information and the confidence of the second terrain pattern information to determine the target information;

[0106] Step a4, determining vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle.

[0107] The vehicle-side data of the preceding vehicle is uploaded to the cloud as basic data support.

[0108] The cloud processes the uploaded vehicle-side data. Based on the processed vehicle-side data, the cloud extracts terrain features based on the NN (Neural Network) algorithm, trains the terrain perception model, and updates the terrain perception model to the vehicle-side through OTA. The vehicle-side ADAS system (Advanced Driving Assistance System) calculates the first terrain mode information based on the terrain perception model and the real-time perception data obtained by the camera and radar.

[0109] Based on the information of the current position of the vehicle, the cloud obtains the second terrain mode information reflecting the current terrain from the historical terrain data pre-stored in the cloud server, and sends the second terrain mode information to the vehicle end.

[0110] In the ADAS system, the first terrain mode information is compared with the second terrain mode information. If the two are consistent, the first terrain mode information is sent to the vehicle control system for subsequent processing by the vehicle control system to determine the vehicle driving parameters.

[0111] If the two are inconsistent, arbitration is performed based on the confidence of the two, and the first terrain mode information or the second terrain mode information that succeeds in arbitration is sent to the vehicle control system. In other examples, when the two are inconsistent, the vehicle-side ADAS perception results and the data sent from the cloud will be uploaded to the cloud as abnormal data to analyze the cause of the error and further train the terrain perception model to make the overall terrain mode switching strategy safer and more reliable.

[0112] In some embodiments, step a4 may further include:

[0113] Step a41, determining a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle based on the target information;

[0114] Step a42, based on the vehicle driving parameters, controlling the driving of the vehicle to adapt to the current terrain.

[0115] The vehicle control system determines the target terrain mode from the vehicle's all-terrain mode according to the first terrain mode information or the second terrain mode information after arbitration, switches the vehicle's terrain mode type, and sends instructions to the vehicle-side execution unit so that the vehicle-side execution unit performs work based on the current terrain mode information (that is, the first terrain mode information or the second terrain mode information after successful arbitration). It should be noted that the vehicle-side execution unit can be, but is not limited to, a suspension system, a chassis control system, a vehicle control system, a steering system, etc.

[0116] Specifically, different terrain modes correspond to pre-set vehicle driving parameter adjustment schemes. The execution unit determines corresponding vehicle driving parameters based on the target terrain mode and controls the driving of the vehicle to adapt to the current terrain.

[0117] After making corresponding parameter adjustments, the new parameters are uploaded to the cloud as new vehicle-side data, forming a data closed loop.

[0118] In some examples, different terrain modes correspond to different driving modes. By automatically switching terrain modes, the vehicle is controlled to drive in the corresponding driving mode to adapt to complex road conditions.

[0119] For example, the current terrain mode information indicates the sand mode: corresponding to soft sand and desert road conditions, to deal with wheel slippage and sand sinking problems, increase engine torque, change the driving force distribution strategy, increase the suspension height, and cancel coasting energy recovery to avoid unexpected deceleration.

[0120] The current terrain mode information indicates mud mode: corresponding to muddy and slippery roads or swampy areas and other road conditions, it enhances vehicle traction and increases the stability of the suspension system.

[0121] The current terrain mode information indicates Snow mode: power output and braking performance are adjusted, and the accelerator pedal response will be more cautious to ensure driving stability on ice and snow covered roads.

[0122] In summary, by accurately extracting, analyzing and identifying the key features of the current road conditions, they are effectively mapped to the terrain mode that is suitable for them, and then the switching strategy and implementation plan of the current terrain mode are clarified to ensure the optimization and adaptation of the vehicle's driving performance.

[0123] In some embodiments, in order to improve the accuracy of terrain pattern recognition, the recognition results of the above-mentioned terrain pattern information are combined with the results of terrain pattern information analyzed based on vehicle dynamic parameters, and the recognition results are arbitrated to achieve complementarity and obtain more reliable recognition results.

