A vehicle control method and vehicle

Abstract

This application discloses a vehicle control method and a vehicle, relating to the field of vehicle control technology. The method includes: acquiring perception information of the target vehicle while it is in motion, the perception information including first information representing the vehicle's inertial motion state, second information representing the wheel motion state, and third information representing the suspension height state; based on the first, second, and third information, removing interference components generated by the vehicle's own motion from the first information to obtain a gravity direction representation quantity; generating a slope estimate of the road where the target vehicle is located based on the first information and the gravity direction representation quantity; and controlling the target vehicle based on the slope estimate to adapt to the current road slope conditions. This application can improve the accuracy of obtaining road slope.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a vehicle control method and a vehicle. Background Technology

[0002] Road gradient is a crucial parameter in vehicle control systems, directly impacting the safety and comfort of functions such as active suspension adjustment, hill start assist, energy recovery management, and traction control. Accurately acquiring road gradient information during vehicle operation is fundamental to achieving these functions.

[0003] Currently, road slope estimation mainly relies on inertial measurement units (IMUs). Commonly used methods fall into two categories: one is to integrate the angular velocity to obtain the attitude angle and then calculate the slope. However, this method is easily affected by sensor noise and suspension motion, causing the angular velocity integration result to accumulate errors over time, resulting in integration divergence and failing to guarantee the long-term stability of the slope estimation. The other method is to separate gravitational acceleration from acceleration to solve the slope. However, under intense driving conditions such as rapid acceleration and sharp turns, it is difficult to accurately separate the vehicle body motion acceleration, leading to a decrease in the accuracy of slope estimation.

[0004] Therefore, improving the accuracy and stability of road slope estimation has become an urgent problem to be solved. Summary of the Invention

[0005] In view of the above problems, this application provides a vehicle control method and vehicle that overcomes or at least partially solves the above problems, and the technical solution is as follows: A vehicle control method, the method comprising: When the target vehicle is in motion, the perception information of the target vehicle is acquired. The perception information includes first information for representing the inertial motion state of the vehicle, second information for representing the wheel motion state, and third information for representing the suspension height state. Based on the first information, the second information, and the third information, the interference component generated by the vehicle's own motion is removed from the first information to obtain the gravity direction characterization quantity; Based on the first information and the gravity direction representation quantity, an estimated slope value of the road where the target vehicle is located is generated; Based on the slope estimate, the target vehicle is controlled to adapt to the current road slope conditions.

[0006] This application embodiment acquires first, second, and third information including the vehicle's inertial motion state, wheel motion state, and suspension height state. Based on the first, second, and third information, it removes interference components generated by the vehicle's own motion from the first information to obtain a pure, interference-free gravity direction representation. Then, it generates a slope estimate based on the first information and the gravity direction representation. Finally, it performs vehicle control based on the slope estimate to adapt to the current road slope conditions. Thus, this application can accurately estimate the road slope and achieve slope adaptive control without increasing additional hardware costs, effectively improving the vehicle's driving stability, safety, and ride comfort under slope conditions.

[0007] Optionally, the step of removing the interference component generated by the vehicle's own motion from the first information based on the first information, the second information, and the third information to obtain the gravity direction characterization quantity includes: calculating the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle posture gravity acceleration generated by the vehicle's own motion based on the first information, the second information, and the third information to generate the interference component; obtaining the initial acceleration from the first information, and removing the interference component from the initial acceleration to obtain the gravity direction characterization quantity.

[0008] This alternative implementation calculates and integrates various interference components generated by the vehicle's own motion, and then uniformly removes them from the initial acceleration. This achieves comprehensive and accurate elimination of vehicle motion interference in the initial acceleration, effectively eliminating multiple errors caused by eccentric installation, translational conditions, and changes in vehicle body posture. It significantly improves the purity and reliability of the gravity direction representation, providing a solid foundation for accurate estimation of subsequent road slope.

[0009] Optionally, the step of calculating the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle attitude gravitational acceleration generated by the vehicle's own motion based on the first information, the second information, and the third information to generate the interference components includes: calculating the eccentric acceleration based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit; the installation position offset includes the offset of the installation position of the inertial measurement unit installed on the target vehicle from the vehicle's center of gravity in multiple directions; calculating the longitudinal motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information; calculating the lateral motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information, combined with vehicle dynamics parameters; and calculating the vehicle attitude gravitational acceleration based on the suspension height in the third information.

[0010] This optional implementation method classifies, models, and calculates four types of disturbance components—eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle attitude gravity acceleration—to achieve systematic identification and quantification of the vehicle's own motion disturbance in the initial acceleration. It effectively covers various error sources caused by installation deviations, translational conditions, and changes in vehicle attitude, providing accurate and comprehensive calculation basis for subsequent disturbance removal. This significantly improves the purity of the gravity direction representation and the accuracy and robustness of subsequent slope estimation.

[0011] Optionally, calculating the eccentric acceleration based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit includes: determining first intermediate data based on the angular acceleration and the offset; the first intermediate data is used to characterize the tangential eccentric acceleration caused by the angular acceleration; determining second intermediate data based on the angular velocity and the offset; the second intermediate data is used to characterize the normal eccentric acceleration caused by the angular velocity; and generating the eccentric acceleration based on the first intermediate data and the second intermediate data.

[0012] This optional implementation calculates the tangential eccentric acceleration caused by angular acceleration and the normal eccentric acceleration caused by angular velocity separately, and then integrates them to generate a complete eccentric acceleration. This achieves a systematic and comprehensive modeling of the Coriolis effect interference caused by the IMU installation position and the vehicle's center of gravity offset. It accurately quantifies the eccentric acceleration interference under rotational conditions, effectively avoids the gravity signal extraction deviation caused by incomplete eccentric acceleration calculation, provides a comprehensive and reliable basis for subsequent interference stripping, and improves the purity of the gravity direction representation and the robustness of subsequent slope estimation.

[0013] Optionally, calculating the longitudinal motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information includes: determining the longitudinal speed of the target vehicle based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information; and determining the longitudinal motion acceleration of the target vehicle based on the longitudinal speed at adjacent time points.

[0014] This optional implementation determines the longitudinal speed of the target vehicle by using wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal. Based on the difference in longitudinal speed between adjacent moments and a preset calculation period, it calculates the longitudinal motion acceleration, achieving accurate quantification of translational interference components under vehicle acceleration and deceleration conditions. It effectively isolates the longitudinal translational acceleration coupled with the road slope gravity signal, avoiding its influence on subsequent gravity direction characterization quantity extraction and slope calculation, and improving the stability and accuracy of slope estimation under dynamic conditions.

[0015] Optionally, the step of calculating the lateral acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information, and in combination with vehicle dynamics parameters, includes: determining the longitudinal speed of the target vehicle based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information; and calculating the lateral acceleration of the target vehicle in combination with the longitudinal speed and preset vehicle dynamics parameters.

[0016] This optional implementation determines the longitudinal speed of the target vehicle by using wheel speed, steering wheel angle, yaw rate, and driving or braking status signals. It then calculates the lateral acceleration by combining the longitudinal speed with preset vehicle dynamics parameters. This achieves accurate modeling and quantification of lateral translational interference components under conditions such as steering and lane changing. It effectively eliminates lateral translational acceleration coupled with road slope gravity signals, avoiding its interference with the extraction of gravity direction representation quantities and slope calculation, and improving the stability and accuracy of slope estimation under dynamic steering conditions.

[0017] Optionally, calculating the vehicle body attitude gravitational acceleration based on the suspension height in the third information includes: determining the vehicle body pitch angle and vehicle body roll angle based on the suspension height in the third information; and determining the vehicle body attitude gravitational acceleration components based on the vehicle body pitch angle and vehicle body roll angle, combined with the gravitational acceleration constant.

[0018] This optional implementation calculates the vehicle pitch and roll angles based on the suspension height signal and generates the vehicle attitude gravity acceleration component by combining the gravity acceleration constant. This achieves accurate quantification and modeling of the gravity projection deviation caused by changes in vehicle attitude, effectively eliminating the interference of additional gravity components caused by vehicle pitch and roll, avoiding the slope calculation deviation caused by its coupling with the road slope gravity signal, and further improving the purity of the gravity direction representation and the accuracy of subsequent slope estimation.

[0019] Optionally, generating the slope estimate of the road where the target vehicle is located based on the first information and the gravity direction representation includes: correcting the vehicle attitude angle corresponding to the first information based on the gravity direction representation; and determining the slope estimate of the road where the target vehicle is located based on the corrected vehicle attitude angle.

[0020] This optional implementation accurately corrects the vehicle attitude angle using a pure and interference-free gravity direction representation, effectively offsetting attitude calculation errors caused by various motion disturbances. Then, based on the corrected vehicle attitude angle, it accurately obtains the road slope estimate, significantly reducing the adverse effects of attitude deviation on slope recognition results. It can stably output accurate and reliable slope data under various dynamic driving conditions of the vehicle, providing accurate and effective data support for subsequent vehicle chassis system adaptability control.

[0021] Optionally, the step of correcting the vehicle attitude angle corresponding to the first information based on the gravity direction representation includes: determining the initial attitude angle of the target vehicle according to the angular velocity in the first information; processing the gravity direction representation using an error state Kalman filter to obtain a posterior error state estimate; and correcting the initial attitude angle based on the attitude error and gyroscope bias error in the posterior error state estimate to obtain the corrected vehicle attitude angle of the target vehicle.

[0022] This optional implementation first determines the initial attitude angle of the vehicle based on the angular velocity of the first information, then processes the gravity direction representation quantity by combining the error state Kalman filter to obtain the posterior error state estimate, and uses the attitude error and gyroscope bias error to correct the initial attitude angle. This can effectively suppress the cumulative drift error generated during the gyroscope integration process, improve the accuracy and real-time performance of the vehicle attitude angle calculation, and provide reliable attitude data support for subsequent accurate road slope estimation and vehicle attitude control.

[0023] Optionally, the method further includes: acquiring slope monitoring values ​​of the road where the target vehicle is located based on sensors, and obtaining initial road surface smoothness; acquiring the vertical vibration acceleration of the target vehicle's wheels collected by an unsprung acceleration sensor, and determining the target road surface smoothness based on the vertical vibration acceleration of the wheels and the initial road surface smoothness; determining a weighted fusion coefficient based on the target road surface smoothness, longitudinal motion acceleration, and lateral motion acceleration; and weighting and fusing the slope estimate and the slope monitoring values ​​based on the weighted fusion coefficient to output the target road slope value, so as to control the target vehicle based on the target road slope value.

[0024] This optional implementation combines the road slope monitoring values ​​collected by sensors with the initial road surface smoothness, accurately calibrates the target road surface smoothness using the vertical vibration acceleration of the wheels, and then determines the appropriate weighted fusion coefficient by correlating the longitudinal and lateral motion accelerations. The slope estimate and slope monitoring values ​​are adaptively weighted and fused, which can take into account the advantages of sensor measured data and algorithm estimated data, adapt to the slope calculation needs under different road surface smoothness conditions, effectively reduce the slope error caused by a single data source, improve the accuracy and adaptability of the road slope output results, and thus ensure that the vehicle can achieve stable and accurate vehicle control based on the target road slope.

[0025] Optionally, controlling the target vehicle based on the slope estimate to adapt to the current road slope conditions includes: when the target road slope value indicates an uphill condition, generating an increase in rear suspension stiffness command and / or activating hill-start assist command, and sending the increase in rear suspension stiffness command and / or the activation of hill-start assist command to the chassis execution system corresponding to the target vehicle to adapt to the current road slope conditions; when the target road slope value indicates a downhill condition, generating an increase in energy recovery intensity command and / or a decrease in drive system output torque command, and sending the increase in energy recovery intensity command and / or the decrease in drive system output torque command to the chassis execution system corresponding to the target vehicle to adapt to the current road slope conditions.

[0026] This optional implementation sends different control commands for uphill and downhill conditions. When going uphill, it increases the stiffness of the rear suspension and activates hill-start assist braking. When going downhill, it enhances energy recovery and reduces the drive output torque. By sending various control commands to the chassis execution system to complete the corresponding actions, the vehicle's driving status can be quickly adapted to the current slope and road conditions, effectively improving the vehicle's stability, driving safety, and ride comfort when driving on slopes. At the same time, it optimizes the power output and braking control effect of the vehicle during uphill driving.

[0027] A vehicle control device, comprising: The acquisition unit is used to acquire the perception information of the target vehicle when the target vehicle is in a driving state. The perception information includes first information for representing the inertial motion state of the vehicle, second information for representing the wheel motion state, and third information for representing the suspension height state. The stripping unit is used to strip the interference components generated by the vehicle's own motion from the first information based on the first information, the second information, and the third information, to obtain the gravity direction characterization quantity. The generation unit is used to generate an estimated slope value of the road where the target vehicle is located based on the first information and the gravity direction characterization quantity. A control unit is used to control the target vehicle based on the slope estimate to adapt to the current road slope conditions.

