All-terrain adaptive chassis control methods, systems and computer program products

Abstract

This application relates to the field of vehicle control technology, specifically to an all-terrain adaptive chassis control method, system, and computer program product. The method includes: identifying the current driving terrain and determining the dominant terrain mode; calculating multiple state indices characterizing the current driving condition based on the vehicle's dynamic state; generating multiple control priority indicators based on the dominant terrain mode, multiple state indices, and driver's operational intentions; querying a pre-calibrated strategy parameter mapping table based on the multiple control priority indicators and the dominant terrain mode to generate target instruction sets for each chassis actuator; and distributing the target instruction sets to each chassis actuator for coordinated execution. This method can dynamically balance passability, stability, and comfort under complex and varied terrain conditions, comprehensively improving the overall vehicle performance and driving safety.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, specifically to an all-terrain adaptive chassis control method, system, and computer program product. Background Technology

[0002] With the development of the automotive industry, the driving environment for vehicles has expanded from flat paved roads to a variety of complex road conditions. When driving on snow, mud, sand, and unpaved roads with significant undulations, the vehicle's suspension characteristics, braking distribution, power output, and body posture all need to be matched with the road conditions to maintain an acceptable ride and sufficient off-road capability while ensuring driving safety. However, the requirements for chassis parameters in different terrains are often contradictory. For example, soft roads usually require lower suspension damping to maintain tire contact with the ground, while high-speed cornering requires higher damping and stiffness to suppress body roll.

[0003] Currently, most mass-produced vehicles employ a limited number of preset driving modes in their chassis systems, which drivers manually switch between based on experience and road conditions. This approach relies on the driver's subjective judgment and manual operation, making it difficult to respond instantly to rapidly changing road conditions. Furthermore, the various subsystems within traditional chassis systems are typically managed by independent electronic control units. There is a lack of effective information exchange and joint decision-making mechanisms between subsystems such as suspension, braking, steering, and powertrain. Each subsystem can only adjust parameters within its own local scope, making it difficult to achieve comprehensive performance optimization at the vehicle level. In addition, traditional systems rely on a limited number of vehicle status sensors, lacking sufficient means of sensing the external road environment, and the adjustment of control strategies lacks adequate environmental information support.

[0004] Therefore, how to enable the chassis system to automatically and timely sense changes in the external driving environment, and coordinate and control multiple chassis systems based on real-time multi-dimensional vehicle status information, so as to dynamically balance the mutually restrictive performance requirements such as passability, stability and comfort under different driving conditions, is a technical problem that urgently needs to be solved in the field of chassis control. Summary of the Invention

[0005] In view of this, embodiments of this application provide an all-terrain adaptive chassis control method, system, and computer program product to dynamically balance passability, stability, and comfort under complex and varied terrain conditions, thereby comprehensively improving the overall performance and driving safety of the vehicle.

[0006] The first aspect of this application provides an all-terrain adaptive chassis control method, including: Identify the current driving terrain and determine the dominant terrain mode; Based on the vehicle's dynamic state, calculate multiple state indices that characterize the current driving conditions; Based on the dominant terrain pattern, the multiple state indices, and the driver's operational intent, multiple control priority indicators are generated; Based on the multiple control priority indicators and the dominant terrain pattern, a pre-calibrated strategy parameter mapping table is queried to generate the target instruction set for each actuator of the chassis; The target instruction set is distributed to each actuator in the chassis for coordinated execution.

[0007] A second aspect of this application provides an all-terrain adaptive chassis control system, comprising: The global perception module is used to collect and identify the current driving terrain and obtain the relevant data required for the vehicle's dynamic status. A central decision control unit, configured to perform the method as described in the first aspect of the embodiments of the present invention; The execution module is used to receive the target instruction set and execute it collaboratively.

[0008] A third aspect of this application provides a computer program product including a computer program that, when run, causes the method described in the first aspect of this application to be executed.

[0009] The first aspect of this application provides an all-terrain adaptive chassis control method that identifies the current driving terrain and determines the dominant terrain mode. It calculates multiple state indices based on the vehicle's dynamic state, generates control priority indicators by combining these with the driver's operational intentions, and then queries a pre-calibrated strategy parameter mapping table to generate target instruction sets for each chassis actuator and executes them collaboratively. This constructs a complete intelligent closed-loop control link from perception to decision-making to execution. This scheme achieves accurate perception of the driving environment through multi-source sensor data fusion, achieves a dynamic balance between passability, stability, and comfort through fuzzy logic reasoning, and realizes deep collaborative control of various chassis subsystems through a pre-calibrated strategy parameter mapping table, thereby comprehensively improving the vehicle's overall performance and driving safety under complex terrain conditions.

[0010] It is understood that the beneficial effects of the second and third aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0012] Figure 1This is a schematic flowchart of an embodiment of the all-terrain adaptive chassis control method provided in this application; Figure 2 This is a schematic flowchart of an all-terrain adaptive chassis control method provided in another embodiment of this application; Figure 3 This is a schematic diagram of the structure of an all-terrain adaptive chassis control system provided in an embodiment of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0015] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0016] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0017] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0019] like Figure 1 As shown, the all-terrain adaptive chassis control method provided in this application includes the following steps S101 to S105 executed by the central decision control unit: Step S101: Identify the current driving terrain and determine the dominant terrain mode.

[0020] In application, the central decision control unit receives multi-source sensor data from the global perception module, classifies and identifies the current driving surface using a pre-trained terrain recognition model, determines the category of the current driving terrain, and further maps this category to a dominant terrain mode for control decision-making. The dominant terrain mode is an abstract description of the physical characteristics of the current driving terrain, used to guide subsequent control strategy selection. Different dominant terrain modes correspond to different chassis control strategy preferences; for example, the highway mode emphasizes comfort and high-speed stability, the snow mode emphasizes traction control and stability under low-traction conditions, and the off-road mode emphasizes passability and high ground clearance. By converting the terrain recognition results into a limited number of control modes, the complexity of subsequent decision-making and control can be effectively reduced, while retaining the ability to respond specifically to different terrains.

[0021] Step S102: Calculate multiple state indices representing the current driving conditions based on the vehicle's dynamic state.

[0022] In application, the central decision control unit calculates multiple state indices in real time based on vehicle dynamic state data. These indices quantify the vehicle's current driving conditions from different dimensions. The vehicle dynamic state data includes, but is not limited to, information such as vehicle speed, acceleration, yaw rate, body posture angle, and the working status of each wheel. The multiple state indices are designed to comprehensively evaluate the vehicle's current state from multiple performance dimensions, such as stability, comfort, and passability, providing a quantitative basis for subsequent control decisions. The thresholds and parameters used in calculating these state indices can be dynamically adjusted according to the current dominant terrain pattern to adapt to the differentiated requirements of various performance indicators under different terrain conditions.

[0023] Step S103: Generate multiple control priority indicators based on the dominant terrain pattern, the multiple state indices, and the driver's operating intention.

