Vehicle intelligent gear control method and system based on multi-state cooperative judgment

Abstract

The invention discloses a vehicle intelligent gear control method and system based on multi-state cooperative judgment. A vehicle multi-source state sensing layer is constructed, and multi-source state data such as a physical gear operation signal, the whole vehicle Ready state, the rotating speed of a driving motor, the opening degree of a brake pedal and a current logic gear are obtained in real time. And on the basis, the physical gear operation state of the driver and the current state of the logic gear are judged, and a corresponding logic gear shifting strategy is constructed. Furthermore, a Markov state transition matrix for characteristic scheduling of each data is established by analyzing the scheduling frequency and the scheduling path of the gear shifting strategy for the multi-source state data. And finally, optimizing and determining a data acquisition strategy of the multi-source state sensing layer according to the transfer matrix. According to the invention, intelligent gear decision and data scheduling optimization under multi-state collaborative judgment are realized, and the accuracy, real-time performance and system efficiency of gear shifting control are improved.

Description

Technical Field

[0001] This invention relates to the field of automotive control technology, and in particular to a vehicle intelligent gear control method and system based on multi-state collaborative judgment. Background Technology

[0002] As vehicles become increasingly intelligent, their gear shift control systems face higher demands for safety, smoothness, and responsiveness. Traditional gear shift control methods primarily rely on physical gear position signals and a few parameters such as vehicle speed and engine speed for judgment, which has limitations in complex driving scenarios. These methods lack sufficient perception of the driver's overall operational intentions and fail to fully utilize the synergistic effect of multiple sources of state information, such as the vehicle's readiness status, brake pedal depth, and real-time motor speed. This can easily lead to unexpected gear shifts under boundary conditions, creating safety hazards.

[0003] Currently, some improvement solutions attempt to incorporate more sensor information to optimize decision-making, but most focus on upper-level control algorithms, lacking systematic optimization of the underlying multi-source data acquisition and scheduling mechanisms. The physical gear position signals, bus status, and analog sensor data generated by the vehicle during operation have varying acquisition characteristics, update frequencies, and transmission paths. Using a fixed data acquisition strategy is difficult to adapt to dynamically changing gear shifting decision requirements, potentially leading to delays in critical data updates, low system resource utilization efficiency, and consequently affecting the real-time performance and accuracy of gear shifting decisions.

[0004] Especially in electric and hybrid vehicles, the drive motors have a fast response and significantly different operating characteristics from traditional internal combustion engines, placing more stringent demands on multi-state collaborative judgment and data synchronization during gear shifting. Current technologies have not yet been able to effectively construct a control system that can adaptively optimize data scheduling and deeply integrate multi-source state perception and intelligent decision-making, making it difficult to achieve a smooth and efficient gear shifting experience while ensuring safety.

[0005] Therefore, there is an urgent need for an innovative intelligent gear control method that can improve the overall performance and reliability of the system through multi-state collaborative judgment and dynamic optimization data scheduling mechanism. Summary of the Invention

[0006] To address at least one of the aforementioned technical problems, this invention proposes a vehicle intelligent gear control method and system based on multi-state collaborative judgment.

[0007] The first aspect of this invention provides a vehicle intelligent gear control method based on multi-state collaborative judgment, comprising: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

[0008] In this solution, the construction of a multi-source state perception layer for the target vehicle, and the acquisition of multi-source state data of the target vehicle based on the multi-source state perception layer, the multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data, specifically: The data features required for the gear control of the target vehicle are obtained, including physical gear operation features, vehicle Ready status features, drive motor speed features, brake pedal opening features, and current logical gear features. In the vehicle control system, a multi-source state perception layer for the target vehicle to perceive the data features is constructed. The data features are pre-acquired based on the multi-source state perception layer, and the data acquisition features of each data feature are recorded. The data acquisition features include the debouncing time constant of the digital input port corresponding to the physical gear operation signal, the sampling rate and resolution of the analog signal corresponding to the brake pedal opening, the pulse capture timer counting period corresponding to the drive motor speed, and the identifier and periodic transmission interval of the CAN message containing the vehicle's Ready status and the current logical gear. Based on the data acquisition characteristics, determine the maximum update frequency of each data feature and the task scheduling cycle in the vehicle control system, and based on the maximum update frequency and the task scheduling cycle, determine the maximum allowable data acquisition delay time for each data feature. The data acquisition thread priority and data cache depth of the multi-source state perception layer for acquiring each data feature are determined based on the maximum allowable data acquisition delay time. The target vehicle's multi-source status data is obtained based on the data acquisition thread priority and data cache depth. The multi-source status data includes the vehicle's physical gear operation signal, vehicle Ready status, drive motor speed, brake pedal opening, and current logical gear data.

