Load estimation system based on multi-sensor fusion and vehicle
By using a multi-sensor fusion system for load estimation, combined with dynamic and static modes and a self-learning module, the problems of low efficiency and insufficient accuracy in traditional truck load measurement are solved, and high-precision load monitoring is achieved.
Patent Information
- Application Number
- CN202510886989.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-30
- Publication Date
- 2025-10-28
AI Technical Summary
Traditional truck load measurement relies on weighbridges, which are inefficient and cannot be monitored in real time. Existing tire pressure load measurement technology is not accurate enough and is easily affected by ambient temperature, road slope and tire aging.
A multi-sensor fusion system is adopted, including tire pressure sensor, tilt sensor and inertial measurement unit. Combining dynamic and static modes and self-learning module, it performs dual correction through temperature-pressure relationship curve and tilt angle-tire force distribution model to achieve high-precision dynamic monitoring.
It achieves high-precision dynamic monitoring during driving, with a load error of less than 2%, eliminates temperature and slope interference, and significantly improves the accuracy of load estimation and anti-interference capability.
Smart Images
Figure CN120846472A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a load estimation system and vehicle based on multi-sensor fusion, belonging to the field of vehicle load monitoring technology. Background Technology
[0002] Traditional truck load measurement mainly relies on weighbridges, which requires the vehicle to be stationary and not engaged in transport, resulting in low efficiency and the inability to achieve real-time monitoring. Existing tire pressure-based load estimation technologies largely ignore the impact of factors such as ambient temperature, road inclination, and tire deformation hysteresis on measurement accuracy, thus lacking precision. To address these issues, a high-precision, interference-resistant automated weighing solution is urgently needed. Summary of the Invention
[0003] The purpose of this invention is to provide a load estimation system and vehicle based on multi-sensor fusion, which can solve the problems of traditional weighbridges requiring the vehicle to stop and be removed from its task, and the inaccuracy of existing tire pressure load technology being easily affected by temperature / slope / tire aging interference. By combining a multi-source compensation mechanism (temperature-air pressure correction + tilt angle-load distribution) with dynamic and static dual modes, high-precision dynamic monitoring during driving is achieved, and the self-learning closed-loop optimization of model parameters is used to significantly improve long-term anti-interference capability.
[0004] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.
[0005] In a first aspect, the present invention provides a load estimation system based on multi-sensor fusion, characterized in that it includes: The data acquisition module includes: Tire pressure sensor is used to acquire tire pressure and tire temperature data; Tilt sensor, used to acquire vehicle slope data; An inertial testing unit is used to monitor vehicle vibration and identify windows of smooth driving conditions. The data processing module includes: The dynamic compensation unit is used to perform dual correction on the tire pressure data and tire temperature data in the window of the smooth driving section according to the pre-calibrated temperature-pressure relationship curve and the tilt angle-tire force distribution model, so as to obtain the equivalent corrected tire pressure value of each tire. The load calculation unit is used to query the pre-calibrated load-pressure mapping table based on the equivalent corrected tire pressure value of each tire and output the vehicle load weight. The self-learning module is used to compare the load weight with the load weight measured by the weighbridge, and dynamically correct the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the tilt angle-tire force distribution model based on the comparison results, so as to update the load-pressure mapping table.
[0006] In conjunction with the first aspect, the pre-calibration process of the temperature-pressure relationship curve further includes: In the temperature-controlled test chamber, the tire type, initial tire pressure P0 and load status are determined, and the temperature inside the chamber is adjusted according to the preset temperature steps. Tire pressure data is collected after the temperature stabilizes at each step. Repeat the process of collecting tire pressure data after each step of constant temperature stabilization to obtain datasets under no-load, half-load, and full-load conditions respectively; Based on the dataset, a temperature-pressure relationship curve was established.
