Vehicle load prediction method, device, storage medium and vehicle

Abstract

The present application provides a vehicle load prediction method, device, storage medium, and vehicle, relating to the field of vehicle technology. The method comprises: obtaining parameter information of a current vehicle belonging to a target category, and using the parameter information belonging to the target category as characteristic information of the current vehicle; inputting the characteristic information into a load prediction model to obtain the load of the current vehicle output by the load prediction model; wherein the load prediction model is obtained by training a preset model using the characteristic information of a test vehicle as a training sample. Since the characteristic information of the current vehicle is information related to the load of the current vehicle, when the characteristic information of the current vehicle and the load prediction model are used to predict the load of the current vehicle, the prediction result has high reliability and accuracy, and real-time prediction can be achieved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to a vehicle load prediction method, device, storage medium and vehicle. Background Art

[0002] To ensure road transportation safety, strict regulations are in place regarding vehicle load capacities. Overloading poses not only a serious safety hazard to vehicles but also a significant threat to human life. Traffic accidents caused by overloaded cargo vehicles are common, resulting in significant losses of life and property. Currently, two methods are commonly used to measure vehicle load capacities.

[0003] Solution 1: Install a sensor at the vehicle suspension position to detect the deformation of the suspension and calculate the vehicle load.

[0004] Option 2: By collecting vehicle information, using vehicle acceleration and vehicle driving force, and based on Newton's second law, the vehicle load can be derived.

[0005] However, Option 1 requires adding sensors to the suspension, creating a complex hardware solution that increases vehicle weight and cost, as well as complexity. Existing technology cannot meet mass production requirements. It is closely tied to suspension system tuning; adjustments to suspension performance require recalibration, impacting consistency. When driving on harsh roads (such as gravel and bumpy roads), the suspension system's displacement fluctuates significantly, resulting in significant errors. Option 2 suffers from large errors in driving force calculations due to uncontrollable vehicle power loss. Furthermore, without information on slope, road roughness, or environmental conditions, calculation data fluctuates significantly under harsh road conditions, leading to significant errors.

[0006] It can be seen that the two existing measurement schemes have large measurement errors and cannot be mass-produced. Summary of the Invention

[0007] The present application provides a vehicle load prediction method, device, storage medium and vehicle to solve the problem that existing measurement solutions have large errors and cannot be mass-produced.

[0008] In order to solve the above problems, in a first aspect, the present application discloses a vehicle load prediction method, which includes:

[0009] Obtain parameter information of the current vehicle belonging to the target category, and use the parameter information belonging to the target category as feature information of the current vehicle;

[0010] Inputting characteristic information of the current vehicle into the load prediction model to obtain the load of the current vehicle output by the load prediction model;

[0011] Among them, the load prediction model is obtained by training the preset model using the characteristic information of the test vehicle as a training sample.

[0012] In an optional embodiment, the target category is determined based on a correlation coefficient between a parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle. The method for determining the target category includes:

[0013] Obtaining parameter values corresponding to the parameter information of the test vehicle and the load of the test vehicle;

[0014] Determining a correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load based on the parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle;

[0015] The parameter information of the test vehicle is screened according to the correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load, and the target category of the parameter information to be obtained is determined.

[0016] In an optional embodiment, the parameter information of the target category includes at least: road slope, road roughness, and vehicle steering information.

[0017] In an optional embodiment, the road slope is obtained by one or more of sensors, electronic maps, and satellite positioning data.

[0018] In an optional embodiment, when the road surface slope is obtained by a sensor, the road surface slope information is calculated according to equations (1), (2), and (3):

[0019] axvRoadSlope=axvSensorRA-axvRaw Formula (1)

[0020] axvSensorRA=axvSensor+yawRate 2 *ISPdxvRearAxis2Axsensor Formula (2)

[0021] axvRaw=(vxvRef-vxvRefK1) / T Formula (3)

[0022] Where axvRoadSlope is the road slope, axvSensorRA is the total acceleration of the vehicle, axvRaw is the acceleration of the tire, axvSensor is the longitudinal acceleration of the vehicle, yawRate is the yaw rate of the vehicle, ISPdxvRearAxis2Axsensor is the longitudinal distance from the vehicle test point to the yaw rate of the vehicle, vxvRef is the tire speed at time T1, vxvRefK1 is the tire speed at time T2; and T is the time interval between time T1 and time T2.

