Vehicle mass estimation method based on multi-sensor fusion

Abstract

The application belongs to the field of vehicle mass estimation, and specifically discloses a vehicle mass estimation method based on multi-sensor fusion, which comprises the following steps: acquiring a stroke sensor signal, a tire pressure signal and a vehicle running state signal of a vehicle; calculating a first mass estimation value m1 according to the stroke sensor signal, calculating a second mass estimation value m2 based on the tire pressure signal, and calculating a third mass estimation value m3 based on a vehicle longitudinal dynamics equation and using a Kalman filtering algorithm according to the vehicle running state signal; dynamically adjusting a first weight coefficient w1, a second weight coefficient w2 and a third weight coefficient w3 used for fusing the first mass estimation value m1, the second mass estimation value m2 and the third mass estimation value m3; and performing weighted fusion to obtain a final estimation value m of the vehicle mass, wherein m=w1m1+w2m2+w3m3. The application can maintain high-precision and high-stability vehicle mass estimation under various working conditions.

Description

Technical Field

[0001] This application belongs to the field of vehicle mass estimation, and more specifically, relates to a vehicle mass estimation method based on multi-sensor fusion. Background Technology

[0002] Vehicle weight estimation and overload alarm are core technologies for ensuring vehicle driving safety and extending vehicle service life. Currently, various estimation schemes exist, but each has its limitations. For example, tire pressure sensor-based schemes are significantly affected by temperature and lack a clear temperature-pressure correspondence, resulting in poor load estimation accuracy. Schemes based on stroke sensors require additional sensors installed on the vehicle's load axles, increasing vehicle manufacturing costs. Schemes based on longitudinal motion equations and using Kalman filtering algorithms can only be used when the vehicle is in motion; they cannot estimate loads when the vehicle is static, and errors are significant at low speeds. While strain gauge-based methods offer high accuracy, they are complex to install and costly. These schemes typically rely on a single sensing principle or a fixed calculation model, making it difficult to maintain high accuracy and stability under different operating conditions (such as static, dynamic, and different temperature environments).

[0003] Therefore, how to achieve a vehicle mass estimation method that can maintain high accuracy and stability under various operating conditions is an urgent problem to be solved. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the purpose of this application is to provide a vehicle mass estimation method based on multi-sensor fusion, which can maintain high accuracy and high stability in vehicle mass estimation under various operating conditions.

[0005] To achieve the above objectives, in a first aspect, this application provides a vehicle mass estimation method based on multi-sensor fusion, comprising the following steps:

[0006] S10: Acquire vehicle travel sensor signals, tire pressure signals, tire temperature signals, and vehicle operating status signals; S20, a first mass estimate m1 is calculated based on the travel sensor signal through the correspondence between axle load and travel; a basic load estimate is determined based on the tire pressure signal through the correspondence between tire pressure and load; and the basic load estimate is temperature-corrected using the tire temperature signal to obtain a second mass estimate m2; and a third mass estimate m3 is calculated based on the vehicle operating state signal using the vehicle longitudinal dynamics equation and a Kalman filter algorithm. S30, when the vehicle is started but not yet moving, a reference value m0 for the vehicle mass is determined based on the first mass estimate m1 and the second mass estimate m2; S40, during vehicle operation, based on the tire temperature signal and the deviation between the third mass estimate m3 and the reference value m0, dynamically adjust the first weighting coefficient w1, the second weighting coefficient w2 and the third weighting coefficient w3 used to fuse the first mass estimate m1, the second mass estimate m2 and the third mass estimate m3; S50, based on the adjusted first weighting coefficient w1, second weighting coefficient w2, and third weighting coefficient w3, the first mass estimate m1, the second mass estimate m2, and the third mass estimate m3 are weighted and fused to obtain the final estimated value m of the vehicle mass, where m = w1. m1+w2 m2+w3 m3.

[0007] As a further preferred embodiment, the step of calculating the first mass estimate value m1 based on the stroke sensor signal includes: Based on the voltage value corresponding to the travel sensor signal installed on the front axle of the vehicle, the estimated value of the front axle load m1f is obtained through the preset correspondence between the front axle load and the voltage. Based on the voltage value corresponding to the travel sensor signal installed on the rear axle of the vehicle, the estimated value of the rear axle load m1r is obtained through the preset correspondence between the rear axle load and the voltage. The first mass estimate m1 is the sum of the front axle load estimate m1f and the rear axle load estimate m1r, i.e., m1 = m1f + m1r.