[0024] In application, the central decision control unit takes the dominant terrain pattern, multiple state indices, and the driver's operational intentions as inputs. Through a built-in decision reasoning mechanism, it comprehensively analyzes the urgency of various performance requirements under the current operating conditions and outputs multiple control priority indicators. Driver operational intentions refer to the driving intentions reflected in the driver's pedal and steering wheel operations, such as rapid acceleration, emergency braking, or sharp steering. Control priority indicators quantify the proportion of control resources that should be allocated to passability, stability, and comfort under the current operating conditions. When the vehicle is on a low-traction surface and a decreasing stability trend is detected, stability-related priority indicators will receive higher values; when the vehicle is traveling on a large, undulating unpaved surface with insufficient ground clearance, passability-related priority indicators will be increased. Through this dynamic priority allocation mechanism, the system can find a suitable balance point among multiple mutually constraining performance objectives for the current operating conditions.

[0025] Step S104: Based on the multiple control priority indicators and the dominant terrain mode, query the pre-calibrated strategy parameter mapping table to generate the target instruction set for each actuator of the chassis.

[0026] In application, the central decision control unit uses multiple control priority indicators and the current dominant terrain mode as indexes to look up the corresponding actuator target parameters in a pre-defined strategy parameter mapping table, generating a complete target instruction set. The strategy parameter mapping table is a multi-dimensional lookup table pre-determined during the system development phase through extensive real-vehicle testing and simulation optimization. This table records the target operating parameters that each actuator should achieve under different priority combinations and terrain modes. The target instruction set contains the specific operating target values ​​of all controlled actuators in the chassis, such as suspension height target value, damping coefficient target value, stabilizer bar stiffness target value, electronic stability system operating mode, transmission shift strategy, and front-to-rear axle torque distribution ratio. Generating the instruction set through table lookup ensures both response speed and consistency between the targets of each actuator.

[0027] Step S105: Distribute the target instruction set to each actuator in the chassis for coordinated execution.

[0028] In application, the central decision control unit simultaneously distributes the target values ​​from the target instruction set to the corresponding actuator control units via the vehicle communication bus. Each actuator adjusts its parameters accordingly based on the received target values. Multiple actuators receive and execute their respective target values ​​within the same control cycle, achieving coordinated response of the chassis system. This coordinated execution method breaks the traditional situation where each subsystem in the chassis system operates independently, enabling the suspension, braking, and power subsystems to be adjusted as a whole, fully leveraging the performance advantages of the system level.

[0029] The above method achieves fully automatic adaptive adjustment of the chassis system to complex terrain environments by constructing a complete control link from terrain perception to state assessment, from decision reasoning to collaborative execution. It can automatically optimize the working parameters of each subsystem of the chassis according to the real-time changes in driving terrain and vehicle status without the need for manual driver intervention, effectively improving the overall driving performance and active safety of the vehicle under various complex terrain conditions.

[0030] In one embodiment, prior to identifying the current driving terrain, the method further includes: The system collects at least one of the following data: visual image data of the road ahead, obstacle distance data, and point cloud data through environmental perception sensors; collects at least one of the following data: three-axis angular velocity, three-axis acceleration, wheel speed, steering angle, and positioning information through vehicle status sensors; and collects suspension height and / or travel data of each wheel through chassis interaction sensors. The collected multi-source sensor data are fused to generate comprehensive vehicle-environment situational information in a unified spatiotemporal coordinate system; The identification of the current driving terrain and the calculation of multiple state indices representing the current driving conditions are both performed based on the vehicle-environment integrated situation information.

[0031] In the application, the global perception module includes three types of sensor units. The environmental perception unit includes a forward-looking multi-function camera, millimeter-wave radar, and lidar, used to acquire visual images of the road ahead, obstacle distances, and contour information, thereby identifying geometric and textural features such as road surface type, pothole depth, and slope. The vehicle status perception unit includes a high-precision inertial measurement unit (IMU), wheel speed sensors, steering angle sensors, and a GPS receiver. The IMU measures the vehicle's three-axis angular velocity and three-axis acceleration, the wheel speed sensors detect the rotational speed of each wheel, the steering angle sensors detect the driver's steering input, and the GPS receiver provides the vehicle's positioning information. The chassis interaction perception unit includes suspension height sensors and travel sensors installed at each wheel location, used to acquire real-time suspension height and travel data for each wheel.

[0032] The aforementioned multi-source sensor data have different acquisition frequencies, data formats, and coordinate systems. The central decision control unit uses the Extended Kalman Filter (EKF) algorithm to perform timestamp synchronization, coordinate system unification, and noise filtering on these heterogeneous data, ultimately generating a unified vehicle-environment integrated situational information system in a unified spatiotemporal coordinate system. This integrated vehicle-environment situational information system is a high-confidence dataset containing the type and degree of terrain undulation ahead, the current vehicle attitude (roll angle, pitch angle, height), estimated values ​​of the adhesion coefficients of each wheel, and the vehicle's dynamic state (speed, acceleration, yaw rate, etc.). The data acquisition and fusion frequency is no less than 100Hz to ensure that the system can respond promptly to changes in road conditions.

[0033] By fusing multi-source sensor data into unified comprehensive situational information, subsequent terrain identification and state assessment are based on the same high-quality information source. This avoids decision-making contradictions that may result from different modules using asynchronous or inconsistent data sources, and improves the reliability and robustness of perception results through the complementarity of multi-sensor information.

[0034] In one embodiment, the comprehensive situational information also includes estimated values ​​of the road surface adhesion coefficient for each wheel; The estimated values ​​of the road surface adhesion coefficient for each wheel are obtained by fusing data from the inertial measurement unit and wheel speed sensor.

[0035] In applications, the road adhesion coefficient is a key parameter characterizing the frictional properties between the tire and the road surface, directly affecting the vehicle's braking distance, traction limit, and lateral stability. Since the road adhesion coefficient cannot be directly measured by a single sensor, the system employs an online estimation method based on a vehicle dynamics model. This method fuses the vehicle acceleration information measured by the inertial measurement unit with the wheel speed information detected by the wheel speed sensors. By comparing the difference between the theoretical and actual speeds of each wheel, and combining this with the longitudinal and lateral acceleration information of the vehicle body, the system calculates the road adhesion coefficient at each wheel position in real time. For example, when a wheel exhibits significant slippage or drift, the estimated adhesion coefficient for that wheel will decrease significantly. Different road surface types have different characteristic values ​​for their adhesion coefficients. The adhesion coefficient for dry asphalt roads is typically between 0.7 and 0.9, for snow-covered roads it is approximately 0.2 to 0.4, while for ice it can be as low as 0.05 to 0.15.

[0036] By incorporating the road surface adhesion coefficient of each wheel into the comprehensive situational information, the system can obtain additional road surface physical characteristics information as a reference when performing terrain identification. At the same time, it can make more accurate judgments based on the actual road surface grip conditions when calculating the state index and determining control priorities.

[0037] In one embodiment, identifying the current driving terrain and determining the dominant terrain pattern includes: Based on the vehicle-environment integrated situational information, multimodal fusion features are obtained, and the multimodal fusion features are input into a pre-trained convolutional neural network model to obtain the probability distribution of the current terrain belonging to each predefined terrain category; The terrain category with the highest probability in the probability distribution is taken as the current basic terrain category, and the basic terrain category is converted into the current dominant terrain pattern through a preset terrain-pattern mapping table; The multimodal fusion features include at least two of the following: image features, vibration spectrum features, slope information, road surface adhesion coefficient estimate, vehicle body attitude parameters, and vehicle dynamic state parameters.

