A vehicle dynamic model modeling method based on a simplified vehicle dynamics model

Abstract

The application discloses a vehicle dynamic model modeling method based on a simplified vehicle dynamics model, comprising the following steps: S10, identifying a vehicle suspension feature point, performing real-time data derivation on a geometric motion state of the vehicle suspension, and obtaining an estimated suspension motion state; S20, collecting through a posture sensor, correcting data through the estimated suspension motion state, and generating an accurate suspension motion state; S30, synthesizing the accurate suspension motion state and other collected vehicle motion information and wheel speed information, and obtaining vehicle posture data; S40, obtaining vehicle motion information collected by various sensors in the vehicle, and obtaining a more accurate tire slip angle by comparing vehicle overall motion information and wheel speed sensor information; S50, synthesizing driver input information and pre-existing vehicle information, synthesizing the above data in the vehicle model, and finally forming a vehicle dynamic model. The application is simple and practical, the prediction result is accurate, and is suitable for multiple different vehicle models.

Description

Technical Field

[0001] This invention belongs to the field of vehicle detection technology, and in particular relates to a method for modeling vehicle dynamics based on a simplified vehicle dynamics model. Background Technology

[0002] Vehicle dynamics modeling is a data-derived form that uses real-world vehicle parameters to reconstruct the vehicle's dynamic behavior in a computer, either in real-time or non-real-time. It is widely used in vehicle chassis development, autonomous driving development, racing simulation software, and vehicle dynamics engineering software. Its current technical approaches mainly fall into the following categories:

[0003] ① Simplified Vehicle Dynamics Modeling Based on Single-Point Tire Model: This modeling method, primarily used in racing simulation software and other applications with high real-time requirements, is based on a pre-set simplified tire model (typically consisting of ① a tire sidewall considered as a spring-damping system, ② a tire reference μ value, and ③ a tire surface stress detection point). By applying external forces to ① and ③, the system calculates the tire's change in ② under external forces based on empirical curves. Then, based on the calculated results, it outputs the frictional force generated by the tire under its own external forces. This is combined with factors such as load transfer and suspension forces calculated from the vehicle's acceleration to ultimately output a relatively accurate dynamic simulation result of the vehicle. However, because it is mainly used in racing simulation software with high real-time requirements, this simulation method primarily uses a simplified vehicle dynamics model that is relatively inaccurate and prone to cumulative errors. It lacks methods to control cumulative errors, and because it heavily relies on pre-entered empirical curves and vehicle data, its simulation accuracy is highly correlated with the entered information.

[0004] ② Vehicle Dynamics Modeling Based on Multi-Point Complex Tire Models: This modeling method is mainly used for industrial-level vehicle simulation and development modeling. It is based on a complete suspension mechanical model, including external force models and internal stress models, usually provided by the manufacturer or derived from extremely detailed dynamic stress analysis of the suspension system itself. A complete tire characteristic model (usually provided by the tire company) is also included to perform a complete and accurate simulation of the vehicle's dynamics. Because it is primarily used for industrial-level vehicle simulation, this method consumes enormous amounts of computing power, making real-time simulation almost impossible. It also places extremely high demands on the hardware's computing power and costs, and still heavily relies on tire models provided by tire manufacturers. This makes it almost unusable for ordinary users and virtually impossible for most users to use in areas requiring real-time simulation of vehicle dynamics, such as subjective driving experience.

[0005] ③ Vehicle dynamic modeling based on vehicle attitude and GG diagram: This modeling method is mainly used for vehicle control of its own motion mode. It is based on simple lateral and longitudinal G-values ​​and angle sensors fixed to the chassis suspension components to perform simple modeling of vehicle attitude and acceleration / deceleration states, providing an inaccurate approximate simulation of the vehicle state. Since it is mainly used for the vehicle's own onboard four-wheel drive system, although this simulation method has low computational requirements and high real-time performance, its dynamic simulation method is too simplistic. Furthermore, because its output judgment method is highly dependent on pre-tuning and empirical curves, it has no practical reference value for any hardware other than highly integrated and fully tuned existing hardware such as the vehicle's four-wheel drive system. Summary of the Invention

[0006] To address the aforementioned issues, this invention proposes a vehicle dynamics modeling method based on a simplified vehicle dynamics model. This method is relatively simple, practical, and provides accurate prediction results, making it suitable for various vehicle models.

