Self-adaptive adjusting method and system for suspension rigidity and transmission output torque

Abstract

The invention relates to a suspension rigidity and transmission output torque self-adaptive adjustment method, which comprises the following steps of: acquiring multi-source time sequence data in a vehicle driving process, constructing a feature vector for judging a driving intention of a driver based on the multi-source time sequence data, matching the feature vector with a preset intention template library, and determining the driving intention of the driver according to the matched intention template library. Performing behavior deduction by adopting a Bayesian network model in combination with time sequence characteristics in a continuous time window, outputting a driver intention probability vector, and determining a current driving intention of the driver based on the intention probability vector; and executing a branch adjustment strategy based on the current driving intention: if the intention is the first intention. According to the method, the current driving intention is recognized, and then the target suspension rigidity and / or the target transmission torque are directly calculated based on the recognized form intention, so that cooperative adjustment of the suspension rigidity and the transmission torque is realized, and dynamic response, driving stability and riding comfort can be considered under different driving working conditions.

Description

Technical Field

[0001] This invention relates to the field of vehicle suspension and torque control technology, specifically to a method and system for adaptive adjustment of suspension stiffness and transmission output torque. Background Technology

[0002] With the rapid development of new energy vehicles, users' comprehensive demands for vehicle power, handling stability and ride comfort are increasing day by day. As core components affecting the above performance, the chassis air suspension system and power transmission system directly determine the overall driving quality of the vehicle through their coordinated control capabilities.

[0003] In related technologies, air suspension systems and powertrain systems typically employ independent control strategies. The suspension system primarily adjusts stiffness and damping based on road conditions and vehicle posture to enhance ride comfort; while the powertrain system primarily distributes output torque based on the driver's throttle and braking operations and vehicle speed to ensure power response. There is a lack of effective information exchange and coordination mechanisms between the two. Therefore, because the suspension and transmission systems are adjusted independently, they cannot coordinate under dynamic conditions such as acceleration, steering, and braking. This makes it difficult to balance power and comfort, and to balance multiple objectives such as power, stability, and comfort under different driving conditions. Moreover, existing technologies usually only make responsive adjustments based on sensor signals at the current moment, that is, the system only makes corresponding adjustments after the driver's operation, lacking the ability to predict the driver's operation trend, resulting in a lag in adjustment response. Therefore, there is an urgent need for an adaptive adjustment method that can achieve coordinated control of the suspension and transmission systems and has the ability to predict the driver's intentions, in order to meet the needs of power, stability, and comfort under different driving conditions. Summary of the Invention

[0004] This application provides a method and system for adaptive adjustment of suspension stiffness and transmission output torque. It identifies the driver's intention probability vector through multi-source time-series data, and outputs an adaptation score by combining operating condition parameters. After calculating the target value based on the intention, the predictive control algorithm corrects the target value according to the intention probability vector and the adaptation score, thereby realizing the adaptive adjustment of suspension stiffness and transmission torque under different driving conditions.

[0005] In a first aspect, embodiments of this application provide a method for adaptive adjustment of suspension stiffness and transmission output torque, comprising the following steps: The system acquires multi-source time-series data during vehicle operation, constructs a feature vector based on the multi-source time-series data to determine the driver's driving intention, matches the feature vector with a preset intention template library, and uses a Bayesian network model combined with time-series features within a continuous time window to perform behavior inference, outputting a driver intention probability vector, and determining the driver's current driving intention based on the intention probability vector. Based on the current driving intention, a branch adjustment strategy is executed: if it is the first intention, the target transmission torque is calculated; if it is the second intention, the target suspension stiffness is calculated; if it is the third intention, both the target transmission torque and the target suspension stiffness are calculated. The calculated target suspension stiffness and / or target transmission torque are output to the actuator.

[0006] In conjunction with the first aspect, in one embodiment, the multi-source time-series data includes steering wheel angle, accelerator pedal opening, brake pressure, vehicle speed, and lateral acceleration, and the time-series features include the rate of change of steering wheel angle, accelerator pedal opening, and brake pressure within a continuous time window.

[0007] In conjunction with the first aspect, in one implementation, matching the feature vector with a preset intent template library specifically includes: The cosine similarity algorithm is used to trigger initial matching when the similarity is higher than a preset threshold.