[0124] As an example, the target information includes target terrain pattern information, and further referring to Figure 2, the step a41 may specifically include the following steps:

[0125] Step a411, obtaining third terrain pattern information obtained by identifying the current terrain according to the vehicle power parameters of the vehicle; the vehicle power parameters are used to reflect the movement of the vehicle;

[0126] Step a412, based on the confidence of the third terrain pattern information and the confidence of the target information, arbitrate the third terrain pattern information and the target information to determine the target terrain pattern information;

[0127] Step a413, determining a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle according to the target terrain mode information.

[0128] Vehicle-side dynamic parameters are a series of key indicators used to characterize the dynamics of the vehicle during driving and reflect the interaction characteristics between the vehicle and the ground. They may include, but are not limited to, wheel-end slip rate, adhesion coefficient, and friction coefficient between the tire and the ground.

[0129] The vehicle control system analyzes the vehicle power parameters based on the vehicle power parameters it has read and determines the third terrain mode information, arbitrates the third terrain mode information and the target information (i.e. the first terrain mode information or the second terrain mode information after the above arbitration is successful), and switches the terrain mode type according to the arbitrated terrain mode information. If the confidence level of the target information is lower than the threshold, the terrain matched according to the slip rate, adhesion coefficient curve and friction coefficient is used as the final terrain mode information.

[0130] It should be noted that the arbitration process also uses the confidence of the third terrain pattern information and the confidence of the target information to arbitrate the third terrain pattern information and the target information to determine the target terrain pattern information. The basic embodiment is similar to the arbitration content of this part of the above embodiment, which will not be repeated here.

[0131] Through the above-mentioned embodiments, while the vehicle side obtains the first terrain mode information through ADAS sensing the terrain type, the cloud also sends the second terrain mode information to the vehicle side, and the arbitration is carried out through the confidence of both parties. The arbitration result is then compared with the third terrain mode information obtained by the vehicle side power parameters. After multiple arbitrations, the accuracy of information judgment is improved, and then the accuracy of the terrain mode finally switched is improved. At the same time, the group intelligence of end-cloud fusion relies on the vehicle-side data uploaded by the previous vehicle to train the terrain perception model, and then continuously supplements and corrects it through the vehicle-side data uploaded by the subsequent vehicles, so that the terrain perception model used by the vehicle-side ADAS system can be dynamically updated based on the vehicle-side data. The algorithm is evolutionary, and the accuracy and precision are continuously improved, which can adapt to complex outdoor road conditions. Accurately identify and switch all-terrain modes to effectively improve driving comfort and safety.

[0132] In other embodiments, a vehicle driving control method is provided based on automatic suspension adjustment.

[0133] The vehicle-side perception information includes first terrain feature information, and the cloud-side reference information includes second terrain feature information.

[0134] Reference Figure 3 As shown, Figure 3 The vehicle driving control method comprises:

[0135] Step b1, during the driving process of the vehicle, sensing the current terrain and obtaining first terrain feature information;

[0136] Step b2, based on the current position of the vehicle, obtaining second terrain feature information reflecting the current terrain from historical terrain data pre-stored in the cloud server;

[0137] Step b3, when the first terrain feature information and the second terrain feature information are inconsistent, arbitrating the first terrain feature information and the second terrain feature information based on the confidence level of the first terrain feature information and the confidence level of the second terrain feature information to determine target information;

[0138] Step b4, determining vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle.

[0139] First, the vehicle uploads the vehicle data of the previous vehicle to the cloud as basic data support. The vehicle data includes but is not limited to vehicle status data, driving control data, vehicle location information, chassis image data and ADAS system perception data.

[0140] The cloud processes the vehicle data based on wavelet transform, Fourier transform, clustering and other algorithms. The cloud maps the bumpy features based on the processed vehicle data, generates a road bumpy feature model, and updates it to the vehicle through OTA (Over-the-Air Technology).

[0141] As an example, the method for constructing the road bump feature model includes the following steps:

[0142] Acquire vehicle-side data when a preceding vehicle passes through a bumpy road, the vehicle-side data including vehicle-side status data and terrain feature information, and vehicle suspension system parameters obtained by analyzing the vehicle-side status data and terrain feature information;

[0143] The acquired vehicle-side data is subjected to correlation analysis to create a road bump feature model.