[0009] In this solution, the step of determining the driver's operational state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing a shifting strategy for the logical gear based on the operational state of the physical gear and the current state of the logical gear, specifically involves: Based on the multi-source state data, the driver's operation status of the physical gear and the current status of the logical gear are determined. When the vehicle is in a non-Ready state, the logical gear is forcibly set to N gear. When the vehicle is in Ready state and the current logical gear is N, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds a preset threshold. If it does not exceed the threshold, it keeps the logical gear in N and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it further checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D. Otherwise, it keeps the vehicle in N and prompts the user to stop and shift gears. When the vehicle is in Ready state and the current logical gear is D, if the physical gear is switched to R, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in D and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to R; otherwise, it switches to N and prompts the user to stop and shift gears. If a rising edge signal is generated when the physical gear is in the D position, the logical gear will be switched to S position.

[0010] In this solution, the step of determining the driver's operational state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing the shifting strategy of the logical gear based on the operational state of the physical gear and the current state of the logical gear, further includes: When the vehicle is in Ready state and the current logical gear is S gear, if the physical gear is switched to R gear, it first checks whether the brake pedal opening exceeds a preset threshold. If it does not exceed the threshold, the logical gear remains in S gear and the instrument prompts the user to press the brake to shift gears. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, the logical gear is switched to R gear; otherwise, it is switched to N gear and the user is prompted to stop and shift gears. If the physical gear generates a D gear position switching signal, the logical gear is switched back to D gear. When the vehicle is in Ready state and the current logical gear is R, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in R and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D; otherwise, it switches to N and prompts the user to stop and shift gears. For the operation of switching the physical gear to N gear, the logical gear is directly set to N gear in any current logical gear state, which constitutes the shifting strategy of the target vehicle's logical gear.

[0011] In this scheme, the step of obtaining the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and constructing a Markov data scheduling state transition matrix for each data feature of the multi-source state data based on the scheduling frequency and scheduling path, specifically involves: Based on the historical records of the target vehicle executing the shifting strategy within a preset time period, the timestamp sequence of the multi-source state data being called and the corresponding data identifiers are extracted during each shifting decision process. The number of times each data feature is accessed per unit time is counted to obtain the scheduling frequency. Based on the order in which data is called during the gear shifting decision process, a dependency graph between multi-source state data is constructed to obtain the scheduling path; Each data feature of the multi-source state data is used as a data scheduling state node. The scheduling transfer direction between data scheduling state nodes is determined according to the scheduling path. The number of times a data scheduling state node is transferred from one data scheduling state node to another within a continuous sampling period is calculated based on the scheduling frequency. The total number of times each state node transfers to other state nodes is counted. The state transition probability is obtained by dividing the number of times a data scheduling state node transitions from one data scheduling state node to another by the total number of transitions of that data scheduling state node. The scheduling transition probabilities between all state nodes are arranged in rows and columns to construct the Markov data scheduling state transition matrix.

[0012] In this scheme, the step of determining the data acquisition strategy for multi-source state data by the multi-source state perception layer based on the Markov data scheduling state transition matrix specifically involves: Based on the Markov data scheduling state transition matrix, the probability of each data scheduling state node being invoked in the next sampling period and the highest probability state transition path are identified. Based on the highest probability state transition path, determine the set of core data features with temporal dependencies in the gear shifting decision process from the multi-source state data; Using the data feature with the highest scheduling frequency in the core data feature set as the key scheduling node, calculate the shortest predicted path length from the key scheduling node to other data scheduling state nodes in the Markov data scheduling state transition matrix. The priority of the data acquisition threads for the corresponding data features in the multi-source state perception layer is adjusted according to the shortest predicted path length, and the data acquisition frequency of each data feature of each multi-source state data in the next sampling period is determined according to the probability that each data scheduling state node is called in the next sampling period. Adjust the corresponding data cache depth according to the steady-state probability distribution of each data scheduling state node in the state transition matrix; Based on the adjusted data acquisition thread priority, data cache depth configuration, and data acquisition frequency of each data feature in the next sampling period, the multi-source state perception layer is driven to acquire and cache multi-source state data, forming a data acquisition strategy.