[0008] In conjunction with the first aspect, in one implementation, the first intention is an intention to accelerate or overtake, the second intention is an intention to steer or brake, and the third intention is an intention to drive smoothly.

[0009] In conjunction with the first aspect, in one embodiment, the step of outputting the calculated target suspension stiffness and / or target transmission torque to the actuator further includes: Obtain the current operating parameters of the vehicle, determine the ideal suspension stiffness and ideal transmission torque under the current operating conditions based on the operating parameters and the current driving intention, compare the deviation between the current suspension stiffness and the ideal suspension stiffness, and the deviation between the current transmission torque and the ideal transmission torque, and output the adaptation score. Based on the intent probability vector and the adaptation score, the target suspension stiffness and / or target transmission torque are corrected by a predictive control algorithm to generate a corrected adjustment command and output it to the actuator.

[0010] In conjunction with the first aspect, in one embodiment, the operating parameters include vehicle speed and road roughness, and the ideal suspension stiffness and ideal transmission torque are calculated using the following formulas: ; ; in, The reference stiffness corresponding to the current driving intention. and The road condition coefficient corresponding to the current driving intention. For road surface roughness, and The speed coefficient corresponding to the current driving intention. For vehicle speed, This is the reference torque corresponding to the current driving intention; The adaptation score is calculated by weighting the deviation between the current suspension stiffness and the ideal suspension stiffness, and the deviation between the current transmission torque and the ideal transmission torque.

[0011] In conjunction with the first aspect, in one implementation, if the intention is first, then the target transmission torque is calculated; if the intention is second, then the target suspension stiffness is calculated; if the intention is third, then both the target transmission torque and the target suspension stiffness are calculated, specifically including: If the current driving intention is to accelerate or overtake, the target transmission torque is calculated based on the rate of change of the accelerator pedal. If the current driving intention is a steering or braking intention, the target suspension stiffness is calculated based on the steering angle and braking intensity. If the current driving intention is to drive smoothly, then the target transmission torque and target suspension stiffness are calculated based on the road roughness.

[0012] In conjunction with the first aspect, in one implementation, the predictive control algorithm optimizes the target suspension stiffness and / or target transmission torque by minimizing a cost function, the cost function including an intent matching term, a state deviation term, and a strategy smoothing term.

[0013] In conjunction with the first aspect, in one embodiment, after generating the modified adjustment command and outputting it to the actuator, the method further includes: Collect and adjust feedback data to characterize the vehicle's operating status. If the feedback data does not meet the preset feedback standard, then readjust the target suspension stiffness and / or target transmission torque.

[0014] Secondly, embodiments of this application provide a system based on an adaptive adjustment method for suspension stiffness and transmission output torque, comprising: The data acquisition module is used to acquire multi-source time-series data during vehicle operation. The intent recognition module is used to construct a feature vector based on the multi-source time-series data, match the feature vector with a preset intent template library, and use a Bayesian network model combined with the time-series features within a continuous time window to perform behavior inference, output a driver intent probability vector, and determine the driver's current driving intent based on the intent probability vector. The calculation module is used to execute a branch adjustment strategy based on the current driving intention: if the intention is the first intention, the target transmission torque is calculated; if the intention is the second intention, the target suspension stiffness is calculated; if the intention is the third intention, the target transmission torque and the target suspension stiffness are calculated, and the calculated target suspension stiffness and / or target transmission torque are output to the actuator.

[0015] The beneficial effects of the technical solutions provided in this application include: 1. This application identifies the current driving intention and then directly calculates the target suspension stiffness and / or target transmission torque based on the identified intention. If the intention is to accelerate or overtake, the transmission torque is adjusted first to ensure power response; if the intention is to turn or brake, the suspension stiffness is adjusted first to suppress body roll or pitching and ensure driving stability; if the intention is to drive smoothly, the suspension stiffness and transmission torque are simultaneously fine-tuned to improve ride comfort. This achieves coordinated adjustment of suspension stiffness and transmission torque, enabling it to balance power response, driving stability and ride comfort under different driving conditions.

[0016] 2. This application acquires multi-source time-series data and uses a Bayesian network model combined with time-series features within a continuous time window to perform behavior inference. It can accurately predict the driver's intentions and output the intention probability vector in advance during operation, solving the problem of delayed response in existing technologies and improving the initiative and response speed of regulation.