[0023] In an optional embodiment, the road surface roughness is obtained through image information collected by a camera, wherein the camera is installed at the rearview mirror on the upper part of the windshield of the car, and / or,

[0024] The road roughness is obtained by measuring the acceleration and angular acceleration in the X, Y, and Z directions using the vehicle's inertial sensors, and / or

[0025] The road roughness is obtained through the longitudinal distance from the wheel center to the wheel arch output by the air suspension.

[0026] In an optional embodiment, the load prediction model includes any one of a decision tree model, an XGBoost model, and a random forest model.

[0027] In a second aspect, the present application discloses a vehicle load prediction device, the device comprising:

[0028] a determination module, configured to determine a target category of the parameter information to be acquired based on a correlation coefficient between a parameter value corresponding to the parameter information of the test vehicle and a load of the test vehicle; the parameter information of the test vehicle includes at least information representing an engine operating condition and a chassis operating condition;

[0029] An acquisition module, configured to acquire parameter information of the current vehicle belonging to a target category, and use the parameter information belonging to the target category as feature information of the current vehicle;

[0030] The acquisition module is used to input the characteristic information of the current vehicle into the load prediction model to obtain the load of the current vehicle output by the load prediction model; wherein the load prediction model is obtained by training the preset model with the characteristic information of the test vehicle as the training sample.

[0031] In a third aspect, the present application discloses a computer-readable storage medium storing a vehicle load prediction program. When the vehicle load prediction program is executed by a processor, the steps of the vehicle load prediction method of any one of the above-mentioned first aspects are implemented.

[0032] In a fourth aspect, the present application discloses a vehicle, the vehicle load of which is predicted by any vehicle load prediction method according to the first aspect.

[0033] Compared with the prior art, this application has the following advantages:

[0034] Using the vehicle load prediction method provided in the embodiment of the present application, the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample. Since the characteristic information of the test vehicle and the load of the test vehicle are known, the characteristic information of the test vehicle and the load of the test vehicle are used to train the preset model, and the obtained load prediction model is also a definite model, which can well reflect the change law between the characteristic signal of the vehicle and the load of the vehicle. When it is necessary to predict the load of the current vehicle, it is only necessary to input the characteristic information of the current vehicle into the load prediction model to predict the load of the current vehicle. Since the characteristic information of the test vehicle is information related to the load of the test vehicle, the accuracy of the load prediction model can be improved when the model is trained using the characteristic information. Compared with the existing measurement scheme, when the load prediction model is used to predict the load of the current vehicle, the prediction result has higher reliability and accuracy, and real-time prediction can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of a vehicle load prediction method according to an embodiment of the present application;

[0036] Figure 2 This is a structural block diagram of obtaining characteristic information of the current vehicle according to an embodiment of the present application;

[0037] Figure 3 This is a flowchart of determining a target category of parameter information according to an embodiment of the present application;

[0038] Figure 4 This is a structural block diagram of obtaining parameter information of a test vehicle according to an embodiment of the present application;

[0039] Figure 5 This is a flow chart of a load prediction model for screening according to an embodiment of the present application;

[0040] Figure 6 This is a structural block diagram of a vehicle load prediction device according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0042] Reference Figure 1 , Figure 1 This is a flow chart of a vehicle load prediction method according to an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0043] S101, obtaining parameter information of a current vehicle belonging to a target category, and using the parameter information belonging to the target category as feature information of the current vehicle;

[0044] Specifically, in an embodiment of the present invention, a carrier (hardware capable of performing data operations, such as a vehicle-side controller or cloud server) obtains parameter information of the current vehicle belonging to a target category and uses this parameter information as the characteristic information of the current vehicle. Specifically, the characteristic information of the current vehicle refers to the parameter information of the current vehicle belonging to the target category, which is obtained by removing information unrelated to the load from the parameter information of the current vehicle and filtering out information related to the load. The parameter information of the current vehicle is information that characterizes the engine and chassis operating conditions of the current vehicle, and the characteristic information of the current vehicle is a portion of the parameter information of the current vehicle.