[0008] As a further preferred embodiment, the step of calculating the first mass estimate value m1 based on the stroke sensor signal includes: Based on the voltage value corresponding to the travel sensor signal installed on the front axle of the vehicle, the estimated value of the front axle load m1f is obtained through the preset correspondence between the front axle load and the voltage. Based on the voltage value corresponding to the travel sensor signal installed on the rear axle of the vehicle, the estimated value of the rear axle load m1r is obtained through the preset correspondence between the rear axle load and the voltage. The first mass estimate m1 is the sum of the front axle load estimate m1f and the rear axle load estimate m1r, i.e., m1 = m1f + m1r.

[0009] As a further preferred embodiment, the reference value m0 is determined by averaging the first mass estimate m1 and the second mass estimate m2, i.e., m0 = 0.5 m1 + 0.5 m2.

[0010] As a further preferred embodiment, the dynamic adjustment of the first weighting coefficient w1, the second weighting coefficient w2, and the third weighting coefficient w3 includes: When the tire temperature indicated by the tire temperature signal is within the preset temperature range [T0, T1], and the relative deviation between the third mass estimation value m3 and the reference value m0 is not greater than a preset threshold, the first weighting coefficient w1, the second weighting coefficient w2 and the third weighting coefficient w3 are adjusted to the first preset value, the second preset value and the third preset value, respectively. When the tire temperature indicated by the tire temperature signal is outside the preset temperature range [T0, T1], the second weighting coefficient w2 is set to zero, and the first weighting coefficient w1 and the third weighting coefficient w3 are adjusted to the fourth preset value and the fifth preset value, respectively. The preset temperature range [T0, T1] is -5 degrees Celsius to 50 degrees Celsius, and the preset threshold is 20%.

[0011] As a further preferred embodiment, the first preset value is 0.48, the second preset value is 0.32, the third preset value is 0.2, the fourth preset value is 0.8, and the fifth preset value is 0.2.

[0012] As a further preferred embodiment, during vehicle operation, the method further includes a step of determining the reliability of the calculations of the first mass estimate m1, the second mass estimate m2, and the third mass estimate m3: Within the first preset time after the vehicle starts, if the vehicle speed error is less than a preset speed error threshold, the mass estimation value obtained in the current calculation cycle is determined to be reliable and used in subsequent steps; otherwise, the mass estimation value obtained in the current calculation cycle is discarded. The first preset duration is 30 seconds, and the preset speed error threshold is 5 kilometers per hour.

[0013] As a further preferred embodiment, for vehicles whose load does not change during operation, the reference value m0 is the mass of the vehicle in this operating cycle.

[0014] As a further preferred option, for vehicles whose load changes during operation, steps S40 and S50 are continuously executed throughout the entire vehicle operation cycle to perform real-time estimation of the vehicle mass.

[0015] Secondly, this application provides a vehicle weight estimation device, including a processor and a memory, the memory storing a computer program, the processor executing the computer program to implement the steps of the method as described in any one of the above.

[0016] This application achieves the following advantages: by parallel computing of three mass estimates based on travel, tire pressure (temperature corrected), and vehicle longitudinal dynamics, and innovatively introducing a dynamic weight adjustment mechanism based on real-time tire temperature and dynamic estimation deviation, it realizes adaptive weighted fusion of multiple estimates. This method overcomes the inherent limitations of single-sensor or single-model schemes under specific operating conditions. For example, static schemes cannot be used when the vehicle is in motion, tire pressure schemes are significantly affected by temperature, and dynamic schemes have large errors at low speeds. Through the complementarity of multi-source information, when the vehicle is stationary, travel and tire pressure estimations can be mainly relied upon, while dynamic estimations can be introduced when the vehicle is in motion, and the reliability judgment of tire pressure estimation can be corrected in real time by temperature signals. This core mechanism of dynamically adjusting the fusion weights based on operating condition information enables the final mass estimation result to adapt to various operating conditions, such as vehicle movement from stationary to moving, from low speed to high speed, and different ambient temperatures, thereby maintaining high estimation accuracy and stability under various complex operating conditions. Attached Figure Description