[0007] To achieve the above objectives, the technical solution adopted by this invention is: a vehicle dynamic modeling method based on a simplified vehicle dynamics model, comprising the following steps:

[0008] S10, Identify vehicle suspension feature points, perform real-time data derivation of the geometric motion state of the vehicle suspension, and obtain an estimated suspension motion state.

[0009] S20 generates an accurate suspension motion state by collecting data through attitude sensors and estimating the suspension motion state through data correction.

[0010] S30, after accurately combining the suspension motion state with other collected vehicle motion information and wheel speed information, obtains vehicle attitude data;

[0011] S40 obtains vehicle motion information collected by various sensors in the vehicle, and obtains a more accurate tire slip angle by comparing the overall vehicle motion information with wheel speed sensor information;

[0012] S50 integrates the driver's input information and pre-given vehicle information, synthesizes the above data in the vehicle model, and finally forms a vehicle dynamic model.

[0013] Furthermore, by using a ToF vision sensor to identify specific feature points on the suspension, real-time data can be derived from the specific geometric motion state of the vehicle suspension.

[0014] Furthermore, the vehicle motion information obtained from multiple sensors in the vehicle includes vehicle motion information and wheel speed information obtained by combining data from wheel speed sensors and GPS-inertial navigation sensors.

[0015] Furthermore, the process of building a vehicle dynamic model includes the following steps:

[0016] S501, based on the suspension motion data obtained by the sensor when the vehicle is in motion, the geometric position of the inner wall of the wheel and the wheel size, and based on the geometric appearance of the suspension, calculate the wheel ground contact midpoint position P_CP, and obtain the instantaneous velocity center P_IC and the suspension rotation center P_RC for any motion state of the suspension.

[0017] S502, by integrating the suspension rotation center P_RC, the roll rotation axis of the vehicle at any time is obtained; based on the roll rotation axis and the real-time acceleration of the vehicle, the original angular momentum of the vehicle at any time is calculated.

[0018] S503, based on the suspension motion data, the actual lateral and longitudinal motion inertia are brought into the suspension motion data; since the spring value of the suspension and the geometric position of the suspension system are known, the effective damping value of the suspension under any motion condition and the suspension speed threshold of the fast and slow valve opening are obtained.

[0019] S504, Constructing a vehicle suspension model: Based on the real-time yaw and pitch data of the vehicle itself, perform real-time stress analysis on the vehicle suspension geometry to obtain the internal stress of the suspension system. Through the internal stress of the suspension and the deformation tracking of the suspension bushings achieved by visual tracking, perform independent stress analysis on each bushing, and then calculate the equivalent stiffness value of the suspension flexible component based on its deformation under different external forces.

[0020] S505 uses measured suspension motion data, raw angular momentum, suspension speed threshold of fast / slow valve switching, and equivalent stiffness value to determine all forces acting on the tires at any given time, including forces caused by the transfer of the vehicle's center of gravity, forces caused by the mutual traction of rigid components within the suspension system, and forces caused by the deformation of flexible suspension components.

[0021] S506, perform a first derivative on the constructed vehicle suspension model, and obtain the real-time tire load W by combining the force analysis obtained from the suspension geometry. transfer By inputting the vehicle's acceleration on the X, Y, and Z axes and the vehicle's slip angle obtained from the vehicle's motion state, the overall grip force vector of the vehicle is decomposed to obtain the real-time grip force estimate on each wheel. Based on the difference between the wheel speed sensor and the calculated ground speed, the tire slip ratio κ is calculated.

[0022] S507, based on the total load W on the tires within the effective range. Transfer,sum With four-wheel lateral grip F y Based on the corresponding relationship, an equivalent grip force μ corresponding to the wheel deformation is derived, where...

[0023]

[0024] The first part of forming a simplified tire model is the empirical model of tire horizontal grip.

[0025] S508. Since the suspension system load and wheel deformation are known, the wheel as a whole is regarded as two longitudinal spring-damping structures connected by the tread. Based on the external force applied by the suspension to the tire and the actual deformation of the tire itself, the equivalent tire shape variable tire force correspondence is calculated to form the second part of the simplified tire model, namely the tire force empirical model.