[0038] In application, the central decision control unit first extracts multimodal fusion features from the vehicle-environment integrated situational information. Image features are texture and color features extracted from road surface images captured by the forward-facing camera. Different road surface types have different visual texture features; for example, asphalt roads exhibit a uniform dark gray texture, snow-covered roads exhibit a uniform white texture, and gravel roads exhibit irregular granular textures. Vibration spectrum features are extracted after frequency domain analysis of vehicle vibration data measured by the inertial measurement unit (IMU). Different road surface roughness and materials will produce vibration responses of different frequencies and amplitudes on the vehicle body. Slope information is the current longitudinal and lateral slope of the road surface calculated through the fusion of IMU and GPS data. The road adhesion coefficient estimate is the adhesion coefficient of each wheel obtained from the aforementioned fusion calculation. Vehicle attitude parameters include the vehicle's current roll angle, pitch angle, and other attitude information. Vehicle dynamic state parameters include the vehicle's speed, acceleration, and yaw rate.

[0039] The features from at least two of the above modalities are fused and then input into a pre-trained deep learning convolutional neural network (CNN) model. A convolutional neural network is a type of deep learning model specifically designed to process data with spatial structure, automatically extracting hierarchical feature representations of the input data through convolutional layers. This CNN model is pre-trained on a large number of labeled terrain datasets, and its output is the probability distribution of the current terrain belonging to various predefined basic physical terrain categories. The predefined basic physical terrain categories include: dry asphalt, wet asphalt, compacted snow, ice, mud, gravel, and rugged rock.

[0040] The system selects the terrain category with the highest probability in the probability distribution as the current base terrain category. Then, it uses a built-in terrain-pattern mapping table to convert this base terrain category into the dominant terrain pattern used for control decisions. The mapping table's correspondence is shown below: The aforementioned mapping table can be expanded or refined according to actual needs. By introducing a two-layer decoupled architecture of basic terrain recognition and dominant driving mode mapping, the CNN model in the perception layer can focus on the refined classification of the physical world, while the decision control layer can focus on abstract control logic strategies. This architecture ensures the robustness of the system and facilitates the addition of new basic terrain categories or adjustment of mapping relationships through remote upgrades without modifying the underlying control logic, thus improving the scalability of the system.

[0041] In one embodiment, the plurality of state indices include a vehicle stability index, a ride comfort index, and a passability risk level; The vehicle stability index is used to characterize the current dynamic stability state of the vehicle and is calculated based on lateral acceleration and yaw rate. The ride comfort index is used to characterize the comfort level of the vehicle's vertical vibration and is calculated based on the vertical acceleration power spectral density. The passability risk level is used to characterize the passability risk of the current terrain, and is calculated based on the current actual ground clearance, the recommended minimum ground clearance under the current terrain type, the undulation of the terrain ahead, and the maximum allowable ground clearance.

[0042] In application, the three state indices mentioned above are used to quantitatively evaluate the current driving conditions from three dimensions: stability, comfort, and passability.

[0043] The Vehicle Stability Index (VSI) is a dimensionless index with a value ranging from [0,1]. A lower value indicates poorer dynamic stability. Lateral acceleration reflects the lateral load transfer of the vehicle under cornering or crosswind conditions, while yaw rate reflects the angular velocity of the vehicle's rotation around its vertical axis. Both together reflect the risk of the vehicle skidding or fishtailing. When both lateral acceleration and yaw rate are within threshold ranges, the VSI is close to 1, indicating that the vehicle is in a stable state. When either index significantly exceeds the threshold, the VSI will drop rapidly.

[0044] The Ride Comfort Index (RCI) is a dimensionless index with a value ranging from [0,1]. A lower value indicates poorer ride comfort. This index is based on the power spectral density of vertical acceleration. Vertical acceleration power spectral density is the energy distribution characteristic obtained after frequency domain analysis of the vehicle's vertical vibration signal, comprehensively reflecting the vibration energy level of the vehicle in different frequency ranges. The human body is most sensitive to vertical vibrations in the 1-12Hz frequency range; the greater the vibration energy in this frequency band, the greater the discomfort experienced by the occupants.

[0045] The Terrain Clearance Risk (TCR) is a dimensionless index ranging from [0,1]. A higher value indicates a greater risk to clearing the terrain. This index considers the difference between the vehicle's current ground clearance and the minimum ground clearance required for the current terrain, as well as the undulation of the terrain ahead. When the vehicle's actual ground clearance is lower than the recommended minimum ground clearance for the current terrain, or when there are significant undulations in the terrain ahead, the TCR will increase, indicating to the system that measures such as raising the vehicle's height are needed to ensure passability.

[0046] By simultaneously monitoring the status indices of the above three dimensions, the system can comprehensively grasp the current driving status of the vehicle, providing sufficient quantitative basis for subsequent priority decisions.

[0047] In one embodiment, the formula for calculating the Vehicle Stability Index (VSI) is: in, For lateral acceleration, The lateral acceleration threshold, The yaw rate is angular velocity. The yaw rate threshold; The formula for calculating the Ride Comfort Index (RCI) is as follows: in, PSD ( f ) represents the vertical acceleration power spectral density, and T represents the comfort tolerance threshold corresponding to the terrain type; The formula for calculating the passability risk level (TCR) is as follows: in, H req This is the recommended minimum ground clearance for the current terrain type. H current Δ represents the current actual ground clearance. h front This refers to the degree of terrain undulation ahead. H clearance This is the maximum permissible ground clearance.

[0048] In application, the VSI calculation formula intuitively reflects the degree to which the vehicle's lateral dynamics deviate from the stability limit by adding the ratio of lateral acceleration to its threshold and the ratio of yaw rate to its threshold and then subtracting 1. When the sum of the two ratios is close to or exceeds 1, the VSI is close to or below 0, indicating that the vehicle is in a highly unstable state or has exceeded the stability limit.

[0049] The RCI (Right Tolerance Capacity) is calculated by comparing the integral value of the vertical acceleration power spectral density with the comfort tolerance threshold T corresponding to the current terrain type to assess comfort. The comfort tolerance threshold T reflects the degree of tolerance to vibration levels under different terrain conditions. For example, on off-road terrain, occupants have a higher tolerance for vibration, resulting in a larger T value; while on paved roads, occupants are more sensitive to vibration, resulting in a correspondingly smaller T value.

[0050] The TCR calculation formula includes two components: the first reflects the degree of inadequacy of the current ground clearance relative to the recommended minimum ground clearance, and the second reflects the additional threat to passability posed by terrain undulations ahead. The recommended minimum ground clearance... H req The results are obtained by looking up a table based on the terrain identification results; different terrain types correspond to different... H req Values, such as those for rugged rocky terrain. H req The value is significantly higher than that of a smooth asphalt road surface. The terrain undulation ahead is Δ... h front This information is derived from camera and radar detection of the road surface ahead. Maximum permissible ground clearance. H clearance It is the maximum ground clearance that the vehicle's suspension system can provide, which is determined by the vehicle's hardware specifications.

[0051] In one embodiment, the thresholds used in calculating the vehicle stability index, ride comfort index, and passability risk level are dynamically adjusted by the dominant terrain model and / or obtained by querying a pre-calibrated terrain-threshold lookup table.