[0013] A second aspect of the present invention also provides a vehicle intelligent gear shift control system based on multi-state collaborative judgment. The system includes a memory and a processor. The memory includes a vehicle intelligent gear shift control method program based on multi-state collaborative judgment. When the processor executes the vehicle intelligent gear shift control method program based on multi-state collaborative judgment, it performs the following steps: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

[0014] This invention discloses a vehicle intelligent gear control method and system based on multi-state collaborative judgment. By constructing a multi-source state perception layer for the vehicle, it acquires multi-source state data in real time, including physical gear operation signals, vehicle Ready status, drive motor speed, brake pedal opening, and the current logical gear. Based on this, it determines the driver's physical gear operation state and the current state of the logical gear, and constructs a corresponding logical gear shifting strategy. Furthermore, by analyzing the scheduling frequency and path of this shifting strategy for multi-source state data, a Markov state transition matrix for scheduling each data feature is established. Finally, based on this transition matrix, the data acquisition strategy of the multi-source state perception layer is optimized and determined. This invention achieves intelligent gear decision-making and data scheduling optimization under multi-state collaborative judgment, improving the accuracy, real-time performance, and system efficiency of gear shift control. Attached Figure Description

[0015] Figure 1 A flowchart of a vehicle intelligent gear control method based on multi-state collaborative judgment according to the present invention is shown; Figure 2 A flowchart illustrating gear switching when the logical gear position of the present invention is N gear is shown; Figure 3 A flowchart illustrating gear switching when the logical gear position of the present invention is D gear is shown; Figure 4 A flowchart illustrating gear switching when the logical gear position of the present invention is S gear is shown; Figure 5 A flowchart illustrating gear switching when the logical gear position of the present invention is R gear is shown; Figure 6 A block diagram of a vehicle intelligent gear control system based on multi-state collaborative judgment according to the present invention is shown. Detailed Implementation

[0016] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0018] Figure 1 A flowchart of a vehicle intelligent gear control method based on multi-state collaborative judgment according to the present invention is shown.

[0019] like Figure 1As shown, the first aspect of the present invention provides a vehicle intelligent gear control method based on multi-state collaborative judgment, comprising: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

[0020] It should be noted that by constructing a dedicated multi-source state perception layer, multi-dimensional real-time data, including physical operation signals, vehicle readiness status, drive motor speed, brake pedal opening, and current logical gear, is systematically integrated. This lays a solid data foundation for accurately judging driving intentions and vehicle status, significantly enhancing the system's completeness of perception under complex operating conditions. Secondly, based on this collaboratively perceived data, a shifting strategy matching the driver's physical operations and the vehicle's current logical state is dynamically constructed. This achieves closed-loop intelligent decision-making from driver intention recognition to safe and smooth gear execution, effectively preventing misoperation and improving shifting safety and the rationality of human-machine interaction. Furthermore, by analyzing the historical scheduling frequency and path of the shifting strategy for multi-source state data and constructing a Markov data scheduling state transition matrix, this method can profoundly reveal the inherent correlation and calling rules of different data features in the shifting decision-making process, thereby achieving quantitative modeling and prediction of system data flow patterns. Ultimately, based on this state transition matrix, the data acquisition strategy is intelligently determined, enabling the multi-source state perception layer to adaptively optimize the thread priority, cache depth, and sampling frequency of data acquisition. It dynamically allocates limited computing and communication resources to the data that is most likely to be called and is most critical to decision-making, thereby significantly improving the efficiency and real-time performance of data scheduling at the system level and reducing decision-making delays.

[0021] According to an embodiment of the present invention, the construction of a multi-source state perception layer for the target vehicle, and the acquisition of multi-source state data of the target vehicle based on the multi-source state perception layer, wherein the multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data, specifically: The data features required for the gear control of the target vehicle are obtained, including physical gear operation features, vehicle Ready status features, drive motor speed features, brake pedal opening features, and current logical gear features. It should be noted that the vehicle's Ready status characteristics include a series of status signals and data indicators indicating that the vehicle is ready for driving, such as the high-voltage system having completed power-on, the low-voltage control system having passed self-test, and the drive system having no faults and allowing torque output. The physical gear position is the state of the gear shift lever when the driver operates it; the logical gear position is the vehicle system control gear position obtained based on the actual vehicle state and driver operation, and after judging the rationality of the driver's gear shifting. The vehicle travels using the logical gear position.

