An automatic driving longitudinal comfort adaptive control method and system

Abstract

This invention provides an adaptive longitudinal comfort control method and system for autonomous driving. The method includes: acquiring the current load state of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load; determining basic longitudinal control parameters corresponding to the current load state from a pre-stored set of control parameters, based on the load range in which the current load state is located; acquiring the current tire pressure state of each tire of the vehicle; determining a correction coefficient corresponding to the current tire pressure state; multiplying the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters; and generating drive or braking control commands for the vehicle based on the final longitudinal control parameters. This method, through innovative multi-state fusion and parameter adaptive mechanisms, solves the control inaccuracy problem caused by dynamic changes in vehicle load and tire pressure. It not only improves the smoothness of acceleration and braking under different load and tire pressure conditions but also enhances the robustness and practicality of the system.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and specifically to an adaptive control method and system for longitudinal comfort in autonomous driving. Background Technology

[0002] With the rapid development of autonomous driving technology, the comfort of vehicle longitudinal control (i.e., acceleration and braking control) has become one of the core indicators for evaluating the performance of autonomous driving systems. A smooth and stable acceleration and deceleration experience is key to improving passenger acceptance and satisfaction. However, current mainstream autonomous driving longitudinal control strategies still show significant shortcomings in comfort performance when dealing with various state changes during actual vehicle operation. Specifically, existing technologies mainly suffer from the following deficiencies: First, the control strategy is insufficiently adaptable to dynamic changes in vehicle load. Traditional longitudinal control algorithms typically calibrate models and set parameters based on fixed vehicle mass and center of gravity positions. However, in actual use, vehicle loads (including the number of passengers and luggage weight) change frequently. From a driver-only configuration to a fully loaded vehicle with passengers and luggage, the total mass and center of gravity position of the vehicle change significantly. This causes deviations in the actual execution of control commands calculated based on a fixed model. For example, under full load conditions, using the braking pressure calibrated under no-load conditions will result in insufficient braking force, deceleration that does not reach the expected level, and a longer braking distance; conversely, using the braking pressure calibrated under full load conditions under no-load conditions will cause over-braking, resulting in a severe "nose-diving" phenomenon, which seriously affects ride comfort.

[0003] Secondly, the impact of tire pressure on longitudinal dynamics is generally overlooked. Tires are the sole medium for force transmission between the vehicle and the road surface, and their pressure directly determines their longitudinal stiffness, rolling resistance, and force transmission characteristics. Abnormal tire pressure leads to increased tire deformation, slower and more non-linear longitudinal force response, which can easily cause hysteresis, deceleration fluctuations, and even vehicle vibration during braking or acceleration. Currently, when tire pressure deviates from the standard value, the control system still operates according to the model under ideal tire pressure, inevitably leading to decreased control accuracy. This can cause potential safety and comfort risks, especially during emergency braking or precise following.

[0004] Third, existing finite adaptive methods often only address isolated factors, lacking coordinated control through multi-state fusion. Some advanced studies or systems consider load changes and adjust braking force distribution or energy recovery intensity accordingly; for abnormal tire pressure, they adjust traction control or tire slip ratio thresholds. However, in real-world driving scenarios, load changes and tire pressure fluctuations often occur simultaneously and jointly affect vehicle dynamics. For example, a fully loaded vehicle with insufficient tire pressure exhibits significantly different dynamic characteristics than an unloaded vehicle with normal tire pressure. Adjusting parameters for only a single factor cannot achieve globally optimal longitudinal comfort control; the coupled effects of these two factors may even produce negative consequences due to incoordination in the correction strategies.

[0005] Fourth, the lack of a smooth transition mechanism for control parameters when conditions change can potentially cause new discomfort. Even if the system can adjust control parameters according to changes in load or tire pressure, if the parameter switching process is instantaneous or abrupt, it can lead to abrupt changes in the vehicle's longitudinal response. For example, when passengers get on or off the vehicle, causing a sudden change in load, if the control parameters change immediately, the vehicle may experience a jerk during acceleration or braking. This dynamic shock introduced by the adjustment of control parameters itself contradicts the original intention of pursuing comfort.

[0006] In summary, existing autonomous driving longitudinal comfort control technologies fail to comprehensively consider the dynamic impact of two key state variables—load and tire pressure—and lack multi-factor coordination and smooth transition mechanisms, making it difficult to provide a consistently high-quality and smooth experience in various real-world and complex driving environments. Summary of the Invention

[0007] The purpose of this invention is to provide an adaptive longitudinal comfort control method and system for autonomous driving, which aims to solve the technical problem that existing autonomous driving longitudinal control strategies cannot adaptively adjust control parameters to achieve smooth acceleration and braking when vehicle load and tire pressure change dynamically at the same time, resulting in a decrease in ride comfort.