[0017] Figure 1 This is a flowchart of the vehicle mass estimation method based on multi-sensor fusion provided in this application; Figure 2 This is a schematic diagram of the installation position of the stroke sensor provided in an embodiment of this application; Figure 3 This is a schematic diagram showing the relationship between the stroke sensor signal voltage and the rear axle load provided in an embodiment of this application; Figure 4 This is a schematic diagram illustrating the relationship between tire pressure and wheel end load provided in an embodiment of this application; Figure 5 This is a flowchart of an algorithm for estimating mass m3 based on the longitudinal dynamic equation and Kalman filter algorithm provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] like Figure 1 As shown, this application provides a vehicle mass estimation method based on multi-sensor fusion, including steps S10 to S50, which are detailed below: Step S10: Acquire the vehicle's travel sensor signal, tire pressure signal, tire temperature signal, and vehicle operating status signal.

[0020] This step, by simultaneously acquiring multi-source signals from the travel sensor, tire pressure sensor, and vehicle operating system, provides comprehensive data input reflecting the vehicle's suspension deformation, tire condition, and kinematic state, laying a data foundation for subsequent mass estimation using different principles.

[0021] Step S20: Calculate the first mass estimate m1 based on the correspondence between axle load and stroke using the stroke sensor signal; determine the basic load estimate based on the correspondence between tire pressure and load using the tire pressure signal; perform temperature correction on the basic load estimate using the tire temperature signal to obtain the second mass estimate m2; and calculate the third mass estimate m3 based on the vehicle operating status signal using the vehicle longitudinal dynamics equation and the Kalman filter algorithm.

[0022] This step allows for the parallel execution of three independent mass estimation processes. Specifically, calculating m1 based on the axle load and travel relationship provides a relatively accurate estimate when the vehicle is stationary or moving smoothly; calculating m2 based on tire pressure and temperature correction comprehensively utilizes tire pressure information and compensates for the effects of temperature changes; and calculating m3 based on the longitudinal dynamic equations and Kalman filtering provides an effective estimate when the vehicle is in motion. This parallel processing approach ensures that at least one estimation method can leverage its advantages under different operating conditions.

[0023] Step S30: When the vehicle is started but not yet moving, determine the reference value m0 of the vehicle mass based on the first mass estimate m1 and the second mass estimate m2.

[0024] This step establishes a mass baseline value using m1 and m2, which can be reliably calculated without requiring the vehicle to be in motion, under static conditions. This provides a stable and reliable initial reference anchor point for judging the magnitude of mass changes and adjusting the estimation strategy during subsequent vehicle operation.

[0025] Step S40: During vehicle operation, based on the tire temperature signal and the deviation between the third mass estimate m3 and the reference value m0, dynamically adjust the first weighting coefficient w1, the second weighting coefficient w2, and the third weighting coefficient w3 used to fuse the first mass estimate m1, the second mass estimate m2, and the third mass estimate m3.

[0026] This step introduces an adaptive weight adjustment mechanism. This mechanism assesses the reliability of the tire pressure estimate based on real-time tire temperature, and evaluates the credibility of the current motion state estimate based on the deviation between the dynamic estimate m3 and the static baseline m0. Based on this, it dynamically assigns weights to the three estimates in the final result. This mechanism allows the fusion strategy to flexibly respond to changes in ambient temperature and vehicle motion state.

[0027] Step S50: Based on the adjusted first weighting coefficient w1, second weighting coefficient w2 and third weighting coefficient w3, the first mass estimate m1, the second mass estimate m2 and the third mass estimate m3 are weighted and fused to obtain the final estimated value m of the vehicle mass.

[0028] This step generates a comprehensive quality estimation result in real time by applying dynamically adjusted weighting coefficients to weight and sum the three estimates. Theoretically, this result can combine the stability of static estimation methods with the responsiveness of dynamic estimation methods. Furthermore, through adaptive weight allocation, it automatically reduces the impact of potentially erroneous estimation methods under specific operating conditions, thereby improving the robustness of the overall output.