[0007] In conjunction with the first aspect, further, the process of performing dual correction on the tire pressure data and tire temperature data based on the pre-calibrated temperature-pressure relationship curve and the camber-tire force distribution model to obtain the equivalent corrected tire pressure value for each tire includes: Obtain the current total vehicle weight W and the distance from the rear axle to the center of gravity L. rear The wheelbase L, center of gravity height h, and track width T are input along with the slope data into a pre-built camber-tire force distribution model to obtain the load increment of each tire; wherein, the slope data includes pitch angle θ and roll angle φ. The load increment is converted into a tire pressure compensation amount, and the tire pressure compensation amount is superimposed on the tire pressure P1 after loading to obtain the tire pressure data after camber compensation. The tire pressure data after camber compensation is then subjected to temperature compensation to obtain the equivalent corrected tire pressure value for each tire after camber and temperature compensation.
[0008] In conjunction with the first aspect, the expression for the load increment of each tire is further as follows: Front axle load increment: ; Where, ΔF front Indicates the front axle load increment; W represents the total vehicle weight; θ represents the pitch angle; L represents the wheelbase; L rear Indicates the distance from the rear axle to the center of gravity; Rear axle load increment: ; Where, ΔF rear Indicates the rear axle load increment; Left tire load increment: ; Where, ΔF left Indicates the load increment on the left tire; Indicates the roll angle; h represents the center of gravity height; TL represents the track width; Right tire load increment: ; Where, ΔF right This indicates the load increment on the right tire.
[0009] In conjunction with the first aspect, further, the step of performing temperature compensation on the camber-compensated tire pressure data includes: (a) Static mode When the vehicle comes to a complete stop, the initial tire pressure P0 and initial tire temperature T0 before loading are collected by the tire pressure sensor, and the tire pressure P1 and temperature T1 after loading are collected. Combined with the compensation parameters, the effective tire pressure change is calculated. The effective tire pressure change is superimposed with the initial tire pressure P0 to output the temperature-corrected tire pressure value in static mode. (b) Dynamic mode When the vehicle is in motion, the inertial measurement unit identifies a smooth road segment window within a preset time period, and filters the tilt-compensated tire pressure data within the smooth road segment window to obtain filtered tire pressure data. Calculate the mean of filtered tire pressure data and the mean of tire temperature data for all sampling points; The difference between the average tire temperature data and the initial tire temperature is compared. If the difference is less than a preset temperature threshold, a linear temperature compensation model is used to calculate the tire pressure offset; if the difference is greater than the preset temperature threshold, a nonlinear temperature compensation model is used to calculate the tire pressure offset. The temperature-corrected tire pressure value in dynamic mode is obtained by subtracting the tire pressure offset from the average tire pressure value.
[0010] In conjunction with the first aspect, the expression for the effective pressure change is further as follows: △P 有效 = (P1-P0) - k(T1-T0) Among them, △P 有效 P0 represents the effective tire pressure change; P1 represents the initial tire pressure; T0 represents the tire pressure after loading; T1 represents the initial tire temperature; and k represents the temperature compensation coefficient.
[0011] In conjunction with the first aspect, the expression for calculating the tire pressure offset using the linear temperature compensation model is as follows: ΔP = k⋅ΔT; Where ΔP represents the pressure deviation caused by temperature change; k represents the linear temperature compensation coefficient; and ΔT represents the temperature change. The expression for calculating tire pressure offset using the nonlinear temperature compensation model is as follows: P(T) = P0 + aT + bT 2 ; Where P(T) represents the tire pressure value after temperature compensation; P0 represents the initial tire pressure under calibrated load; a represents the first nonlinear temperature compensation coefficient; T represents the temperature value; and b represents the second nonlinear temperature compensation coefficient.