[0008] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In a first aspect, the present invention provides an adaptive control method for longitudinal comfort in autonomous driving, comprising the following steps: S1. Obtain the current load status of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load. S2. Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; S3. Obtain the current tire pressure status of each tire of the vehicle; determine the correction coefficient corresponding to the current tire pressure status based on the current tire pressure status; S4. Multiply the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters; S5. Generate driving or braking control commands for the vehicle based on the final longitudinal control parameters.

[0009] Based on the above scheme, step S2 further includes: The total mass range of the vehicle is divided into at least two load ranges; the load ranges include an unloaded range, a half-loaded range, and a fully loaded range. A set of basic longitudinal control parameters is pre-calibrated for each load interval to form the control parameter set; Determine the load range to which the total mass calculated from the current load state belongs, and select the corresponding basic longitudinal control parameters.

[0010] Furthermore, when the total mass is at or between the boundaries of two load intervals, determining the basic longitudinal control parameters corresponding to the current load state includes: Obtain two sets of foundation longitudinal control parameters corresponding to the two load intervals adjacent to the total mass; By interpolation calculation, basic longitudinal control parameters that match the total mass are generated.

[0011] Furthermore, step S3, which involves determining a correction coefficient corresponding to the current tire pressure state, includes: The tire pressure deviation is obtained by comparing the current tire pressure value of each tire with the standard tire pressure value. The tire pressure condition level is determined based on the degree of tire pressure deviation. Based on the tire pressure condition level and the location of the tire where the tire pressure deviation occurred, a correction factor is determined to correct the basic longitudinal control parameter.

[0012] Furthermore, the tire pressure status levels include normal, slightly abnormal, and significantly abnormal; wherein, When the tire pressure is slightly abnormal or significantly abnormal, the correction factor is assigned different weights based on at least one of the attributes that the tire is a drive wheel, driven wheel, front wheel, or rear wheel.

[0013] Furthermore, the basic longitudinal control parameters mentioned in step S4 include at least one of the following: a braking pressure curve for achieving the target deceleration, a driving torque curve for achieving the target acceleration, and the maximum allowable acceleration value; The maximum allowable jerk value is dynamically adjusted according to the load range in which the current load state is located; the larger the load, the smaller the maximum allowable jerk value.

[0014] Furthermore, when the current load state or the current tire pressure state changes, the method further includes: Within a preset time window, the correction coefficient is gradually changed from its original value to its changed value to achieve a smooth transition of the final longitudinal control parameter.

[0015] Secondly, the present invention also provides a system for compressing vehicle controller software to implement the above method, comprising: The perception and estimation module is used to obtain the current load status of the vehicle and the current tire pressure status of each tire. The parameter adaptive module is used for: Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; Based on the current tire pressure status, determine the correction factor corresponding to the tire pressure status; and The final longitudinal control parameters are determined by multiplying the basic longitudinal control parameters by the correction coefficient. The longitudinal control execution module is used to generate drive or braking control commands for the vehicle based on the final longitudinal control parameters.

[0016] Thirdly, the present invention also provides an electronic device, including at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the at least one processor implements the autonomous driving longitudinal comfort adaptive control method as described in the first aspect by executing the instructions stored in the memory.

[0017] Fourthly, the present invention also provides a computer-readable storage medium storing at least one instruction or at least one program, wherein the at least one instruction or at least one program is loaded and executed by a processor to implement the autonomous driving longitudinal comfort adaptive control method as described in the first aspect.

[0018] The adaptive longitudinal comfort control method for autonomous driving provided by this invention is the first to systematically integrate the real-time load state and tire pressure state of the vehicle and directly apply them to the calculation of longitudinal control parameters. This effectively solves the comfort problem caused by traditional autonomous driving control strategies neglecting dynamic changes in vehicle state. This method can achieve smooth, stable, and consistent longitudinal acceleration control under different occupant loads, luggage weights, and tire pressures, significantly improving the ride quality of autonomous driving. Specifically, it achieves the following significant technical effects: 1. Precise longitudinal control based on load status is achieved, ensuring consistency and stability of comfort under different loads. The system first determines the current total load based on the number of occupants and luggage weight, and then matches or calculates the basic control parameters for the corresponding load range. This fundamentally changes the traditional approach of using fixed parameters, enabling the vehicle to automatically apply matching drive torque or braking pressure to achieve the same target acceleration or deceleration when unloaded, half-loaded, and fully loaded. For example, when fully loaded, the braking force is automatically increased to maintain the expected deceleration, while when unloaded, the braking force is automatically reduced to prevent "nose-diving," thus making the longitudinal dynamic response consistent under different load conditions and eliminating abrupt acceleration or braking shocks caused by load changes.