[0045] When using a load prediction model to determine the current vehicle's load, if too much parameter information is input into the model, the model's operation efficiency will be slow. To improve the model's operation efficiency, before performing calculations, it is necessary to remove parameter information unrelated to the current vehicle's parameter information and use parameter information related to the load as feature information. However, since the current vehicle's load is unknown, it is impossible to use the current vehicle's parameter information to filter feature information. However, since the test vehicle's load and parameter information are both known, the test vehicle's load and parameter information can be used to filter parameter information related to the load from the test vehicle's parameter information, and the category to which the parameter information related to the load belongs is used as the target category for the parameter information. The target category of the test vehicle's parameter information is the same as the target category of the current vehicle's parameter information. Therefore, when using the load prediction model to determine the current vehicle's load, only the parameter information of the current vehicle belonging to the target category needs to be input into the model.

[0046] It should be noted that in the embodiments of this application, the test vehicle refers to the vehicle participating in parameter information screening and preset model training, and the parameter information and load of the test vehicle are known; the current vehicle refers to the vehicle for which load prediction is required, and the parameter information of the current vehicle is known, but the load is unknown. The load of the current vehicle can be predicted using the parameter information (feature information) of the current vehicle. The test vehicle and the current vehicle can be the same vehicle or the same type of vehicle.

[0047] S102. Input the characteristic information of the current vehicle into the load prediction model to obtain the load of the current vehicle output by the load prediction model; wherein the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample.

[0048] Specifically, in an embodiment of the present invention, the load prediction model is a function that characterizes the change pattern between the characteristic information of a vehicle and the load of the vehicle. The load prediction model is obtained by training a preset model using the characteristic information of a test vehicle as a training sample. The preset model is any one of a decision tree model, an XGBoost (EXtreme Gradient Boosting) model, and a random forest model. The load prediction model is a preset model after training. The characteristic information of the test vehicle here refers to the parameter information of the test vehicle belonging to the target category, and the characteristic information of the test vehicle is part of the parameter information of the test vehicle. The characteristic information of the current vehicle is input into the load prediction model, and the load of the current vehicle can be obtained through the load prediction model.

[0049] Using the vehicle load prediction method provided in the embodiment of the present application, the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample. Since the characteristic information of the test vehicle and the load of the test vehicle are known, the characteristic information of the test vehicle and the load of the test vehicle are used to train the preset model, and the obtained load prediction model is also a definite model, which can well reflect the change law between the characteristic signal of the vehicle and the load of the vehicle. When it is necessary to predict the load of the current vehicle, it is only necessary to input the characteristic information of the current vehicle into the load prediction model to predict the load of the current vehicle. Since the characteristic information of the test vehicle is information related to the load of the test vehicle, the accuracy of the load prediction model can be improved when the model is trained using the characteristic information. Compared with the existing measurement scheme, when the load prediction model is used to predict the load of the current vehicle, the prediction result has higher reliability and accuracy, and real-time prediction can be achieved. Real-time prediction of the current vehicle load helps to determine the vehicle's load level when the vehicle load changes, making it easier for the driver to choose a more fuel-efficient driving method based on the load. It is also beneficial for the review and supervision of regulatory authorities, effectively avoiding overloading. On the other hand, it helps with fleet management. Fleet managers can directly and effectively judge the load conditions of vehicles in the fleet based on the real-time uploaded vehicle load information, facilitating statistics and supervision.