[0022] In the vehicle control system, a multi-source state perception layer for the target vehicle to perceive the data features is constructed. The data features are pre-acquired based on the multi-source state perception layer, and the data acquisition features of each data feature are recorded. The data acquisition features include the debouncing time constant of the digital input port corresponding to the physical gear operation signal, the sampling rate and resolution of the analog signal corresponding to the brake pedal opening, the pulse capture timer counting period corresponding to the drive motor speed, and the identifier and periodic transmission interval of the CAN message containing the vehicle's Ready status and the current logical gear. Based on the data acquisition characteristics, determine the maximum update frequency of each data feature and the task scheduling cycle in the vehicle control system, and based on the maximum update frequency and the task scheduling cycle, determine the maximum allowable data acquisition delay time for each data feature. The data acquisition thread priority and data cache depth of the multi-source state perception layer for acquiring each data feature are determined based on the maximum allowable data acquisition delay time. It should be noted that, based on the maximum allowable data acquisition delay time, the system can quantify the urgency of the real-time requirements of different data features. Data features with stricter latency requirements are assigned higher acquisition thread priority to ensure that they can be scheduled and processed in a timely manner. At the same time, by combining the data update frequency and task cycle, an appropriate cache depth is calculated, which can both prevent data from being overwritten and lost due to untimely processing and prevent invalid cache from consuming too many resources.

[0023] The target vehicle's multi-source status data is obtained based on the data acquisition thread priority and data cache depth. The multi-source status data includes the vehicle's physical gear operation signal, vehicle Ready status, drive motor speed, brake pedal opening, and current logical gear data.

[0024] It should be noted that in the actual engineering implementation of intelligent gear shift control in vehicles, the multi-source state data originates from different physical interfaces and communication buses, resulting in significant differences in their inherent acquisition characteristics. For example, digital switch signals require hardware debouncing, analog signals are limited by sampling rate and resolution, pulse signals depend on timer capture accuracy, and CAN bus data follows a specific message cycle. These underlying hardware physical characteristics and protocol constraints directly determine the theoretical maximum update frequency of each data feature. If these constraints are not precisely quantified and uniformly managed, and a fixed, empirical data acquisition strategy is adopted, the system will be unable to accurately coordinate the timeliness of different data streams. Some data with high real-time requirements may miss the effective sampling window due to improper thread scheduling or insufficient buffering, while slower-updating data may waste system resources due to oversampling. By determining the maximum allowable data acquisition delay time for each data feature through data acquisition characteristics, and setting thread priorities and buffer depth accordingly, the physical limitations of the hardware and the characteristics of the communication protocol are precisely and quantitatively correlated with the real-time requirements of the upper-level gear shifting decision for the first time, starting from the system's underlying layer. By calculating the theoretical maximum allowable latency for each data type, the system can scientifically and differentiatedly allocate acquisition resources. Higher thread priority is assigned to critical signals with low latency tolerance and high real-time requirements, and appropriate buffer depths are configured to handle transient data streams, thereby ensuring that critical status information can be captured and transmitted to the decision-making level in a timely and reliable manner.

[0025] According to an embodiment of the present invention, the step of determining the driver's operation state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing a shifting strategy for the logical gear based on the operation state of the physical gear and the current state of the logical gear, specifically includes: Based on the multi-source state data, the driver's operation status of the physical gear and the current status of the logical gear are determined. When the vehicle is in a non-Ready state, the logical gear is forcibly set to N gear. like Figure 2 As shown, when the vehicle is in the Ready state and the current logical gear is N, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in N and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it further checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D. Otherwise, it keeps the N gear and prompts the user to stop and shift gears. like Figure 3As shown, when the vehicle is in the Ready state and the current logical gear is D, if the physical gear is switched to R, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in D and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to R; otherwise, it switches to N and prompts the user to stop and shift gears. If a rising edge signal is generated when the physical gear is in the D position, the logical gear will be switched to S position.

[0026] According to an embodiment of the present invention, the step of determining the driver's operation state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing a shifting strategy for the logical gear based on the operation state of the physical gear and the current state of the logical gear, further includes: like Figure 4 As shown, when the vehicle is in the Ready state and the current logical gear is S gear, if the physical gear is switched to R gear, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, the logical gear is kept in S gear and the instrument prompts the user to press the brake to shift gears. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, the logical gear is switched to R gear; otherwise, it is switched to N gear and the user is prompted to stop and shift gears. If the physical gear generates a D gear position switching signal, the logical gear is switched back to D gear. like Figure 5 As shown, when the vehicle is in the Ready state and the current logical gear is R, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in R and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D; otherwise, it switches to N and prompts the user to stop and shift gears. For the operation of switching the physical gear to N gear, the logical gear is directly set to N gear in any current logical gear state, which constitutes the shifting strategy of the target vehicle's logical gear.