[0019] 2. A dynamic tire pressure compensation mechanism has been introduced, improving the system's control robustness and comfort under various tire operating conditions. The system acquires tire pressure information in real time and generates correction coefficients. This design incorporates tire pressure, a key variable long neglected in longitudinal comfort control, into the control closed loop. When tire pressure is abnormal (e.g., insufficient tire pressure), the system can automatically adjust control commands (e.g., increase braking pressure) through the correction coefficients to compensate for force transmission losses caused by changes in tire longitudinal stiffness. This ensures that even under non-ideal tire pressure conditions, the vehicle's actual acceleration response closely follows the expected value, avoiding discomfort such as braking lag and acceleration fluctuations caused by abnormal tire pressure, and enhancing the system's adaptability to changes in external conditions.

[0020] 3. Through parametric fusion of multi-source information, globally optimized adaptive comfort control is achieved. Cooperative optimization involves multiplying the basic parameters from the load with the correction coefficient from tire pressure to generate the final longitudinal control parameters. This method overcomes the limitations of existing technologies that only independently correct for a single factor, achieving comprehensive compensation for the coupled effects of load and tire pressure. Through this fusion, control commands can simultaneously reflect changes in vehicle mass inertia and tire contact characteristics, thus enabling the calculation of globally optimal control quantities under various complex operating conditions (such as full load and insufficient tire pressure), providing a smoother driving experience.

[0021] 4. A clear and universally applicable control architecture has been developed, demonstrating high practicality and integrability. The five steps constitute a complete adaptive control closed loop. This scheme is independent of specific vehicle platforms or actuator types, and its "perception-judgment-calculation-execution" framework exhibits good versatility. Furthermore, this method can be implemented using existing onboard sensors without requiring additional special hardware, facilitating software integration and deployment within existing autonomous driving domain controllers. It also boasts low engineering implementation costs and is easy to promote and apply.

[0022] In summary, the adaptive longitudinal comfort control method for autonomous driving provided by this invention systematically solves the control inaccuracy problem caused by dynamic changes in vehicle load and tire pressure through an innovative multi-state fusion and parameter adaptation mechanism. It not only significantly improves the smoothness of acceleration and braking under different load and tire pressure conditions, but also enhances the robustness and practicality of the system through a clear and efficient algorithm architecture, providing an effective technical solution for improving the core user experience of high-level autonomous driving systems. Attached Figure Description

[0023] Figure 1 A schematic diagram of an adaptive control method for longitudinal comfort in autonomous driving provided in an embodiment of the present invention; Figure 2 A schematic diagram of the structure of the adaptive control system for longitudinal comfort of autonomous driving provided in an embodiment of the present invention; Figure 3 A schematic diagram of the hardware structure of a possible electronic device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided in an embodiment of the present invention. Detailed Implementation

[0024] The implementation methods of this solution will be described in further detail below. Obviously, the described embodiments are only a part of the embodiments of this solution, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this solution can be combined with each other.

[0025] Example 1 This invention provides an adaptive longitudinal comfort control method for autonomous driving, aiming to change the traditional approach of longitudinal control based on a fixed vehicle model and parameters. Instead, it dynamically adjusts control parameters by real-time sensing of two key dynamic states of the vehicle—load state and tire pressure state—and comprehensively processing these two states to achieve better longitudinal comfort. Figure 1 As shown, it includes the following steps: Step S1: Obtain the current load status of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load.

[0026] Specifically, it includes: The aforementioned current load status refers to the actual load condition of the vehicle during operation, and is a comprehensive reflection of the vehicle's total mass and center of gravity position. This step aims to obtain at least two pieces of information: the number of occupants and the weight of luggage load.

[0027] The number of occupants can be obtained in various ways, such as, but not limited to: detecting whether there are occupants sitting by pressure sensors built into the seats; using seat belt buckle status sensors for auxiliary judgment; or using in-vehicle cameras for visual recognition and headcount.

