Vehicle-mounted tire pressure dynamic adjusting system based on multi-mode road condition recognition and self-adaptive control method

Abstract

The invention discloses a vehicle-mounted tire pressure dynamic adjustment system based on multi-mode road condition recognition and a self-adaptive control method, and relates to the technical field of automobile control, the system comprises a multi-mode road condition recognition module, a driving style acquisition module, a tire pressure calculation module and a dynamic adjustment execution module; the multi-mode road condition recognition module recognizes the road condition type through fusion data of a millimeter wave radar, a camera, a vehicle body accelerometer and a tire deformation sensor. Multi-sensor fusion road condition recognition is adopted, the dependence on driver experience is eliminated, the recognition precision is greatly improved, meanwhile, dynamic inflation and deflation in the driving process can be achieved without parking, different road condition changes can be accurately responded in real time through the two-process design of partitioned polling inflation and vibration auxiliary deflation, and the driving safety is improved. And the adaptability of the vehicle to a complex environment is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of automotive control technology, specifically to an on-board tire pressure dynamic adjustment system and adaptive control method based on multimodal road condition recognition. Background Technology

[0002] During driving, different road conditions place different demands on tire pressure. For example, sandy and snowy terrain requires lower tire pressure to increase the contact area between the tire and the ground, improving grip; while rocky roads require higher tire pressure to prevent damage to the tires from sharp objects. At the same time, driving style also affects tire pressure requirements; aggressive driving may require more stable tire pressure. However, traditional tire pressure adjustment methods are difficult to meet these complex needs. With the development of automotive chassis control technology, achieving real-time, precise, and intelligent tire pressure adjustment has become an important research direction in the industry to improve vehicle adaptability and driving safety.

[0003] Traditional tire pressure monitoring systems (TPMS) have numerous shortcomings in terms of adjustment methods, environmental perception, adjustment precision, and safety protection, failing to meet the adaptability requirements of modern vehicles to complex road conditions. Regarding adjustment methods, traditional TPMS are mostly manual or require stopping for adjustment. This means they cannot respond in real-time to changes in road conditions while driving. When a vehicle transitions from one road condition to another, the driver needs to stop and manually adjust the tire pressure, which is not only time-consuming but may also affect driving continuity and safety. In terms of environmental perception, traditional systems rely heavily on the driver's experience to judge road conditions, making accuracy difficult to guarantee. Different drivers may have different judgments of road conditions, and drivers may not be able to perceive changes in road conditions in a timely and accurate manner, leading to untimely or inaccurate tire pressure adjustments. Regarding adjustment precision, traditional systems have a fixed threshold, failing to achieve fine-tuning. This coarse adjustment method may not allow tires to reach the optimal tire pressure under different road conditions, affecting vehicle performance and safety. Furthermore, traditional TPMS lacks protection against misoperation, posing a risk of over-inflation and over-deflation. Over-inflation may lead to tire blowout, while over-deflation affects vehicle handling and passability, increasing driving safety hazards. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an on-board tire pressure dynamic adjustment system and adaptive control method based on multimodal road condition recognition. It integrates multiple sensors, including millimeter-wave radar, cameras, vehicle accelerometers, and tire deformation sensors, and employs multi-sensor fusion technology to achieve accurate road condition recognition. Simultaneously, it collects driving operation data in real time through the on-board CAN bus and vehicle control system, generates driving style correction items, and calculates the target tire pressure using a road condition-tire pressure mapping model, thus achieving tire pressure adjustment. In terms of adjustment methods, it adopts a dual-process design of zoned polling inflation and vibration-assisted deflation, enabling dynamic adjustment of tire pressure during driving and real-time response to road condition changes without stopping. Furthermore, the system has a safety locking mechanism that disables the tire pressure adjustment function when the vehicle is detected to be in high-speed, large steering angle, or low adhesion coefficient conditions, ensuring vehicle driving safety.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On the one hand, an on-board tire pressure dynamic adjustment system based on multimodal road condition recognition, the system comprising: a multimodal road condition recognition module, a driving style acquisition module, a tire pressure calculation module, and a dynamic adjustment execution module; The multimodal road condition recognition module identifies road condition types by fusing data from millimeter-wave radar, cameras, vehicle accelerometers, and tire deformation sensors. The driving style acquisition module obtains operation data through the vehicle CAN bus, and combines the number of rapid accelerations and brakings recorded by the vehicle control system to generate driving style labels. The tire pressure calculation module uses a reference tire pressure as a basis, combined with fixed correction coefficients for different road conditions, temperature compensation based on the difference between tire temperature and 25°C, and driving style correction items, to calculate the target tire pressure through a road condition-tire pressure mapping model. The dynamic adjustment execution module includes an on-board air pump, a high-speed solenoid valve, and a tire sidewall vibrator. It performs inflation and deflation operations according to the target tire pressure and has a safety locking function.