[0027] It should be noted that the shifting strategy systematically integrates multiple data sources, including vehicle readiness status, current logical gear, physical gear operation, brake pedal opening, and drive motor speed, to achieve safe, precise, and driver-intention-compliant intelligent shifting control. Through multi-layered condition judgments, a rigorous safety interlock logic is established for different gear shifting paths, effectively preventing unexpected gear shifts in dangerous scenarios such as when the vehicle is not ready, while driving, or due to driver error, significantly improving driving safety. Simultaneously, this strategy, combined with a clear prompt mechanism, guides the driver to perform correct operations and ensures that shifting between forward and reverse gears only occurs when the brakes are effective and the motor speed drops to a safe threshold, avoiding shocks and wear on the powertrain and enhancing shifting smoothness and system durability.

[0028] According to an embodiment of the present invention, the step of obtaining the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and constructing a Markov data scheduling state transition matrix for each data feature of the multi-source state data based on the scheduling frequency and scheduling path, specifically involves: Based on the historical records of the target vehicle executing the shifting strategy within a preset time period, the timestamp sequence of the multi-source state data being called and the corresponding data identifiers are extracted during each shifting decision process. The number of times each data feature is accessed per unit time is counted to obtain the scheduling frequency. Based on the order in which data is called during the gear shifting decision process, a dependency graph between multi-source state data is constructed to obtain the scheduling path; Each data feature of the multi-source state data is used as a data scheduling state node. The scheduling transfer direction between data scheduling state nodes is determined according to the scheduling path. The number of times a data scheduling state node is transferred from one data scheduling state node to another within a continuous sampling period is calculated based on the scheduling frequency. The total number of times each state node transfers to other state nodes is counted. The state transition probability is obtained by dividing the number of times a data scheduling state node transitions from one data scheduling state node to another by the total number of transitions of that data scheduling state node. The scheduling transition probabilities between all state nodes are arranged in rows and columns to construct the Markov data scheduling state transition matrix.

[0029] It is important to note that in real-time decision-making for intelligent gear shifting, different multi-source state data do not exist in isolation. The frequency and order in which they are accessed reflect the inherent patterns of gear shifting logic and the driver's operating habits. However, traditional methods typically employ fixed-priority or polling data acquisition strategies, failing to effectively mine and utilize the temporal dependencies and statistical patterns among these data calls. This often leads to a disconnect between system resource allocation and actual needs. Either oversampling of unnecessarily needed data results in resource waste, or delays are introduced at critical decision moments due to untimely updates of key data. By constructing a Markov state transition matrix, the dynamic and uncertain data scheduling process is transformed into a predictable system behavior model that can be described by a probabilistic model. This matrix can quantify the probability of transitioning from one data characteristic state to another within a continuous decision-making cycle. This transforms the temporal dependencies and access frequencies of data into a mathematical basis for predicting future data demands, providing a core foundation for subsequent adaptive and forward-looking intelligent data scheduling.

[0030] According to an embodiment of the present invention, the step of determining the data acquisition strategy of the multi-source state perception layer for multi-source state data based on the Markov data scheduling state transition matrix specifically includes: Based on the Markov data scheduling state transition matrix, the probability of each data scheduling state node being invoked in the next sampling period and the highest probability state transition path are identified. Based on the highest probability state transition path, determine the set of core data features with temporal dependencies in the gear shifting decision process from the multi-source state data; Using the data feature with the highest scheduling frequency in the core data feature set as the key scheduling node, calculate the shortest predicted path length from the key scheduling node to other data scheduling state nodes in the Markov data scheduling state transition matrix. The priority of the data acquisition threads for the corresponding data features in the multi-source state perception layer is adjusted according to the shortest predicted path length, and the data acquisition frequency of each data feature of each multi-source state data in the next sampling period is determined according to the probability that each data scheduling state node is called in the next sampling period. Adjust the corresponding data cache depth according to the steady-state probability distribution of each data scheduling state node in the state transition matrix; Based on the adjusted data acquisition thread priority, data cache depth configuration, and data acquisition frequency of each data feature in the next sampling period, the multi-source state perception layer is driven to acquire and cache multi-source state data, forming a data acquisition strategy.