[0028] The estimated weight of luggage load can be obtained through the suspension height sensor on the rear axle or the pressure sensor of the air suspension system. By measuring the change in suspension height or pressure when the vehicle is stationary, the increased load mass can be indirectly calculated. Adding the estimated total passenger mass (which can be calculated based on the standard average weight per person), the luggage mass, and the vehicle's curb weight gives the current total mass of the vehicle. This total mass is one of the key quantitative indicators characterizing the load status.

[0029] Step S2: Based on the load range of the current load state, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set.

[0030] The control parameter set in this step is a pre-calibrated database stored in the vehicle control system (such as the autonomous driving domain controller). This database establishes a mapping relationship between different "load states" and "ideal control parameters". Here, "load state" is divided into "load ranges", that is, based on the range of total mass, the possible load conditions of the vehicle are divided into several typical ranges, such as "light load range", "normal range", and "heavy load range".

[0031] This step first determines the load range based on the total mass calculated by S1. Then, it directly searches for and selects a set of "basic longitudinal control parameters" corresponding to this load range from the pre-stored control parameter set. This set of parameters is optimized for this typical load condition and aims to provide a smooth longitudinal control baseline. Through this step, the control strategy achieves adaptation to different load conditions, laying the foundation for subsequent precise control.

[0032] More specifically, the steps include: The total mass of the vehicle is divided into at least two load ranges; these load ranges include an unloaded range, a half-loaded range, and a fully loaded range. Specifically, based on the vehicle's design load capacity and common usage scenarios, the continuous range of total mass from minimum (e.g., driver only) to maximum (e.g., fully loaded with passengers and luggage) can be divided into at least two discrete load ranges. As a preferred, easily calibrated, and implementable approach, three typical ranges can be defined: an unloaded range, a half-loaded range, and a fully loaded range. The unloaded range corresponds to the mass range of only the driver or the driver plus a small amount of luggage; the half-loaded range corresponds to the mass range of 2-3 passengers plus moderate luggage; and the fully loaded range corresponds to the mass range of 4-5 passengers plus heavy luggage. The boundary values ​​of the ranges can be determined experimentally based on the specific vehicle model. A set of basic longitudinal control parameters is pre-calibrated for each load interval to form the control parameter set; specifically, for each load interval defined above, an optimal set of "basic longitudinal control parameters" is pre-calibrated. The calibration work is completed during the vehicle development stage. Under each typical load condition (such as unloaded, half-loaded, and fully loaded counterweight states), through real vehicle testing, the control parameters that can achieve the most comfortable acceleration and braking experience are calibrated to form the "control parameter set". The system determines the load range to which the total mass calculated from the current load state belongs and selects the corresponding basic longitudinal control parameters. Specifically, during actual vehicle operation, the system determines which predefined load range the real-time total mass estimated in step S1 falls into and directly selects the pre-calibrated parameters corresponding to that range as the current "basic longitudinal control parameters".

[0033] It should be noted that when the total mass is at or between the boundaries of two load ranges, the determination of the basic longitudinal control parameters corresponding to the current load state includes: Obtain two sets of foundation longitudinal control parameters corresponding to the two load intervals adjacent to the total mass; By interpolation calculation, basic longitudinal control parameters that match the total mass are generated.

[0034] Specifically, when the total mass M is calculated in real time total If the value is exactly equal to the boundary value of a certain load interval, or clearly falls within a certain defined interval, then the parameter set of that interval is directly selected. However, when M... total When the mass is between two adjacent load ranges (e.g., the mass is greater than the upper limit M of the unloaded range), low However, it is less than or equal to the lower limit M of the half-load interval. medium To avoid abrupt changes in vehicle dynamic response caused by minute changes in mass between two different parameter sets, this implementation method uses interpolation calculation.

[0035] A specific example is: The system first obtains the current total mass M. total Two sets of foundation longitudinal control parameters corresponding to two adjacent load intervals (e.g., unloaded interval and half-loaded interval). Then, based on M total Relative to the positions of these two interval boundary values, a new set of values ​​is calculated using linear interpolation, which is consistent with the current precise mass M. total Matching basic longitudinal control parameters. For example, for each point on the braking pressure curve, its pressure value P can be obtained using the formula P = P empty +(P half - P empty ) (M total - M low ) / (M medium - M low To calculate, where P empty and P half These are the pressure values ​​corresponding to the no-load and half-load ranges, respectively.

[0036] By introducing an inter-interval interpolation algorithm, the control parameters can change continuously and smoothly with the total mass of the vehicle, completely eliminating the risk of step changes in control parameters due to small load changes. This ensures extremely smooth longitudinal response of the vehicle during continuous load changes, improving the precision of control and the consistency of the riding experience.