[0028] A vehicle includes a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to implement the vehicle control method as described above.

[0029] An electronic device, characterized in that it comprises: a memory for storing a computer program; and a processor for executing the computer program to implement the vehicle control method as described above.

[0030] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the vehicle control method described in any of the preceding claims.

[0031] A computer program product includes a computer program that, when executed by a processor, implements the steps of any of the above-described vehicle control methods.

[0032] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0033] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This application provides an illustrative flowchart of a vehicle control method according to an embodiment. Figure 1 ; Figure 2 This application provides an illustrative flowchart of a vehicle control method according to an embodiment. Figure 2 ; Figure 3 This application provides an illustrative flowchart of a vehicle control method according to an embodiment. Figure 3 ; Figure 4 This diagram illustrates the relative positions of the vehicle coordinate system and the inertial measurement unit in a vehicle control method provided in an embodiment of this application. Figure 4 This application provides an illustrative flowchart of a vehicle control method according to an embodiment. Figure 5 ; Figure 5 This paper shows one of the parameter relationship diagrams in a vehicle control method provided in an embodiment of this application; Figure 6 This illustrates a second parameter relationship diagram in a vehicle control method provided in an embodiment of this application; Figure 6 This paper shows a schematic diagram of the system framework corresponding to a vehicle control method provided in an embodiment of this application; Figure 7 A schematic structural diagram of a vehicle control device provided in an embodiment of this application is shown; Figure 7 A schematic structural diagram of a vehicle provided in an embodiment of this application is shown. Detailed Implementation

[0034] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0035] In verifying the cross-vehicle adaptability and robustness of intelligent driving algorithms, it is necessary to convert the multi-view videos collected by the original vehicle's sensors into target vehicle's view videos to provide data support for algorithm simulation.

[0036] Currently, road slope estimation mainly relies on vehicle motion information collected by inertial measurement units (IMUs). Commonly used slope estimation methods in existing technologies fall into two categories: One approach relies solely on integrating angular velocity data measured by an IMU (Integrated Measurement Unit) angular velocity sensor to calculate the vehicle's attitude angle change and thus the road slope. However, due to measurement noise and zero-bias instability inherent in the IMU sensor, and the additional interference introduced by the vibration and deformation of the suspension system during vehicle operation, the angular velocity integration results accumulate errors over time, leading to integration divergence and compromising the long-term stability of the slope estimation.

[0037] Another approach attempts to directly separate the gravitational acceleration component from the acceleration measured by the IMU accelerometer and calculate the road slope by extracting pure gravitational acceleration. The basic idea is to subtract the vehicle's motion acceleration from the total acceleration measured by the IMU to obtain the gravitational acceleration. However, in practical applications, vehicle driving conditions are complex and varied, especially under aggressive driving conditions such as rapid acceleration, emergency braking, and sharp turns. The amplitude of the vehicle's motion acceleration is large and changes rapidly, making it extremely difficult to accurately separate the motion acceleration, resulting in a significant decrease in the reliability of the slope estimation results.

[0038] To overcome or at least partially solve the above problems, this application considers that the initial acceleration collected by the vehicle inertial measurement unit is mixed with various interference components, including eccentric acceleration caused by the sensor installation position deviating from the center of mass, longitudinal and lateral motion acceleration caused by wheel acceleration, deceleration and steering, and vehicle attitude gravity acceleration components caused by vehicle pitch and roll. Therefore, it is necessary to remove the above interference components from the initial acceleration to obtain a pure and interference-free gravity direction characterization quantity.

[0039] Specifically, the eccentric acceleration component can be calculated using the first information (including angular velocity, angular acceleration, and initial acceleration collected by the inertial measurement unit); the longitudinal motion acceleration and lateral motion acceleration can be calculated using the second information (including wheel speed, steering wheel angle, yaw rate, driving or braking status signals, and vehicle dynamics parameters); and the vehicle body attitude gravity acceleration can be calculated using the third information (including suspension height).

[0040] Furthermore, by systematically eliminating all kinds of interfering accelerations from the initial acceleration components in the first information, a pure gravity direction representation quantity is separated. This gravity direction representation quantity does not change with the vehicle's own motion state. Then, based on this gravity direction representation quantity, the road slope is accurately obtained through state estimation methods to implement slope adaptive control of the chassis system.

[0041] Furthermore, refer to Figure 8 As shown, Figure 8 This is a flowchart of a vehicle control method provided in this application embodiment, which specifically includes the following steps S101-S104: S101. When the target vehicle is in motion, acquire the perception information of the target vehicle.

[0042] The perceived information includes first information representing the vehicle's inertial motion state, second information representing the wheel motion state, and third information representing the suspension height state.

[0043] Specifically, while the target vehicle is in motion, the system acquires the target vehicle's perception information. This perception information includes first information, second information, and third information.

[0044] The first information is collected by an inertial measurement unit (IMU). The IMU is usually installed near the vehicle's center of gravity and is used to collect the vehicle's angular velocity, angular acceleration, and initial acceleration in real time. It should be noted that the angular velocity, angular acceleration, and initial acceleration all include components corresponding to the three axes, namely the components in the X, Y, and Z axes of the vehicle coordinate system. The above data together constitute the first information, which is used to represent the inertial motion state of the vehicle in space and is the basic data for subsequent removal of interference components.

[0045] The second information is provided by wheel speed sensors, steering wheel angle sensors, yaw rate sensors, and the vehicle CAN bus.

[0046] In some embodiments, wheel speed sensors are integrated at the wheel hubs of each wheel to collect the rotational speed signals of the four wheels: left front, right front, left rear, and right rear. A steering wheel angle sensor is integrated into the steering column to collect the steering wheel angle signal. A yaw rate sensor can be integrated into the IMU or installed independently to collect the vehicle's yaw rate signal. Drive or braking status signals are acquired via the CAN bus, including accelerator pedal opening signals and brake pedal opening signals. These data collectively constitute the second information, used to represent the vehicle's wheel motion state and the driver's operational intentions.

[0047] The third information is provided by height sensors; these sensors are installed in each suspension system, specifically including a left front suspension height sensor, a right front suspension height sensor, a left rear suspension height sensor, and a right rear suspension height sensor, which are used to collect suspension height data at four positions: left front, right front, left rear, and right rear. This data collectively constitutes the third information, used to indicate the suspension height status, in order to calculate the vehicle's pitch and roll angles.

[0048] S102. Based on the first information, the second information and the third information, the interference component generated by the vehicle's own motion is removed from the first information to obtain the gravity direction characterization quantity.

[0049] In some embodiments, the initial acceleration component acquired by the inertial measurement unit is not pure gravitational acceleration, but rather contains various interference components generated by the vehicle's own motion, mainly including: Eccentric acceleration component: Because the inertial measurement unit (IMU) is not installed at the vehicle's center of gravity, an additional eccentric acceleration will be generated at the IMU when the vehicle rotates. This eccentric acceleration is related to the angular velocity, angular acceleration, and installation position offset, and can be calculated using the angular velocity, angular acceleration, and preset installation position offset from the first information.

[0050] Longitudinal acceleration: During vehicle operation, acceleration and deceleration generate longitudinal (X-axis) acceleration. This acceleration can be calculated using the wheel speed, steering wheel angle, yaw rate, and driving or braking status signals from the second information.

[0051] Lateral acceleration: During vehicle operation, steering will generate lateral acceleration along the vehicle's side (Y-axis direction). This acceleration can be calculated using the wheel speed, steering wheel angle, yaw rate, and driving or braking status signals from the second information, combined with vehicle dynamics parameters.

[0052] Vehicle attitude gravitational acceleration component: When a vehicle is traveling on an incline or experiencing pitch or roll, gravitational acceleration will produce an additional component in the vehicle coordinate system. This component can be obtained by calculating the vehicle pitch and roll angles using the suspension height from the third information, and then combining it with the gravitational acceleration constant.

[0053] Based on this, the aforementioned interfering accelerations are systematically removed from the initial acceleration components in the first set of information. Specifically, the eccentric acceleration component, longitudinal motion acceleration, lateral motion acceleration, and vehicle attitude gravity acceleration component are subtracted from the initial acceleration components, thereby separating a pure gravity direction representation. This gravity direction representation is only related to gravity and does not change with the vehicle's own motion state, serving as a reliable basis for subsequent calculations of road slope.

[0054] S103. Based on the first information and the gravity direction representation quantity, generate an estimated value of the slope of the road where the target vehicle is located.

[0055] In this step, the current attitude angle of the target vehicle is initially predicted based on the angular velocity data in the first information. Since the attitude angle obtained solely by integrating the angular velocity will drift over time, the gravity direction representation obtained in S102 needs to be used to correct the initially predicted attitude angle.

[0056] Since the gravity direction representation is a pure gravitational acceleration component after eliminating various motion disturbances, its direction always points towards the Earth's center. In the vehicle coordinate system, the distribution of this gravity direction representation along each axis has a definite geometric correspondence with the vehicle's current attitude angles (pitch and roll angles). Therefore, by analyzing the component proportions of the gravity direction representation along each axis of the vehicle coordinate system, the vehicle's current pitch and roll angles can be calculated.

[0057] Furthermore, when a vehicle is on a level road, the gravity direction representation is entirely distributed along the vertical axis, with zero longitudinal and lateral components. When a vehicle is traveling on a slope, the gravity direction representation produces a non-zero component along the longitudinal axis, the magnitude of which has a trigonometric function relationship with the road slope angle. Based on this geometric relationship, the estimated slope of the road where the vehicle is currently located can be determined by the ratio of the longitudinal to the vertical components of the gravity direction representation.

[0058] This step directly calculates the road slope using the gravity direction representation quantity, which is unaffected by vehicle motion. This avoids the drift problem caused by relying solely on angular velocity integrals, and improves the accuracy and stability of slope estimation.

[0059] S104. Based on the slope estimate, control the target vehicle to adapt to the current road slope conditions.

[0060] In this embodiment of the application, after obtaining the slope estimate based on the above S103, the slope condition type of the current road can be identified based on the slope estimate, such as uphill, downhill, gentle slope or steep slope, etc.; then, based on different slope condition types, corresponding vehicle control strategies are generated, including but not limited to the coordinated control of suspension parameters, braking system, drive system and steering system.

[0061] For example, in uphill conditions, the vehicle's climbing ability can be improved by adjusting the power output of the drive system and optimizing the suspension posture; in downhill conditions, the vehicle can be prevented from rolling backward or losing control by intervening in the braking system in advance and adjusting the suspension damping; in steep curve conditions, the vehicle's lateral stability can be improved by adjusting the steering assist and body roll control; thus, the vehicle can maintain a stable and controllable driving state under various slope conditions, effectively avoiding the control lag or over-control problems caused by inaccurate slope estimation in traditional control methods, and significantly improving the vehicle's driving safety and ride comfort in complex slope environments.

[0062] This application embodiment acquires first, second, and third information including the vehicle's inertial motion state, wheel motion state, and suspension height state. Based on the first, second, and third information, it removes interference components generated by the vehicle's own motion from the first information to obtain a pure, interference-free gravity direction representation. Then, it generates a slope estimate based on the first information and the gravity direction representation. Finally, it performs vehicle control based on the slope estimate to adapt to the current road slope conditions. Thus, this application can accurately estimate the road slope and achieve slope adaptive control without increasing additional hardware costs, effectively improving the vehicle's driving stability, safety, and ride comfort under slope conditions.

[0063] As an extension and refinement of the above embodiments, refer toFigure 9 As shown, Figure 9 This application provides an embodiment of another vehicle control method, which specifically includes the following steps S201-S206: S201. When the target vehicle is in motion, acquire the perception information of the target vehicle.

[0064] The perceived information includes first information representing the vehicle's inertial motion state, second information representing the wheel motion state, and third information representing the suspension height state.

[0065] S202. Based on the first information, the second information, and the third information, calculate the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle body attitude gravity acceleration generated by the vehicle's own motion to generate the interference components.

[0066] It should be noted that the initial acceleration in the first information collected by the IMU during vehicle movement is a three-axis acceleration component. This includes not only the desired gravitational acceleration but also various interfering accelerations caused by the vehicle's own motion, and therefore cannot accurately reflect the effective gravity signal of the road slope. This is because dynamic conditions such as vehicle acceleration and deceleration, steering, and suspension undulations generate longitudinal and lateral translational acceleration interference; the IMU's installation position being off-center from the vehicle's center of gravity also introduces eccentric acceleration interference due to the Coriolis effect; and changes in the vehicle's pitch and roll attitude also generate additional gravitational component interference.