[0026] S509, based on the suspension motion data measured in S501, the original angular momentum obtained in S502, the suspension speed threshold for the fast and slow valve switching obtained in S503, the equivalent stiffness value obtained in S504, the empirical model of tire horizontal grip obtained in S507, and the empirical model of tire force obtained in S508, the vehicle dynamic model is obtained by combining them.

[0027] Furthermore, the vehicle dynamic model includes a tire model, comprising the steps of: acquiring vehicle chassis data and assumed data, and acquiring the vehicle motion state; combining these two parts to solve the vehicle chassis data and produce a GG tire model.

[0028] Furthermore, in the vehicle dynamics model, the vehicle is then subjected to a series of intense driving maneuvers on a closed road to obtain the vehicle's motion state under these conditions. Subsequently, the four-wheel grip is modeled using a GG tire model scatter plot through data decoupling and continuously corrected.

[0029] The beneficial effects of adopting this technical solution are:

[0030] Compared to simplified vehicle dynamic modeling based on single-point tire models, this invention uses a backtracking modeling approach, which is less prone to cumulative errors and improves accuracy. Furthermore, the tire grip envelope model based on GG scatter plots does not rely on simplified single-point tire models that are highly dependent on experience or debugging time, thus reducing the professional threshold and debugging time. It can also be used for actual vehicle control and modeling.

[0031] Compared to vehicle dynamic modeling based on complex multi-point tire models, this invention is based on a simplified vehicle dynamics model, which greatly reduces the computational requirements of the model itself. Furthermore, this invention has real-time self-correction capabilities, which not only simplifies the vehicle development process but also significantly reduces costs.

[0032] Compared to vehicle dynamic modeling based on vehicle attitude and GG diagrams for four-wheel drive systems, this invention can not only be used for actual vehicle control and modeling, but is also more accurate and reliable than the two models mentioned above. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of a vehicle dynamic modeling method based on a simplified vehicle dynamics model according to the present invention.

[0034] Figure 2 This is a schematic diagram illustrating the establishment of the tire model in the vehicle dynamic model of this invention. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings.

[0036] In this embodiment, see Figure 1 As shown, this invention proposes a vehicle dynamics modeling method based on a simplified vehicle dynamics model, including the following steps:

[0037] S10, Identify vehicle suspension feature points, perform real-time data derivation of the geometric motion state of the vehicle suspension, and obtain an estimated suspension motion state.

[0038] S20 generates an accurate suspension motion state by collecting data through attitude sensors and estimating the suspension motion state through data correction.

[0039] S30, after combining the accurate suspension motion state with other collected vehicle attitude information, obtains vehicle attitude data.

[0040] S40 obtains vehicle motion information collected by various sensors in the vehicle, and obtains a more accurate tire slip angle by comparing the overall vehicle motion information with wheel speed sensor information;

[0041] S50 integrates the driver's input information and pre-given vehicle information, synthesizes the above data in the vehicle model, and finally forms a vehicle dynamic model.

[0042] Preferably, a ToF vision sensor is used to identify specific feature points on the suspension and derive real-time data on the specific geometric motion state of the vehicle suspension.

[0043] Preferably, the vehicle motion information obtained from multiple sensors in the vehicle includes: vehicle motion information obtained by combining data from wheel speed sensors and GPS-inertial navigation sensors; or vehicle motion information synthesized from other sensors that can provide similar information, such as an accelerometer array or a 360° vehicle camera.

[0044] As an optimization of the above embodiments, the process of constructing the vehicle dynamic model includes the following steps:

[0045] S501, based on the suspension motion data obtained by the sensor when the vehicle is in motion, the geometric position of the inner wall of the wheel and the wheel size, and based on the geometric appearance of the suspension, calculate the wheel ground contact midpoint position P_CP, and obtain the instantaneous velocity center P_IC and the suspension rotation center P_RC for any motion state of the suspension.

[0046] S502, by integrating the suspension rotation center P_RC, the roll rotation axis of the vehicle at any time is obtained; based on the roll rotation axis and the real-time acceleration of the vehicle, the original angular momentum of the vehicle at any time is calculated.