[0052] In applications, the dynamic characteristics of vehicles and driving expectations vary significantly under different terrain conditions. Therefore, the thresholds used in calculating each state index need to be dynamically adjusted according to changes in terrain patterns. The central decision control unit stores a terrain-threshold lookup table, which records the threshold parameters required for calculating each state index under each dominant terrain pattern.

[0053] Taking the Vehicle Stability Index (VSI) as an example, in highway mode, the lateral acceleration threshold... Yaw angular velocity threshold The threshold is set according to the standard value. In snow mode, due to the reduced road adhesion coefficient, the lateral force required for the vehicle to reach its stability limit is smaller. Therefore, the system lowers the threshold by about 20%, making the VSI more sensitive to changes in vehicle condition under snow conditions and able to trigger stability intervention in the early stages of vehicle condition deterioration. In ice mode, the threshold is further reduced by about 30% to meet the more stringent stability requirements under extremely low adhesion conditions.

[0054] Similarly, the comfort tolerance threshold T of the Ride Comfort Index (RCI) takes different values ​​in different terrain modes. In off-road mode and mud / gravel mode, the T value is set larger because occupants have higher psychological expectations of vibrations from unpaved roads; on snow and gravel roads, the system uses a wider frequency band for evaluation; in highway mode, the T value is set smaller to maintain a higher comfort standard.

[0055] Recommended minimum ground clearance based on the sexual risk level TCR H req The minimum ground clearance also varies depending on the terrain mode. For example, the minimum ground clearance required for rugged rocky terrain (off-road mode) is much higher than that for flat paved roads (highway mode). By looking up the corresponding value in the terrain-threshold lookup table, the system can accurately assess whether the current ground clearance meets the passability requirements.

[0056] This terrain-pattern-based dynamic threshold adjustment mechanism enables each state index to maintain reasonable sensitivity and judgment accuracy under different terrain conditions, avoiding the problems of oversensitivity or underresponse under certain terrain conditions that may result from using a fixed threshold.

[0057] In one embodiment, generating multiple control priority indicators based on the dominant terrain pattern, the multiple state indices, and the driver's operational intent includes: The dominant terrain pattern, the multiple state indices, and the driver's operating intention are used as input variables for the fuzzy logic controller, wherein the driver's operating intention includes at least one of the accelerator pedal opening, brake pedal opening, and steering wheel angle. Based on a preset fuzzy rule base, fuzzy reasoning is performed on the input variables to output a control priority index that includes passability priority, stability intervention intensity, and comfort priority.

[0058] In application, the fuzzy logic controller is an intelligent control method based on fuzzy set theory and fuzzy inference rules, suitable for handling complex decision-making problems with uncertainty and multivariate coupling characteristics. In this scheme, the input variables of the fuzzy logic controller include the dominant terrain pattern, vehicle stability index (VSI), ride comfort index (RCI), passability risk level (TCR), and driver's operating intention. The driver's operating intention is characterized by signals such as accelerator pedal opening, brake pedal opening, and steering wheel angle, reflecting the driver's current acceleration, deceleration, or steering needs.

[0059] The output variables of the fuzzy logic controller include three control priority indices, all of which are normalized values ​​ranging from [0,1]: passability priority P traction This indicates the priority of passability requirements under the current operating conditions; stability intervention intensity Pstability This indicates the required intensity of stability control intervention under the current operating conditions; comfort priority P comfort This indicates the priority of ride comfort requirements under the current operating conditions.

[0060] The fuzzy rule base contains a large number of pre-defined IF-THEN inference rules, which were developed by chassis tuning engineers based on their professional experience and test data. For example, if the terrain is snowy and the VSI is below 0.2 and the TCR is above 0.2, then the priority for passability is high, and the intervention for stability is strong; if the terrain is ice and the VSI is close to 0 and the driver has a moderate steering intention, then the intervention for stability is extremely strong, and the priority for comfort is low. The fuzzy rule base covers typical operating conditions under various combinations of terrain patterns and state indices, enabling the system to generate reasonable priority allocation schemes in various complex scenarios.

[0061] In one embodiment, the chassis actuators include at least one of an air suspension system, a continuously damped adjustable shock absorber, an active stabilizer bar system, a vehicle electronic stability system, a transmission control unit, and a powertrain controller; The strategy parameter mapping table records the mapping relationship between the target parameters of each actuator and multiple control priority indicators and the dominant terrain mode.

[0062] In application, the various actuators of the chassis constitute the physical execution layer of the system, each performing different functions: the air suspension system adjusts the vehicle height by controlling the inflation and deflation of the air springs to meet the ground clearance requirements under different terrain conditions; the continuously variable damping shock absorber (CDC shock absorber) changes the damping force of each shock absorber by adjusting the current of the solenoid valve, realizing real-time adjustment of the vehicle's vibration characteristics, and the shock absorber at each wheel position can be controlled independently; the active stabilizer bar system adjusts the equivalent stiffness of the stabilizer bar through motor drive to suppress the body roll motion when the vehicle is cornering; the electronic stability system (ESP / TCS) corrects the vehicle's unstable motion state by applying braking force to individual wheels and adjusting the engine output torque, including different working modes such as high intervention mode and traction control mode; the transmission control unit (TCU) optimizes the power transmission characteristics under different driving conditions by adjusting the shift points; and the powertrain controller improves the drive force distribution strategy by adjusting the torque distribution ratio between the front and rear axles.

[0063] The strategy parameter mapping table pre-defines the mapping relationships between the target parameters of each actuator, control priority indicators, and dominant terrain modes. It serves as the core interface between the decision-making and execution layers of the system. The calibration parameters in this mapping table are pre-determined through extensive real-vehicle testing and simulation optimization, stored in the non-volatile memory of the central control unit, and can be updated via remote upgrade technology. The specific form of its mapping rules and examples of calibration parameters are shown in the table below: The mapping table described above transforms abstract priority indicators into specific target parameters for each executor, achieving a seamless connection from the decision-making layer to the physical execution layer. Furthermore, since the target parameters for all executors are generated based on the same set of priority indicators, the coordination and consistency of actions among the executors are guaranteed.

[0064] In one embodiment, the mapping relationship is as follows: H target =H base +ΔH max ×P traction ; Among them, H target H is the target height for the air suspension. base As the reference height, ΔH max P represents the maximum elevation change corresponding to the current terrain pattern. traction Prioritize passability; S target =S max ×P stability And not less than S min ; Among them, S target S is the target stiffness value for the active stabilizer bar. max For maximum stiffness, S min For minimum stiffness, P stability The intensity of the stability intervention.

[0065] In application, the mapping formula for the target height of the air suspension is determined by the sum of the reference height and the maximum lift weighted by passability priority. Wherein, the reference height H... base This is the vehicle's design ground clearance under standard driving conditions, for example, 180mm. Maximum lift / relief ΔH max Based on the current dominant terrain pattern: ΔH in highway mode max The value is 30mm because paved roads have a smaller range of ground clearance requirements; in snow mode, ΔH max The thickness is 50mm to address the potential obstruction caused by snow cover on the vehicle's underside; in ice mode, ΔH max The thickness is 20mm. Since ice surfaces are typically flat, and lowering the center of gravity appropriately improves stability; in off-road mode, ΔH... max It is 50mm to provide maximum ground clearance to handle terrain undulations. When passability priority P traction At higher altitudes, the suspension target height is significantly increased to ensure the vehicle has sufficient ground clearance to traverse complex terrain.