[0067] The aforementioned disturbances can directly cause slope calculations to drift and become distorted, making them unsuitable for dynamic and complex road conditions. Therefore, it is necessary to classify, calculate, and remove the various disturbance accelerations contained in the initial acceleration.

[0068] Based on this, this step calculates the interference components according to the first information, the second information and the third information. The interference components include the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration and vehicle body posture gravity acceleration generated by the vehicle's own motion.

[0069] Among them, eccentric acceleration, also known as triaxial Coriolis acceleration, is mainly caused by the offset between the IMU installation position and the vehicle's center of gravity, combined with the vehicle's attitude rotation motion. It is used to characterize the acceleration interference caused by the Coriolis effect and directly affects the extraction accuracy of the gravity signal in the initial acceleration.

[0070] Longitudinal acceleration is the translational acceleration generated along the longitudinal direction of the vehicle body during driving. It is mainly related to the acceleration and deceleration conditions of the vehicle and can reflect the changes in the longitudinal driving state of the vehicle. It is one of the main translational disturbance components in the initial acceleration and can cause longitudinal deviations in the gravity direction representation.

[0071] Lateral motion acceleration is the translational acceleration generated along the side of the vehicle body when the vehicle is turning or driving. It is closely related to the vehicle's turning state and driving speed and is used to characterize the acceleration interference caused by the vehicle's lateral motion. It can interfere with the accurate extraction of gravity signals in the lateral direction.

[0072] Vehicle attitude gravitational acceleration is an additional component interference in the three-axis detection direction of the IMU caused by changes in vehicle attitude such as pitch and roll. It is not a new acceleration generated by the vehicle's own translation or rotation, but rather an attitude projection deviation of gravitational acceleration. It can reflect the influence of vehicle attitude changes on the gravitational component in the initial acceleration and is an important interference factor that causes deviations in slope calculation.

[0073] Specifically, refer to Figure 10 As shown, the specific calculation process of S202 (calculating the eccentric acceleration, longitudinal acceleration, lateral acceleration, and vehicle attitude gravitational acceleration generated by the vehicle's own motion based on the first information, the second information, and the third information) includes the following steps S2021-S2024: S2021. Based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit, calculate the eccentric acceleration.

[0074] The installation position offset includes the offset of the installation position of the inertial measurement unit installed on the target vehicle from the vehicle's center of gravity in multiple directions.

[0075] In some embodiments, the inertial measurement unit (IMU) is typically installed in the vehicle chassis area. Due to vehicle structural layout limitations, the actual installation location of the IMU is difficult to perfectly coincide with the vehicle's center of gravity, thus resulting in offsets in multiple directions.

[0076] Specifically, the multiple directions include the X-axis (vehicle longitudinal direction), Y-axis (vehicle lateral direction), and Z-axis (vehicle vertical direction) of the vehicle coordinate system. The origin of the vehicle coordinate system is the vehicle's center of gravity. Therefore, in each axis, the installation position of the inertial measurement unit has a fixed offset distance relative to the vehicle's center of gravity. These offset distances in the three axes together constitute the installation position offset described in this application. It should be noted that this offset is an inherent parameter of the vehicle, which can be pre-calibrated and stored in the vehicle controller, and its value does not change with the vehicle's driving state.

[0077] Reference Figure 10The diagram shows the vehicle coordinate system and inertial measurement unit (IMU) installation offset provided in this embodiment. The vehicle coordinate system, with the vehicle's center of gravity 41 as its origin, has the X-axis representing the vehicle's longitudinal forward direction, the Y-axis representing the vehicle's lateral rightward direction, and the Z-axis representing the vehicle's vertical upward direction. 42 represents the actual installation position of the IMU, which has a three-axis offset from the vehicle's center of gravity. This represents the offset in the X-axis direction. This is the offset in the Y-axis direction. The offset in the Z-axis direction, the , , The offsets of the IMU installation position relative to the center of mass on the three axes are used to calculate the eccentric acceleration when the vehicle rotates, providing a basis for subsequently removing interference from the initial acceleration components, purifying the pure gravitational acceleration components, and solving the road slope.

[0078] Furthermore, during vehicle operation, the acceleration signal collected by the IMU is affected by its installation position. When the vehicle rotates or changes attitude, an IMU mounted off-center will generate additional eccentric acceleration due to the Coriolis effect. This component is an invalid interference signal and will be directly superimposed on the original acceleration data, causing deviations in subsequent gradient calculations. Therefore, it is necessary to calculate the eccentric acceleration to provide accurate parameters for subsequent interference removal and ensure the purity of the original acceleration data.

[0079] S2022. Based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking status signal in the second information, calculate the longitudinal motion acceleration.

[0080] Specifically, during vehicle operation, when the vehicle drives, brakes, or accelerates / decelerates, the longitudinal translational acceleration is directly superimposed on the original acceleration signal and coupled with the gravity component corresponding to the road slope. If this interference is not removed, it will lead to significant deviations when extracting the pure gravity component from the acceleration and calculating the road slope, especially under dynamic conditions of frequent acceleration and deceleration, where the error will be further amplified. Therefore, it is necessary to calculate the longitudinal motion acceleration to provide data for subsequently removing longitudinal translational interference from the initial acceleration component, ensuring the purity of the effective gravity signal.

[0081] Specifically, in this step, the following parameters from the second information are needed: Wheel speed is the real-time rotational speed signal of each wheel of the target vehicle, used to calculate the vehicle's basic driving speed. Steering wheel angle signal represents the vehicle's steering angle, used to correct speed calculation errors under steering conditions. Yaw rate signal represents the vehicle's rotational rate around the Z-axis, used for speed correction in the vehicle dynamics model. Driving or braking status signal indicates whether the vehicle is currently driving, braking, or coasting, used to distinguish between acceleration and deceleration conditions.

[0082] Furthermore, the longitudinal vehicle speed is first calculated using the aforementioned data. This speed is the real-time speed of the target vehicle along the X-axis and forms the basis for calculating the longitudinal acceleration. Then, the longitudinal acceleration is calculated based on the longitudinal vehicle speed. This acceleration is the translational acceleration component generated by the vehicle's longitudinal acceleration and deceleration, and it represents interference signals that need to be removed.

[0083] S2023. Based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking status signal in the second information, and in combination with vehicle dynamics parameters, calculate the lateral motion acceleration.

[0084] Specifically, during vehicle operation, significant lateral translational acceleration is generated when the vehicle turns, changes lanes, or curves. This component is directly superimposed on the original acceleration signal and coupled with the gravity component corresponding to the road slope. If this interference is not removed, it will lead to substantial deviations when extracting the pure gravity component from the acceleration and calculating the road slope, especially under dynamic conditions of continuous turning and frequent lane changes, where the error will be further amplified. Therefore, it is necessary to calculate the lateral acceleration accurately to provide data for subsequent interference removal and ensure the purity of the effective gravity signal.

[0085] Specifically, in this step, the parameters required in the second information are the same as those in S2022, and will not be repeated here. Unlike S2022 for calculating longitudinal acceleration, this step also requires vehicle dynamics parameters when calculating lateral acceleration, including front axle lateral stiffness, rear axle lateral stiffness, front wheel slip angle, rear wheel slip angle, center of gravity slip angle, and vehicle mass. These parameters reflect the lateral mechanical characteristics of the tires and the basic physical properties of the vehicle, and are key parameters for describing the lateral dynamic response of the vehicle.

[0086] Furthermore, based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information, the longitudinal speed of the target vehicle is determined; then, by combining the longitudinal speed with vehicle dynamics parameters, the accurate lateral acceleration is obtained.

[0087] Through the above method, this step can accurately calculate the lateral motion acceleration, providing reliable data for removing lateral translational interference from the initial acceleration components, ensuring the purity of the gravity direction representation, and thus improving the accuracy of road slope estimation. Especially under dynamic conditions such as continuous turning and frequent lane changes, it can effectively suppress the impact of lateral interference on slope calculation.

[0088] S2024. Based on the suspension height in the third information, calculate the vehicle body attitude gravitational acceleration.

[0089] It should be noted that during vehicle operation, when the vehicle pitches or rolls due to suspension compression or extension, or road bumps, the changes in vehicle attitude will cause an additional component of gravitational acceleration in the vehicle coordinate system. This component is highly similar in characteristics to the gravitational component induced by road slope. Failure to distinguish between them will directly lead to distortion in slope calculation. Therefore, this step needs to calculate the gravitational acceleration component corresponding to the vehicle attitude to restore all the gravitational interference caused by the vehicle attitude.

[0090] Specifically, suspension height is the suspension travel height signal corresponding to each wheel of the target vehicle, reflecting the vertical displacement of the vehicle body relative to the wheels in real time, and is the core data source for calculating the vehicle body attitude. Furthermore, based on the suspension height, the vehicle body pitch angle and roll angle are calculated. The vehicle body pitch angle represents the vehicle's tilt angle around the X-axis (longitudinal direction), used to distinguish the vehicle body pitch attitude from the longitudinal slope of the road; the vehicle body roll angle represents the vehicle's tilt angle around the Y-axis (lateral direction), used to distinguish the vehicle body roll attitude from the lateral slope of the road.

[0091] Then, the gravitational acceleration constant is used for standard conversion of gravitational acceleration components. Furthermore, by using the relationship between the vehicle pitch angle, vehicle roll angle and gravitational acceleration constant, the vehicle attitude gravitational acceleration components are calculated. These vehicle attitude gravitational acceleration components can be understood as gravitational acceleration components induced by the vehicle's own pitch and roll attitudes, and are interference signals that need to be eliminated.

[0092] This application embodiment classifies, models, and calculates four types of interference components—eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle attitude gravity acceleration—to achieve systematic identification and quantification of the vehicle's own motion interference in the initial acceleration. This effectively covers various error sources caused by installation deviations, translational conditions, and changes in vehicle attitude, providing accurate and comprehensive calculation basis for subsequent interference stripping. As a result, it significantly improves the purity of the gravity direction representation and the accuracy and robustness of subsequent slope estimation.

[0093] S203. Obtain the initial acceleration from the first information, and remove the interference component from the initial acceleration to obtain the gravity direction characterization quantity.

[0094] Specifically, the initial acceleration is extracted from the first information, namely the raw triaxial acceleration data collected by the IMU. This initial acceleration contains various interference components caused by the vehicle's own motion, and cannot be directly used to reflect the direction of gravity and the slope. Subsequently, combined with the interference components calculated above, these interference components are removed one by one from the initial acceleration to completely eliminate the influence of factors such as the vehicle's own motion and installation deviations on the gravity signal. The final gravity direction representation contains only signals related to the action of gravity, which can accurately reflect the direction of gravity and can be directly used for subsequent slope calculation, attitude calibration and other processes, ensuring the accuracy and reliability of subsequent related calculations and avoiding problems such as slope calculation deviation and inappropriate control strategies caused by residual interference.

[0095] Based on the above S202-S203, by calculating and integrating various interference components generated by the vehicle's own motion, and then uniformly separating them from the initial acceleration, the interference of vehicle motion in the initial acceleration is comprehensively and accurately eliminated. This effectively eliminates multiple errors caused by eccentric installation, translational conditions and changes in vehicle body posture, significantly improves the purity and reliability of the gravity direction representation, and provides a solid foundation for the accurate estimation of subsequent road slope.

[0096] S204. Based on the gravity direction representation quantity, correct the vehicle attitude angle corresponding to the first information.

[0097] It should be noted that vehicle attitude angles typically include pitch, roll, and yaw; among them, pitch and roll are directly related to road slope. Therefore, the vehicle attitude angles in this embodiment mainly refer to pitch and roll.

[0098] Specifically, the gravity direction representation is the pure gravitational acceleration component after eliminating various motion disturbances, and its direction always points towards the Earth's center. In the vehicle coordinate system, the distribution of this gravity direction representation in each axis has a definite geometric correspondence with the vehicle's current pitch and roll angles: when the vehicle is on a level road, the gravity direction representation is completely distributed on the vertical axis, and the longitudinal and lateral components are zero; when the vehicle is traveling on a slope or the vehicle body is tilted, the gravity direction representation will produce non-zero components on the longitudinal axis and / or the lateral axis, and the magnitude of the components has a definite trigonometric function relationship with the pitch and roll angles.

[0099] Therefore, by analyzing the proportions of the components of the gravity direction representation in each axis of the vehicle coordinate system, the vehicle's current true pitch and roll angles can be calculated. By fusing this result with the attitude angle prediction obtained from angular velocity integration, and using the gravity direction representation to correct the prediction in real time, the cumulative drift caused by angular velocity integration can be effectively suppressed, resulting in accurate and stable vehicle attitude angles.