[0037] Step S3: Obtain the current tire pressure status of each tire of the vehicle; determine the correction coefficient corresponding to the current tire pressure status.

[0038] The step of determining the correction coefficient corresponding to the current tire pressure status, based on the current tire pressure status, includes: The tire pressure deviation is obtained by comparing the current tire pressure value of each tire with the standard tire pressure value. The tire pressure condition level is determined based on the degree of tire pressure deviation. Based on the tire pressure condition level and the location of the tire where the tire pressure deviation occurred, a correction factor is determined to correct the basic longitudinal control parameter.

[0039] The tire pressure status level includes normal, slightly abnormal, and significantly abnormal; wherein, when the tire pressure status is slightly abnormal or significantly abnormal, the correction coefficient is assigned different weights based on at least one of the attributes of the tire being a drive wheel, driven wheel, front wheel, or rear wheel.

[0040] This step uses the vehicle's existing tire pressure monitoring system (TPMS) to acquire the current tire pressure values ​​of all four tires in real time. The "tire pressure status" refers not only to the pressure value itself, but also to the degree of deviation from the standard value and its potential impact on vehicle dynamics. Based on the acquired current tire pressure, the system calculates a "correction factor." This correction factor is a scaling factor used to fine-tune or compensate for the basic longitudinal control parameters obtained in step S2. Its purpose is to counteract changes in tire longitudinal stiffness, rolling resistance, and force transmission characteristics caused by tire pressure deviations from the standard value, ensuring that the accuracy of control commands is not affected by tire pressure fluctuations.

[0041] The specific process for determining the correction factor based on the current tire pressure is as follows: 1. Calculate tire pressure deviation: Obtain the real-time tire pressure value P for each tire (usually front left, front right, rear left, and rear right). i Compare this real-time tire pressure value with the standard tire pressure value P set by the vehicle manufacturer. ref Compare the tire pressures and calculate the tire pressure deviation ΔP for each tire. i =P i - P ref .

[0042] 2. Determine Tire Pressure Status Levels: Not all minor tire pressure deviations require control and compensation. This implementation classifies tire pressure status into different levels based on the absolute value of the deviation. For example, a deviation threshold range can be set, where |ΔP| < ΔP / ΔP. i When |ΔP| is less than the first threshold (e.g., 0.2 bar), it is judged as "normal" and its impact on longitudinal dynamics is considered negligible; when ... i | When |ΔP| falls between the first threshold and a larger second threshold (e.g., 0.5 bar), it is classified as a "minor abnormality" level; when .... i When the value exceeds the second threshold, it is classified as "significantly abnormal". This classification makes the control strategy more robust and targeted.

[0043] 3. Determine the correction coefficient: Based on the tire pressure status level determined in step 2, and more importantly, based on the tire location where the pressure deviation occurs, determine the final correction coefficient. Different tire locations (e.g., front / rear wheels, drive / driven wheels) have different weights in their impact on the vehicle's longitudinal dynamics. For example, abnormal front tire pressure has a greater impact on braking performance, while abnormal drive tire pressure has a greater impact on acceleration performance. The system will preset different correction coefficient mapping tables or calculation rules for different locations and levels of tire pressure abnormalities.

[0044] By quantifying tire pressure status into two dimensions—"degree of deviation" and "location of occurrence"—and determining correction coefficients accordingly, tire pressure compensation becomes more scientific and precise. This avoids a "one-size-fits-all" approach to compensation and can more accurately reflect the actual impact of specific tire pressure anomalies on the longitudinal dynamics of the entire vehicle, thereby enabling more effective compensation control and further improving comfort under abnormal tire pressure conditions.

[0045] Furthermore, the weights assigned to the tire pressure condition level and correction factor are shown in the following example: Tire pressure status can be clearly divided into three levels: normal, slightly abnormal, and significantly abnormal. The judgment criteria can be as described above, for example, using a deviation of 0.2 bar and 0.5 bar as the boundary.

[0046] When tire pressure is determined to be "slightly abnormal" or "significantly abnormal," the system considers the tire's functional attributes when determining the correction factor. Specifically, the correction factor is assigned different weighting factors based on at least one of the following attributes: whether the tire is a drive wheel or a driven wheel, or a front wheel or a rear wheel. For example, a weighting matrix can be established: For braking control: if the left front tire pressure is significantly abnormal, its weighting factor may be 1.15 (i.e., the baseline braking force needs to be increased by 15%); if the right rear tire pressure is slightly abnormal, its weighting factor may be only 1.02.