[0066] The target stiffness value of the active stabilizer bar is determined by multiplying the maximum stiffness by the stability intervention strength, while a minimum stiffness lower limit is set to ensure basic anti-roll capability. For example, when S... max 1000 Nm / rad, S min When the stability intervention intensity P is 200 Nm / rad, stability If it is 0.9, then S target =1000×0.9=900Nm / rad, providing strong anti-rolling moment; if P stability If S is 0.1, then S target =1000×0.1=100Nm / rad, but because it is lower than S min =200Nm / rad, the actual value is 200Nm / rad, retaining the basic roll suppression capability.

[0067] In one embodiment, the mapping relationship is as follows: When the dominant terrain mode is snow or off-road and the passability priority is greater than the first preset threshold and the comfort priority is less than the second preset threshold, the damping target value is set to the minimum damping value. Otherwise, the target damping value is calculated based on the basic damping coefficient and the stability correction coefficient; The basic damping coefficient is obtained by linear interpolation between the minimum and maximum damping values ​​based on comfort priority, and the stability correction coefficient is calculated based on the correction gain and stability intervention intensity corresponding to the terrain pattern.

[0068] In application, the damping mapping rules of continuously damped adjustable shock absorbers are quite complex because the damping characteristics directly affect the vehicle's comfort, stability, and tire grip, requiring fine control by considering multiple factors.

[0069] The damping mapping rule includes a mandatory condition check: when the dominant terrain mode is snow or off-road, and the passability priority P... traction When the damping target value is greater than the first preset threshold (e.g., 0.8) and the comfort priority is less than the second preset threshold (e.g., 0.4), the system will forcibly set the damping target value to the minimum damping value. (For example, 30%). The engineering meaning of this mandatory condition is that on soft or undulating surfaces such as snow or off-road terrain, when the system determines that the need for passability is extremely high and the need for comfort is low, using a fully soft damping setting can maximize the contact between each wheel and the road surface, improve tire grip, and thus improve traction output and braking performance.

[0070] When the above mandatory conditions are not met, the target damping value is calculated in the following two steps. The first step is to calculate the minimum damping value based on comfort priority. and maximum damping value Linear interpolation is performed between them to obtain the base damping coefficient. : when The higher the level (the stronger the comfort requirement), the closer the base damping coefficient is to... (A softer damping setting is beneficial for absorbing road vibrations); when The lower the value, the closer the base damping coefficient is to... (A stiffer damping setting is beneficial for vehicle body attitude control).

[0071] The second step is based on the stability intervention intensity. Calculate the stability correction coefficient : in, This is the correction gain coefficient corresponding to the terrain mode; different values ​​are used for different terrain modes: Highway mode =1.0, Snow Mode =2.0, Ice Surface Mode =3.0. The larger the value, the more significant the effect of stability intervention intensity on damping correction. (Ice surface mode) The value is the highest because, under extremely low adhesion conditions, the damping characteristics are more sensitive to the impact on vehicle stability, requiring a stronger correction force.

[0072] Final damping target value for: The clip function restricts the calculation result to and Within the specified range, prevent exceeding the physical adjustment range of the shock absorber.

[0073] In practice, the shock absorbers of each wheel can be controlled independently. For example, when driving in a curve, the damping of the outer wheel can be further increased based on the above calculations, while the damping of the inner wheel can be appropriately reduced to better suppress body roll.

[0074] In one embodiment, the mapping relationship is as follows: The operating mode of the vehicle electronic stability system is determined based on the comparison results of the stability intervention intensity and the passability priority thresholds. When the stability intervention intensity exceeds the preset intervention threshold, it enters the high intervention mode; when the passability priority exceeds the preset escape threshold, it enters the escape mode. The upshift delay coefficient and downshift advance coefficient of the transmission control unit are determined according to the passability priority and stability intervention intensity, respectively. The final shift point is calculated based on the shift point of the basic shift map and the corresponding coefficient. The front and rear axle torque distribution ratio of the powertrain controller is determined based on the torque adjustment coefficient corresponding to the passability priority and terrain mode.

[0075] In application, the Electronic Stability Program (ESP / TCS) employs a discrete mode selection strategy: when the stability intervention intensity P... stability When the threshold for intervention is exceeded (e.g., 0.8), the system enters a high-intervention mode. In this mode, ESP will more actively intervene in the vehicle's braking and torque distribution, applying greater braking force to wheels showing signs of instability and strictly limiting engine torque output to maximize vehicle stability. When the passability priority P... traction When the preset traction threshold (e.g., 0.8) is exceeded, the system enters traction control mode. In this mode, TCS allows the drive wheels to slip to a certain extent, preventing premature traction intervention that would limit the output of driving force and helping the vehicle gain greater traction on soft or slippery surfaces. When neither condition is met, the system maintains standard terrain mode operation.

[0076] The shift logic of the transmission control unit is based on the upshift delay factor. Downshift advance coefficient Adjustments are needed. The formula for calculating the upshift delay coefficient is as follows: =1+γ×P traction Where γ is the upshift adjustment coefficient corresponding to the terrain mode, for example, γ=0.5 for snow mode and γ=0.2 for ice mode. When passability is a higher priority, the upshift delay coefficient is larger, causing the transmission to delay upshifting and maintain a lower gear to provide greater driving force output. The formula for calculating the downshift advance coefficient is: ,in This refers to the downshift adjustment factor corresponding to the terrain mode, such as the snow mode. =0.8, Ice Surface Mode =1.2. When the stability intervention intensity is high, the downshift advance coefficient is large, causing the transmission to downshift earlier and utilizing engine braking effect to enhance the vehicle's deceleration stability. The final shift point is obtained by multiplying the shift point of the basic shift diagram by the corresponding coefficient.

[0077] The powertrain controller distributes torque between the front and rear axles based on off-road capability priority. Adjustments are made, and the calculation formula is as follows: , .in, This refers to the torque adjustment coefficient corresponding to the terrain mode, such as snow mode. =20%, Ice Surface Mode =10%. When When the torque distribution is greater than 0.5, the system favors rear-wheel drive, enhancing the driving force output of the rear wheels to improve passability; the torque distribution ratio is limited to the range of 40:60 to 60:40 to prevent extreme torque distribution from causing unbalanced driving force.

[0078] In one embodiment, the all-terrain adaptive chassis control method further includes closed-loop adjustment of the target instruction set based on execution feedback, specifically including: Collect actual status feedback signals of each actuator, wherein the actual status feedback signals include at least one of actual suspension height, actual body roll angle and actual wheel speed of each wheel; The actual state feedback signal is compared with the target value in the target instruction set to obtain the execution deviation. The PID controller generates compensation instructions based on the execution deviation, and the target values ​​of each actuator are fine-tuned in real time.

[0079] In applications, due to factors such as the response delay of the mechanical system, external disturbances, and the accuracy limitations of the actuators themselves, the actual output state of each actuator may deviate from the target value issued by the central decision control unit. To eliminate this deviation and improve control accuracy, the system introduces a closed-loop feedback regulation mechanism based on a PID controller at the command execution layer.