[0012] In conjunction with the first aspect, further, the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the camber-tire force distribution model are dynamically corrected, including: Based on the vehicle's driving conditions in different climate zones, tire pressure data, tire temperature, and load status are recorded in real time at a sampling frequency of 1Hz to generate a validation dataset. The validation dataset is input into the initial temperature-pressure relationship curve to obtain the tire pressure prediction error value; If the tire pressure prediction error value is greater than the preset first error threshold, then the temperature compensation coefficient in the temperature-pressure relationship curve is adjusted in segments. When a vehicle passes over the weighbridge, the actual load weight measured by the weighbridge is automatically acquired and recorded. The vehicle load weight output by the load calculation unit is compared with the load weight measured by the weighbridge, and the error between the two is calculated. If the error exceeds a preset second error threshold, the mechanical parameters in the camber-tire force distribution model are corrected.
[0013] In conjunction with the first aspect, the data processing module further includes a signal filtering unit for performing Kalman filtering on the tire pressure data within the smooth driving section window.
[0014] Secondly, a vehicle equipped with the multi-sensor fusion-based load estimation system described in the first aspect.
[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: This invention uses a tire pressure sensor to acquire tire pressure and temperature data in real time, a tilt sensor to collect vehicle slope data, and an inertial measurement unit to monitor vibration and identify stable driving windows. The data processing module uses a pre-calibrated temperature-pressure curve and a tilt-tire force distribution model to perform dual temperature and tilt correction on the tire pressure / temperature data within the stable driving window, outputting the equivalent corrected tire pressure value for each tire. The load calculation unit queries the load-pressure mapping table based on the corrected tire pressure value to output the vehicle weight. The self-learning module dynamically corrects the temperature compensation coefficient and tilt model mechanical parameters (such as the center of gravity position L) by comparing with the actual weight measured on a weighbridge. rear (Height h), closed-loop update mapping table, thereby realizing dynamic monitoring during driving, with an error of <2%; temperature / slope interference is eliminated through multi-source compensation, and the self-learning mechanism suppresses the impact of tire aging, which is significantly better than the traditional tire pressure solution with 15% accuracy. Attached Figure Description
[0016] Figure 1 The figure shown is a block diagram of a load estimation system based on multi-sensor fusion provided by an embodiment of the present invention; Figure 2 The figure shown is an installation diagram of the tire pressure sensor provided in an embodiment of the present invention; Figure 3 The diagram shown is an algorithm flowchart for a stationary vehicle mode provided in an embodiment of the present invention. Figure 4 The figure shown is a schematic diagram of the dynamic compensation model provided in an embodiment of the present invention; Figure 5 The diagram shown is a structural schematic of a heavy-duty truck provided in an embodiment of the present invention; In the diagram: 1. Tire pressure sensor; 2. Tire angle sensor; 3. Inertial testing unit; 4. Data processing model; 5. Self-learning module. Detailed Implementation
[0017] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.
[0018] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Example 1
[0019] See Figure 1 and Figure 5 A dynamic estimation system based on multi-sensor fusion includes: Tire pressure sensor 1 is used to collect tire pressure and tire temperature data in real time. Tilt sensor 2 is used to synchronously monitor the vehicle's slope data (pitch angle and roll angle data). Inertial testing unit 3 is used to monitor vehicle vibration and identify windows of smooth driving sections; Data processing module 4 includes: The signal filtering unit is used to perform Kalman filtering on the tire pressure data within the window of a smooth driving section to eliminate driving vibration noise; The dynamic compensation unit is used to perform dual correction on the original tire pressure data and tire temperature data in the window of the smooth driving section according to the pre-calibrated temperature-pressure relationship curve and the tilt angle-tire force distribution model, so as to obtain the equivalent corrected tire pressure value of each tire. The load calculation unit is used to look up the pre-calibrated load-pressure mapping table based on the equivalent corrected tire pressure value of each tire, and directly output the vehicle load weight. Self-learning module 5 is used to compare the load weight with the actual load weight measured by the weighbridge, and dynamically correct the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the tilt angle-tire force distribution model based on the comparison results, so as to update the load-pressure mapping table.