[0047] S503, based on the suspension motion data, the actual lateral and longitudinal motion inertia are brought into the suspension motion data; since the spring value of the suspension and the geometric position of the suspension system are known, the effective damping value of the suspension under any motion condition and the suspension speed threshold of the fast and slow valve opening are obtained.

[0048] S504, Constructing a vehicle suspension model: Based on the real-time yaw and pitch data of the vehicle itself, perform real-time stress analysis on the vehicle suspension geometry to obtain the internal stress of the suspension system. Through the internal stress of the suspension and the deformation tracking of the suspension bushings achieved by visual tracking, perform independent stress analysis on each bushing, and then calculate the equivalent stiffness value of the suspension flexible component based on its deformation under different external forces.

[0049] S505 uses measured suspension motion data, raw angular momentum, suspension speed threshold of fast / slow valve switching, and equivalent stiffness value to determine all forces acting on the tires at any given time, including forces caused by the transfer of the vehicle's center of gravity, forces caused by the mutual traction of rigid components within the suspension system, and forces caused by the deformation of flexible suspension components.

[0050] S506, perform a first derivative on the constructed vehicle suspension model, and obtain the real-time tire load W by combining the force analysis obtained from the suspension geometry. transfer By inputting the vehicle's acceleration on the X, Y, and Z axes and the vehicle's slip angle obtained from the vehicle's motion state, the overall grip force vector of the vehicle is decomposed to obtain the real-time grip force estimate on each wheel. Based on the difference between the wheel speed sensor and the calculated ground speed, the tire slip ratio κ is calculated.

[0051] S507, based on the total load W on the tires within the effective range. Transfer,sum With four-wheel lateral grip F y Based on the corresponding relationship, an equivalent grip force μ corresponding to the wheel deformation is derived, where...

[0052]

[0053] The first part of forming a simplified tire model is the empirical model of tire horizontal grip.

[0054] S508. Since the suspension system load and wheel deformation are known, the wheel as a whole is regarded as two longitudinal spring-damping structures connected by the tread. Based on the external force applied by the suspension to the tire and the actual deformation of the tire itself, the equivalent tire shape variable tire force correspondence is calculated to form the second part of the simplified tire model, namely the tire force empirical model.

[0055] S509, based on the suspension motion data measured in S501, the original angular momentum obtained in S502, the suspension speed threshold for the fast and slow valve switching obtained in S503, the equivalent stiffness value obtained in S504, the empirical model of tire horizontal grip obtained in S507, and the empirical model of tire force obtained in S508, the vehicle dynamic model is obtained by combining them.

[0056] As an optimization of the above embodiments, such as Figure 2 As shown, the vehicle dynamic model includes a tire model, and includes the following steps: acquiring vehicle chassis data and assumed data, and acquiring vehicle motion state; combining these two parts to solve the vehicle chassis data and produce a GG tire model.

[0057] The system first makes assumptions about other unknown factors besides tires based on the data provided in the vehicle model; for example, for electric vehicles equipped with dual motors or more on a single axle, the system assumes that there is a known open mechanical differential.

[0058] Subsequently, the vehicle was subjected to a series of aggressive driving tests on a closed road to obtain its motion state under these conditions. Then, the four-wheel grip was modeled using a GG tire model scatter plot through data decoupling and continuously corrected.