[0080] Among them, the PID controller (proportional-integral-derivative controller) is a classic feedback control algorithm. It achieves precise control of the controlled object by using the proportional term to respond instantly to the current deviation, the integral term to eliminate steady-state error, and the derivative term to predict the trend of deviation change. In this scheme, each controlled actuator is equipped with a corresponding PID controller, which independently compensates for its own execution deviation.

[0081] Taking air suspension height control as an example, the height sensor collects the actual height of the suspension in real time and compares it with the target height. If the actual height is lower than the target height, the PID controller outputs a positive compensation signal to drive the solenoid valve to increase the air intake of the air spring; if the actual height is higher than the target height, the PID controller outputs a negative compensation signal to drive the solenoid valve to exhaust air. Similarly, for the control of the vehicle body roll angle, the IMU monitors the actual roll angle of the vehicle body in real time. When the actual roll angle exceeds the target value, the PID controller outputs a compensation command to increase the stiffness of the stabilizer bar or increase the damping of the outer shock absorber.

[0082] This closed-loop feedback mechanism enables the system to maintain the accuracy and consistency of the actuator output under various disturbance conditions. For example, in an ice curve, if the IMU detects that the actual roll angle reaches 2.5° while the target value should be less than 2°, the PID controller will output an additional signal to enhance the stiffness of the stabilizer bar and increase the damping of the outer shock absorber until the roll angle converges to the target range.

[0083] In one embodiment, the all-terrain adaptive chassis control method further includes: Sensor data, decision commands, and execution results are anonymized and then uploaded to the cloud. Based on cloud-based big data analysis, the calibration parameters of the terrain recognition model, control rule base, and the strategy parameter mapping table are optimized and updated; The optimized and updated model, rule base, and calibration parameters are pushed to the vehicle via remote upgrade.

[0084] In application, the system continuously records multi-source sensor data, decision commands generated by the central decision control unit, and the actual execution results of each actuator during daily operation. This data, after being anonymized, is uploaded to a cloud-based big data platform via the vehicle network. Anonymization refers to removing sensitive information that could identify a specific vehicle and driver before uploading, retaining only technical data related to vehicle driving status and chassis control to protect user privacy.

[0085] The cloud-based big data platform aggregates a large amount of vehicle operation data and continuously optimizes existing control strategies through data analysis and machine learning algorithms. For example, by comparing the differences between the actual control response and the ideal control model, it can be found that the VSI calculation threshold setting is not reasonable under certain terrain conditions, or that the inference result of a certain fuzzy rule is not accurate enough in a specific scenario. Based on this, the platform can optimize the parameters of the CNN terrain recognition model, adjust the rule weights in the fuzzy rule base, and update the calibration parameters in the strategy parameter mapping table.

[0086] The optimized and updated model, rule base, and calibration parameters are pushed to the vehicle via OTA (Over-The-Air) remote upgrade. The central decision control unit automatically loads the updated content after downloading and verifying it. This continuous optimization and self-evolution mechanism enables the system's control strategy to continuously improve with the accumulation of operational data, demonstrating stronger adaptability when facing unforeseen new terrain conditions or boundary conditions.

[0087] The following describes the complete working process of the all-terrain adaptive chassis control method provided in this embodiment of the invention, using a specific driving scenario as an example.

[0088] In one embodiment, the vehicle initially travels in a straight line at 100 km / h on a dry highway, then enters a mountain road covered with compacted snow. The road surface adhesion coefficient drops sharply, and the road begins to curve continuously. Upon entering the snow-covered section, a sharp bend covered with a thin layer of ice appears ahead, and the driver begins to turn the steering wheel and apply moderate braking. Figure 2 As shown, it includes the following steps: Step S201: Multi-source data acquisition and fusion.

[0089] The forward-facing camera captures an image of the road ahead, initially showing clear, dry asphalt texture, then transitioning to a uniform white snow texture, followed by reflective areas indicating ice. Millimeter-wave radar detects the undulations in the road surface. LiDAR generates a point cloud, identifying a snow thickness of approximately 5cm and smooth ice areas. The IMU measures the vehicle's three-axis acceleration and angular velocity in real time. Wheel speed sensors detect slight slippage in the drive wheels. GPS, combined with map data, confirms the vehicle has entered a mountainous area.

[0090] EKF fusion algorithm output: Road surface types within 20m ahead of the current location are, in order: dry asphalt (0-50m), compacted snow (50-120m), and thin ice (120-150m, curve area). Current vehicle status: speed 98km / h, roll angle 0.2°, pitch angle 0.1°, ground clearance 180mm. Estimated coefficient of adhesion: asphalt 0.8, snow 0.3, ice 0.1. Driver operation: throttle opening 20%, steering wheel angle begins to increase.

[0091] Step S202: Intelligent identification and status assessment.

[0092] The terrain recognition CNN model outputs different recognition results at different locations: at 50m, it outputs 95% dry asphalt, and after mapping, the dominant mode is highway mode; at 60m, it outputs 91% compacted snow, and after mapping, the dominant mode is snow mode; at 130m, it outputs 87% ice surface, and after mapping, the dominant mode is ice surface mode.

[0093] The condition index is updated in real time according to the road section changes. On a dry road section, the lateral acceleration is 0.05g and the yaw rate is 0.02rad / s. Substituting into the formula VSI=1-(0.05 / 0.8+0.02 / 0.1)=1-0.2625≈0.74 (within the stable range), the vertical vibration PSD integral value is low, the RCI is about 0.95 (comfortable), the ground clearance of 180mm is much greater than the required 120mm for this road section, and the TCR is about 0 (no risk).

[0094] Upon entering a snow-covered section, the terrain correction threshold decreases by 20%, lateral acceleration is 0.15g, yaw rate is 0.06 rad / s, and VSI is calculated as 1 - (0.15 / 0.64 + 0.06 / 0.08) = 0.016 (extremely unstable). Vertical vibration increases, PSD integral value is 0.4, terrain tolerance threshold T = 0.6, RCI = 1 - 0.4 / 0.6 = 0.33 (poor comfort). A ground clearance of 200mm is required, currently 180mm, with a 5cm undulation ahead, TCR = (200-180) / 200 + 50 / 250 = 0.3 (low to medium risk).

[0095] Upon entering the ice curve, the terrain correction threshold is reduced by 30%, lateral acceleration is 0.4g, yaw rate is 0.12 rad / s, VSI calculation result is negative, and amplitude limit is 0 (risk of complete instability). RCI remains at 0.5. Required ground clearance is 160mm, currently 180mm, no undulations ahead, TCR=0.

[0096] Step S203: Fuzzy logic decision-making and priority generation.

[0097] The fuzzy controller outputs corresponding priority indicators based on the terrain pattern, state index, and driver's operational intentions for each road segment. Dry road segment: =0.2, =0.3, =0.8. Snow-covered road sections: =0.9, =0.9, =0.2. Ice surface curve: =0.7, =1.0, =0.1.

[0098] Step S204: Generating executor target instructions.

[0099] The central decision-making unit queries the mapping table based on priority combinations and terrain patterns to calculate the specific instruction set for each road segment.