[0047] For acceleration control: If the left rear drive tire pressure is significantly abnormal, its weighting factor may be 1.10 (i.e., the base drive torque needs to be increased by 10%).

[0048] Finally, the weighting factors of the four tires are combined (e.g., taking the maximum value, average value, or position-based weighted average) to obtain the global correction coefficients applied to the basic longitudinal control parameters.

[0049] Step S4: Multiply the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters.

[0050] Specifically, this step is used to achieve multi-state information fusion. It performs mathematical operations (multiplication) on the "basic longitudinal control parameters" derived from the load state and the "correction coefficient" derived from the tire pressure state to generate a "final longitudinal control parameter" that integrates the current vehicle load and the actual tire state. This fusion method enables the final control command to respond simultaneously to the dynamic changes in vehicle mass inertia (determined by the load) and tire contact characteristics (affected by tire pressure), achieving coordinated optimization control for complex operating conditions.

[0051] The basic longitudinal control parameters include at least one of the following: a braking pressure curve for achieving the target deceleration, a driving torque curve for achieving the target acceleration, and a maximum allowable jerk value; the maximum allowable jerk value is dynamically adjusted according to the load range in which the current load state is located, wherein the larger the load, the smaller the maximum allowable jerk value.

[0052] Specifically, the aforementioned basic longitudinal control parameters are selected from the control parameter set and are a set of key parameters aimed at achieving comfortable longitudinal control. Their specific forms may include, but are not limited to: Braking pressure curve for achieving target deceleration: A mapping curve that specifies the braking pressure value required to achieve a certain target deceleration.

[0053] Drive torque curve for achieving target acceleration: A mapping curve that specifies the required drive torque value to achieve a certain target acceleration.

[0054] Maximum permissible jerk (Jerk) value: Jerk is the rate of change of acceleration. Limiting the maximum jerk is a direct means of controlling the impact and ensuring smoothness. This value limits the drastic degree of acceleration change.

[0055] Specifically, the maximum allowable jerk value is not a fixed value, but is dynamically adjusted according to the load range of the current load state. The adjustment principle is: the greater the load, the smaller the maximum allowable jerk value is set. This is because, under the same jerk, a vehicle with a larger mass experiences a greater change in inertial force, resulting in a stronger impact felt by the occupants. Therefore, in the fully loaded range, the system uses a smaller maximum jerk limit value, making the acceleration and braking process smoother; while in the unloaded range, a relatively larger limit value can be used to achieve a more agile response while ensuring comfort.

[0056] Step S5: Generate driving or braking control commands for the vehicle based on the final longitudinal control parameters.

[0057] Specifically, the final longitudinal control parameters calculated in step S4 are sent to the vehicle's drive control system (such as the motor controller) or braking control system (such as the Electronic Stability Program (ESP) or brake-by-wire system) to be converted into specific drive torque requests or braking pressure requests, thereby controlling the vehicle to perform acceleration or deceleration actions.

[0058] In summary, by synchronously incorporating vehicle load and tire pressure status into the control closed loop and employing a fusion architecture of "basic parameters + correction coefficients," this invention can automatically adapt to common load changes and tire pressure fluctuations encountered in daily vehicle use. Compared to traditional fixed-parameter control methods, this method can maintain stable and smooth longitudinal acceleration response across a wider range of operating conditions, effectively mitigating "nodding" or "lurching" phenomena caused by load changes, as well as braking lag caused by abnormal tire pressure, significantly improving ride comfort under various real-world conditions for autonomous driving.

[0059] Based on the above steps, the method provided in this embodiment further includes: when the current load state or the current tire pressure state changes, within a preset time window, gradually changing the correction coefficient from the value before the change to the value after the change, so as to achieve a smooth transition of the final longitudinal control parameter.

[0060] Specifically, during the control process, when the system detects a change in the current load state (such as a sudden decrease in total mass due to occupants exiting the vehicle) or the current tire pressure state (such as a rapid drop in tire pressure due to a tire puncture), the correction coefficient calculated based on the new state will change. If the old correction coefficient is switched abruptly to the new correction coefficient, it will cause a jump in the final longitudinal control parameters, which may lead to unexpected jerking of the vehicle.

[0061] To avoid this problem, this method adds a smooth transition step after the parameter update is triggered by the state change. Specifically, the system does not immediately apply the new correction coefficient. Instead, within a preset time window (e.g., 3 or 5 seconds), the correction coefficient to be calculated gradually transitions from the old value before the change to the new value in a linear (or other smooth function) manner. During this period, the correction coefficient used to calculate the final longitudinal control parameter in step S4 is a continuously changing intermediate value, thus ensuring that the final control command also changes smoothly.