[0050] In an optional embodiment, when the carrier obtains the characteristic information of the current vehicle, the characteristic information of the current vehicle can be obtained directly or indirectly, and the characteristic information of the current vehicle includes: the vehicle's longitudinal acceleration, the vehicle's lateral acceleration, the yaw angular velocity, the engine speed, the engine status, the vehicle speed, the brake pressure, the accelerator pedal position, the brake pedal position, the instantaneous fuel consumption, the remaining fuel, the engine output torque, the engine loss torque, the four-wheel speed value, the transmission input torque, the steering wheel angle, the steering wheel angle direction, the steering wheel angle velocity, the steering wheel angle velocity direction, the driver's steering torque, the gear position, the transmission ratio, the drag coefficient, the road slope and the road roughness.

[0051] Engine speed refers to the number of crankshaft revolutions per unit time. Engine status refers to the engine's starting, stopping (shutdown), and operating under various loads. Accelerator pedal position refers to the force applied by the driver to the accelerator. When the accelerator pedal is at zero, the vehicle is not accelerating. Brake pedal position refers to the force applied by the driver to the brake pedal. When the brake pedal is at zero, the vehicle is not decelerating. Instantaneous fuel consumption refers to the engine's fuel consumption at a given moment. Remaining fuel refers to the remaining fuel in the tank. Engine output torque refers to the torque output from the engine's crankshaft. Engine loss torque refers to the torque lost during the engine's torque output. Vehicle speed refers to the vehicle's speed, generally derived from the four-wheel speed values. Brake pressure refers to the pressure applied by the master cylinder to the brake discs. The four-wheel speed values refer to the tire speeds. Transmission input torque refers to the torque applied to the transmission from the engine. Gear refers to the transmission's P, N, D, R, or M gears. The steering wheel angle refers to the angle at which the steering wheel is turned, the steering wheel angle direction refers to the direction in which the steering wheel is turned, the steering wheel angle velocity refers to the speed at which the steering wheel is turned, the steering wheel angle velocity direction refers to the current direction of the steering wheel, and the driver steering torque refers to the torque applied to the steering wheel by the driver.

[0052] In this embodiment, the steering wheel angle, steering wheel angle direction, steering wheel angle velocity, steering wheel angle velocity direction, and driver steering torque are collectively referred to as vehicle steering information, while road surface roughness and road slope are collectively referred to as road condition information. By incorporating vehicle steering information and road condition information into the calculation of the current vehicle load, the calculation structure becomes more accurate. For example, the inclusion of slope can eliminate the changes in gravity caused by the vehicle traveling up and downhill.

[0053] It's important to note that when using the current vehicle's characteristic information to determine its load, the validity of that information must be considered. When the characteristic information is invalid (e.g., due to ESP, EPS, engine, or transmission failures), the load prediction model outputs a marked load, which is considered invalid and should be discarded. The current vehicle load here refers to the weight carried by the vehicle, typically the sum of the weight of the passengers, cargo, and remaining fuel. Including the remaining fuel weight can prevent errors caused by fuel consumption during long driving periods.

[0054] See also Figure 2 , shows a block diagram of a structure for obtaining characteristic information of the current vehicle according to one embodiment of the present application. In this embodiment, characteristic information of the current vehicle is obtained through ABM, ESP, EPS, transmission, engine, inertial sensor, air suspension, camera, etc., and the characteristic signal of the current vehicle is sensed to the carrier via the CAN bus (Controller Area Network).

[0055] Specifically, the vehicle's longitudinal acceleration, lateral acceleration, and yaw rate are collected by the ABM and transmitted to the carrier via the CAN bus;

[0056] The engine speed, engine status, accelerator pedal position, brake pedal position, instantaneous fuel consumption, remaining fuel, engine output torque and engine loss torque are collected by the engine and transmitted to the carrier via the CAN bus.

[0057] Vehicle speed, brake pressure and four-wheel speed values are collected by ESP and transmitted to the carrier via the CAN bus.

[0058] The transmission collects the input torque and gear position, which are then transmitted to the carrier via the CAN bus. Transmission input torque refers to the torque applied to the transmission from the engine, while gear position refers to the transmission's P, N, D, R, and M gears.

[0059] The steering wheel angle, steering wheel angle direction, steering wheel angle speed, steering wheel angle speed direction and driver steering torque are collected by the EPS and transmitted to the carrier via the CAN bus.