[0031] It's important to note that by identifying the calling probability and the highest-probability state transition path of each data node in the next sampling period, the system predicts which data features are most likely to be needed at the upcoming decision-making moment and their calling order. This allows the multi-source state awareness layer to prepare in advance, effectively reducing decision delays caused by waiting for data to be ready. By identifying the core data feature set and calculating the shortest prediction path length centered on key scheduling nodes, the system quantifies the criticality and urgency of different data in achieving the complete decision-making process. This adjusts the priority of data acquisition threads, ensuring that the data with the greatest impact on the decision-making chain and the most urgent needs receive priority in acquisition and processing resources, giving higher acquisition priority to data features on the shortest prediction path. By setting the acquisition frequency of each data feature based on the calling probability of the next period, on-demand sampling is achieved. This ensures the real-time availability of high-frequency demand data while avoiding unnecessary sampling of low-probability data, reducing the overall system load and communication overhead. Simultaneously, the data cache depth is adjusted based on the steady-state probability distribution of each data node in the state transition matrix. This ensures that data with high access probability has sufficient buffer space to prevent loss, allowing for larger cache space allocation for data features with high steady-state probability. It also avoids allocating excessive redundant cache for data with very low access, thus optimizing memory resource utilization efficiency. The shortest prediction path length refers to the distance to another data scheduling state node in the constructed Markov data scheduling state transition matrix, starting from a specific key scheduling node (data feature) and following the most likely call sequence described by the state transition probability, with the minimum number of prediction steps.

[0032] Figure 6 A block diagram of a vehicle intelligent gear control system based on multi-state collaborative judgment according to the present invention is shown.

[0033] A second aspect of the present invention also provides a vehicle intelligent gear shift control system based on multi-state collaborative judgment. The system includes: a memory 601, a processor 602, and a communication interface 603. The memory includes a vehicle intelligent gear shift control method program based on multi-state collaborative judgment. The communication interface is used for data connection and communication between the memory and the processor. When the processor executes the vehicle intelligent gear shift control method program based on multi-state collaborative judgment, it performs the following steps: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

[0034] According to an embodiment of the present invention, it further includes: The frequency of physical gear switching between forward and reverse gears per unit time is identified based on the operation signal of the physical gear and the switching record of the logical gear, and this frequency is defined as the shift frequency index. At the same time, the average opening level of the brake pedal during the continuous switching process is calculated based on the brake pedal opening signal. The shift frequency index is compared with a first preset threshold, and the average opening level of the brake pedal is compared with a second preset threshold. When the shift frequency index exceeds the first preset threshold and the average opening level exceeds the second preset threshold, the vehicle is determined to have entered a high-frequency parking maneuver. Under the high-frequency shifting condition, a behavior prediction model based on the brake pedal opening and motor speed of the most recent shift cycles is constructed based on the current multi-source state data. The behavior prediction model predicts the vehicle state time window after the next shift request is issued by analyzing the delay time characteristics from when the brake pedal opening drops below the allowable threshold to when the motor speed rises above the limit threshold during historical shifts. Based on the vehicle state time window, the data sampling scheduling priority and judgment timing of the brake pedal opening and drive motor speed are dynamically adjusted. Based on the output of the behavior prediction model, when a physical gear shifting request signal is received, the vehicle state data within the prediction time window is used first for collaborative judgment. If the prediction data indicates that the driver's operation meets the safe gear shifting conditions, the logical gear shifting is executed immediately and the generation of conventional safety prompt information is suppressed. At the same time, the system uses the time series data of the actual collected brake opening, motor speed and physical gear operation as incremental learning samples to update the behavior prediction model, so that the model can adapt to the operating habits of different drivers under this working condition. When the shift frequency index is lower than the first preset threshold for a predetermined time, or when the average brake pedal opening level is lower than the second preset threshold, the high-frequency parking maneuvering condition is determined to have ended. The system then closes the fast collaborative judgment mode and reverts to the standard collaborative judgment and prompting logic, thus completing the control optimization for the frequent parking scenario.

[0035] It's important to note that in scenarios involving frequent back-and-forth maneuvering, such as reversing into a parking space or parallel parking, skilled drivers operate at an extremely fast pace. Existing intelligent gear control methods, due to the need for strict, sequential verification of multiple safety conditions such as brake opening and motor speed, introduce unavoidable delays in their fixed-sequence data acquisition, transmission, and decision-making processes. This results in a "half-beat slower" effect between gear shift response and the driver's rapid operation. Furthermore, even if the driver's operation is compliant, the system still mechanically triggers prompts such as "Please apply the brakes" during each verification process, causing frequent flashing and beeping of these prompts within a short period. This creates severe visual and auditory interference, significantly impacting the driving experience and potentially distracting the driver's already heightened attention in confined spaces, increasing the risk of misoperation. By recognizing typical parking maneuvering patterns involving high-frequency, deep braking in real time, the system can automatically switch to a rapid collaborative judgment mode, proactively adapting to the driver's rapid operational rhythm in this scenario. In this mode, the system builds a behavior prediction model based on historical operation data, proactively judging the trend of vehicle status changes. This allows for advance authorization of gear shifts based on high-confidence predictions before the data collection and processing process is complete. This effectively eliminates the response delay caused by fixed-sequence verification chains, making gear shifting actions almost synchronous with the driver's operational intentions. Simultaneously, by intelligently suppressing unnecessary prompts generated during compliant operation, it completely avoids the interference caused by frequent prompts to the driver, ensuring focus during operation in confined spaces. The entire system can continuously adapt to the habits of different drivers through incremental learning, ultimately achieving instant gear shift response while ensuring that safety conditions (braking depth, vehicle speed) are substantially met.