[0144] Specifically, when creating a road bump feature model, the sample data mainly comes from the data of the previous vehicle passing through the bumpy road. These data include road feature information (such as whether the road surface is asphalt, cement or gravel, the flatness of the road surface, whether there are cracks, potholes, etc.), elevation classification information (such as whether the road surface is flat or has a slope change, whether it is a small hilly terrain or an undulating road surface in a mountainous terrain, etc.), and vehicle-side state data when the vehicle is driving on these roads (such as suspension height changes, vehicle acceleration, body pitch angle and roll angle, etc.). At the same time, the damping coefficient, spring stiffness and other parameter settings used by the vehicle suspension system at that time were also recorded.

[0145] The sample data of multiple bumpy roads collected are superimposed and analyzed to find the intrinsic connection between the road feature information, elevation classification information and vehicle-side status data, as well as the relationship between the vehicle-side status data and the damping coefficient and spring stiffness, to obtain the road bump feature model / logic algorithm.

[0146] After the current vehicle recognizes the road feature information and elevation classification information, the parameters that need to be adjusted can be calculated through this model / algorithm, and these parameters can effectively improve the driving state of the vehicle on bumpy roads.

[0147] Next, the cloud sends target information to the vehicle based on the current vehicle location information. The target information may include but is not limited to road feature information and elevation classification information.

[0148] The vehicle-side ADAS system is based on a terrain perception model and calculates road feature information and elevation classification information using real-time perception data obtained by cameras and radars. If the road feature information and elevation classification information are inconsistent with those sent from the cloud, a comparison is performed based on the confidence level of the two, and the result of the successful arbitration is sent to the chassis system.

[0149] In some cases, if the road feature information and elevation classification information perceived by the ADAS system on the vehicle side are inconsistent with the road feature information and elevation classification information sent from the cloud, the current data is considered abnormal data and needs to be sent back to the cloud to analyze the cause of the abnormality. The terrain perception model is further trained to improve the model accuracy, and OTA is sent to the vehicle side to form a new cycle.

[0150] Finally, the chassis system determines vehicle driving parameters adapted to the current terrain based on the road bump feature model and target information (the first terrain feature information or the second terrain feature information determined after arbitration) for controlling the driving of the vehicle.

[0151] As an example, the step b4 may specifically include the following steps:

[0152] Step b41, obtaining a road bump feature model from the cloud server;

[0153] Step b42, determining the vehicle driving parameters required for the vehicle to pass through the current terrain based on the road bump feature model and the target information data;

[0154] Step b43, adjusting the suspension system of the vehicle based on the vehicle driving parameters to control the vehicle to travel on the current terrain.

[0155] After the current vehicle recognizes the target information (road surface feature information and elevation classification information), it inputs this information into the created road bump feature model. Based on the input information, the model calculates the damping coefficient and spring stiffness that the current vehicle needs to adjust according to the predetermined rules and mathematical models. Then, the vehicle's suspension system adjusts parameters based on these calculation results to adapt to the bumpy road that is about to pass.

[0156] Furthermore, the updated suspension system parameters will be uploaded to the cloud as new vehicle-side data, forming a closed data loop.

[0157] Through the above embodiments, the chassis system automatically adjusts the suspension based on the road bump feature model according to the road surface feature information and elevation classification information, without the need for manual adjustment by the driver, thereby improving the driving experience such as safety and comfort. In addition, the group intelligence of end-cloud fusion relies on the vehicle-side data uploaded by the previous vehicle to train the road perception model and generate a road bump feature model, which is then continuously supplemented and corrected by the vehicle-side data uploaded by the subsequent vehicle, so that the road perception model and the road bump feature model can be dynamically updated based on the vehicle-side data. The algorithm is evolutionary, continuously improving accuracy and precision, and can adapt to complex outdoor road conditions. It effectively solves the problems existing in the existing suspension adjustment technology, helps drivers better adapt to the current road conditions, and effectively improves driving comfort and safety.