[0017] 3. This application improves the multi-objective balance capability of power, stability and comfort under different driving conditions by performing intention recognition and state assessment, using adaptation scores and intention probability vectors to optimize and correct target values ​​through predictive control algorithms, and then using feedback data for closed-loop adjustment after execution. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram illustrating the main steps of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0021] Example 1: Please see Figure 1 and Figure 2 This application provides a method for adaptive adjustment of suspension stiffness and transmission output torque, including the following steps: S1. Acquire multi-source time-series data during vehicle driving, construct a feature vector based on the multi-source time-series data to determine the driver's driving intention, match the feature vector with a preset intention template library, and use a Bayesian network model combined with the time-series features within a continuous time window to perform behavior inference, output the driver's intention probability vector, and determine the driver's current driving intention based on the intention probability vector. S101, Data Preprocessing: This embodiment acquires multi-source time-series data in real time through an onboard sensor network. In one specific embodiment, the following signals are simultaneously acquired at a sampling rate of 100Hz: including steering wheel angle. Accelerator pedal opening Braking pressure Speed and lateral acceleration ; After collecting the above data, this embodiment also includes a preprocessing step. In a specific implementation of this embodiment, this includes: noise reduction using Kalman filtering, removal of outlier data using a 3σ statistical thresholding method based on a sliding window, and construction of the feature vector for the current time step after data normalization. ; S102, Intent template library matching: In this embodiment, based on the differences in the adjustment strategies for suspension stiffness and transmission torque, the driver's intentions are divided into three categories: the first intention (acceleration or overtaking intention), which corresponds to prioritizing the adjustment of transmission torque to ensure power response; the second intention (steering or braking intention), which corresponds to prioritizing the adjustment of suspension stiffness to ensure driving stability; and the third intention (smooth driving intention), which corresponds to simultaneously adjusting suspension stiffness and transmission torque to improve ride comfort. The above classification method fully considers the different needs for power, stability and comfort in different driving scenarios. The above classification method is only an example. In actual application, it can be adaptively adjusted according to the vehicle tuning style or driving mode setting. For example, in sport mode, a category of aggressive driving intention can be added, and in off-road mode, a category of getting out of trouble can be added.

[0022] To achieve the recognition of the aforementioned intentions, this embodiment pre-constructs an intention template library, which stores reference feature vectors corresponding to the three types of intentions mentioned above, such as acceleration or overtaking intention templates, turning or braking intention templates, and smooth driving intention templates. In a specific example of this embodiment, the pre-constructed intention template library includes: Intention to accelerate or overtake: ; Steering or braking intention: ; Intent for smooth driving: ; It should be noted that the specific values ​​of the template vectors listed in this embodiment are merely exemplary values ​​for ease of understanding. In actual applications, they can be flexibly set according to actual vehicle calibration data, driving style, or specific working conditions, and do not constitute a limitation on the scope of protection of this application.

[0023] This embodiment uses the cosine similarity algorithm to calculate the similarity between the current feature vector and each template: ; When the similarity is higher than a preset threshold, preliminary matching is triggered. In this embodiment, the preset threshold is preferably 0.8, which can be adaptively adjusted according to vehicle model tuning, driving mode or actual application scenario. For example, the threshold can be appropriately reduced in sports mode to improve the sensitivity of intent recognition, and the threshold can be appropriately increased in comfort mode to reduce false triggering. The specific value does not constitute a limitation on the scope of protection of this application.

[0024] This step outputs a preliminary matching set of intent candidates, which are one or more intent categories with similarity higher than a preset threshold. If the similarity of multiple intents is higher than the preset threshold, multiple candidate intents are output for further refinement based on subsequent behavior deduction. If the similarity of all intents is lower than the preset threshold, no candidate intents are output, and at this time, it can be directly determined as the third intent (smooth driving intent).

[0025] S103, Behavioral Deduction: The feature vector of the current moment output in step S101, the intent candidate set output in step S102, and the historical multi-source time series data within the continuous time window are jointly input into the Bayesian network model. In this embodiment, the historical multi-source time series data is preferably from the past 3 seconds, which can be adjusted according to actual needs. For example, the window can be appropriately extended under high-speed conditions to improve the stability of intent recognition, and the window can be appropriately shortened under urban conditions to improve response sensitivity. The Bayesian formula is as follows: ; in, For intent category, This is the prior probability (obtained through training with historical driving data). Let be the likelihood function.