[0100] Preferably, the above-mentioned fusion correction process can be implemented using a Kalman filter. Specifically, the angular velocity data in the first information is used as the basis for state prediction, and the vehicle attitude angle is predicted through the time update equation of the Kalman filter; simultaneously, the gravity direction representation is used as the observation, and the predicted value is corrected through the measurement update equation of the Kalman filter. The Kalman filter can dynamically adjust the Kalman gain according to the uncertainty of prediction and observation, achieve optimal estimation of attitude angle, effectively suppress the drift caused by angular velocity integral, and maintain high-frequency response capability to attitude changes.

[0101] S205. Based on the corrected vehicle attitude angle, determine the estimated slope of the road where the target vehicle is located.

[0102] Furthermore, based on the corrected vehicle attitude angles, the corrected pitch and roll angles are obtained. The estimated road slope includes two components: longitudinal slope and lateral slope, which are then combined to describe the road's tilt state.

[0103] Specifically, the longitudinal slope is represented by the pitch angle. When the pitch angle is positive, it indicates that the vehicle is in an uphill condition; when the pitch angle is negative, it indicates that the vehicle is in a downhill condition; and when the pitch angle is zero or close to zero, it indicates that the vehicle is in a level road condition.

[0104] Lateral slope is characterized by the roll angle. When the roll angle is positive or negative, it means that the vehicle is on a laterally inclined road surface with one side higher than the other. When the roll angle is zero or close to zero, it means that the vehicle is on a road surface with no lateral inclination.

[0105] Based on the above S204-S205, the vehicle attitude angle is accurately corrected by a pure and interference-free gravity direction representation quantity, which effectively offsets the attitude measurement error caused by various motion interferences. Then, the road slope estimate is accurately obtained based on the corrected vehicle attitude angle, which greatly reduces the adverse effects of attitude deviation on the slope recognition result. It can stably output accurate and reliable slope data under various dynamic driving conditions of the vehicle, and provide accurate and effective data support for the subsequent vehicle chassis system adaptability control.

[0106] S206. Based on the slope estimate, control the target vehicle to adapt to the current road slope conditions.

[0107] Specifically, the above-mentioned control of the target vehicle based on the slope estimate to adapt to the current road slope conditions can be further divided into the following two conditions: (1) When the target road slope value indicates an uphill condition, generate an increase rear suspension stiffness command and / or activate hill start assist command, and send the increase rear suspension stiffness command and / or activate hill start assist command to the chassis execution system corresponding to the target vehicle to adapt to the current road slope condition.

[0108] During uphill driving, the vehicle experiences a backward drag effect due to the rearward component of gravity along the slope. Simultaneously, the rear suspension bears a greater load due to the rearward shift of the vehicle's center of gravity, causing the vehicle to pitch backward, affecting ride comfort and drive wheel traction. Therefore, when the target road gradient indicates an uphill condition, a command to increase rear suspension stiffness is generated. This increases the support force of the rear suspension through the active suspension system, suppressing vehicle pitch and maintaining vehicle stability. Simultaneously, a command to activate hill start assist is generated, ensuring that the braking system maintains a certain braking pressure even after the driver releases the brake pedal, preventing the vehicle from rolling backward when starting on an incline and ensuring safe starting.

[0109] The generated instructions are then transmitted via an in-vehicle communication network (such as CAN bus or Ethernet) to the chassis execution system corresponding to the target vehicle. The chassis execution system includes, but is not limited to, an active suspension system, an electronic stability control system, a hill start assist system, a motor controller, and a battery management system. Each execution system automatically adjusts its operating parameters based on the received instructions, enabling the vehicle to smoothly and safely adapt to road conditions with varying inclines.

[0110] (2) When the target road slope value indicates a downhill condition, generate an instruction to increase the energy recovery intensity and / or a instruction to reduce the output torque of the drive system, and send the instruction to increase the energy recovery intensity and / or the instruction to reduce the output torque of the drive system to the chassis execution system corresponding to the target vehicle to adapt to the current road slope condition.

[0111] During downhill driving, the vehicle experiences a natural acceleration due to the component of gravity acting forward along the slope. Relying solely on traditional mechanical braking to control speed can easily lead to overheating and brake fade in the braking system, while also wasting energy. Therefore, when the target road gradient indicates a downhill condition, a command to increase energy recovery intensity is generated. This converts the vehicle's potential energy during the downhill descent into electrical energy stored in the battery via the motor control system, achieving energy recovery while assisting in speed control. Simultaneously, a command to reduce the output torque of the drive system is generated to prevent the drive system from continuing to output driving force and causing excessive speed. If necessary, negative torque auxiliary braking can be actively applied to achieve stable downhill speed control.

[0112] Then, the generated instructions are sent to the chassis execution system corresponding to the target vehicle to adapt to the current road slope conditions.

[0113] Specifically, the generated instructions are sent to the chassis execution system corresponding to the target vehicle via an in-vehicle communication network (such as CAN bus or Ethernet). The chassis execution system includes, but is not limited to, an active suspension system, an electronic stability control system, a hill start assist system, a motor controller, and a battery management system. Each execution system automatically adjusts its operating parameters according to the received instructions, enabling the vehicle to smoothly and safely adapt to road conditions with varying inclines.

[0114] Through the above-mentioned slope adaptability control, the embodiments of this application can actively adjust the suspension characteristics, braking pressure distribution, energy recovery intensity and drive torque output in the vehicle chassis system according to the real-time estimated road slope, so as to realize the adaptive control of the vehicle on the slope road conditions, effectively prevent the vehicle from rolling back on the slope, suppress the vehicle body pitch, optimize energy recovery, and comprehensively improve the vehicle's driving safety, smoothness and economy under slope conditions.

[0115] In some embodiments, the above-described S2021 (calculating the eccentric acceleration based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit) can be further refined into the following steps: Step 11: Determine the first intermediate data based on the angular acceleration and the offset.

[0116] The first intermediate data is used to characterize the tangential eccentric acceleration caused by angular acceleration. The tangential eccentric acceleration refers to the additional acceleration component along the rotational tangential direction induced by angular acceleration at the IMU mounting position when the vehicle rotates around the center of mass, which is due to the offset of the IMU mounting position relative to the center of mass.

[0117] Specifically, the first intermediate data The following formula can be used as a reference:

[0118] in, The vector (column vector) representing angular acceleration comes from attitude data and characterizes the rate of change of the vehicle's rotational speed around the X, Y, and Z axes. , , These are the angular acceleration components of the vehicle around the X, Y, and Z axes, respectively. The vector (column vector) representing the offset is the fixed distance between the IMU mounting position and the centroid. , , These represent the offset distances of the IMU's installation position relative to the centroid on the X, Y, and Z axes, respectively; T is the matrix transpose symbol, used to convert row vectors into column vectors, conforming to the vector operation specifications in the vehicle coordinate system.

[0119] Furthermore, through the vector cross product of the angular acceleration and the offset described above, the tangential additional acceleration induced by the vehicle's angular acceleration at the eccentric mounting position of the IMU is quantified, reflecting the eccentric acceleration generated at the eccentric mounting position of the IMU due to the change in rotational angular acceleration.

[0120] Step 12: Determine the second intermediate data based on the angular velocity and the offset.

[0121] The second intermediate data is used to characterize the normal eccentric acceleration caused by angular velocity. The normal eccentric acceleration refers to the additional centripetal acceleration component induced by the angular velocity at the IMU installation position and directed towards the center of mass when the vehicle is continuously rotating around the center of mass. This is because the IMU installation position is offset relative to the center of mass. Its magnitude is proportional to the square of the angular velocity.

[0122] Specifically, the second intermediate data The following formula can be used as a reference:

[0123] in, The vector (column vector) representing the three-axis angular velocities comes from attitude data and characterizes the angular rates of the vehicle's rotation about the X, Y, and Z axes. , , These are the real-time rotational angular rates of the vehicle around the X, Y, and Z axes, respectively, derived from attitude data.

[0124] It should be noted that the second intermediate data here... The calculation process involves a double vector cross product operation, which requires two steps: first, the inner cross product within the parentheses is calculated, and then the outer cross product is executed. The matrix expansion of the double vector cross product is then used to accurately quantify the coupled components of centrifugal acceleration and Coriolis acceleration induced by the continuous rotation of the vehicle at the eccentric mounting position of the IMU.

[0125] Step 13: Generate the eccentric acceleration based on the first intermediate data and the second intermediate data.

[0126] Furthermore, in obtaining the first intermediate data Second intermediate data Then, the eccentric acceleration can be obtained by referring to the following formula:

[0127] in, Represents eccentric acceleration (column vector). The X-axis (longitudinal) component of the eccentric acceleration characterizes the additional acceleration disturbance in the longitudinal direction of the vehicle caused by the IMU's installation eccentricity. The Y-axis (lateral) component of eccentric acceleration characterizes the additional acceleration disturbance generated by the IMU in the lateral direction of the vehicle due to the installation eccentricity. The Z-axis (vertical) component of the eccentric acceleration characterizes the additional acceleration disturbance in the vertical direction of the vehicle caused by the IMU's installation eccentricity.

[0128] Then, by summing the first and second intermediate data obtained in the above two steps according to the above calculation formula, the final eccentric acceleration can be synthesized completely. This component fully reflects all the additional acceleration interference introduced by the IMU installation position being off-center from the vehicle's center of gravity, providing an accurate calculation basis for subsequently removing this interference from the original acceleration and purifying the effective gravity signal.

[0129] Based on steps 11-12 above, by calculating the tangential eccentric acceleration caused by angular acceleration and the normal eccentric acceleration caused by angular velocity respectively, and then integrating them to generate a complete eccentric acceleration, a systematic and comprehensive modeling of the Coriolis effect interference caused by the IMU installation position and the vehicle's center of gravity offset is achieved. The eccentric acceleration interference under rotational conditions is accurately quantified, effectively avoiding the gravity signal extraction deviation caused by incomplete eccentric acceleration calculation. This provides a comprehensive and reliable basis for subsequent interference stripping, and improves the purity of the gravity direction representation and the robustness of subsequent slope estimation.

[0130] In some embodiments, the above-mentioned S2022 (calculating the longitudinal motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information) can be further refined into the following steps: Step 21: Based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking status signal in the second information, determine the longitudinal speed of the target vehicle.

[0131] It should be noted that under different operating conditions such as steering, driving acceleration, and braking deceleration, the wheels are prone to slippage or rotation. Directly calculating the vehicle speed by averaging the four wheel speeds will produce significant errors, leading to distortion in subsequent longitudinal acceleration calculations. Therefore, it is necessary to first determine whether the vehicle is in a large-angle steering condition based on the steering wheel angle and yaw rate, and to determine whether the vehicle is in a severe acceleration or deceleration condition based on driving or braking status signals. The vehicle speed calculation method should be adaptively switched to ensure a stable and accurate longitudinal vehicle speed under various dynamic driving conditions, providing a reliable foundation for subsequent longitudinal acceleration calculations.

[0132] Specifically, the optimal longitudinal speed calculation method for different working conditions includes the following: Condition 1: When the steering wheel angle δ is greater than the preset steering threshold and the yaw rate r is greater than the preset yaw threshold, the vehicle is determined to be in a large-angle steering condition. It should be noted that the preset steering threshold and preset yaw threshold are preset condition judgment thresholds, which can be adjusted according to vehicle type and driving scenario. At this point, due to the large steering angle of the front wheels, the wheel speed is prone to slippage. Therefore, only the average rear wheel speed is used to calculate the longitudinal speed. The specific formula is as follows:

[0133] in, The longitudinal speed of the target vehicle; and The linear velocities of the left and right rear wheels, respectively, can be calculated by multiplying the wheel speed by the wheel rolling radius.

[0134] Under this condition, the rear wheels are less affected by steering slippage. Using the average value of the rear wheels can effectively correct the deviation in vehicle speed calculation under steering conditions and avoid vehicle speed distortion caused by front wheel slippage.

[0135] Condition 2: When the drive pedal opening AccP is greater than the preset drive threshold, or the brake pedal opening BrakP is greater than the preset brake threshold, the vehicle is determined to be in a drive acceleration or brake deceleration condition.

[0136] At this point, the drive wheels are prone to slippage, and the brake wheels are prone to slippage. Directly calculating the vehicle speed using wheel rotation speed will result in a significant error. Therefore, a recursive prediction method is used to calculate the longitudinal vehicle speed. The specific formula is as follows:

[0137] in, The longitudinal speed of the target vehicle at the current moment; The longitudinal speed of the vehicle at the previous moment; The longitudinal acceleration of the vehicle can be obtained from the original acceleration of the IMU after preliminary interference removal, or it can be estimated from the torque of the power system. The preset calculation period of the vehicle controller in this embodiment is in seconds (s), for example, 0.01s, 0.02s, etc., which is the time interval between two adjacent vehicle speed calculations. It should be noted that the time interval between the "previous moment" and the "current moment" is the same as the preset calculation period of the vehicle controller in this embodiment. seconds, for example, if =0.01s, then the above recursive formula is executed once every 0.01 seconds to update the longitudinal vehicle speed.