[0158] Based on the above embodiment, the cloud performs data processing based on the previous vehicle-side data, extracts terrain features, forms a terrain layer and / or a bumpy layer, and then sends it to the vehicle-side. The vehicle-side navigation system overlays the terrain layer and the bumpy layer to achieve map rendering, and displays annotation information on the navigation map when passing through special roads such as potholes and speed bumps. Figure 4 , Figure 4 Render link graphs for navigation maps. Here are three methods for rendering navigation maps.

[0159] As an example, after step 400, the method further includes:

[0160] Step c11, obtaining a layer for representing current terrain features from the cloud server, wherein the layer is determined by the cloud server from a plurality of pre-stored layers according to the terrain features reflected by the cloud reference information;

[0161] Step c12, superimposing the layer with the original map data of the vehicle, and rendering the original map to display the features of the current terrain.

[0162] The layer includes at least one of a terrain layer and a bumpy layer, and the required layer is selected according to actual terrain features.

[0163] When the cloud receives the request sent by the vehicle, it processes the data based on the previous vehicle data or cloud reference information, analyzes the terrain features in the previous vehicle data or cloud reference information in detail, extracts the terrain features, and forms a terrain layer and a bumpy layer.

[0164] The vehicle receives the terrain layer and bump layer sent from the cloud, and overlays the vehicle navigation system with the terrain layer and bump layer to achieve map rendering. When overlaying, the system accurately matches the current terrain layer with the original map data based on the geographic coordinate information. This requires calibration and conversion of the coordinate systems of the two data sources to ensure that the terrain features in the layer can accurately correspond to the corresponding positions on the map. For example, matching the contour layer representing mountainous terrain with the road direction on the map allows the driver to intuitively see how the road crosses the mountainous terrain.

[0165] Once the overlay is completed, the rendering operation will be performed. The rendering process will display the information in the current terrain feature layer on the map interface in the form of intuitive graphics, colors, symbols, etc. according to the preset display rules and visual effect parameters. For example, different color gradients are used to represent the altitude changes of mountains, specific icons are used to mark dangerous terrain such as rocky sections or steep slopes, and textures are used to display special roads such as sand or wetlands. In this way, by viewing the navigation map, the driver can clearly understand the terrain features in front of the vehicle and make driving preparations in advance, such as adjusting the speed, choosing the appropriate driving mode or route, etc., thereby improving driving safety and comfort.

[0166] As an example, after step 300, the method further includes:

[0167] Step c21, determining a layer for representing current terrain features from a plurality of pre-stored layers according to the terrain features reflected by the target information;

[0168] Step c22, superimposing the layer with the original map data of the vehicle, and rendering the original map to display the features of the current terrain.

[0169] The vehicle side pre-stores multiple different types of layers, which are built based on a large amount of terrain data and geographic information. For example, there may be terrain layers specifically used to represent different terrain features (such as mountains, plains, hills, deserts, wetlands, etc.), road condition layers for road conditions (such as flat roads, bumpy roads, potholes, etc.), and element layers that reflect special terrain elements (such as rivers, bridges, steep slopes, etc.). Each layer is stored in a specific data format and structure, and contains rich geographic coordinate information and corresponding terrain feature description information.

[0170] After the target information is acquired, the layer or layer combination that can most accurately represent the current terrain features is finally determined from multiple different types of layers pre-stored on the vehicle side according to the information of the terrain features reflected by the target information.

[0171] When overlaying, the system will accurately match the current terrain layer with the original map data based on the geographic coordinate information. Once the overlay is completed, the rendering operation will be performed. The rendering process will display the information in the current terrain feature layer on the map interface in the form of intuitive graphics, colors, symbols, etc. according to the preset display rules and visual effect parameters.

[0172] As an example, after step 300, the method further includes:

[0173] Step c31, generating a layer for representing current terrain features according to the target information;

[0174] Step c32, superimposing the layer with the original map data of the vehicle, and rendering the original map to display the features of the current terrain.

[0175] The vehicle side conducts in-depth analysis of the target information, extracts key data about terrain features, and constructs a layer representing the current terrain features based on the extracted data. These data include but are not limited to terrain type, road conditions, and special terrain signs.