[0060] The road surface slope is obtained through one or more of the following methods: sensors, electronic maps, and satellite positioning data. Specifically, the road surface slope of the target road section can be obtained through an electronic map, satellite positioning data, or sensors on a vehicle. Alternatively, the road surface slope can be determined using a combination of electronic maps, satellite positioning data, and vehicle sensor data. It should be noted that the slope in the embodiments of the present invention refers to the longitudinal slope.

[0061] In a specific embodiment, when the road surface slope is obtained by a sensor, the method includes: calculating the road surface slope according to equations (1), (2), and (3):

[0062] axvRoadSlope=axvSensorRA-axvRaw Formula (1)

[0063] axvSensorRA=axvSensor+yawRate 2 *ISPdxvRearAxis2Axsensor Formula (2)

[0064] axvRaw=(vxvRef-vxvRefK1) / T Formula (3)

[0065] Where axvRoadSlope is the road slope, axvSensorRA is the total acceleration of the vehicle, axvRaw is the acceleration of the tire, axvSensor is the longitudinal acceleration of the vehicle, yawRate is the yaw rate of the vehicle, ISPdxvRearAxis2Axsensor is the longitudinal distance from the vehicle test point to the yaw rate of the vehicle, vxvRef is the tire speed at time T1, vxvRefK1 is the tire speed at time T2; and T is the time interval between time T1 and time T2.

[0066] Specifically, the vehicle's longitudinal acceleration is obtained through the vehicle's longitudinal acceleration sensor installed on the airbag, the vehicle's yaw angular velocity is obtained through the yaw angular velocity sensor installed on the airbag, and the tire speed at times T1 and T2 is obtained through the speed sensor installed on the airbag. The longitudinal distance from the vehicle test point to the vehicle's yaw angular velocity can be understood as the distance from the center of the rear axle to the airbag, which is usually a fixed value.

[0067] It should be noted that the calculation structure of Equation (3) is an approximation. That is, it roughly assumes that the difference between the total acceleration of the vehicle and the acceleration of the tires is equal to the road slope. The road slope is the change in height of the vehicle for every 1 meter of travel. For example, a slope of 0.25 means that the height changes by 25 meters for every 100 meters of travel.

[0068] For example, when the longitudinal acceleration of the vehicle is 0.25m / s 2 , the yaw rate is 1 rad / s, the distance from the rear axle center to the airbag is 0.5 m, and from formula (2), the total acceleration of the vehicle is 0.75 m / s 2 When the tire speed at time T1 is 30 km / h, and the tire speed at time T2, which is 0.5 minutes apart from time T1, is 84 km / h, it can be seen from formula (3) that the tire acceleration is 0.5 m / s2 Substituting the total acceleration of the vehicle and the acceleration of the tire into equation (1) for calculation, we can see that the slope of the road is 0.25.

[0069] Road roughness is obtained by: obtaining it through image information collected by a camera, where the camera is installed on the rearview mirror above the car's windshield; and / or obtaining it through acceleration and angular acceleration information in the X, Y, and Z directions measured by the vehicle's inertial sensors; and / or obtaining it through the longitudinal distance from the wheel center to the wheel arch output by the air suspension.

[0070] Specifically, there are several schemes for obtaining the road surface roughness: (1) A camera installed at the rearview mirror on the upper part of the car windshield collects the current vehicle road surface information, calculates the road surface roughness using image processing technology, and inputs the calculated road surface roughness into the carrier; (2) The vehicle inertial sensor can measure the acceleration and angular velocity of the vehicle in the X, Y, and Z directions, and obtains the road surface roughness using the change in acceleration and angular velocity; (3) The air suspension can output the longitudinal distance from the wheel center to the wheel arch, and obtains the road surface roughness using the change in the longitudinal distance from the wheel center to the wheel arch; (4) The road surface roughness obtained by scheme (1) and scheme (2) are integrated to obtain the road surface roughness, such as integrating scheme (1) and scheme (2) ) and take the average value as the road surface roughness; (5) fuse the road surface roughness obtained by scheme (1) and scheme (3) to obtain the road surface roughness, such as averaging the results of scheme (1) and scheme (3) and taking the average value as the road surface roughness; (6) fuse the road surface roughness obtained by scheme (2) and scheme (3) to obtain the road surface roughness, such as averaging the results of scheme (2) and scheme (3) and taking the average value as the road surface roughness; (7) fuse the road surface roughness obtained by scheme (1), scheme (2) and scheme (3) to obtain the road surface roughness, such as averaging the results of scheme (1), scheme (2) and scheme (3) and taking the average value as the road surface roughness.