[0020] In summary, by employing both static and dynamic modes (rapid measurement while parked / dynamic monitoring while in motion) and self-learning capabilities, we can overcome the dependence on weighbridges, improve logistics efficiency, reduce long-term maintenance costs, achieve continuous high-precision weighing during operation (final load error <2%), and eliminate errors of over 15% caused by temperature, slope, and vibration in uncompensated solutions. Example 2
[0021] Based on the system architecture and functional modules (including tire pressure sensor, tilt sensor, data processing module, and self-learning module) described in Example 1, this example further details the specific hardware deployment method of the system, the collaborative working mechanism between sensors, and the pre-calibration process of the key model on which the system relies to achieve high-precision estimation, as follows: See Figure 1 , Figure 2 and Figure 5 The system hardware adopts a distributed layout: a high-precision tire pressure sensor 1 is installed at the valve position of each tire to achieve independent monitoring of all wheels; the tilt sensor 2 is fixed on the center axis of the vehicle chassis frame to ensure that the attitude detection reference is aligned with the vehicle's center of gravity axis; the inertial measurement unit 3 (IMU), the data processing module 4 and the self-learning module are integrated into the on-board ECU in the driver's cab.
[0022] The sensors ensure data coordination through heterogeneous network communication: the tire pressure sensor transmits data to the vehicle gateway via Bluetooth Low Energy, while the tilt sensor, IMU, and self-learning module interact with the data processing module in real time via a CAN bus, thus ensuring data synchronization and transmission reliability in complex electromagnetic environments. This deployment method significantly reduces the risk of single-point failures and provides stable multi-source input for dynamic compensation algorithms.
[0023] Furthermore, to support the operation of the dynamic compensation unit in the data processing module, the system relies on pre-calibrated temperature-pressure relationship curves and camber-tire force distribution models; the pre-calibration process of the temperature-pressure relationship curves and the calibration process of the camber-tire force distribution models include: (1) Calibration of temperature-pressure relationship curve In a temperature-controlled test chamber, the tire type (e.g., steel-belted tire, nylon tire), initial tire pressure P0 (setting the standard tire pressure value when unloaded, such as 80 psi), and load status (unloaded / half-loaded / fully loaded) are determined. The chamber temperature is adjusted according to a preset temperature gradient, and tire pressure data is collected after each gradient stabilizes. The ambient temperature range is -20℃ to 60℃. For example, the tire is placed in the temperature-controlled chamber to simulate different environmental conditions for testing. The temperature is increased / decreased in 5℃ increments from -20℃ to 60℃, and tire pressure data is recorded after each stage stabilizes for 30 minutes. Repeat the above process to obtain datasets under no-load, half-load, and full-load conditions, and establish temperature-pressure relationship curves based on these datasets.
[0024] (2) Calibration of the camber angle-tire force distribution model: The vehicle is parked on an adjustable tilt platform, and the pitch angle (-10° to +10°) and roll angle (-5° to +5°) are adjusted step by step. At each angle, the force on each tire and the corresponding roll angle are recorded. A high-precision inclinometer is used to accurately measure the roll angle of the vehicle. A six-axis force sensor is installed on the suspension system of the vehicle body, which can directly measure the force on the tires and dynamically weigh the total weight of the vehicle using a weighbridge.
[0025] See Figure 4 The tire pressure and tire temperature data are double-calibrated using a pre-calibrated temperature-pressure relationship curve and a camber-tire force distribution model to obtain the equivalent corrected tire pressure value for each tire. This process includes the following steps: Step S31: Obtain the current total vehicle weight W and the distance L from the rear axle to the center of gravity. rear The wheelbase L, center of gravity height h, and track width T are input along with the slope data into a pre-built camber-tire force distribution model to obtain the load increment of each tire; the slope data includes pitch angle θ and roll angle φ. The expression for the load increment of each tire is: Front axle load increment: ; Where, ΔF front Indicates the front axle load increment; W represents the total vehicle weight; θ represents the pitch angle; L represents the wheelbase; L rear Indicates the distance from the rear axle to the center of gravity; Rear axle load increment: ; Where, ΔF rear Indicates the rear axle load increment; Left tire load increment: ; Where, ΔF left Indicates the load increment on the left tire; Indicates the roll angle; h represents the center of gravity height; TL represents the track width; Right tire load increment: ; Where, ΔF right This indicates the load increment on the right tire.