[0176] The generated layer is accurately superimposed with the original map data according to the geographic coordinates. Once the superposition is completed, the rendering operation will be performed. The rendering process will display the information in the current terrain feature layer on the map interface in the form of intuitive graphics, colors, symbols, etc. according to the preset display rules and visual effect parameters.

[0177] In other examples, during the overlay process, the display priority of different layers is set.

[0178] Specifically, the basic road network, place names and other information in the original map data will maintain a higher display priority to ensure that the driver can clearly see the driving route and important places, so as not to distract the driver from the road information. The special terrain signs and road condition prompts in the generated terrain layer and bumpy layer will be set with corresponding display priorities according to their importance and relevance to driving.

[0179] Through the above-mentioned embodiments, complex road conditions are visualized, and the cloud sends terrain layers and bumpy layers to the vehicle side and overlays them with the navigation map. When passing through special roads, the navigation map can reflect the annotation information to enhance the driver's visualization experience.

[0180] The present application provides a vehicle driving control method, device and computer-readable storage medium. Compared with the current poor driving experience, during the vehicle driving process, the current terrain is sensed to obtain vehicle-side perception information, and based on the current position of the vehicle, cloud reference information of the current terrain is obtained from the cloud, and the accuracy of vehicle-side perception is improved by arbitrating the vehicle-side perception information and the cloud reference information. After determining the target information after arbitration, the vehicle driving parameters adapted to the current terrain are determined based on the accurate target information, helping the driver to better adapt to the current road conditions and effectively improving driving comfort and safety.

[0181] Based on the same application concept as the above method, the embodiment of the present application also proposes a vehicle driving control device, such as Figure 5 shown.

[0182] The device comprises:

[0183] The vehicle-side perception module 502 is used to perceive the current terrain during the vehicle's driving process and obtain vehicle-side perception information;

[0184] A cloud recognition module 504, based on the current position of the vehicle, obtains cloud reference information reflecting the current terrain from the historical terrain data pre-stored in the cloud server;

[0185] The arbitration module 506 arbitrates the vehicle-side perception information and the cloud-side reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud-side reference information to determine the target information when the vehicle-side perception information and the cloud-side reference information are inconsistent.

[0186] The parameter adjustment module 508 determines the vehicle driving parameters adapted to the current terrain based on the target information, so as to control the driving of the vehicle.

[0187] The implementation process of the functions and effects of each module / submodule / unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and the same technical effect can be achieved, so it will not be repeated here.

[0188] Figure 6 An example of a physical structure diagram of a vehicle driving control device is shown in FIG. Figure 6 As shown, the vehicle driving control device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the vehicle driving control method.

[0189] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0190] On the other hand, the present application also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the vehicle driving control methods provided by the above methods.

[0191] On the other hand, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the vehicle driving control method provided by the above-mentioned methods.

[0192] It should be noted that the technical solutions or technical features described in the above embodiments can be combined or supplemented with each other without causing conflicts. The scope of protection of this application is not limited to the precise structures described in the above embodiments and shown in the drawings; all modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application should be included in the scope of protection of this application.

Claims

1. A vehicle driving control method, characterized in that: Applied to a vehicle-side controller, the method comprises: During the driving process of the vehicle, the current terrain is sensed to obtain vehicle-side perception information; Based on the current position of the vehicle, obtaining cloud reference information reflecting the current terrain from historical terrain data pre-stored in the cloud server; In the case where the vehicle-side perception information and the cloud-side reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud-side reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud-side reference information to determine the target information; Based on the target information, a vehicle driving parameter adapted to the current terrain is determined for controlling the travel of the vehicle.

2. The vehicle driving control method according to claim 1, characterized in that: During the driving process of the vehicle, the current terrain is sensed to obtain vehicle-side sensing information, including: During the driving process of the vehicle, the current terrain is perceived based on the terrain perception model to obtain vehicle-side perception information; The terrain perception model is trained based on historical terrain data pre-stored in the cloud server.