[0062] This step, by introducing a parameter gradual change mechanism during state changes, ensures that the output of the control system is continuous and smooth at all times. This effectively prevents step changes in control parameters caused by events such as passengers getting in and out of the vehicle or sudden changes in tire pressure, thereby completely eliminating the longitudinal dynamic shock to the vehicle that may be caused by such state changes, and further ensuring a seamless and consistent comfort experience.

[0063] Further explanation will be provided with specific examples: Taking commercially available electric SUVs as an example Example 1 compares braking performance under empty and full load conditions. When the vehicle is empty (driver only, total mass 2.0 tons), the automatic driving system brakes at a traffic light using the braking pressure curve calibrated for the empty load range, resulting in a smooth stop without any nose-diving. When the vehicle is fully loaded (5 passengers plus luggage, total mass 2.5 tons), the system recognizes the load as being within the full load range and automatically switches to the braking pressure curve for the full load range. The braking pressure increases accordingly, and the vehicle again stops smoothly with the same deceleration as when empty, and rear passengers do not experience any impact.

[0064] Example 2 illustrates interpolation processing when the load falls within a certain range. The vehicle has three passengers (total mass 2.2 tons), and the load range falls between the unloaded range (2.0-2.1 tons) and the half-loaded range (2.1-2.3 tons). The system uses linear interpolation to calculate a braking curve suitable for a 2.2-ton load from the unloaded and half-loaded parameters. During braking, the vehicle response is smooth, with no noticeable parameter jumps.

[0065] Example 3 illustrates correction for insufficient tire pressure. One day, the pressure of the left front tire dropped from the standard 2.5 bar to 1.8 bar, a significant anomaly. After detecting the tire pressure deviation, the system adjusted the brake pressure correction factor to 1.15 based on the position of the left front wheel (the critical braking wheel). During subsequent braking, the system applied 15% higher brake pressure than the standard value, compensating for the decrease in braking force caused by insufficient tire pressure and maintaining deceleration consistent with normal tire pressure. Simultaneously, the acceleration limit was increased from 2.0 m / s². 3 Adjusted to 1.7 m / s 3 This makes the braking process smoother and avoids the jolts that may be caused by abnormal tire pressure.

[0066] Example 4 illustrates a combined operating condition and transition process. The vehicle is fully loaded with a slightly low right rear tire pressure. The system uses the full-load range parameters plus tire pressure correction. During operation, two passengers and some luggage disembark, reducing the load from 2.5 tons to 2.2 tons. Upon detecting the load change, the system smoothly transitions the control parameters from the full-load range to the half-load range within 3 seconds, while maintaining the tire pressure correction coefficient unchanged. Throughout the entire process, the driver and remaining passengers do not experience any dynamic changes in the vehicle due to the parameter switching.

[0067] Example 2 See Figure 2 , Figure 2 A system for implementing the adaptive longitudinal comfort control method for autonomous driving described in Embodiment 1, provided by an embodiment of the present invention, specifically includes: The perception and estimation module is used to obtain the current load status of the vehicle and the current tire pressure status of each tire. The parameter adaptive module is used for: Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; Based on the current tire pressure status, determine the correction factor corresponding to the tire pressure status; and The final longitudinal control parameters are determined by multiplying the basic longitudinal control parameters by the correction coefficient. The longitudinal control execution module is used to generate drive or braking control commands for the vehicle based on the final longitudinal control parameters.

[0068] Example 3 See Figure 3 , Figure 3 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 3 As shown, this embodiment of the invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following steps: S1. Obtain the current load status of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load. S2. Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; S3. Obtain the current tire pressure status of each tire of the vehicle; determine the correction coefficient corresponding to the current tire pressure status based on the current tire pressure status; S4. Multiply the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters; S5. Generate driving or braking control commands for the vehicle based on the final longitudinal control parameters.

[0069] Example 4 See Figure 4 , Figure 4 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by the present invention. (See diagram below.) Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it performs the following steps: S1. Obtain the current load status of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load. S2. Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; S3. Obtain the current tire pressure status of each tire of the vehicle; determine the correction coefficient corresponding to the current tire pressure status based on the current tire pressure status; S4. Multiply the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters; S5. Generate driving or braking control commands for the vehicle based on the final longitudinal control parameters.