[0036] This invention discloses a vehicle intelligent gear control method and system based on multi-state collaborative judgment. By constructing a multi-source state perception layer for the vehicle, it acquires multi-source state data in real time, including physical gear operation signals, vehicle Ready status, drive motor speed, brake pedal opening, and the current logical gear. Based on this, it determines the driver's physical gear operation state and the current state of the logical gear, and constructs a corresponding logical gear shifting strategy. Furthermore, by analyzing the scheduling frequency and path of this shifting strategy for multi-source state data, a Markov state transition matrix for scheduling each data feature is established. Finally, based on this transition matrix, the data acquisition strategy of the multi-source state perception layer is optimized and determined. This invention achieves intelligent gear decision-making and data scheduling optimization under multi-state collaborative judgment, improving the accuracy, real-time performance, and system efficiency of gear shift control.

[0037] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0038] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A vehicle intelligent gear control method based on multi-state collaborative judgment, characterized in that, Includes the following steps: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

2. The vehicle intelligent gear control method based on multi-state collaborative judgment according to claim 1, characterized in that, The construction of a multi-source state perception layer for the target vehicle involves acquiring multi-source state data of the target vehicle based on this layer. This multi-source state data includes the vehicle's physical gear shift signal, vehicle Ready status, drive motor speed, brake pedal opening, and current logical gear data. Specifically: The data features required for the gear control of the target vehicle are obtained, including physical gear operation features, vehicle Ready status features, drive motor speed features, brake pedal opening features, and current logical gear features. In the vehicle control system, a multi-source state perception layer for the target vehicle to perceive the data features is constructed. The data features are pre-acquired based on the multi-source state perception layer, and the data acquisition features of each data feature are recorded. The data acquisition features include the debouncing time constant of the digital input port corresponding to the physical gear operation signal, the sampling rate and resolution of the analog signal corresponding to the brake pedal opening, the pulse capture timer counting period corresponding to the drive motor speed, and the identifier and periodic transmission interval of the CAN message containing the vehicle's Ready status and the current logical gear. Based on the data acquisition characteristics, determine the maximum update frequency of each data feature and the task scheduling cycle in the vehicle control system, and based on the maximum update frequency and the task scheduling cycle, determine the maximum allowable data acquisition delay time for each data feature. The data acquisition thread priority and data cache depth of the multi-source state perception layer for acquiring each data feature are determined based on the maximum allowable data acquisition delay time. The target vehicle's multi-source status data is obtained based on the data acquisition thread priority and data cache depth. The multi-source status data includes the vehicle's physical gear operation signal, vehicle Ready status, drive motor speed, brake pedal opening, and current logical gear data.

3. The vehicle intelligent gear control method based on multi-state collaborative judgment according to claim 1, characterized in that, The step of determining the driver's operational state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing a shifting strategy for the logical gear based on the operational state of the physical gear and the current state of the logical gear, specifically involves: Based on the multi-source state data, the driver's operation status of the physical gear and the current status of the logical gear are determined. When the vehicle is in a non-Ready state, the logical gear is forcibly set to N gear. When the vehicle is in Ready state and the current logical gear is N, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds a preset threshold. If it does not exceed the threshold, it keeps the logical gear in N and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it further checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D. Otherwise, it keeps the vehicle in N and prompts the user to stop and shift gears. When the vehicle is in Ready state and the current logical gear is D, if the physical gear is switched to R, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in D and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to R; otherwise, it switches to N and prompts the user to stop and shift gears. If a rising edge signal is generated when the physical gear is in the D position, the logical gear will be switched to S position.