[0026] It should be noted that the prior probabilities and likelihood functions of the Bayesian network are trained using historical driving data collected from real vehicles. Different model parameters will be trained using datasets of different vehicle models and driving styles. This embodiment does not limit the specific training data.

[0027] Extracting temporal features from a continuous time window of the past 3 seconds, this embodiment preferably uses the rate of change of steering wheel angle, accelerator pedal opening, and brake pressure; For example, if the rate of change of the accelerator pedal over three consecutive frames is +20% / s, +30% / s, and +20% / s, and shows a continuous upward trend, it can be inferred that the driver intends to accelerate or overtake.

[0028] For each intent in the candidate set, its posterior probability is calculated, and the final output intent probability vector P= ; In this embodiment, the dimension with the highest probability is randomly selected as the current driving intention: .

[0029] S2. Execute branch adjustment strategy based on the current driving intention: If it is the first intention, calculate the target transmission torque; if it is the second intention, calculate the target suspension stiffness; if it is the third intention, calculate both the target transmission torque and the target suspension stiffness. In this embodiment, the specific calculation formula for the branch adjustment strategy is as follows: ① If S1 determines the primary intention (acceleration or overtaking intention): then prioritize power response and calculate the target transmission torque: ; in, Where k is the current base torque, and k is the dynamic gain coefficient. This represents the rate of change of the accelerator pedal. It should be noted that the dynamic gain coefficient k is determined through actual vehicle calibration. In this embodiment, the example value is 1 (corresponding to a throttle change rate in units of % / s). Its specific value can be adjusted according to the vehicle's power characteristics, driving mode, etc. For example, in sport mode, the value of k can be appropriately increased to improve response sensitivity, while in economy mode, the value of k can be appropriately decreased to reduce energy consumption.

[0030] ②If S1 is determined to be the secondary intention (steering or braking intention): then prioritize ensuring driving stability and calculate the target suspension stiffness: ; in, Given the current suspension base stiffness, For adjustment coefficients, This is the absolute value of the steering wheel angle. This represents the rate of change of brake pressure. It should be noted that the adjustment coefficient g is determined through actual vehicle calibration; in this embodiment, the example value is 0.2. In practical applications, it can be adjusted according to the characteristics of the suspension system, the vehicle's center of gravity, etc.

[0031] ③ If S1 determines the third intention (smooth driving intention): then comfort must also be considered, therefore the two systems need to be fine-tuned in coordination: ; ; in, This is the torque adjustment amount. This is the stiffness adjustment amount. This refers to the road roughness coefficient (0-1, provided by road surface sensors or map data). It should be noted that... and Based on actual vehicle calibration, the example value in this embodiment is determined to be... =5Nm, =10N / mm, which can be adjusted according to the vehicle's positioning (comfort, sport, etc.) in practical applications.

[0032] S3. Output the calculated target suspension stiffness and / or target transmission torque to the actuator.

[0033] Based on the driver's current driving intention determined by S1, the target suspension stiffness and / or target transmission torque calculated by S2 are sent to the corresponding actuator (transmission system controller or air suspension controller) via CAN bus, and then the adjustment is completed. This achieves coordinated adjustment of suspension stiffness and transmission torque, enabling it to balance power response, driving stability and ride comfort under different driving conditions.

[0034] S4. State Assessment and Collaborative Correction: In addition, this embodiment includes a correction step before S3, which specifically includes: S401: Obtain the current operating parameters of the vehicle, determine the ideal suspension stiffness and ideal transmission torque under the current operating conditions based on the operating parameters and the current driving intention, compare the deviation between the current suspension stiffness and the ideal suspension stiffness, and the deviation between the current transmission torque and the ideal transmission torque, and output the adaptation score. In this embodiment, the vehicle's current operating parameters include vehicle speed V and road roughness coefficient. ; Ideal suspension stiffness and ideal transmission torque are obtained using the following formulas: ; ; in, The reference stiffness corresponding to the current driving intention. and The road condition coefficient corresponding to the current driving intention. For road surface roughness coefficient, and The speed coefficient corresponding to the current driving intention. For vehicle speed, This is the reference torque corresponding to the current driving intention; It should be noted that the above coefficients are all determined through actual vehicle calibration. Different intention categories correspond to different coefficient values. This embodiment does not limit the specific values. In actual applications, calibration can be performed according to the characteristics of the vehicle model.