[0029] The vehicle mass estimation method based on multi-sensor fusion provided in this application has the following advantages: It achieves adaptive weighted fusion of multiple estimates by parallelly calculating three mass estimates based on travel, tire pressure (temperature corrected), and vehicle longitudinal dynamics, and innovatively introduces a dynamic weight adjustment mechanism based on real-time tire temperature and dynamic estimation deviation. This method overcomes the inherent limitations of single-sensor or single-model schemes under specific operating conditions. For example, static schemes cannot be used when the vehicle is in motion, tire pressure schemes are significantly affected by temperature, and dynamic schemes have large errors at low speeds. Through the complementarity of multi-source information, travel and tire pressure estimations can be mainly relied upon when the vehicle is stationary, while dynamic estimation can be introduced when the vehicle is in motion, and the reliability judgment of tire pressure estimation can be corrected in real time by temperature signals. This core mechanism of dynamically adjusting the fusion weights according to operating condition information enables the final mass estimation result to adapt to various operating conditions, including vehicle movement from stationary to moving, low speed to high speed, and different ambient temperatures, thus maintaining high estimation accuracy and stability under various complex operating conditions.

[0030] In one embodiment, the technical solution to achieve the above objective can specifically be as follows: This embodiment provides a vehicle mass estimation method based on multi-sensor fusion, including the following steps: 1. For example Figure 2 As shown, travel sensors are installed on the front and rear axles of the vehicle. There is a linear relationship between the axle load and the travel sensor readings; different travel distances correspond to different voltages, and different voltages correspond to different axle loads. Taking the rear axle as an example, when the rear axle load changes, the travel sensor signal voltage changes synchronously, as shown... Figure 3 As shown, different voltages correspond to different rear axle loads.

[0031] 2. Similarly, different values ​​of the front axle load sensor voltage also correspond to different front axle loads m1f, and the total vehicle mass can be m1 = m1r + m1f.

[0032] 3. For vehicles equipped with tire pressure sensors, when the front tire pressure is normal, different vehicle weights will correspond to different tire pressures. Taking a specific wheel as an example, different tire pressure values ​​correspond to different wheel-end loads, such as... Figure 4 As shown, different tire pressures correspond to different loads.

[0033] The above diagram can be calibrated when the vehicle is stationary, with different loads applied and tire pressure recorded simultaneously to create a linear table. In actual vehicle applications, the corresponding load can be obtained by looking up the table based on the tire pressure sensor values. Assuming the load corresponding to the front left tire is m2f1, the front right tire is m2f2, the rear left tire is m2r1, and the rear right tire is m2r2, the total vehicle mass can be represented as m2 = a(m2r + m2f), where 'a' is a temperature check coefficient. This temperature coefficient represents the relationship between tire pressure and temperature and can be provided by the tire pressure sensor supplier.

[0034] 4. When the vehicle is moving, m3 can be obtained by using Kalman filtering of the longitudinal dynamic equation. The algorithm flowchart is as follows: Figure 5 As shown.

[0035] 5. After the vehicle is ignited and started, but before starting, perform calculations 1 and 2. These are m1 and m2 respectively. m0 = 0.5m1 + 0.5m2 is used as the baseline mass. For vehicles whose load does not change during operation, this value is the mass of the vehicle for this operating cycle.

[0036] 6. For vehicles whose load changes during operation, such as sanitation vehicles, the vehicle's mass changes continuously during operation and needs to be estimated in real time. After the vehicle starts, three calculation methods are performed simultaneously. Within a time range t1 (generally 30 seconds, calibrable), if the vehicle speed error is less than v0 (generally 5 km / h, calibrable), the above calculations have 100% reliability; otherwise, they are discarded. For [(m3 - m0) / m0] <= 20%, m3 is accepted. For the tire temperature range T0-T1 (generally -5 to 50 degrees Celsius), m2 has 100% feasibility. Therefore, the real-time vehicle mass is: m = 0.8 (0.6 m1 + 0.4 m2) +0.2 m3; for tires whose temperature range is not within T0-T1, m = 0.8 m1 + 0.2 m3.