4. The vehicle intelligent gear control method based on multi-state collaborative judgment according to claim 3, characterized in that, The step of determining the driver's operational state of the physical gear and the current state of the logical gear based on the multi-source state data, and constructing a shifting strategy for the logical gear based on the operational state of the physical gear and the current state of the logical gear, further includes: When the vehicle is in Ready state and the current logical gear is S gear, if the physical gear is switched to R gear, it first checks whether the brake pedal opening exceeds a preset threshold. If it does not exceed the threshold, the logical gear remains in S gear and the instrument prompts the user to press the brake to shift gears. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, the logical gear is switched to R gear; otherwise, it is switched to N gear and the user is prompted to stop and shift gears. If the physical gear generates a D gear position switching signal, the logical gear is switched back to D gear. When the vehicle is in Ready state and the current logical gear is R, if the physical gear is switched to D, it first checks whether the brake pedal opening exceeds the preset threshold. If it does not exceed the threshold, it keeps the logical gear in R and prompts the user to press the brake to shift gears via the instrument panel. If it exceeds the threshold, it checks whether the drive motor speed is lower than the shift allowable speed threshold. If it is lower than the shift allowable speed threshold, it switches the logical gear to D; otherwise, it switches to N and prompts the user to stop and shift gears. For the operation of switching the physical gear to N gear, the logical gear is directly set to N gear in any current logical gear state, which constitutes the shifting strategy of the target vehicle's logical gear.

5. The vehicle intelligent gear control method based on multi-state collaborative judgment according to claim 1, characterized in that, The process of obtaining the target vehicle's shifting strategy's scheduling frequency and path for multi-source state data, and constructing a Markov data scheduling state transition matrix for each data feature of the multi-source state data based on the scheduling frequency and path, specifically involves: Based on the historical records of the target vehicle executing the shifting strategy within a preset time period, the timestamp sequence of the multi-source state data being called and the corresponding data identifiers are extracted during each shifting decision process. The number of times each data feature is accessed per unit time is counted to obtain the scheduling frequency. Based on the order in which data is called during the gear shifting decision process, a dependency graph between multi-source state data is constructed to obtain the scheduling path; Each data feature of the multi-source state data is used as a data scheduling state node. The scheduling transfer direction between data scheduling state nodes is determined according to the scheduling path. The number of times a data scheduling state node is transferred from one data scheduling state node to another within a continuous sampling period is calculated based on the scheduling frequency. The total number of times each state node transfers to other state nodes is counted. The state transition probability is obtained by dividing the number of times a data scheduling state node transitions from one data scheduling state node to another by the total number of transitions of that data scheduling state node. The scheduling transition probabilities between all state nodes are arranged in rows and columns to construct the Markov data scheduling state transition matrix.

6. The vehicle intelligent gear control method based on multi-state collaborative judgment according to claim 1, characterized in that, The step of determining the data acquisition strategy for multi-source state data by the multi-source state perception layer based on the Markov data scheduling state transition matrix is ​​as follows: Based on the Markov data scheduling state transition matrix, the probability of each data scheduling state node being invoked in the next sampling period and the highest probability state transition path are identified. Based on the highest probability state transition path, determine the set of core data features with temporal dependencies in the gear shifting decision process from the multi-source state data; Using the data feature with the highest scheduling frequency in the core data feature set as the key scheduling node, calculate the shortest predicted path length from the key scheduling node to other data scheduling state nodes in the Markov data scheduling state transition matrix. The priority of the data acquisition threads for the corresponding data features in the multi-source state perception layer is adjusted according to the shortest predicted path length, and the data acquisition frequency of each data feature of each multi-source state data in the next sampling period is determined according to the probability that each data scheduling state node is called in the next sampling period. Adjust the corresponding data cache depth according to the steady-state probability distribution of each data scheduling state node in the state transition matrix; Based on the adjusted data acquisition thread priority, data cache depth configuration, and data acquisition frequency of each data feature in the next sampling period, the multi-source state perception layer is driven to acquire and cache multi-source state data, forming a data acquisition strategy.

7. A vehicle intelligent gear control system based on multi-state collaborative judgment, characterized in that, The vehicle intelligent gear shift control system based on multi-state collaborative judgment includes a storage device and a processor. The storage device includes a vehicle intelligent gear shift control method program based on multi-state collaborative judgment. When the processor executes the vehicle intelligent gear shift control method program based on multi-state collaborative judgment, it performs the following steps: A multi-source state perception layer for the target vehicle is constructed, and multi-source state data of the target vehicle is obtained based on the multi-source state perception layer. The multi-source state data includes the vehicle's physical gear operation signal, the vehicle's Ready state, the drive motor speed, the brake pedal opening, and the current logical gear data. The driver's operation status of the physical gear and the current status of the logical gear are determined based on the multi-source state data, and a shifting strategy for the logical gear is constructed based on the operation status of the physical gear and the current status of the logical gear. Obtain the scheduling frequency and scheduling path of the target vehicle's shifting strategy for multi-source state data, and construct a Markov data scheduling state transition matrix for scheduling each data feature of the multi-source state data based on the scheduling frequency and scheduling path; The data acquisition strategy for multi-source state data by the multi-source state perception layer is determined based on the Markov data scheduling state transition matrix.

Images