[0071] Figure 3 This is a flow chart of determining the target category of parameter information according to an embodiment of the present application. Figure 3 As shown, in one embodiment, in step S101, the target category is determined based on the correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle. The method for determining the target category includes:

[0072] S201, obtaining parameter values corresponding to parameter information of a test vehicle and a load of the test vehicle;

[0073] See also Figure 4In the embodiment of the present application, parameter information of the test vehicle is obtained through ABM (Air Bag Module), ESP (Electronic Stability Program), EPS (Electronic Power Steering), transmission, engine, inertial sensor, air suspension, camera, etc. The parameter information of the test vehicle at least includes: information characterizing the engine working condition and the chassis working condition.

[0074] Specifically, the ABM is used to obtain the motion information of the test vehicle (such as the vehicle's lateral acceleration, the vehicle's yaw angular velocity, the vehicle's longitudinal acceleration, etc.), the ESP is used to obtain the test vehicle's braking and tire information (such as brake pressure, tire speed, tire acceleration, tire cornering force, vehicle speed, tire temperature, tire pressure, etc.), the transmission is used to obtain the test vehicle's transmission information (such as gear position, transmission output torque, etc.), the engine is used to obtain the test vehicle's power information (such as engine torque, engine speed, remaining fuel, accelerator pedal position, etc.), the EPS is used to obtain the test vehicle's steering information (such as steering wheel angle, steering wheel angle direction, steering wheel angle velocity, steering wheel angle velocity direction, driver steering torque, etc.), the camera at the interior rearview mirror above the car's windshield, one or more combinations of inertial sensors and air suspension are used to obtain road surface roughness; and the road surface slope is obtained using one or more methods selected from sensors, electronic maps and satellite positioning data. Since the drag coefficient, drag area and transmission ratio are fixed parameters of the vehicle, they can be input into the carrier; the road surface roughness and road surface slope are obtained indirectly through other means, so Figure 4 The method for obtaining the drag coefficient, drag area, transmission ratio, road surface roughness and road surface slope is not shown.

[0075] S202: determining a correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load based on the parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle;

[0076] Specifically, in an embodiment of the present invention, based on the parameter values corresponding to the parameter information of the test vehicle and the load of the test vehicle, the correlation coefficient between the parameter values corresponding to the parameter information of the test vehicle and the load is determined using the Pearson correlation coefficient method. According to the characteristics of the Pearson correlation coefficient method, the Pearson correlation coefficient is meaningful only when the variances of the two variables are not zero, and the value range of the Pearson correlation coefficient is [-1,1]. The Pearson correlation coefficient describes the strength of the linear correlation between the two variables. If the Pearson correlation coefficient is greater than zero, it means that the two variables are positively correlated, that is, the larger the value of one variable, the larger the value of the other variable; if the Pearson correlation coefficient is less than zero, it means that the two variables are negatively correlated, that is, the larger the value of one variable, the smaller the value of the other variable. The larger the absolute value of the Pearson correlation coefficient, the stronger the correlation. If the Pearson correlation coefficient is zero, it means that the two variables are not linearly related, but may be related in other ways.

[0077] S203 : Screen the parameter information of the test vehicle according to the correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load, and determine the target category of the parameter information to be acquired.