[0026] Step S32: Convert the load increment into tire pressure compensation amount, and add the tire pressure compensation amount to the tire pressure P1 after loading to obtain the tire pressure data after camber compensation. Step S33: Perform temperature compensation on the tire pressure data after camber compensation to obtain the equivalent corrected tire pressure value of each tire after camber and temperature compensation.
[0027] For details, see Figure 3 The system automatically switches operating modes based on the vehicle's operating status, including static mode and dynamic mode.
[0028] (a) Static mode When the vehicle comes to a complete stop, the initial tire pressure P0 and initial tire temperature T0 before loading are collected by the tire pressure sensor, and the tire pressure P1 and temperature T1 after loading are collected. Combined with compensation parameters (such as the temperature compensation coefficient k), the effective tire pressure change is calculated, and its expression is as follows: △P 有效 = (P1-P0) - k(T1-T0); Among them, △P 有效 P0 represents the effective tire pressure change; P1 represents the initial tire pressure; T0 represents the tire pressure after loading; T1 represents the initial tire temperature; and k represents the temperature compensation coefficient.
[0029] The effective pressure change is superimposed with the initial tire pressure P0 to output the temperature-corrected tire pressure value in static mode.
[0030] (b) Dynamic mode When the vehicle is in motion, the inertial measurement unit identifies a window of stable road sections within a preset time period, and filters the tire pressure data after tilt compensation within the window of stable road sections to obtain filtered tire pressure data. Calculate the mean of filtered tire pressure data and the mean of tire temperature data for all sampling points; The difference between the average tire temperature data and the initial tire temperature is compared. If the difference is less than a preset temperature threshold, a linear temperature compensation model is used to calculate the tire pressure deviation; if the difference is greater than the preset temperature threshold, a nonlinear temperature compensation model is used to calculate the tire pressure deviation. The expressions for calculating the tire pressure deviation using the linear / nonlinear temperature compensation models are as follows: Linear model (small temperature range) ΔP = k⋅ΔT; Where ΔP represents the pressure deviation caused by temperature change; k represents the linear temperature compensation coefficient; and ΔT represents the temperature change.
[0031] Nonlinear model (wide temperature range) P(T) = P0 + aT + bT 2 ; Where P(T) represents the tire pressure value after temperature compensation; P0 represents the initial tire pressure under calibrated load; a represents the first nonlinear temperature compensation coefficient; T represents the temperature value; and b represents the second nonlinear temperature compensation coefficient.
[0032] The temperature-corrected tire pressure value in dynamic mode is obtained by subtracting the tire pressure offset from the average tire pressure value.
[0033] Simultaneously, the system continuously optimizes compensation parameters through a self-learning module, dynamically correcting the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the camber-tire force distribution model, including: Based on the vehicle's driving conditions in different climate zones (such as driving in high-temperature deserts / low-temperature areas and on slopes or curves), tire pressure data, tire temperature data, and load status are recorded in real time at a sampling frequency of 1Hz to generate a validation dataset. The validation dataset is input into the temperature-pressure relationship curve (temperature-controlled experimental chamber construction) to obtain the tire pressure prediction error value; If the tire pressure prediction error is greater than the preset first error threshold, the temperature compensation coefficient in the temperature-pressure relationship curve will be adjusted in segments. Specifically, in this embodiment of the invention, the first error threshold is set to 5%. The tire pressure data obtained from the temperature control test chamber is compared with the tire pressure data under real conditions. If the error exceeds 5%, the model is corrected, for example, the k value in the high temperature zone is increased by 10% to improve the prediction accuracy of the model in this range.