[0035] The fit score is calculated using the following formula: ; in, Given the current suspension stiffness, The current transmission torque, , Weighting coefficients ( =1), the example value in this embodiment is 1). All are 0.5. In practical applications, they can be adjusted according to vehicle tuning preferences (such as comfort orientation emphasizing stiffness weight, and sport orientation emphasizing torque weight).

[0036] S402: Based on the intent probability vector and the adaptation score, the target suspension stiffness and / or target transmission torque are corrected through a predictive control algorithm to generate a corrected adjustment command and output it to the actuator.

[0037] The pre-adjustment strategy output by the intention probability vector P, the fit score S, and S2. The input predictive control algorithm solves for the optimal regulation strategy by minimizing the cost function. ; Where P is the intent probability vector P output in step S103. ; This is a reference intent vector, used to represent the expected distribution of driving intentions. It is predetermined through real vehicle calibration. In this embodiment, the example value is [value to be filled in]. =[0.9, 0.05, 0.05], which can be dynamically adjusted according to the driving mode (Sport / Comfort / Economy) in practical applications; S is the adaptation score output in step S401; The variables represent the adjustment strategy to be optimized. , , These are weighting coefficients, used to balance intent matching degree, state deviation tolerance, and policy smoothness, and are predetermined through real vehicle calibration; in this embodiment... , , Example value 0.4 0.4 The value is 0.2; in practical applications, it can be dynamically adjusted according to the driving mode (Sport / Comfort / Economy). For example, it can be appropriately increased in Sport mode. To prioritize meeting the driver's intentions, the comfort mode can be appropriately increased. Prioritizing improved passenger comfort; The expected fit score is used to characterize the expected system state matching degree. It is predetermined through real vehicle calibration. In this embodiment, the example value is 0.8. In actual applications, it can be dynamically adjusted according to the driving mode (sport / comfort / economy). The pre-adjustment strategy output in step S2 is the target transmission torque and / or target suspension stiffness. It should be noted that the above parameters , The specific values ​​of α, β, and γ are all exemplary values ​​in this embodiment. In actual applications, they can be flexibly calibrated and adjusted according to vehicle characteristics, driving modes, control precision requirements, etc., and do not constitute a limitation on the scope of protection of this application.

[0038] Solve iteratively using the gradient descent method. ; Iterative update Until the convergence threshold satisfy Output the corrected strategy Output to the actuator.

[0039] in, The convergence threshold is predetermined through actual vehicle calibration. In this embodiment, the example value is 0.01. In actual applications, it can be adjusted according to the optimization accuracy and speed requirements.

[0040] In summary, by using the intention probability vector P and the adaptation score S to optimize and correct the target suspension stiffness and / or target transmission torque through predictive control algorithms, the adjustment strategy can be pre-adjusted according to the current operating conditions before execution, avoiding problems such as excessive power response or insufficient stability caused by mismatch in conditions, thus improving the accuracy and adaptability of the adjustment. At the same time, through the strategy smoothing term in the cost function, abrupt changes in the adjustment amount are effectively prevented, ensuring the smoothness of the adjustment process and further improving driving comfort.

[0041] S5, Closed-loop feedback: In addition, after applying the modified target suspension stiffness and / or target transmission torque, this embodiment also includes: Collect and adjust feedback data to characterize the vehicle's operating status. If the feedback data does not meet the preset feedback standard, then readjust the target suspension stiffness and / or target transmission torque. Specifically, the feedback data in this embodiment includes, but is not limited to, actual suspension stiffness, actual transmission torque, actual vehicle speed, and actual acceleration. After collecting the above data, the adaptation effect error is calculated using the following formula: ; in, For the actual feedback data collected, For the corresponding preset target value, Number of feedback dimensions; If the adaptation effect error E is greater than the preset threshold, return to step S402 and re-execute the collaborative correction, that is, optimize and correct the target suspension stiffness and / or target transmission torque again through the predictive control algorithm until the adaptation effect error E is less than or equal to the preset threshold. In this embodiment, the preset threshold is preferably 0.1, but it can be adjusted according to the control accuracy requirements in actual applications.