[0138] Furthermore, by deriving the integral calculation of the vehicle speed and acceleration at the previous moment through this formula, the wheel slippage and rotation errors under acceleration and deceleration conditions are effectively avoided, ensuring the continuity and accuracy of vehicle speed calculation.

[0139] Condition 3: When the vehicle does not meet the above-mentioned steering, acceleration, and deceleration conditions, it is determined that the vehicle is in a steady-state straight-line driving and constant-speed driving condition. At this time, there is no obvious wheel slippage or rotation; therefore, the longitudinal speed is calculated using the average speed of all four wheels. The specific formula is as follows:

[0140] in, Let the longitudinal speed of the target vehicle be denoted as . These are the linear velocities of the left and right front wheels, respectively. These are the linear velocities of the left and right rear wheels, respectively.

[0141] By using the four-wheel average, random fluctuations in wheel speed can be effectively smoothed out, improving the robustness and accuracy of vehicle speed calculation under steady-state conditions.

[0142] Furthermore, the embodiments of this application use adaptive calculation logic based on different working conditions to select the optimal longitudinal speed calculation method for different working conditions.

[0143] Step 22: Determine the longitudinal motion acceleration based on the difference between the longitudinal vehicle speed at the current moment and the longitudinal vehicle speed at the previous moment, and the preset calculation period.

[0144] It should be noted that longitudinal acceleration can be understood as the rate of change of the vehicle's longitudinal speed over time, directly reflecting the magnitude of the translational acceleration caused by driving and braking. This acceleration is superimposed on the raw acceleration signal acquired by the IMU and is considered an interference component independent of road gradient.

[0145] Therefore, by using the difference between the longitudinal vehicle speed at the current moment and the previous moment, and combining it with a preset calculation period to perform differential calculation, the longitudinal acceleration of the vehicle can be accurately calculated. The specific calculation formula is as follows:

[0146] in, Represents longitudinal acceleration. This indicates the longitudinal speed of the vehicle at the current moment; This indicates the longitudinal speed of the vehicle at the previous moment.

[0147] Based on steps 21-22 above, the longitudinal speed of the target vehicle is determined by wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal. The longitudinal acceleration is calculated based on the difference in longitudinal speed between adjacent moments and a preset calculation period. This achieves accurate quantification of translational interference components under vehicle acceleration and deceleration conditions, effectively isolates the longitudinal translational acceleration coupled with road slope gravity signal, avoids its influence on subsequent gravity direction representation quantity extraction and slope calculation, and improves the stability and accuracy of slope estimation under dynamic conditions.

[0148] In some embodiments, the above-mentioned S2023 (calculating the lateral acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information, and in combination with vehicle dynamics parameters) can be further refined into the following steps: Step 31: Based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking status signal in the second information, determine the longitudinal speed of the target vehicle.

[0149] This step can be referred to the explanation of step 21 above, and will not be repeated here.

[0150] Step 32: Calculate the lateral acceleration of the target vehicle by combining the longitudinal vehicle speed with the preset vehicle dynamics parameters.

[0151] Specifically, it is necessary to first obtain the front axle lateral stiffness, rear axle lateral stiffness, front wheel slip angle, rear wheel slip angle, center of gravity slip angle, and vehicle mass from the vehicle dynamics parameters.

[0152] Among them, the front axle lateral stiffness and rear axle lateral stiffness are inherent dynamic parameters of the target vehicle, representing the magnitude of the lateral force of the axle under a unit slip angle. They are vehicle calibration parameters and are pre-stored in the vehicle controller. The front wheel slip angle and rear wheel slip angle represent the angle between the actual driving direction of the wheel and the wheel orientation, and are calculated from real-time data such as longitudinal speed, steering wheel angle, and yaw rate. The center of gravity slip angle represents the angle between the velocity direction of the vehicle's center of gravity and the longitudinal direction of the vehicle, and is a core state parameter of the vehicle's lateral dynamics. The vehicle mass is the curb weight of the target vehicle, which is an inherent parameter of the vehicle and is used for the conversion of lateral force to acceleration.

[0153] Then, the sum of the product of the front axle lateral stiffness and the front wheel sideslip angle, and the product of the rear axle lateral stiffness and the rear wheel sideslip angle, divided by the vehicle mass, yields the third intermediate data. Finally, the fourth intermediate data is obtained by multiplying the longitudinal vehicle speed and the center of gravity sideslip angle.

[0154] Specifically, third intermediate data This can be understood as the translational acceleration component corresponding to the lateral force of the vehicle, and its calculation formula can be found as follows:

[0155] in, For front axle lateral stiffness, This refers to the front wheel slip angle; For rear axle lateral stiffness, This refers to the rear wheel slip angle; For vehicle quality.

[0156] According to Newton's second law, the lateral net force of a vehicle is equal to its mass multiplied by its lateral acceleration. Therefore, by summing the lateral forces (lateral stiffness × slip angle) of the front and rear axles and dividing by the mass, we can obtain the lateral translational acceleration component induced by the lateral force of the tires. This component is the core lateral disturbance term during vehicle steering.

[0157] Fourth intermediate data This can be understood as the centripetal acceleration component resulting from the coupling of the vehicle's longitudinal velocity and the sideslip angle. The calculation formula is as follows:

[0158] in, The longitudinal vehicle speed obtained in step 21; It is the centroid sideslip angle.

[0159] When a vehicle has a sideslip angle, the longitudinal vehicle speed will generate an additional component in the lateral direction of the vehicle coordinate system. This component is also a lateral motion disturbance and needs to be removed from the original acceleration.

[0160] Furthermore, based on the third and fourth intermediate data obtained above, the lateral acceleration can be calculated. The specific calculation formula is as follows:

[0161] in, It represents lateral acceleration.

[0162] Based on steps 31-32 above, the longitudinal speed of the target vehicle is determined by wheel speed, steering wheel angle, yaw rate, and driving and braking status signals. The lateral acceleration is calculated by combining the longitudinal speed with preset vehicle dynamic parameters. This achieves accurate modeling and quantification of lateral translational interference components under conditions such as steering and lane changing. It effectively eliminates lateral translational acceleration coupled with road slope gravity signals, avoids its interference with the extraction of gravity direction representation quantities and slope calculation, and improves the stability and accuracy of slope estimation under dynamic steering conditions.

[0163] In some embodiments, the above-mentioned S2024 (calculating the vehicle body attitude gravitational acceleration based on the suspension height in the third information) can be further refined into the following steps: Step 41: Based on the suspension height in the third information, determine the vehicle's pitch angle and roll angle.

[0164] In this embodiment of the application, the vehicle pitch angle needs to be determined based on the difference between the average front axle suspension height and the average rear axle suspension height, as well as the vehicle wheelbase. Then, the vehicle roll angle is determined based on the difference between the left front suspension height and the right front suspension height, the difference between the left rear suspension height and the right rear suspension height, as well as the front track and the rear track.

[0165] Specifically, the formula for calculating the vehicle body pitch angle is as follows:

[0166] in, The vehicle body pitch angle; This refers to the height of the left front suspension. This refers to the height of the right front suspension. This refers to the height of the left rear suspension. The height is the right rear suspension; L is the vehicle wheelbase.

[0167] By taking the difference between the average suspension height of the front axle and the average suspension height of the rear axle, and combining it with the vehicle wheelbase, the pitch angle of the vehicle body around the lateral axis (i.e., the Y-axis of the vehicle coordinate system) is further obtained based on the arctangent trigonometric function. This angle only reflects the vehicle body attitude and does not represent the road slope.

[0168] The formula for calculating the vehicle body roll angle is as follows:

[0169] in, This refers to the vehicle body roll angle; For the left front suspension height, For the right front suspension height, For the left rear suspension height, This refers to the height of the right rear suspension. This refers to the vehicle's wheelbase.

[0170] By calculating the difference in height between the left and right front suspensions, the difference in height between the left and right rear suspensions, and the vehicle track width, the roll angles of the front and rear axles are calculated and averaged. The roll angle of the vehicle body around the longitudinal axis (i.e., the X-axis of the vehicle coordinate system) is obtained by using the arctangent trigonometric function. This angle is also considered a vehicle attitude disturbance.

[0171] Step 42: Determine the vehicle attitude gravitational acceleration based on the vehicle pitch angle and vehicle roll angle, combined with the gravitational acceleration constant.

[0172] Furthermore, based on the trigonometric relationship between the vehicle pitch angle and the gravitational acceleration constant, the longitudinal attitude gravitational acceleration component is determined, and based on the trigonometric relationship between the vehicle roll angle and the gravitational acceleration constant, the lateral attitude gravitational acceleration component is determined.

[0173] Specifically, when the vehicle pitches, gravitational acceleration introduces an additional component in the vehicle's longitudinal direction. This component is considered interference and needs to be removed. This is because vehicle pitch causes a relative tilt between the vehicle coordinate system and the geodetic coordinate system. The originally vertically downward gravitational acceleration vector is projected onto the vehicle's longitudinal (X-axis) and vertical (Z-axis) axes. At this point, the longitudinal acceleration measured by the IMU includes both the actual longitudinal translational acceleration generated by the vehicle's motion and the gravitational projection component caused by the vehicle's attitude tilt. If this component is not removed, it will directly interfere with the subsequent purification of the pure road slope gravity component, leading to a systematic bias in the slope estimation results. The calculation formula for the longitudinal attitude gravitational acceleration component is as follows:

[0174] in, This represents the longitudinal attitude gravitational acceleration component; 9.8 is the vehicle pitch angle calculated in step 41; 9.8 is the gravitational acceleration constant.

[0175] When the vehicle body tilts, gravitational acceleration introduces an additional component in the lateral direction. This component is also considered interference and needs to be removed. This is because vehicle tilt causes a lateral shift between the vehicle coordinate system and the geodetic coordinate system. The originally vertically downward gravitational acceleration vector is projected onto the vehicle's lateral (Y-axis) and vertical (Z-axis) axes. At this point, the lateral acceleration measured by the IMU includes both the lateral translational acceleration generated by the actual vehicle motion and the gravitational projection component caused by the vehicle's attitude shift. If this component is not removed, it will directly interfere with the subsequent purification of the pure road slope gravity component, leading to deviations in the slope estimation results. The formula for calculating the lateral attitude gravitational acceleration component is as follows:

[0176] in, This represents the lateral attitude gravitational acceleration component; 9.8 is the vehicle roll angle calculated in step 41; 9.8 is the gravitational acceleration constant.

[0177] Furthermore, the longitudinal attitude gravitational acceleration components obtained above and lateral attitude gravitational acceleration components This refers to the vehicle's gravitational acceleration due to its posture.

[0178] Based on steps 41-42 above, the vehicle pitch and roll angles are calculated based on the suspension height signal, and the vehicle attitude gravity acceleration components are generated by combining the gravity acceleration constant. This achieves accurate quantification and modeling of the gravity projection deviation caused by changes in vehicle attitude, effectively eliminating the interference of additional gravity components caused by vehicle pitch and roll, avoiding the slope calculation deviation caused by its coupling with the road slope gravity signal, and further improving the purity of the gravity direction representation and the accuracy of subsequent slope estimation.

[0179] In this embodiment, after obtaining the data corresponding to each interfering component, the stripping process in S203 of the above embodiment can be performed by referring to the following formula to calculate the gravity direction characterization quantity:

[0180]

[0181]

[0182] in, The longitudinal acceleration component in the gravity direction representation is the longitudinal acceleration that retains only the gravity component related to the road slope after removing all longitudinal interferences. The lateral acceleration component in the gravity direction representation is the lateral acceleration that retains only the gravity component related to road slope after removing all lateral interferences. The vertical acceleration component in the gravity direction representation is the vertical acceleration that retains only the gravity component related to road slope after eliminating vertical eccentricity interference.

[0183] The longitudinal acceleration component in the initial acceleration is represented by the original longitudinal acceleration signal acquired by the IMU. The lateral acceleration component is the initial acceleration, and the original lateral acceleration signal is acquired by the IMU. The vertical acceleration component is the initial acceleration, and the original vertical acceleration signal is acquired by the IMU.

[0184] It is the longitudinal acceleration; This refers to the longitudinal eccentric acceleration component in the eccentric acceleration; This refers to the longitudinal attitude gravity acceleration component within the vehicle body attitude gravity acceleration component.