[0034] When a vehicle passes over the weighbridge, the actual load weight measured by the weighbridge is automatically acquired and recorded. The vehicle load weight output by the load calculation unit is compared with the load weight measured by the weighbridge, and the error between the two is calculated. If the error exceeds the preset second error threshold (e.g., the critical value is also set to 5%), then the mechanical parameters in the camber-tire force distribution model are corrected.
[0035] It should be noted that as tires age during long-term use, the response characteristics of air pressure to temperature changes may change, such as a decrease in the temperature compensation coefficient k. Therefore, it is necessary to record operational data regularly through the vehicle system and update the model's parameters, such as a and b, based on this operational data to ensure the model's accuracy and reliability. Example 3
[0036] A vehicle is equipped with a load estimation system based on multi-sensor fusion as described in Examples 1 and 2.
[0037] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.
Claims
1. A load estimation system based on multi-sensor fusion, characterized in that, include: The data acquisition module includes: Tire pressure sensor (1) is used to acquire tire pressure data and tire temperature data; Tilt sensor (2) is used to acquire vehicle slope data; An inertial testing unit (3) is used to monitor the vibration state of the vehicle and identify windows of smooth driving sections; The data processing module (4) includes: The dynamic compensation unit is used to perform dual correction on the tire pressure data and tire temperature data in the window of the smooth driving section according to the pre-calibrated temperature-pressure relationship curve and the tilt angle-tire force distribution model, so as to obtain the equivalent corrected tire pressure value of each tire. The load calculation unit is used to query the pre-calibrated load-pressure mapping table based on the equivalent corrected tire pressure value of each tire and output the vehicle load weight. The self-learning module (5) is used to compare the load weight with the load weight measured by the weighbridge, and dynamically correct the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the tilt angle-tire force distribution model based on the comparison results, so as to update the load-pressure mapping table.
2. The load estimation system based on multi-sensor fusion according to claim 1, characterized in that, The pre-calibration process of the temperature-pressure relationship curve includes: In the temperature-controlled test chamber, the tire type, initial tire pressure P0 and load status are determined, and the temperature inside the chamber is adjusted according to the preset temperature steps. Tire pressure data is collected after the temperature stabilizes at each step. Repeat the process of collecting tire pressure data after each step of constant temperature stabilization to obtain datasets under no-load, half-load, and full-load conditions respectively; Based on the dataset, a temperature-pressure relationship curve was established.
3. The load estimation system based on multi-sensor fusion according to claim 1, characterized in that, The tire pressure and tire temperature data are double-corrected based on a pre-calibrated temperature-pressure relationship curve and a camber-tire force distribution model to obtain equivalent corrected tire pressure values for each tire, including: Obtain the current total vehicle weight W and the distance from the rear axle to the center of gravity L. rear The wheelbase L, center of gravity height h, and track width T are input along with the slope data into a pre-built camber-tire force distribution model to obtain the load increment of each tire; wherein, the slope data includes pitch angle θ and roll angle φ. The load increment is converted into a tire pressure compensation amount, and the tire pressure compensation amount is superimposed on the tire pressure P1 after loading to obtain the tire pressure data after camber compensation. The tire pressure data after camber compensation is then subjected to temperature compensation to obtain the equivalent corrected tire pressure value for each tire after camber and temperature compensation.
4. The load estimation system based on multi-sensor fusion according to claim 3, characterized in that, The expression for the load increment of each tire is: Front axle load increment: ； Where, ΔF front Indicates the front axle load increment; W represents the total vehicle weight; θ represents the pitch angle; L represents the wheelbase; L rear Indicates the distance from the rear axle to the center of gravity; Rear axle load increment: ； Where, ΔF rear Indicates the rear axle load increment; Left tire load increment: ； Where, ΔF left Indicates the load increment on the left tire; Indicates the roll angle; h represents the center of gravity height; TL represents the track width; Right tire load increment: ； Where, ΔF right This indicates the load increment on the right tire.