[0037] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A vehicle mass estimation method based on multi-sensor fusion, characterized in that, Includes the following steps: S10: Acquire vehicle travel sensor signals, tire pressure signals, tire temperature signals, and vehicle operating status signals; S20, a first mass estimate m1 is calculated based on the travel sensor signal through the correspondence between axle load and travel; a basic load estimate is determined based on the tire pressure signal through the correspondence between tire pressure and load; and the basic load estimate is temperature-corrected using the tire temperature signal to obtain a second mass estimate m2; and a third mass estimate m3 is calculated based on the vehicle operating state signal using the vehicle longitudinal dynamics equation and a Kalman filter algorithm. S30, when the vehicle is started but not moving, a reference value m0 of the vehicle mass is determined based on the first mass estimate m1 and the second mass estimate m2; S40, during vehicle operation, based on the tire temperature signal and the deviation between the third mass estimate m3 and the reference value m0, dynamically adjust the first weighting coefficient w1, the second weighting coefficient w2 and the third weighting coefficient w3 used to fuse the first mass estimate m1, the second mass estimate m2 and the third mass estimate m3; S50, based on the adjusted first weighting coefficient w1, second weighting coefficient w2, and third weighting coefficient w3, the first mass estimate m1, the second mass estimate m2, and the third mass estimate m3 are weighted and fused to obtain the final estimated value m of the vehicle mass, where m = w1. m1+w2 m2+w3 m3.

2. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, The calculation of the first mass estimate value m1 based on the travel sensor signal includes: Based on the voltage value corresponding to the travel sensor signal installed on the front axle of the vehicle, the estimated value of the front axle load m1f is obtained through the preset correspondence between the front axle load and the voltage. Based on the voltage value corresponding to the travel sensor signal installed on the rear axle of the vehicle, the estimated value of the rear axle load m1r is obtained through the preset correspondence between the rear axle load and the voltage. The first mass estimate m1 is the sum of the front axle load estimate m1f and the rear axle load estimate m1r, i.e., m1 = m1f + m1r.

3. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, The calculation of the first mass estimate value m1 based on the travel sensor signal includes: Based on the voltage value corresponding to the travel sensor signal installed on the front axle of the vehicle, the estimated value of the front axle load m1f is obtained through the preset correspondence between the front axle load and the voltage. Based on the voltage value corresponding to the travel sensor signal installed on the rear axle of the vehicle, the estimated value of the rear axle load m1r is obtained through the preset correspondence between the rear axle load and the voltage. The first mass estimate m1 is the sum of the front axle load estimate m1f and the rear axle load estimate m1r, i.e., m1 = m1f + m1r.

4. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, The baseline value m0 is determined by averaging the first estimated mass value m1 and the second estimated mass value m2, i.e., m0 = 0.5 m1 + 0.5 m2.

5. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, The dynamic adjustment of the first weighting coefficient w1, the second weighting coefficient w2, and the third weighting coefficient w3 includes: When the tire temperature indicated by the tire temperature signal is within the preset temperature range [T0, T1], and the relative deviation between the third mass estimation value m3 and the reference value m0 is not greater than a preset threshold, the first weighting coefficient w1, the second weighting coefficient w2 and the third weighting coefficient w3 are adjusted to the first preset value, the second preset value and the third preset value, respectively. When the tire temperature indicated by the tire temperature signal is outside the preset temperature range [T0, T1], the second weighting coefficient w2 is set to zero, and the first weighting coefficient w1 and the third weighting coefficient w3 are adjusted to the fourth preset value and the fifth preset value, respectively. The preset temperature range [T0, T1] is -5 degrees Celsius to 50 degrees Celsius, and the preset threshold is 20%.

6. The vehicle mass estimation method based on multi-sensor fusion as described in claim 5, characterized in that, The first preset value is 0.48, the second preset value is 0.32, the third preset value is 0.2; the fourth preset value is 0.8, and the fifth preset value is 0.

2.

7. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, During vehicle operation, the process also includes a step of determining the reliability of the calculations of the first mass estimate m1, the second mass estimate m2, and the third mass estimate m3: Within the first preset time after the vehicle starts, if the vehicle speed error is less than a preset speed error threshold, the mass estimation value obtained in the current calculation cycle is determined to be reliable and used in subsequent steps; otherwise, the mass estimation value obtained in the current calculation cycle is discarded. The first preset duration is 30 seconds, and the preset speed error threshold is 5 kilometers per hour.

8. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, For vehicles whose load does not change during operation, the reference value m0 is the mass of the vehicle in this operating cycle.

9. The vehicle mass estimation method based on multi-sensor fusion as described in claim 1, characterized in that, For vehicles whose load changes during operation, steps S40 and S50 are continuously executed throughout the vehicle's entire operating cycle to perform real-time estimation of the vehicle's mass.

10. A vehicle weight estimation device, characterized in that, It includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method as described in any one of claims 1 to 9.

Images