[0185] This is lateral acceleration; This refers to the lateral eccentric acceleration component in eccentric acceleration; This refers to the lateral attitude gravitational acceleration component within the vehicle body attitude gravitational acceleration component.

[0186] This represents the vertical eccentric acceleration component in the eccentric acceleration.

[0187] Furthermore, the above calculations yield the following results. , , These three factors together constitute the gravity direction representation quantity, laying the data foundation for subsequent attitude filtering fusion and real road slope calculation.

[0188] It should be noted that the above stripping process can be understood as the process of collecting the original total acceleration from the inertial measurement unit. In this process, eccentric acceleration caused by the IMU's installation position deviating from the center of gravity, longitudinal acceleration caused by vehicle acceleration and deceleration, lateral acceleration caused by vehicle steering, and attitude gravity acceleration components caused by vehicle pitch and roll are systematically removed, ultimately retaining only pure acceleration information related to gravity. The essence of this separation operation is to decouple various dynamic disturbances generated by the vehicle's own motion from static gravity acceleration, ensuring that the final gravity direction representation is unaffected by vehicle acceleration and deceleration, steering, suspension motion, and changes in vehicle attitude. This accurately reflects the distribution of gravity along each axis of the vehicle coordinate system, providing a clean and reliable data foundation for subsequent road slope calculations.

[0189] As an extension and refinement of the above embodiments, when executing S204 (correcting the vehicle attitude angle corresponding to the first information based on the gravity direction representation quantity) in the aforementioned embodiments, a Kalman filter can be used. For example, this application embodiment takes an error state Kalman filter as an example for further explanation, specifically including the following steps one to three: Step 1: Determine the initial attitude angle of the target vehicle based on the angular velocity in the first information.

[0190] Specifically, the angular velocity data in the first information reflects the vehicle's rotation rate around three axes; by integrating the angular velocity data over time, the vehicle's attitude angle changes in space can be obtained, thereby determining the vehicle's initial attitude angles, which include pitch angle, roll angle, and yaw angle.

[0191] It should be noted that the initial attitude angle obtained based on the angular velocity integral in the first information has a cumulative drift problem. As time goes by, the integration error will gradually increase, causing the attitude angle to deviate from the true value. Therefore, it is necessary to continue to perform the following step two to correct it using the gravity direction characterization quantity.

[0192] Furthermore, quaternions can be used as a mathematical representation of attitude angles; in the error state Kalman filter, the variation of attitude quaternions with time is described by the following differential equation:

[0193] in, For attitude quaternions, representing the rotation state of the vehicle coordinate system relative to the world coordinate system; The angular velocity measured by the IMU is obtained from the first information; The bias is the gyroscope bias, which characterizes the zero bias error of the gyroscope. Its derivative is 0, which means that the bias is assumed to be constant in a short time. This represents quaternion multiplication.

[0194] To implement discrete-time recursion in the onboard controller, the above differential equation is discretized, resulting in the discrete recursive formula for attitude quaternions:

[0195] in, The preset calculation period is (e.g., 0.01s). For rotation quaternions, representing time... Internal angular velocity The resulting rotation is calculated using the following formula:

[0196] In the above formula, For rotation angle, is the unit vector in the direction of the rotation axis.

[0197] Through the above integral recursion, the initial attitude quaternion of the vehicle in space can be obtained, denoted as . This is used as the quaternion component in the nominal state of the error state Kalman filter. The initial attitude angles of the vehicle are then determined; these initial attitude angles include the initial pitch angle and the initial roll angle, which can be extracted from the quaternions using the following formula:

[0198]

[0199] in, The pitch angle is the initial attitude angle of the vehicle. The roll angle is the initial attitude angle of the vehicle.

[0200] It should be noted that the initial attitude angle obtained based on the angular velocity integral in the first information has a cumulative drift problem. As time goes by, the integration error will gradually increase, causing the vehicle attitude angle to deviate from the true value. Therefore, it is necessary to continue to perform the following step two to correct it using the gravity direction representation quantity.

[0201] Step 2: Using the gravity direction characterization quantity as an observation, input it into the error state Kalman filter to obtain the posterior error state estimate.

[0202] It should be noted that the error state Kalman filter uses the gravity direction representation as a basis to measure and update the error state at the current moment to obtain the posterior error state estimate; the posterior error state estimate includes the attitude estimation error and the gyroscope bias estimation error.

[0203] Among them, the gravity direction characterization quantity is the pure acceleration component obtained after removing all interferences, which only contains the pure gravity acceleration characteristics related to the road slope.

[0204] Specifically, the Error State Kalman Filter (ESKF) is a state estimation algorithm running in the vehicle controller. It is used to fuse multi-source sensor data, suppress measurement noise and accumulated errors, and achieve accurate estimation of vehicle attitude.

[0205] The error state at the current moment mentioned above refers to state quantities including vehicle attitude error, gyroscope bias error, etc., which represent the deviation between the nominal state and the actual state; the posterior error state estimate refers to the optimal estimate of the error state obtained after measurement update, which is the core basis for correcting the nominal attitude.

[0206] It should be noted that the error-state Kalman filter is a variant of the Kalman filter for inertial navigation systems. It employs a two-state architecture: a nominal state and an error state. It corrects the deviation from the nominal state (the error state) by filtering. The nominal state is recursively obtained from gyroscope integration (which introduces cumulative drift). The error state characterizes the deviation between the nominal and actual states and is the core object of the filter. After time updates (prediction), the error state, being a priori estimate, needs to be corrected using observations.

[0207] In this embodiment, the filter uses the aforementioned gravitational direction characterization quantity, i.e., the pure gravitational acceleration component, as the measurement value, and establishes the measurement equation as follows:

[0208] in, The measured value is the pure gravitational acceleration component. ; The measurement equation represents the projection of the theoretical gravitational acceleration calculated based on the current estimated attitude onto the carrier coordinate system. The current state vector contains the attitude quaternion q and the gyroscope bias. ; For measuring noise, it follows a Gaussian distribution N(0, V).

[0209] Specifically, measurement equations The specific form is as follows:

[0210] in, This is a rotation matrix used to transform the gravitational acceleration vector in the navigation coordinate system to the vehicle coordinate system; This is the gravitational acceleration vector in the world coordinate system (9.8 m / s² is the gravitational acceleration constant).

[0211] The corresponding expansion of RotMartx is:

[0212] in, , , , These are the four components of the prior attitude quaternion obtained after time update (prediction), representing the rotation state of the currently predicted vehicle coordinate system relative to the world coordinate system.

[0213] Furthermore, the measurement residuals are calculated using the measurement update formula of Kalman filtering. And based on this, correct the prediction error state. First, calculate the Kalman gain:

[0214] Where K is the Kalman gain matrix, used to balance prediction uncertainty (covariance P) and measurement uncertainty (measurement noise covariance V); P is the covariance matrix of the error state, characterizing the uncertainty of the error state estimation; H is the Jacobian matrix of the measurement equation, representing the partial derivative of the measurement with respect to the error state. V is the transpose of the Jacobian matrix; V is the measurement noise covariance matrix, which characterizes the measurement uncertainty of the accelerometer.

[0215] The specific form of the Jacobian matrix H of the measurement equation is as follows:

[0216] in, Let be the error state vector, and δθ be the attitude angle error. This refers to the gyroscope bias error; This is the transformation matrix from error state to quaternion error. Represented as:

[0217] The transformation matrix from attitude error to quaternion error is given below, which converts the attitude angle error δθ into a quaternion error:

[0218] The matrix of partial derivatives of the measured values ​​with respect to quaternions is represented as follows:

[0219] It should be noted that the above and Quaternion components in a matrix , , , These are all prior attitude quaternions obtained after time updates (predictions) at the current moment, i.e., attitude estimates obtained recursively through angular velocity integration and not yet corrected by gravity direction representations. These prior attitude quaternions serve as the linearization operating point for calculating the aforementioned Jacobian matrix; after the measurement update is completed, the calculated attitude error is injected into it to obtain the corrected posterior attitude quaternions.

[0220] Furthermore, the error state is corrected using Kalman gain:

[0221] in, This represents the nominal state estimate at the current moment (typically a quaternion q and gyroscope bias). ), This represents the theoretical measurement calculated based on the current estimated nominal state (i.e., the theoretical projection of pure gravitational acceleration). This represents the measurement residual, which is the difference between the actual measured pure gravitational acceleration and the theoretical projection.

[0222] Simultaneously update the covariance matrix:

[0223] Where P is the covariance matrix of the error state (prior covariance before update, posterior covariance after update), I is the identity matrix, K is the Kalman gain matrix, and H is the Jacobian matrix of the measurement equation; after measurement update, the posterior error state estimate is obtained. .

[0224] in, This is the posterior error state estimate, in rad, representing the deviation between the nominal quaternion and the true quaternion; This is the optimal estimate of the attitude error; This is the optimal estimate of the gyroscope bias error, expressed in rad / s, representing the deviation between the nominal bias and the actual bias.

[0225] Step 3: Based on the attitude error and gyroscope bias error in the posterior error state estimate, the initial attitude angle is corrected to obtain the corrected vehicle attitude angle of the target vehicle.

[0226] It should be noted that this step uses error state Kalman filtering and gravity direction representation as an observation to obtain the optimal estimates of attitude error and gyroscope bias error. Based on these error estimates, the initial attitude angle of the vehicle is corrected to obtain the updated current attitude angle.

[0227] Specifically, because ESKF adopts an architecture that separates the nominal state from the error state, it is based on the nominal state obtained by gyroscope integration (this nominal state includes the quaternion q and the gyroscope bias). In this context, the quaternion q in the nominal state is the one obtained from the initial attitude angle. This state will accumulate drift over time, resulting in errors in the initial attitude angle; corresponding to the nominal state, ESKF also introduces an error state. This is used to characterize the deviation between the nominal state and the actual state. The error state vector is defined as:

[0228] in, The attitude angle error represents the deviation between the nominal quaternion and the actual quaternion. This is the gyroscope bias error, representing the deviation between the nominal bias and the actual bias.

[0229] The error state is the optimal correction obtained through measurement updates. ESKF first estimates the error state a priori using the prediction equation:

[0230] in, Here is the state transition matrix. Let be the covariance matrix of the error state. Let be the process noise covariance matrix.

[0231] Then, using the gravitational direction as an observation, the error state is measured and updated to obtain the posterior error state estimate, which is the optimal correction.

[0232] Therefore, the attitude error in the error state needs to be included. and gyroscope bias error To compensate to the nominal state and obtain the true vehicle attitude, the attitude error δθ needs to be injected into the nominal quaternion q in the form of quaternion multiplication:

[0233] in, This represents quaternion multiplication. Let be the rotation quaternion corresponding to the attitude error, representing . This represents a rotational transformation of the rotation angle.

[0234] Then, bias correction is performed, that is, the gyroscope bias error is corrected. Apply directly to the nominal bias :

[0235] in, This is the gyroscope bias, measured in rad / s, representing the zero-bias error of the gyroscope. This is the optimal estimate of the gyroscope bias error, expressed in rad / s.

[0236] After the above error injection, the nominal state q and The correction has been made. At this point, the nominal quaternion q represents the vehicle's true attitude at the current moment, which is the updated current attitude angle.

[0237] Then, based on the updated current attitude angle, the slope estimate of the road where the target vehicle is located is determined.

[0238] Specifically, vehicle attitude angles include pitch angle, roll angle, and yaw angle. Among them, longitudinal slope usually refers to the pitch angle in the longitudinal direction of the vehicle, reflecting the degree of inclination of the road in the direction of vehicle travel; lateral slope usually refers to the roll angle in the lateral direction of the vehicle, reflecting the degree of inclination of the road in the left and right directions of the vehicle; the angle of rotation of the vehicle around the vertical axis (Z-axis) reflects the degree of deflection of the vehicle's front end relative to the direction of travel, and is mainly used for heading judgment and path tracking, and is not directly used for road slope estimation.

[0239] Furthermore, the updated pitch and roll angles can be extracted from the updated attitude quaternions. The pitch component corresponds to the longitudinal slope estimate, and the roll component corresponds to the lateral slope estimate.

[0240] Specifically, from the corrected quaternion Extracting pitch angle and roll angle The formulas are as follows:

[0241]

[0242] in, It is the arcsine function, used to convert quaternions into angle values.

[0243] Extracted pitch angle and roll angle This allows the corrected vehicle attitude angles of the target vehicle to be formed.

[0244] It should also be noted that after the error is injected into the nominal state, the mean error state... It will be reset to zero because the new pose error will be represented in the local coordinate system relative to the new nominal pose frame. For the ESKF update to be complete, the covariance of the error needs to be updated based on this modification.