5. The load estimation system based on multi-sensor fusion according to claim 3, characterized in that, The step of performing temperature compensation on the tire pressure data after tilt angle compensation includes: Static mode When the vehicle comes to a complete stop, the initial tire pressure P0 and initial tire temperature T0 before loading and the tire pressure P1 and temperature T1 after loading are collected by the tire pressure sensor (1), and the effective air pressure change is calculated in combination with the compensation parameters. The effective tire pressure change is superimposed with the initial tire pressure P0 to output the temperature-corrected tire pressure value in static mode. (b) Dynamic mode When the vehicle is driving, the inertial measurement unit (5) identifies a smooth road section window within a preset time period, and filters the tilt-compensated tire pressure data within the smooth road section window to obtain filtered tire pressure data. Calculate the mean of filtered tire pressure data and the mean of tire temperature data for all sampling points; The difference between the average tire temperature data and the initial tire temperature is compared. If the difference is less than a preset temperature threshold, a linear temperature compensation model is used to calculate the tire pressure offset; if the difference is greater than the preset temperature threshold, a nonlinear temperature compensation model is used to calculate the tire pressure offset. The temperature-corrected tire pressure value in dynamic mode is obtained by subtracting the tire pressure offset from the average tire pressure value.
6. The load estimation system based on multi-sensor fusion according to claim 5, characterized in that, The expression for the effective pressure change is: △P 有效 = (P1-P0) - k(T1-T0) Among them, △P 有效 P0 represents the effective tire pressure change; P1 represents the initial tire pressure; T0 represents the tire pressure after loading; T1 represents the initial tire temperature; and k represents the temperature compensation coefficient.
7. The load estimation system based on multi-sensor fusion according to claim 5, characterized in that, The expression for calculating tire pressure offset using the linear temperature compensation model is as follows: ΔP = k⋅ΔT; Where ΔP represents the pressure deviation caused by temperature change; k represents the linear temperature compensation coefficient; and ΔT represents the temperature change. The expression for calculating tire pressure offset using the nonlinear temperature compensation model is as follows: P(T) = P0 + aT + bT 2 ; Where P(T) represents the tire pressure value after temperature compensation; P0 represents the initial tire pressure under calibrated load; a represents the first nonlinear temperature compensation coefficient; T represents the temperature value; and b represents the second nonlinear temperature compensation coefficient.
8. The load estimation system based on multi-sensor fusion according to claim 1, characterized in that, Dynamically correcting the compensation parameters in the temperature-pressure relationship curve and the mechanical parameters in the camber-tire force distribution model includes: Based on the vehicle's driving conditions in different climate zones, tire pressure data, tire temperature, and load status are recorded in real time at a sampling frequency of 1Hz to generate a validation dataset. The validation dataset is input into the initial temperature-pressure relationship curve to obtain the tire pressure prediction error value; If the tire pressure prediction error value is greater than the preset first error threshold, then the temperature compensation coefficient in the temperature-pressure relationship curve is adjusted in segments. When a vehicle passes over the weighbridge, the actual load weight measured by the weighbridge is automatically acquired and recorded. The vehicle load weight output by the load calculation unit is compared with the load weight measured by the weighbridge, and the error between the two is calculated. If the error exceeds a preset second error threshold, the mechanical parameters in the camber-tire force distribution model are corrected.
9. The load estimation system based on multi-sensor fusion according to claim 1, characterized in that, The data processing module also includes a signal filtering unit, which performs Kalman filtering on the tire pressure data within the window of the smooth driving section.
10. A vehicle, characterized in that, The system is equipped with a load estimation system based on multi-sensor fusion as described in any one of claims 1-9.
Citation Information
Cited By
Method and system for determining automatic metering of mining equipment
CN121384208A