[0078] Specifically, in an embodiment of the present invention, after filtering the test vehicle's parameter information using the Pearson correlation coefficient method, the category of the parameter information related to load is selected as the target category of the parameter information. For example, information such as tire temperature, tire pressure, driver's seat belt warning, and passenger's seat belt warning is deleted, and information such as vehicle longitudinal acceleration, vehicle lateral acceleration, engine speed, engine status, vehicle speed, brake pressure, accelerator pedal position, brake pedal position, instantaneous fuel consumption, remaining fuel, engine output torque, engine loss torque, four-wheel speed value, transmission input torque, steering wheel angle, steering wheel angle direction, steering wheel angle velocity, steering wheel angle velocity direction, driver steering torque, gear position, transmission ratio, drag coefficient, road slope, and road roughness are selected as the target category of the parameter information.

[0079] In an optional embodiment, when a load prediction model is obtained by training a preset model using the characteristic information of a test vehicle as a training sample, the load prediction model is an XGBoost model. The XGBoost model is the optimal model selected from the decision tree model, the XGBoost model, and the random forest model. Figure 5 A flow chart of a load prediction model is shown. Figure 5 As shown, in one embodiment, in step S102, the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample, including:

[0080] S301. Using feature information of the test vehicle as a training sample, train a preset model to obtain multiple candidate models, wherein the preset model includes any one of a decision tree model, an XGBoost (EXtreme Gradient Boosting) model, and a random forest model.

[0081] In this embodiment, the characteristic information of the test vehicle includes the longitudinal acceleration of the test vehicle, the lateral acceleration of the test vehicle, the engine speed, the engine status, the vehicle speed, the brake pressure, the accelerator pedal position, the brake pedal position, the instantaneous fuel consumption, the remaining fuel, the engine output torque, the engine loss torque, the four-wheel speed value, the transmission input torque, the steering wheel angle, the steering wheel angle direction, the steering wheel angle velocity, the steering wheel angle velocity direction, the driver's steering torque, the gear position, the transmission ratio, the drag coefficient, the road surface slope, and the road surface roughness. The above data of the test vehicle are respectively input into the decision tree model, the XGBoost model and the random forest model for training to obtain multiple candidate models. The multiple candidate models here refer to: the trained decision tree model, the trained XGBoost model and the trained random forest model.

[0082] It should be noted that while the test vehicle and the current vehicle have the same feature information, the specific values corresponding to the test vehicle's feature information and the current vehicle's feature information are different. Because the current vehicle's feature information is primarily used to predetermine the vehicle's load, the current vehicle's feature information only has a single value, such as a single speed or a single remaining fuel value. The test vehicle's feature information, on the other hand, is primarily used to train the pre-set model, so it has multiple values, such as M speed values or N remaining fuel values.

[0083] S302. Based on the calculation accuracy and performance of the multiple candidate models, select the trained XGBoost model from the multiple candidate models as the load prediction model.

[0084] Specifically, the characteristic information of the test vehicle is input into multiple candidate models respectively, and the candidate models are used to calculate the load corresponding to the characteristic information. The calculated value is compared with the true value of the test vehicle, and the candidate model with the smallest error is selected as the load prediction model of this embodiment. In this embodiment, the trained XGBoost model is finally selected as the load prediction model. Since the characteristic information of the test vehicle has multiple values, some of them can be used for model training and other parts for model verification; or multiple values can be used for model training and some are selected for model verification. This embodiment does not impose any restrictions on the number of samples selected for training.

[0085] In addition, during the driving process, the vehicle may encounter vehicle failures caused by itself or the environment, resulting in invalid or abnormal driving data. In order to eliminate errors caused by emergencies in parameter information screening and subsequent preset model training, before performing parameter information screening, the parameter information collected by the test vehicle needs to be cleaned. According to the validity and rationality of the data, invalid and abnormal signals in the data are deleted, and valid data is screened out for parameter information screening and subsequent preset model training.

[0086] Based on the same inventive concept, Figure 6 This is a structural block diagram of a vehicle load prediction device according to an embodiment of the present application. Figure 6 As shown, the device includes:

[0087] An acquisition module 101 acquires parameter information of a current vehicle load that belongs to a target category, and uses the parameter information of the target category as characteristic information of the current vehicle;

[0088] The acquisition module 102 is used to input the characteristic information into the load prediction model to obtain the load of the current vehicle output by the load prediction model; wherein the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample.