3. The vehicle driving control method according to claim 1, characterized in that: The vehicle-side perception information includes first terrain pattern information, and the cloud-side reference information includes second terrain pattern information. When the vehicle-side perception information and the cloud reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information to determine the target information includes: arbitrating the first terrain pattern information and the second terrain pattern information based on the confidence of the first terrain pattern information and the confidence of the second terrain pattern information to determine target information; The determining of the vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle, comprises: Based on the target information, determining a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle; Based on the vehicle driving parameters, the driving of the vehicle is controlled to adapt to the current terrain.

4. The vehicle driving control method according to claim 3, characterized in that: The target information includes target terrain pattern information, The determining, based on the target information, a target terrain mode and a vehicle driving parameter corresponding to the target terrain mode from the all-terrain mode of the vehicle comprises: Acquiring third terrain pattern information obtained by identifying the current terrain according to vehicle power parameters of the vehicle; the vehicle power parameters are used to reflect vehicle movement; Based on the confidence level of the third terrain pattern information and the confidence level of the target information, arbitrating the third terrain pattern information and the target information to determine the target terrain pattern information; A target terrain mode and a vehicle driving parameter corresponding to the target terrain mode are determined from the all-terrain mode of the vehicle according to the target terrain mode information.

5. The vehicle driving control method according to claim 1, characterized in that: The vehicle-side perception information includes first terrain feature information, and the cloud-side reference information includes second terrain feature information. When the vehicle-side perception information and the cloud reference information are inconsistent, arbitrating the vehicle-side perception information and the cloud reference information based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information to determine the target information includes: In the case where the first terrain feature information and the second terrain feature information are inconsistent, arbitrating the first terrain feature information and the second terrain feature information based on the confidence level of the first terrain feature information and the confidence level of the second terrain feature information to determine target information; The determining of the vehicle driving parameters adapted to the current terrain based on the target information, for controlling the driving of the vehicle, comprises: Acquire a road bump feature model from the cloud server; Determining vehicle driving parameters required for the vehicle to pass through the current terrain based on the road bump feature model and the target information data; A suspension system of the vehicle is adjusted based on the vehicle driving parameters to control the vehicle to travel on a current terrain.

6. The vehicle driving control method according to claim 5, characterized in that: The method for constructing the road bump feature model comprises the following steps: Acquire vehicle-side data when a preceding vehicle passes through a bumpy road, the vehicle-side data including vehicle-side status data and terrain feature information, and vehicle suspension system parameters obtained by analyzing the vehicle-side status data and terrain feature information; The acquired vehicle-side data is subjected to correlation analysis to create a road bump feature model.

7. The vehicle driving control method according to claim 1, characterized in that: After acquiring cloud reference information reflecting the current terrain from historical terrain data pre-stored in the cloud server based on the current position of the vehicle, the method further includes: Acquire a layer for representing current terrain features from the cloud server, wherein the layer is determined by the cloud server from a plurality of pre-stored layers according to the terrain features reflected by the cloud reference information; The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

8. The vehicle driving control method according to claim 1, characterized in that: When the vehicle-side perception information and the cloud reference information are inconsistent, the vehicle-side perception information and the cloud reference information are arbitrated based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information, and after determining the target information, the method further includes: Determining a layer for representing current terrain features from a plurality of pre-stored layers according to the terrain features reflected by the target information; The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

9. The vehicle driving control method according to claim 1, characterized in that: In the case where the vehicle-side perception information and the cloud reference information are inconsistent, the vehicle-side perception information and the cloud reference information are arbitrated based on the confidence of the vehicle-side perception information and the confidence of the cloud reference information, and after determining the target information, the method further includes: Generate a layer for representing current terrain features according to the target information; The layer is superimposed on the original map data of the vehicle, and the original map is rendered to display the features of the current terrain.

10. The vehicle driving control method according to any one of claims 7 to 9, characterized in that: The layer includes at least one of a terrain layer and a bumpy layer.

11. A vehicle driving control device, characterized in that: It includes a memory, a processor and a vehicle driving control program stored in the memory and executable on the processor, and when the processor executes the vehicle driving control program, the steps of the vehicle driving control method as described in any one of claims 1 to 10 are implemented.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a vehicle driving control program, which, when executed, implements the steps of the vehicle driving control method according to any one of claims 1 to 10.

Images