[0245] Specifically, the error state reset formula is as follows:

[0246] in, G is the error state vector, which includes state variables such as attitude error and gyroscope bias error. It is reset to 0 to adapt to the updated nominal attitude. G is the coordinate transformation matrix, which is approximately the identity matrix I under the assumption of small angle error. It is used to project the error state covariance matrix onto the new nominal attitude coordinate system. P is the error state covariance matrix. After being updated by the transformation matrix, it ensures the continued effectiveness of ESKF under the new attitude framework and prepares for the filtering update in the next cycle.

[0247] Through the above reset operation, the error state is cleared to zero, and the covariance matrix is ​​updated accordingly, preparing for the filtering estimation at the next time step.

[0248] After the error injection is completed, the original error state has been used to correct the nominal attitude. Therefore, the mean of the error state needs to be reset to 0 so that the new error state is redefined based on the updated nominal attitude. At the same time, the continuity and stability of the filtering process in the new local coordinate system are ensured by updating the covariance matrix, avoiding the coupling deviation between the error state and the nominal state, and ensuring the continuous high-precision operation of ESKF under long-term dynamic driving conditions of the vehicle.

[0249] It should also be noted that the error state Kalman filter used in this embodiment is only a preferred implementation. Those skilled in the art can also use other types of Kalman filter estimators (such as extended Kalman filter EKF, unscented Kalman filter UKF, etc.) instead, as long as the vehicle attitude angle can be filtered and estimated based on pure gravitational acceleration observations.

[0250] In this embodiment, the initial attitude angle of the vehicle is first determined based on the angular velocity of the first information. Then, the gravity direction representation is processed by the error state Kalman filter to obtain the posterior error state estimate. The initial attitude angle is corrected by using the attitude error and the gyroscope bias error. This can effectively suppress the cumulative drift error generated during the gyroscope integration process, improve the accuracy and real-time performance of the vehicle attitude angle calculation, and provide reliable attitude data support for subsequent accurate road slope estimation and vehicle attitude control.

[0251] As an extension and refinement of the embodiments, refer to ​ As shown, this application also provides a vehicle control method, which specifically includes the following steps: S51. Based on the sensor, collect the slope monitoring value of the road where the target vehicle is located, and obtain the initial road surface smoothness.

[0252] Specifically, the sensors include intelligent driving perception sensors such as vehicle-mounted cameras and LiDAR; wherein, the intelligent driving controller calculates the longitudinal slope of the road in real time based on image data collected by the vehicle-mounted cameras or point cloud data collected by the LiDAR. and lateral slope This slope monitoring value serves as the external reference input for subsequent weighted fusion.

[0253] Meanwhile, the initial road surface smoothness can be set according to default settings or vehicle network information. For example, when the vehicle network information indicates that the current weather is severe weather such as rain, snow, or fog, the initial road surface smoothness can be set to be low (i.e., the road surface condition is poor) to reflect the impact of severe weather on the perception accuracy of intelligent driving.

[0254] S52. Obtain the vertical vibration acceleration of the target vehicle's wheels collected by the unsprung acceleration sensor, and determine the target road surface smoothness based on the vertical vibration acceleration of the wheels and the initial road surface smoothness.

[0255] Specifically, the unsprung acceleration sensor is installed near the vehicle's wheels, such as at the lower control arm of the suspension or the wheel hub bearing, to collect the vibration acceleration of the wheel in the vertical direction in real time. .right A first-order low-pass filter is performed to remove high-frequency noise; the rate of change of the filtered vertical acceleration is obtained by differentiating the result. .

[0256] Furthermore, according to and The smoothness of the first road surface is determined through a preset mapping relationship. ;reference ​ As shown, ​ The first road surface smoothness The acceleration of the wheel in the vertical direction With respect to the rate of change of vertical acceleration A changing three-dimensional lookup table curve; the surface as a whole exhibits a stepped upward trend, as the wheel's vibration acceleration in the vertical direction... The larger the rate of change of vertical acceleration When the surface roughness is higher, the first road surface smoothness is... The larger the value, the higher the final target road surface smoothness. Through this preset mapping relationship, the current road surface state can be quickly and adaptively calibrated based on unsprung vibration data, providing a reliable basis for determining the subsequent weighted fusion coefficients.

[0257] Initial road surface smoothness based on intelligent driving perception With based and The first road surface smoothness obtained Weighted fusion is performed to obtain the final target road surface smoothness. :

[0258] in, This refers to the fusion coefficient; when the vehicle network information indicates that the current weather is severe, such as rain, snow, or fog, the reliability of intelligent driving perception decreases. Take a smaller value (e.g., 0.3) to reduce The weight; otherwise Set the weight to 0.5 to make the weights of the two consistent.

[0259] S53. Determine the weighted fusion coefficient based on the target road surface smoothness, longitudinal motion acceleration, and lateral motion acceleration.

[0260] Specifically, weighted fusion coefficient This is used to dynamically adjust the confidence level distribution between the slope estimate based on the gravity direction representation and the slope monitoring value based on the intelligent driving sensor during subsequent weighted fusion. The determination of [the value] can be carried out according to the following process: When the target road surface smoothness When the surface roughness is greater (i.e., the road surface is more uneven), vehicles may travel on rough roads such as speed bumps, cobblestone streets, Belgian roads, or potholes. In these conditions, the acceleration signal measured by the IMU is significantly affected by road vibrations, with amplitude fluctuations far exceeding the true value. This leads to increased deviations in the slope estimate calculated based on gravity direction characteristics. Therefore, the weight of the slope estimate should be reduced (i.e., the weight of the slope estimate should be decreased). ), and increase the weight of slope monitoring values.

[0261] Larger amplitudes of longitudinal or lateral acceleration indicate that the vehicle is undergoing intense acceleration, braking, or steering maneuvers. In such cases, the longitudinal and lateral accelerations calculated based on the motion model may contain errors, leading to inaccurate extraction of gravity direction representations. Therefore, the weight of the slope estimate should be reduced (i.e., the weight of the slope should be decreased). ), and increase the weight of slope monitoring values.

[0262] Based on the above process, the target road surface smoothness can be pre-established. Longitudinal motion acceleration, lateral motion acceleration and weighted fusion coefficient The mapping relationship between them (such as the three-dimensional lookup table method). Based on the target road surface smoothness, longitudinal motion acceleration, and lateral motion acceleration at the current moment, the current weighted fusion coefficients can be determined by looking up the table. , refer to ​ As shown, ​ Weighted fusion coefficient A three-dimensional mapping curve showing the relationship between the target road surface roughness (Rou) and the ratio of longitudinal acceleration (axlg) to lateral acceleration (aylg) (axlg / aylg); the surface generally shows a downward trend, with a higher proportion of lateral motion occurring as the road surface becomes more uneven. The smaller the value, the lower the weight of the slope estimate and the higher the weight of the monitoring value to suppress attitude calculation interference; conversely, the larger the value, the lower the weight of the slope estimate and the higher the weight of the monitoring value. The larger the value, the greater the weight of the slope estimate to leverage its accuracy advantage under stable working conditions. This lookup table method enables rapid adaptive calibration of the coefficients, improving the robustness and accuracy of slope calculations under different working conditions.

[0263] S54. The slope estimate and the slope monitoring value are weighted and fused based on the weighted fusion coefficient to output the target road slope value, so as to control the target vehicle according to the target road slope value.

[0264] Specifically, a weighted Kalman filter can be used to fuse the two types of slope information.

[0265] The state vector needs to be defined first. ,in This refers to the longitudinal slope after merging. The lateral slope after fusion; input vector ,in The longitudinal slope is calculated based on the gravitational direction characterization. The lateral slope is calculated based on the gravitational direction characterization; observations ,in Longitudinal slope calculated for intelligent driving sensors. Lateral slope calculated for intelligent driving sensors.

[0266] Furthermore, prediction and correction are performed using Kalman filtering. The prediction process is as follows:

[0267]

[0268] Where A is the state transition matrix (zeroed matrix), B is the input matrix, and Q is the process noise covariance matrix.

[0269] The correction process is as follows:

[0270]

[0271]

[0272]

[0273] Where H is the observation matrix and R is the observation noise covariance matrix. Based on the weighted fusion coefficients... Adjust the observation noise covariance matrix: ;when When the variance is small, the observation noise covariance R is small, and the filter trusts the observations more (i.e., the slope monitoring values ​​calculated by the intelligent driving sensors); when When the value is large, the observation noise covariance R is also large, and the filter trusts the predicted value more (i.e., the slope estimate based on the gravity direction characterization).

[0274] Through the weighted Kalman filter fusion described above, the final output is the fused longitudinal slope PAg and lateral slope RAg, which serve as the target road slope value. This target road slope value can be used for slope-adaptive control of the target vehicle, such as active suspension stiffness adjustment, hill-start assist braking, and energy recovery intensity adjustment, to improve the safety, ride comfort, and economy of the vehicle when driving on sloping roads.

[0275] This application embodiment combines road slope monitoring values ​​collected by sensors with the initial road surface smoothness, accurately calibrates the target road surface smoothness using the vertical vibration acceleration of the wheels, and then determines the appropriate weighted fusion coefficient by associating longitudinal and lateral motion accelerations. The slope estimate and slope monitoring values ​​are adaptively weighted and fused, which can take into account the advantages of sensor measured data and algorithm estimated data, adapt to the slope calculation needs under different road surface smoothness conditions, effectively reduce the slope error caused by a single data source, improve the accuracy and adaptability of road slope output results, and thus ensure that the vehicle can achieve stable and accurate vehicle control based on the target road slope.

[0276] As an extension and refinement of the embodiments, refer to ​ As shown, this application also provides a system framework diagram corresponding to a vehicle control method, which can be deployed in the chassis domain controller or vehicle controller of the target vehicle, specifically including: The sensor acquisition module 81 is used to acquire the current running dataset of the target vehicle, which includes the target vehicle's attitude data, motion state data, and vehicle dynamic parameters.

[0277] The sensor acquisition module 81 specifically includes an inertial measurement unit 811, a height sensor 812, a wheel speed sensor 813, a steering wheel angle sensor 814, and an unsprung sensor 815. The inertial measurement unit 811 outputs angular velocity, angular acceleration, and initial acceleration; the height sensor 812 outputs suspension height data; the wheel speed sensor 813 outputs wheel rotation speed signals; the steering wheel angle sensor 814 outputs steering wheel angle signals and yaw rate signals; and the unsprung sensor 815 outputs wheel vertical vibration acceleration signals, which are used to characterize road surface excitation.

[0278] It should be noted that the first information includes angular velocity, angular acceleration, and initial acceleration collected by the inertial measurement unit 811; the second information includes wheel speed signals collected by the wheel speed sensor 813, steering wheel angle signals and yaw rate signals collected by the steering wheel angle sensor 814, and also includes drive pedal opening signals and brake pedal opening signals obtained from the vehicle CAN bus, used to characterize the vehicle's driving state and the driver's operating intention; the third information includes suspension height data collected by the height sensor 812, used to characterize the vehicle's spatial attitude.

[0279] The vehicle dynamics parameters include pre-calibrated and stored vehicle-specific parameters in the on-board memory, including but not limited to front axle lateral stiffness, rear axle lateral stiffness, front wheel slip angle, rear wheel slip angle, center of gravity slip angle, vehicle mass, wheelbase, track width, and center of gravity height, which are used for subsequent calculations of disturbance components such as lateral motion acceleration.

[0280] The interference component calculation module 82 is used to calculate the interference components based on the data collected by the sensor acquisition module 81 and combined with the vehicle dynamics parameters; wherein, the interference components include eccentric acceleration, vehicle body attitude gravity acceleration, longitudinal motion acceleration and lateral motion acceleration.

[0281] The elimination module 83 is used to receive the interference component and the initial acceleration, and to eliminate the interference component from the initial acceleration to obtain the gravitational direction characterization quantity.

[0282] The Kalman filter module 84 is used to receive the gravity direction representation and angular velocity, and to perform Kalman filtering on the gravity direction representation to calculate the actual road slope.

[0283] The road smoothness calculation module 86 is used to receive the wheel vertical vibration acceleration signal collected by the unsprung sensor 815, and calculate the target road smoothness in combination with the initial road smoothness.

[0284] The weighted coefficient confirmation module 87 is used to receive the target road surface smoothness and determine the weighted fusion coefficient by combining the longitudinal and lateral motion acceleration of the vehicle.

[0285] The slope monitoring module 88 is used to receive suspension height data collected by the height sensor 812 and calculate the slope monitoring value.

[0286] The weighted fusion module 85 is used to receive the slope estimate output by the Kalman filter module 84, the slope monitoring value output by the slope monitoring module 88, and the weighted fusion coefficient output by the weighted coefficient confirmation module 87, and to perform weighted fusion on the slope estimate and the slope monitoring value to output the target road slope value.