[0070] In summary, the autonomous driving longitudinal comfort adaptive control method, system, and device proposed in this invention innovatively integrates and processes the real-time vehicle load status and tire pressure status to construct a complete longitudinal comfort adaptive control framework. This method effectively overcomes the control deviations caused by traditional control strategies neglecting load and tire pressure changes, and solves the limitations of single-factor correction and the problem of unsmooth parameter switching. Its ultimate technical effect is that it enables autonomous vehicles to automatically and smoothly adjust longitudinal control parameters under different passenger loads, luggage weights, and tire pressure conditions, thereby maintaining consistent and excellent acceleration and braking smoothness under various complex real-world operating conditions, significantly improving ride comfort, and enhancing the system's adaptability and robustness, thus possessing significant engineering application value.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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; and these 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 the present invention.

Claims

1. An automatic longitudinal comfort adaptive control method, characterized in that, Includes the following steps: S1. Obtain the current load status of the vehicle, which is determined at least based on the number of occupants and the weight of luggage load. S2. Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; S3. Obtain the current tire pressure status of each tire of the vehicle; Based on the current tire pressure status, determine the correction factor corresponding to the tire pressure status; S4. Multiply the basic longitudinal control parameters by the correction coefficient to determine the final longitudinal control parameters; S5. Generate driving or braking control commands for the vehicle based on the final longitudinal control parameters.

2. The method of claim 1, wherein, Step S2 specifically includes: The total mass range of the vehicle is divided into at least two load ranges; the load ranges include an unloaded range, a half-loaded range, and a fully loaded range. A set of basic longitudinal control parameters is pre-calibrated for each load interval to form the control parameter set; Determine the load range to which the total mass calculated from the current load state belongs, and select the corresponding basic longitudinal control parameters.

3. The method of claim 2, wherein, When the total mass is at or between the boundaries of two load ranges, the determination of the basic longitudinal control parameters corresponding to the current load state includes: Obtain two sets of foundation longitudinal control parameters corresponding to the two load intervals adjacent to the total mass; By interpolation calculation, basic longitudinal control parameters that match the total mass are generated.

4. The method of claim 1, wherein, Step S3, which involves determining the correction factor corresponding to the current tire pressure state, includes: The tire pressure deviation is obtained by comparing the current tire pressure value of each tire with the standard tire pressure value. The tire pressure condition level is determined based on the degree of tire pressure deviation. Based on the tire pressure condition level and the location of the tire where the tire pressure deviation occurred, a correction factor is determined to correct the basic longitudinal control parameter.

5. The method of claim 4, wherein, The tire pressure status levels include normal, slightly abnormal, and significantly abnormal; among which... When the tire pressure is slightly abnormal or significantly abnormal, the correction factor is assigned different weights based on at least one of the attributes that the tire is a drive wheel, driven wheel, front wheel, or rear wheel.

6. The method of claim 1, wherein, The basic longitudinal control parameters mentioned in step S4 include at least one of the following: a braking pressure curve for achieving the target deceleration, a driving torque curve for achieving the target acceleration, and the maximum allowable acceleration value; The maximum allowable jerk value is dynamically adjusted according to the load range in which the current load state is located; the larger the load, the smaller the maximum allowable jerk value.

7. The method according to claim 1, characterized in that, When the current load state or the current tire pressure state changes, the method further includes: Within a preset time window, the correction coefficient is gradually changed from its original value to its changed value to achieve a smooth transition of the final longitudinal control parameter.

8. An adaptive control system for longitudinal comfort in autonomous driving for implementing the method of any one of claims 1 to 7, characterized in that, include: The perception and estimation module is used to obtain the current load status of the vehicle and the current tire pressure status of each tire. The parameter adaptive module is used for: Based on the load range in which the current load state is located, determine the basic longitudinal control parameters corresponding to the current load state from the pre-stored control parameter set; Based on the current tire pressure status, determine the correction factor corresponding to the tire pressure status; as well as The final longitudinal control parameters are determined by multiplying the basic longitudinal control parameters by the correction coefficient. The longitudinal control execution module is used to generate drive or braking control commands for the vehicle based on the final longitudinal control parameters.

9. An electronic device, comprising at least one processor and a memory communicatively connected to said at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which implements the autonomous driving longitudinal comfort adaptive control method as described in any one of claims 1 to 7 by executing the instructions stored in the memory.

10. A computer-readable storage medium storing at least one instruction or at least one program, said at least one instruction or at least one program being loaded and executed by a processor to implement the autonomous driving longitudinal comfort adaptive control method as described in any one of claims 1 to 7.

Images