[0100] Dry road section (highway mode): Height = 180 + 30 × 0.2 = 186 mm (almost constant). Damping is calculated according to the basic formula. =30 + 60 × (1 - 0.8) = 42%, =1 + 1.0 × (0.3 - 0.5) = 0.8, =42 × 0.8 = 34% (slightly soft, comfort-oriented). Stabilizer bar =1000×0.3=300Nm / rad (relatively weak). ESP / TCS selected in standard highway mode. TCU upshift delay coefficient 1.04, downshift advance coefficient 1.15. Torque distribution is 53% front and 47% rear, close to balanced.

[0101] Snow-covered road section (snow mode): Height = 180 + 50 × 0.9 = 225 mm (significant rise). Meets the mandatory conditions for snow mode. >0.8 and <0.4), damping forced setting is =30% (fully soft, ensuring wheel contact with the ground). Stabilizer bar =1000×0.9=900Nm / rad (strong anti-roll). ESP / TCS simultaneously meets the conditions for high intervention and getting out of trouble. The system prioritizes the getting-out-of-trouble mode, allowing for moderate slippage, while high intervention allows for earlier braking engagement. TCU upshift delay coefficient = 1 + 0.5 × 0.9 = 1.45, downshift advance coefficient = 1 + 0.8 × 0.9 = 1.72, significantly delaying upshifts and advancing downshifts, keeping the engine in its high-efficiency range and utilizing engine braking. Torque distribution. =50-20×(0.9-0.5)=42%, then 58% (rear-biased, enhancing passability).

[0102] Ice surface curve (ice surface mode): Height = 180 + 20 × 0.7 = 194mm (slight adjustment, the ice surface is relatively flat and high ground clearance is not required). Forced soft conditions are not met; use the basic formula. =30 + 60 × (1 - 0.1) = 84%, =1 + 3.0 × (1.0 - 0.5) = 2.5, =84×2.5=210%, the limit is =90% (stiffest). In actual operation, based on the lateral acceleration in the corner, the damping of the outer wheel is further increased to 95%, while the damping of the inner wheel is appropriately reduced to 85%. Stabilizer bar =1000 × 1.0 = 1000 Nm / rad (maximum stiffness). ESP / TCS selects high intervention mode, pre-charging braking pressure onto the inner rear wheel to strictly limit torque output. TCU upshift delay coefficient is 1.14, downshift advance coefficient = 1 + 1.2 × 1.0 = 2.2, strongly advancing downshifts to utilize engine braking to stabilize the vehicle. Torque distribution. =50-10×(0.7-0.5)=48%, then 52%, and limit the total torque output.

[0103] Step S205: Instruction distribution and collaborative execution.

[0104] The central control unit simultaneously sends the above commands to each actuator ECU via the CAN FD bus. The air suspension ECU controls the solenoid valves to inflate the air springs, reaching 225mm on snowy roads and 194mm on icy roads, with real-time feedback from the height sensor and PID adjustment ensuring precision. The CDC controller adjusts the current of each shock absorber's solenoid valve according to the damping target value, maintaining a 30% duty cycle (fully soft) on snowy roads, and adjusting the damping of the outer front wheel to 95%, the outer rear wheel to 90%, and the inner wheel to 85% in icy corners. The active stabilizer bar ECU adjusts the motor output torque, providing 900Nm of anti-roll torque on snowy roads and increasing it to 1000Nm in icy corners. The ESP / TCU switches to the corresponding mode.

[0105] Step S206: Closed-loop feedback and fine-tuning.

[0106] In an icy curve, the IMU detected an actual roll angle of 2.5°, while the target was less than 2°. The PID controller output an additional signal to increase the stabilizer bar stiffness to 1050 Nm / rad and increase the damping of the outer shock absorber by 5%. Simultaneously, the wheel speed sensor detected a tendency for the inner rear wheel to lock up, and the ESP immediately released the braking pressure on that wheel and slightly adjusted the torque. After fine-tuning, the vehicle's actual roll angle stabilized at 1.8°, with good matching between the yaw rate and steering wheel angle, and no tendency to understeer or fishtail. This embodiment of the invention demonstrates the system's coordinated adjustment effect on various actuators under different road conditions.

[0107] The above specific scenarios fully demonstrate the working process of the all-terrain adaptive chassis control method provided by the embodiments of the present invention in actual driving. From highways to snow-covered mountain roads and then to sharp bends on icy surfaces, the system does not require manual intervention from the driver throughout the entire process. It automatically identifies terrain changes, assesses vehicle status, generates control decisions, and coordinates the adjustment of various chassis actuators, achieving fully automatic adaptive control for complex and ever-changing terrain conditions.

[0108] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0109] In one embodiment, the all-terrain adaptive chassis system provided by this invention includes: The global perception module is used to collect and identify the current driving terrain and obtain the relevant data required for the vehicle's dynamic status. A central decision control unit is used to execute the methods described in any of the above embodiments; The execution module is used to receive the target instruction set and execute it collaboratively.

[0110] In applications, such as Figure 3As shown, the all-domain perception module is responsible for the system's data acquisition function, including an environmental perception unit (forward-looking multi-function camera, millimeter-wave radar, lidar), a vehicle status perception unit (high-precision inertial measurement unit, wheel speed sensor, steering angle sensor, GPS receiver), and a chassis interaction perception unit (suspension height sensor and travel sensor for each wheel). The central decision control unit uses a high-performance domain controller or an onboard central computer, internally running multi-sensor fusion algorithms, terrain recognition intelligent models, and adaptive control strategy libraries, serving as the core decision center of the entire system. The actuator module includes the air suspension system, continuously damped adjustable shock absorbers, active stabilizer bar system, vehicle electronic stability system, transmission control unit, and powertrain controller. Each actuator is connected to the central decision control unit via a CAN FD bus, receiving and executing the target instruction set from the central decision control unit. The data flow is as follows: the perception module provides multi-source sensor data to the central decision control unit; the central decision control unit generates the target instruction set and distributes it to the actuator module; the execution result is fed back to the perception module and the central decision unit through the vehicle's dynamic status, forming a continuously optimized closed-loop control loop.

[0111] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0112] This application also provides an electronic device, including: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in the above-described method embodiments.

[0113] In applications, the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0114] In applications, memory can be an internal storage unit of an electronic device in some embodiments, such as a hard drive or RAM. In other embodiments, memory can be an external storage device of the electronic device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, memory can include both internal and external storage units of the electronic device. Memory is used to store operating systems, applications, bootloaders, data, and other programs, such as program code for computer programs. Memory can also be used to temporarily store data that has been output or will be output.

[0115] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0117] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps described in the various method embodiments above.

[0118] This application provides a computer program product, including a computer program, which, when run on an electronic device, enables the electronic device to perform the steps described in the various method embodiments above.

[0119] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0120] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0121] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0122] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0123] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0124] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An all-terrain adaptive chassis control method, characterized by, include: Identify the current driving terrain and determine the dominant terrain mode; Based on the vehicle's dynamic state, calculate multiple state indices that characterize the current driving conditions; Based on the dominant terrain pattern, the multiple state indices, and the driver's operational intent, multiple control priority indicators are generated; Based on the multiple control priority indicators and the dominant terrain pattern, a pre-calibrated strategy parameter mapping table is queried to generate the target instruction set for each actuator of the chassis; The target instruction set is distributed to each actuator in the chassis for coordinated execution.