[0287] The vehicle control module 89 is used to adjust the driving control strategy of the target vehicle based on the target road slope value output by the weighted fusion module 85, so as to adapt to the driving conditions of the current road.

[0288] In this embodiment, the vehicle control module can flexibly select the control method according to the actual working conditions: when the road surface is stable and the vehicle's motion is stable, the slope estimate output by the Kalman filter can be directly used for vehicle control; when the road surface is bumpy or the lateral motion is severe, the target road slope value calculated based on the weighted fusion coefficient is used first for vehicle control, thereby achieving adaptive adaptation to different working conditions and ensuring the vehicle's driving stability and control accuracy under various road conditions.

[0289] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0290] like ​ As shown, embodiments of this application also provide a vehicle control device, the device comprising: The acquisition unit 91 is used to acquire the perception information of the target vehicle when the target vehicle is in a driving state. The perception information includes first information for representing the inertial motion state of the vehicle, second information for representing the wheel motion state, and third information for representing the suspension height state. The stripping unit 92 is used to strip the interference components generated by the vehicle's own motion from the first information based on the first information, the second information and the third information, to obtain the gravity direction characterization quantity. The generation unit 93 is used to generate an estimated slope value of the road where the target vehicle is located based on the first information and the gravity direction characterization quantity. Control unit 94 is used to control the target vehicle based on the slope estimate to adapt to the current road slope conditions.

[0291] In one specific embodiment, the stripping unit 92 is specifically used to calculate the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration and vehicle posture gravity acceleration generated by the vehicle's own motion based on the first information, the second information and the third information, so as to generate the interference component; obtain the initial acceleration from the first information, and strip the interference component from the initial acceleration to obtain the gravity direction characterization quantity.

[0292] In one specific embodiment, the stripping unit 92 is specifically used to calculate the eccentric acceleration based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit; the installation position offset includes the offset of the installation position of the inertial measurement unit installed on the target vehicle from the vehicle's center of gravity in multiple directions; calculate the longitudinal motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information; calculate the lateral motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information, combined with vehicle dynamics parameters; and calculate the vehicle body attitude gravitational acceleration based on the suspension height in the third information.

[0293] In one specific embodiment, the stripping unit 92 is specifically used to determine first intermediate data based on the angular acceleration and the offset; the first intermediate data is used to characterize the tangential eccentric acceleration caused by the angular acceleration; determine second intermediate data based on the angular velocity and the offset; the second intermediate data is used to characterize the normal eccentric acceleration caused by the angular velocity; and generate the eccentric acceleration based on the first intermediate data and the second intermediate data.

[0294] In one specific embodiment, the stripping unit 92 is specifically used to determine the longitudinal speed of the target vehicle based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking state signal in the second information; and to determine the longitudinal acceleration of the target vehicle based on the longitudinal speed at adjacent time points.

[0295] In one specific embodiment, the stripping unit 92 is specifically used to determine the longitudinal speed of the target vehicle based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking state signal in the second information; and to calculate the lateral acceleration of the target vehicle by combining the longitudinal speed with preset vehicle dynamics parameters.

[0296] In one specific embodiment, the stripping unit 92 is specifically used to determine the vehicle body pitch angle and vehicle body roll angle of the target vehicle based on the suspension height in the third information; and to determine the vehicle body attitude gravitational acceleration component based on the vehicle body pitch angle and vehicle body roll angle, combined with the gravitational acceleration constant.

[0297] In one specific embodiment, the generation unit 93 is specifically used to correct the vehicle attitude angle corresponding to the first information based on the gravity direction representation quantity; and to determine the slope estimate of the road where the target vehicle is located based on the corrected vehicle attitude angle.

[0298] In one specific embodiment, the generation unit 93 is specifically used to determine the initial attitude angle of the target vehicle based on the angular velocity in the first information; process the gravity direction representation quantity using an error state Kalman filter to obtain a posterior error state estimate; and correct the initial attitude angle based on the attitude error and gyroscope bias error in the posterior error state estimate to obtain the corrected vehicle attitude angle of the target vehicle.

[0299] In one specific embodiment, the control unit 94 is further configured to: acquire slope monitoring values ​​of the road where the target vehicle is located based on sensors, and obtain initial road surface smoothness; acquire the vertical vibration acceleration of the target vehicle's wheels collected by the unsprung acceleration sensor, and determine the target road surface smoothness based on the vertical vibration acceleration of the wheels and the initial road surface smoothness; determine a weighted fusion coefficient based on the target road surface smoothness, longitudinal motion acceleration, and lateral motion acceleration; and perform weighted fusion of the slope estimate and the slope monitoring value based on the weighted fusion coefficient to output the target road slope value, so as to control the target vehicle based on the target road slope value.

[0300] In one specific embodiment, the control unit 94 is specifically configured to generate a command to increase the rear suspension stiffness and / or activate the hill-start assist braking command when the target road slope value indicates an uphill condition; generate a command to increase the energy recovery intensity and / or reduce the output torque of the drive system when the target road slope value indicates a downhill condition; and send the generated commands to the chassis execution system corresponding to the target vehicle to adapt to the current road slope condition.

[0301] Regarding the apparatus in the above embodiments, the specific manner in which each unit performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0302] ​ This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0303] For example, such as​ As shown, the vehicle includes a memory 1001 and a processor 1002. The memory 1001 stores executable program code 1011, and the processor 1002 is used to call and execute the executable program code 1011 to perform a vehicle control method.

[0304] This embodiment can divide the vehicle into functional modules according to the above method example. For example, each function can be assigned to a separate module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods. When dividing each functional module according to its corresponding function, the vehicle may include: an acquisition unit 91, a stripping unit 92, a generation unit 93, a control unit 94, etc. It should be noted that all relevant content of each step involved in the above method embodiment can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0305] The vehicle provided in this embodiment is used to execute the vehicle control method described above, and therefore can achieve the same effect as the above implementation method.

[0306] When using integrated units, the vehicle may include a processing module and a storage module. The processing module is used to control and manage the vehicle's actions. The storage module supports the vehicle in executing program code and data.

[0307] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as disclosed in this application. The processor may also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0308] Embodiments of this application also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the vehicle control method embodiments described above.

[0309] Embodiments of this application also provide a computer-readable storage medium storing a computer program configured to execute the steps in any of the vehicle control method embodiments described above when running.

[0310] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0311] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the vehicle control method embodiments described above.

[0312] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the vehicle control method embodiments described above.

[0313] The beneficial effects of the above embodiments can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0314] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0315] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0316] In the description of this application, it should be understood that if the terms "upper", "lower", "front", "rear", "left" and "right" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the position or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0317] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0318] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A vehicle control method, characterized in that, include: When the target vehicle is in motion, the perception information of the target vehicle is acquired. The perception information includes first information for representing the inertial motion state of the vehicle, second information for representing the wheel motion state, and third information for representing the suspension height state. Based on the first information, the second information, and the third information, the interference component generated by the vehicle's own motion is extracted from the first information to obtain the gravity direction characterization quantity; the interference component includes the eccentric acceleration generated by the vehicle's own motion. Based on the first information and the gravity direction representation quantity, an estimated slope value of the road where the target vehicle is located is generated; Based on the slope estimate, the target vehicle is controlled to adapt to the current road slope conditions. The method specifically includes: The vehicle attitude angle is corrected by the gravity direction characterization quantity, and the road slope estimate is obtained based on the corrected vehicle attitude angle. The slope estimate is obtained by using the angular velocity in the first information as the basis for state prediction, predicting the vehicle attitude angle through the time update equation of the Kalman filter, and using the gravity direction characterization quantity as the observation quantity, correcting the predicted value through the measurement update equation of the Kalman filter. The method further includes: The slope monitoring value of the road where the target vehicle is located is collected based on the sensor, and the initial road surface smoothness is obtained. The vertical vibration acceleration of the target vehicle's wheels, collected by the unsprung acceleration sensor, is obtained, and the target road surface smoothness is determined based on the vertical vibration acceleration of the wheels and the initial road surface smoothness. The weighted fusion coefficients are determined based on the target road surface smoothness, longitudinal motion acceleration, and lateral motion acceleration. The estimated slope value and the monitored slope value are weighted and fused based on the weighted fusion coefficient to output the target road slope value, so as to control the target vehicle according to the target road slope value.

2. The method according to claim 1, characterized in that, The step of extracting the interference component caused by the vehicle's own motion from the first information, based on the first information, the second information, and the third information, to obtain the gravity direction characterization quantity includes: Based on the first information, the second information, and the third information, the eccentric acceleration, longitudinal motion acceleration, lateral motion acceleration, and vehicle body posture gravitational acceleration generated by the vehicle's own motion are calculated to generate the interference components. The initial acceleration is obtained from the first information, and the interference component is removed from the initial acceleration to obtain the gravitational direction characterization quantity.

3. The method according to claim 2, characterized in that, Based on the first information, the second information, and the third information, the eccentric acceleration, longitudinal acceleration, lateral acceleration, and vehicle attitude gravitational acceleration generated by the vehicle's own motion are calculated to generate the interference components, including: Based on the angular velocity and angular acceleration in the first information and the offset of the installation position of the inertial measurement unit, the eccentric acceleration is calculated; The installation position offset includes the offset of the installation position of the inertial measurement unit installed on the target vehicle from the vehicle's center of gravity in multiple directions; Based on the wheel speed, steering wheel angle signal, yaw rate signal and driving or braking status signal in the second information, the longitudinal motion acceleration is calculated. Based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information, and in combination with vehicle dynamics parameters, the lateral motion acceleration is calculated. Based on the suspension height in the third information, the vehicle body's gravitational acceleration is calculated.

4. The method according to claim 3, characterized in that, The calculation of the eccentric acceleration based on the angular velocity and angular acceleration in the first information and the installation position offset of the inertial measurement unit includes: Based on the angular acceleration and the offset, a first intermediate data is determined; the first intermediate data is used to characterize the tangential eccentric acceleration caused by the angular acceleration. Based on the angular velocity and the offset, a second intermediate data is determined; the second intermediate data is used to characterize the normal eccentric acceleration caused by the angular velocity. The eccentric acceleration is generated based on the first intermediate data and the second intermediate data.

5. The method according to claim 3, characterized in that, The calculation of the longitudinal motion acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information includes: Based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information, the longitudinal speed of the target vehicle is determined. The longitudinal acceleration is determined based on the difference between the current longitudinal vehicle speed and the previous longitudinal vehicle speed, as well as a preset calculation period.

6. The method according to claim 3, characterized in that, The calculation of the lateral acceleration based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking state signal in the second information, combined with vehicle dynamics parameters, includes: Based on the wheel speed, steering wheel angle signal, yaw rate signal, and driving or braking status signal in the second information, the longitudinal speed of the target vehicle is determined. The lateral acceleration of the target vehicle is calculated by combining the longitudinal vehicle speed with preset vehicle dynamics parameters.

7. The method according to claim 3, characterized in that, The calculation of the vehicle body attitude gravitational acceleration based on the suspension height in the third information includes: Based on the suspension height in the third information, determine the vehicle's pitch angle and roll angle. Based on the vehicle pitch angle and vehicle roll angle, and combined with the gravitational acceleration constant, the gravitational acceleration components of the vehicle attitude are determined.

8. The method according to claim 1, characterized in that, The step of generating an estimated slope value for the road where the target vehicle is located based on the first information and the gravity direction representation includes: Based on the gravity direction representation, the vehicle attitude angle corresponding to the first information is corrected; Based on the corrected vehicle attitude angle, the estimated slope of the road where the target vehicle is located is determined.

9. The method according to claim 8, characterized in that, The step of correcting the vehicle attitude angle corresponding to the first information based on the gravity direction representation includes: Based on the angular velocity in the first information, determine the initial attitude angle of the target vehicle; The gravity direction characterization quantity is processed by an error state Kalman filter to obtain a posterior error state estimate. Based on the attitude error and gyroscope bias error in the posterior error state estimate, the initial attitude angle is corrected to obtain the corrected vehicle attitude angle of the target vehicle.

10. The method according to claim 1, characterized in that, The step of controlling the target vehicle based on the target road gradient value includes: When the target road slope value indicates an uphill condition, an instruction to increase rear suspension stiffness and / or activate hill start assist braking is generated, and the instruction to increase rear suspension stiffness and / or activate hill start assist braking is sent to the chassis execution system corresponding to the target vehicle. When the target road gradient value indicates a downhill condition, an instruction to increase energy recovery intensity and / or reduce drive system output torque is generated, and the instruction to increase energy recovery intensity and / or reduce drive system output torque is sent to the chassis execution system corresponding to the target vehicle.

11. A vehicle, characterized in that, The system includes a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to implement the vehicle control method as described in any one of claims 1 to 10.

Images