[0042] It should be noted that when returning to step S402 for re-correction, the pre-adjustment strategy output in step S2... The adjustment instructions will be updated to reflect the previous revision. The intention probability vector P and the adaptation score S are also updated synchronously to the real-time values ​​at the current moment to ensure the effectiveness of the closed-loop iteration.

[0043] S6. Intent recognition feature library update; After the adjustment effect meets the target (i.e., the adaptation effect error E in step S5 is less than or equal to the preset threshold), this embodiment also includes synchronously updating the intent recognition feature library: ; in, For learning rate, This refers to the current driving characteristics (including steering wheel angle, accelerator pedal opening, brake pressure, vehicle speed, lateral acceleration, etc.). The feature library stores historical feature vectors. By weighting and fusing current driving features with historical features, the feature library can continuously adapt to changes in the driver's driving habits.

[0044] In this embodiment, the learning rate α is preferably 0.7. In practical applications, it can be adjusted according to the learning speed requirements. For example, when adapting to a new driver, the learning rate can be increased to speed up the learning process. It should be noted that this update mechanism is only executed after the adjustment effect meets the standard, ensuring that the data entered into the database are valid and reasonable driving features, and avoiding abnormal data from polluting the feature database.

[0045] By setting up an intent recognition feature library update mechanism, the current driving characteristics are weighted and fused with historical characteristics after the adjustment effect meets the target. This allows the feature library to continuously learn and adapt to the current driver's driving habits. As usage time increases, the intent recognition module's understanding of driver operation characteristics becomes more accurate, thereby further improving the accuracy of intent recognition, reducing the response delay of subsequent adjustments, and realizing personalized adaptive control. With long-term use, this can effectively improve driving comfort and system response efficiency.

[0046] Example 2: This embodiment provides an adaptive adjustment system for suspension stiffness and transmission output torque, which is used to execute the adaptive adjustment method for suspension stiffness and transmission output torque as described in Embodiment 1.

[0047] Specifically, the system includes: The data acquisition module is used to acquire multi-source time-series data during vehicle operation. This module is connected to the vehicle sensor network and collects signals such as steering wheel angle, accelerator pedal opening, brake pressure, vehicle speed and lateral acceleration in real time. It also performs preprocessing on the collected data, including time synchronization, filtering and noise reduction, anomaly removal and normalization, to provide high-quality data input for subsequent modules. The intent recognition module is used to construct feature vectors based on multi-source time-series data, match the feature vectors with a preset intent template library, and use a Bayesian network model to combine time-series features within a continuous time window to perform behavior inference, outputting a driver intent probability vector. Based on the intent probability vector, the driver's current driving intent is determined. The output of this module includes the intent probability vector and the current driving intent category, which are used for subsequent collaborative correction and branch adjustment, respectively.

[0048] The calculation module is used to execute branch adjustment strategies based on the current driving intention: if the intention is the first intention (acceleration or overtaking intention), the target transmission torque is calculated; if the intention is the second intention (steering or braking intention), the target suspension stiffness is calculated; if the intention is the third intention (smooth driving intention), the target transmission torque and target suspension stiffness are calculated, and the calculated target suspension stiffness and / or target transmission torque are output to the actuator.

[0049] The modules mentioned above interact with each other via an onboard communication bus (such as a CAN bus) to collaboratively achieve adaptive adjustment of suspension stiffness and transmission output torque.

[0050] The system in this embodiment corresponds to the method in Embodiment 1 and has the same beneficial effects, which will not be repeated here.

[0051] In the description of this application, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device 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. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this application can be understood according to the specific circumstances.

[0052] It should be noted that in this application, relational terms such as "first" and "second" are used merely 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. Furthermore, 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 a process, method, article, or apparatus. Without further limitations, 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 said element.