[0059] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for modeling vehicle dynamics based on a simplified vehicle dynamics model, characterized in that, Including the following steps: S10, Identify vehicle suspension feature points, perform real-time data derivation of the geometric motion state of the vehicle suspension, and obtain an estimated suspension motion state. S20 generates an accurate suspension motion state by collecting data through attitude sensors and estimating the suspension motion state through data correction. S30, after accurately combining the suspension motion state with other collected vehicle motion information and wheel speed information, obtains vehicle attitude data; S40 obtains vehicle motion information collected by various sensors in the vehicle, and obtains a more accurate tire slip angle by comparing the overall vehicle motion information with wheel speed sensor information; S50, based on the driver's input information and the pre-given vehicle information, synthesizes the above data in the vehicle model to finally form a vehicle dynamic model; The process of building a vehicle dynamic model includes the following steps: S501, based on the suspension motion data obtained by the sensor when the vehicle is in motion, the geometric position of the inner wall of the wheel and the wheel size, and based on the geometric appearance of the suspension, calculate the wheel ground contact midpoint position P_CP, and obtain the instantaneous velocity center P_IC and the suspension rotation center P_RC for any motion state of the suspension. S502, by integrating the suspension rotation center P_RC, the roll rotation axis of the vehicle at any time is obtained; based on the roll rotation axis and the real-time acceleration of the vehicle, the original angular momentum of the vehicle at any time is calculated. S503, based on the suspension motion data, the actual lateral and longitudinal motion inertia are brought into the suspension motion data; since the spring value of the suspension and the geometric position of the suspension system are known, the effective damping value of the suspension under any motion condition and the suspension speed threshold of the fast and slow valve opening are obtained. S504, Constructing a vehicle suspension model: Based on the real-time yaw and pitch data of the vehicle itself, perform real-time stress analysis on the vehicle suspension geometry to obtain the internal stress of the suspension system. Through the internal stress of the suspension and the deformation tracking of the suspension bushings achieved by visual tracking, perform independent stress analysis on each bushing, and then calculate the equivalent stiffness value of the suspension flexible component based on its deformation under different external forces. S505 uses measured suspension motion data, raw angular momentum, suspension speed threshold of fast / slow valve switching, and equivalent stiffness value to determine all forces acting on the tires at any given time, including forces caused by vehicle center of gravity shift, forces caused by mutual traction between rigid components within the suspension system, and forces caused by deformation of flexible suspension components. S506, perform a first derivative on the constructed vehicle suspension model, and obtain the real-time tire load W by combining the force analysis obtained from the suspension geometry. transfer By inputting the vehicle's acceleration on the X, Y, and Z axes and the vehicle's slip angle obtained from the vehicle's motion state, the overall grip force vector of the vehicle is decomposed to obtain the real-time grip force estimate on each wheel. Based on the difference between the wheel speed sensor and the calculated ground speed, the tire slip ratio κ is calculated. S507, based on the total load W acting on the tires Transfer,sum With four-wheel lateral grip F y Based on the corresponding relationship, an equivalent grip force μ corresponding to the wheel deformation is derived, where... ； The first part of forming a simplified tire model is the empirical model of tire horizontal grip. S508. Since the suspension system load and wheel deformation are known, the wheel as a whole is regarded as two longitudinal spring-damping structures connected by the tread. Based on the external force applied by the suspension to the tire and the actual deformation of the tire itself, the equivalent tire deformation and tire force correspondence is calculated, forming the second part of the simplified tire model, namely the tire force empirical model. S509, based on the suspension motion data measured in S501, the original angular momentum obtained in S502, the suspension speed threshold for the fast and slow valve switching obtained in S503, the equivalent stiffness value obtained in S504, the empirical model of tire horizontal grip obtained in S507, and the empirical model of tire force obtained in S508, the vehicle dynamic model is obtained by combining them.

2. The vehicle dynamic modeling method based on a simplified vehicle dynamics model according to claim 1, characterized in that, By using a ToF vision sensor to identify feature points on the suspension, the specific geometric motion state of the vehicle suspension can be derived in real time.

3. The vehicle dynamic modeling method based on a simplified vehicle dynamics model according to claim 1, characterized in that, The vehicle motion information obtained from multiple sensors in the vehicle includes vehicle motion information and wheel speed information obtained by combining data from wheel speed sensors and GPS-inertial navigation sensors.

4. The vehicle dynamic modeling method based on a simplified vehicle dynamics model according to claim 1, characterized in that, The vehicle dynamic model includes a tire model, and includes the following steps: acquiring vehicle chassis data and assumed data, and acquiring vehicle motion state; combining these two parts to solve the vehicle chassis data and produce a GG tire model.

5. The vehicle dynamic modeling method based on a simplified vehicle dynamics model according to claim 4, characterized in that, In the vehicle dynamics model, the vehicle is then subjected to a series of intense driving maneuvers on a closed road to obtain the vehicle's motion state under these conditions. Subsequently, the four-wheel grip is modeled using a GG tire model scatter plot through data decoupling and continuously corrected.

Images