2. The all-terrain adaptive chassis control method of claim 1, wherein, Before identifying the current driving terrain, the method also includes: The system collects at least one of the following data: visual image data of the road ahead, obstacle distance data, and point cloud data through environmental perception sensors; collects at least one of the following data: three-axis angular velocity, three-axis acceleration, wheel speed, steering angle, and positioning information through vehicle status sensors; and collects suspension height and / or travel data of each wheel through chassis interaction sensors. The collected multi-source sensor data are fused to generate comprehensive vehicle-environment situational information in a unified spatiotemporal coordinate system; The identification of the current driving terrain and the calculation of multiple state indices representing the current driving conditions are both performed based on the vehicle-environment integrated situation information.

3. The all-terrain adaptive chassis control method of claim 2, wherein, The comprehensive situational information also includes estimated values ​​of the road surface adhesion coefficient for each wheel; The estimated values ​​of the road surface adhesion coefficient for each wheel are obtained by fusing data from the inertial measurement unit and wheel speed sensor.

4. The all-terrain adaptive chassis control method of claim 2, wherein, The process of identifying the current driving terrain and determining the dominant terrain pattern includes: Based on the vehicle-environment integrated situational information, multimodal fusion features are obtained, and the multimodal fusion features are input into a pre-trained convolutional neural network model to obtain the probability distribution of the current terrain belonging to each predefined terrain category; The terrain category with the highest probability in the probability distribution is taken as the current basic terrain category, and the basic terrain category is converted into the current dominant terrain pattern through a preset terrain-pattern mapping table; The multimodal fusion features include at least two of the following: image features, vibration spectrum features, slope information, road surface adhesion coefficient estimate, vehicle body attitude parameters, and vehicle dynamic state parameters.

5. The all-terrain adaptive chassis control method of claim 1, wherein, The multiple status indices include vehicle stability index, ride comfort index, and passability risk level; The vehicle stability index is used to characterize the current dynamic stability state of the vehicle and is calculated based on lateral acceleration and yaw rate. The ride comfort index is used to characterize the comfort level of the vehicle's vertical vibration and is calculated based on the vertical acceleration power spectral density. The passability risk level is used to characterize the passability risk of the current terrain, and is calculated based on the current actual ground clearance, the recommended minimum ground clearance under the current terrain type, the undulation of the terrain ahead, and the maximum allowable ground clearance.

6. The all-terrain adaptive chassis control method of claim 5, wherein, The formula for calculating the Vehicle Stability Index (VSI) is as follows: wherein is a lateral acceleration, is a lateral acceleration threshold, is a yaw rate, is a yaw rate threshold; The formula for calculating the Ride Comfort Index (RCI) is as follows: wherein, PSD f is the vertical acceleration power spectral density, and T is the comfort tolerance threshold corresponding to the terrain type.​ The formula for calculating the passability risk level (TCR) is as follows: wherein, H req is the recommended minimum ground clearance for the current terrain type, H current is the current actual ground clearance, h front is the magnitude of the terrain undulation ahead, H clearance is the maximum allowed ground clearance.

7. The all-terrain adaptive chassis control method of claim 5 or 6, wherein, The thresholds used in the calculation of the vehicle stability index, ride comfort index, and passability risk level are dynamically adjusted by the dominant terrain model and / or obtained by querying a pre-calibrated terrain-threshold lookup table.

8. The all-terrain adaptive chassis control method as described in claim 1, characterized in that, The process generates multiple control priority indicators based on the dominant terrain pattern, the multiple state indices, and the driver's operational intentions, including: The dominant terrain pattern, the multiple state indices, and the driver's operating intention are used as input variables for the fuzzy logic controller, wherein the driver's operating intention includes at least one of the accelerator pedal opening, brake pedal opening, and steering wheel angle. Based on a preset fuzzy rule base, fuzzy reasoning is performed on the input variables to output a control priority index that includes passability priority, stability intervention intensity, and comfort priority.

9. The all-terrain adaptive chassis control method as described in claim 1, characterized in that, The chassis actuators include at least one of the following: air suspension system, continuously damped adjustable shock absorber, active stabilizer bar system, electronic stability control system, transmission control unit, and powertrain controller; The strategy parameter mapping table records the mapping relationship between the target parameters of each actuator and multiple control priority indicators and the dominant terrain mode.

10. The all-terrain adaptive chassis control method as described in claim 9, characterized in that, In the mapping relationship: H target =H base +ΔH max ×P traction ; where H target is the air suspension target height, H base is the reference height, ΔH max is the maximum lift amount corresponding to the current terrain mode, P traction is the priority of passability; S target =S max ×P stability , and is not lower than S min ; S target is a stiffness target value of the active stabilizer bar, S max is a maximum stiffness, S min is a minimum stiffness, and P stability is a stability intervention strength.

11. The all-terrain adaptive chassis control method as described in claim 9, characterized in that, In the mapping relationship: When the dominant terrain mode is snow or off-road and the passability priority is greater than the first preset threshold and the comfort priority is less than the second preset threshold, the damping target value is set to the minimum damping value. Otherwise, the target damping value is calculated based on the basic damping coefficient and the stability correction coefficient; The basic damping coefficient is obtained by linear interpolation between the minimum and maximum damping values ​​based on comfort priority, and the stability correction coefficient is calculated based on the correction gain and stability intervention intensity corresponding to the terrain pattern.

12. The all-terrain adaptive chassis control method as described in claim 9, characterized in that, In the mapping relationship: The operating mode of the vehicle electronic stability system is determined based on the comparison results of the stability intervention intensity and the passability priority thresholds. When the stability intervention intensity exceeds the preset intervention threshold, it enters the high intervention mode; when the passability priority exceeds the preset escape threshold, it enters the escape mode. The upshift delay coefficient and downshift advance coefficient of the transmission control unit are determined according to the passability priority and stability intervention intensity, respectively. The final shift point is calculated based on the shift point of the basic shift map and the corresponding coefficient. The front and rear axle torque distribution ratio of the powertrain controller is determined based on the torque adjustment coefficient corresponding to the passability priority and terrain mode.

13. The all-terrain adaptive chassis control method as described in claim 1, characterized in that, It also includes closed-loop adjustment of the target instruction set based on execution feedback, specifically including: Collect actual status feedback signals of each actuator, wherein the actual status feedback signals include at least one of actual suspension height, actual body roll angle and actual wheel speed of each wheel; The actual state feedback signal is compared with the target value in the target instruction set to obtain the execution deviation. The PID controller generates compensation instructions based on the execution deviation, and the target values ​​of each actuator are fine-tuned in real time.

14. The all-terrain adaptive chassis control method as described in claim 1, characterized in that, Also includes: Sensor data, decision commands, and execution results are anonymized and then uploaded to the cloud. Based on cloud-based big data analysis, the calibration parameters of the terrain recognition model, control rule base, and the strategy parameter mapping table are optimized and updated; The optimized and updated model, rule base, and calibration parameters are pushed to the vehicle via remote upgrade.

15. An all-terrain adaptive chassis system, characterized in that, include: The global perception module is used to collect and identify the current driving terrain and obtain the relevant data required for the vehicle's dynamic status. A central decision control unit, configured to perform the method as described in any one of claims 1 to 14; The execution module is used to receive the target instruction set and execute it collaboratively.

16. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method as described in any one of claims 1-14 to be performed.

Images