[0053] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for adaptive adjustment of suspension stiffness and transmission output torque, characterized in that, Includes the following steps: The system acquires multi-source time-series data during vehicle operation, constructs a feature vector based on the multi-source time-series data to determine the driver's driving intention, matches the feature vector with a preset intention template library, and uses a Bayesian network model combined with time-series features within a continuous time window to perform behavior inference, outputting a driver intention probability vector, and determining the driver's current driving intention based on the intention probability vector. Based on the current driving intention, a branch adjustment strategy is executed: if it is the first intention, then the target transmission torque is calculated; If the second intention is to calculate the target suspension stiffness; If it is a third intention, then calculate the target transmission torque and the target suspension stiffness; The calculated target suspension stiffness and / or target transmission torque are output to the actuator.

2. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 1, characterized in that, The multi-source time-series data includes steering wheel angle, accelerator pedal opening, brake pressure, vehicle speed, and lateral acceleration. The time-series features include the rate of change of steering wheel angle, accelerator pedal opening, and brake pressure within a continuous time window.

3. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 1, characterized in that, Matching the feature vector with a preset intent template library specifically includes: The cosine similarity algorithm is used to trigger initial matching when the similarity is higher than a preset threshold.

4. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 1, characterized in that, The first intention is to accelerate or overtake, the second intention is to steer or brake, and the third intention is to drive smoothly.

5. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 4, characterized in that, Before outputting the calculated target suspension stiffness and / or target transmission torque to the actuator, the process further includes: Obtain the current operating parameters of the vehicle, determine the ideal suspension stiffness and ideal transmission torque under the current operating conditions based on the operating parameters and the current driving intention, compare the deviation between the current suspension stiffness and the ideal suspension stiffness, and the deviation between the current transmission torque and the ideal transmission torque, and output the adaptation score. Based on the intent probability vector and the adaptation score, the target suspension stiffness and / or target transmission torque are corrected by a predictive control algorithm to generate a corrected adjustment command and output it to the actuator.

6. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 5, characterized in that, The operating parameters include vehicle speed and road roughness, and the ideal suspension stiffness and ideal transmission torque are calculated using the following formulas: ； ； in, The reference stiffness corresponding to the current driving intention. and The road condition coefficient corresponding to the current driving intention. For road surface roughness, and The speed coefficient corresponding to the current driving intention. For vehicle speed, This is the reference torque corresponding to the current driving intention; The adaptation score is calculated by weighting the deviation between the current suspension stiffness and the ideal suspension stiffness, and the deviation between the current transmission torque and the ideal transmission torque.

7. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 6, characterized in that, If the first intention is to calculate the target transmission torque; If the second intention is to calculate the target suspension stiffness; If the intention is third-party, then the target transmission torque and target suspension stiffness are calculated, specifically including: If the current driving intention is to accelerate or overtake, the target transmission torque is calculated based on the rate of change of the accelerator pedal. If the current driving intention is a steering or braking intention, the target suspension stiffness is calculated based on the steering angle and braking intensity. If the current driving intention is to drive smoothly, then the target transmission torque and target suspension stiffness are calculated based on the road roughness.

8. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 5, characterized in that, The predictive control algorithm optimizes the target suspension stiffness and / or target transmission torque by minimizing a cost function, which includes an intent matching term, a state deviation term, and a strategy smoothing term.

9. The adaptive adjustment method for suspension stiffness and transmission output torque according to claim 5, characterized in that, After generating the corrected adjustment command and outputting it to the actuator, the process also includes: Collect and adjust feedback data to characterize the vehicle's operating status. If the feedback data does not meet the preset feedback standard, then readjust the target suspension stiffness and / or target transmission torque.

10. A system based on the adaptive adjustment method for suspension stiffness and transmission output torque as described in claim 1, characterized in that, include: The data acquisition module is used to acquire multi-source time-series data during vehicle operation. The intent recognition module is used to construct a feature vector based on the multi-source time-series data, match the feature vector with a preset intent template library, and use a Bayesian network model combined with the time-series features within a continuous time window to perform behavior inference, output a driver intent probability vector, and determine the driver's current driving intent based on the intent probability vector. A calculation module is used to execute a branch adjustment strategy based on the current driving intention: if it is the first intention, then calculate the target transmission torque; If the second intention is to calculate the target suspension stiffness; If the intention is third, then the target transmission torque and target suspension stiffness are calculated, and the calculated target suspension stiffness and / or target transmission torque are output to the actuator.

Images