[0089] In an optional embodiment, the device further comprises:

[0090] The determination module is used to determine the target category based on a correlation coefficient between a parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle.

[0091] In an optional embodiment, the determining module includes:

[0092] An acquisition submodule is used to obtain parameter values corresponding to the parameter information of the test vehicle and the load of the test vehicle;

[0093] a first determining submodule, configured to determine a correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load based on the parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle;

[0094] The second determining submodule is used to screen the parameter information of the test vehicle according to the correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load, and determine the target category of the parameter information to be obtained.

[0095] In an optional embodiment, the vehicle load prediction device further includes:

[0096] The training module is used to train the preset model using the characteristic information of the test vehicle as a training sample to obtain a load prediction model.

[0097] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, which stores a vehicle load prediction program. When the vehicle load prediction program is executed by a processor, the steps of the vehicle load prediction method of any of the above embodiments are implemented.

[0098] Based on the same inventive concept, an embodiment of the present application provides a vehicle, which predicts the vehicle load through the vehicle load prediction method of any one of the above embodiments.

[0099] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0100] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0101] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1The function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0105] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0106] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0107] The above is a detailed introduction to a resource allocation method and a resource allocation device provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

[0108] The above is a detailed introduction to the vehicle load prediction method, system and vehicle provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A vehicle load prediction method, characterized in that: The method comprises: Acquiring parameter information of a current vehicle belonging to a target category, and using the parameter information belonging to the target category as feature information of the current vehicle; wherein the parameter information of the target category includes: road slope, road roughness, and vehicle steering information; Inputting the characteristic information of the current vehicle into a load prediction model to obtain the load of the current vehicle output by the load prediction model; The load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample.

2. The vehicle load prediction method according to claim 1, characterized in that: The target category is determined based on a correlation coefficient between a parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle. The method for determining the target category includes: Obtaining parameter values corresponding to the parameter information of the test vehicle and the load of the test vehicle; determining a correlation coefficient between the parameter value corresponding to the parameter information of the test vehicle and the load based on the parameter value corresponding to the parameter information of the test vehicle and the load of the test vehicle; The parameter information of the test vehicle is screened according to a correlation coefficient between a parameter value corresponding to the parameter information of the test vehicle and the load, and a target category of the parameter information to be acquired is determined.

3. The vehicle load prediction method according to claim 1, characterized in that: The road slope is obtained through one or more of sensors, electronic maps, and satellite positioning data.

4. The vehicle load prediction method according to claim 1, characterized in that: The road surface roughness is obtained through image information collected by a camera, wherein the camera is installed at the rearview mirror on the upper part of the windshield of the car, and / or, The road surface roughness is obtained by measuring the acceleration and angular acceleration information in the X, Y, and Z directions by the vehicle inertial sensor, and / or The road surface roughness is obtained by the longitudinal distance from the wheel center to the wheel arch output by the air suspension.

5. The vehicle load prediction method according to any one of claims 1 to 4, characterized in that: The load prediction model includes: Any of the decision tree model, XGBoost model, and random forest model.

6. A vehicle load prediction device, characterized in that: The device comprises: an acquisition module, configured to acquire parameter information of a current vehicle belonging to a target category, and use the parameter information belonging to the target category as characteristic information of the current vehicle; wherein the parameter information of the target category includes: road slope, road roughness, and vehicle steering information; An acquisition module is used to input the characteristic information of the current vehicle into a load prediction model to obtain the load of the current vehicle output by the load prediction model; wherein, the load prediction model is obtained by training a preset model using the characteristic information of the test vehicle as a training sample.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a vehicle load prediction program, which, when executed by a processor, implements the steps of the vehicle load prediction method according to any one of claims 1 to 5.

8. A vehicle, characterized in that: The vehicle load is predicted by the vehicle load prediction method according to any one of claims 1 to 5.

Images