[0006] Furthermore, the multimodal road condition recognition module includes a millimeter-wave radar, a camera, a vehicle accelerometer, a tire deformation sensor, and a road condition change rate monitoring unit. The millimeter-wave radar determines road conditions by detecting the intensity of ground echo scattering; low scattering corresponds to sandy or snowy terrain. The camera determines road conditions by detecting texture features; high edge density corresponds to rocky surfaces. The vehicle accelerometer determines road conditions by detecting the Z-axis vibration frequency spectrum; a peak value of 3-8Hz corresponds to muddy surfaces. The tire deformation sensor determines road conditions by detecting the rate of change of the sideslip angle; a sudden increase in the rate of change of the sideslip angle corresponds to soft ground. Simultaneously, by collecting data on the vehicle's rapid acceleration / deceleration frequency, steering amplitude, and speed fluctuations, the module identifies the driver's driving style.

[0007] Furthermore, the multimodal traffic condition recognition module identifies traffic condition types by fusing data, and its recognition formula is as follows: , in, It is a comprehensive road condition identification index, used to ultimately determine the road condition type. , , , These are the weighting coefficients of the recognition results from each sensor, summing to 1. They are set based on the recognition accuracy and reliability of the sensors. It is a millimeter-wave radar traffic identification index. It is the camera's visual road condition recognition index. It is the vehicle accelerometer road condition recognition index. It is the road condition recognition index of the tire deformation sensor.

[0008] Furthermore, the driving style acquisition module obtains driving operation data through the vehicle CAN bus and vehicle control system, specifically including: accelerator pedal opening change rate, brake pedal travel and application time, and steering wheel angular velocity; it also records the number of rapid accelerations per unit time, defined as accelerator pedal opening > 80% and change rate > 50% / s, and the number of emergency brakings, defined as brake pedal travel > 70% and application time < 0.5s, and generates driving style labels based on this data.

[0009] Furthermore, in the tire pressure calculation module, the fixed correction coefficients for different road conditions include 0.0 kPa for highways, -35.0 kPa for sandy areas, -28.0 kPa for muddy areas, -15.0 kPa for snowy areas, and +20.0 kPa for rocky areas.

[0010] Furthermore, the tire pressure calculation module calculates the target tire pressure using a road condition-tire pressure mapping model, and the calculation formula is as follows: , in, This is the target tire pressure, which is the final tire pressure value that needs to be adjusted to. That is the reference tire pressure. It is the first Correction factors for different road conditions It is the road condition recognition state coefficient. When the first road condition is recognized... When dealing with various road conditions, ,otherwise , This is the temperature compensation coefficient, with a value of 0.2. This is the current tire temperature. The ideal tire temperature is fixed at 25°C. This is the driving style correction factor, determined based on the driving style label. It is a quantitative value for driving style.

[0011] Furthermore, the dynamic adjustment execution module includes an on-board air pump, a high-speed solenoid valve, a tire deformation monitoring device, and an environmental perception center. The on-board air pump is an oil-free silent scroll compressor with a flow rate ≥35L / min and noise <65dB. The high-speed solenoid valve is driven by piezoelectric ceramics with a response time <50ms. The tire deformation monitoring uses a sidewall embedded strain gauge array with an accuracy of ±0.05%. The environmental perception center uses the NVIDIA Orin+ LiDAR point cloud classification algorithm with a processing latency <80ms.

[0012] Furthermore, the dynamic adjustment execution module detects that the current tire pressure is higher than the target tire pressure, initiates the deflation program, opens the high-speed solenoid valve of the corresponding tire and activates the tire sidewall vibrator to accelerate pressure release, monitors the tire pressure in real time, and closes and locks the valve after reaching the target value. When the current tire pressure is detected to be lower than the target tire pressure, the scroll compressor is started, and the solenoid valves of each tire are opened in the order of left front-right front-left rear-right rear. After the first round of inflation lasts 5 seconds, the valve is closed, the current tire pressure is detected and the difference from the target value is calculated. If the difference is >0.5psi, the inflation time of the next round is reduced by 30%, and the adjustment is carried out in turn until the tire pressure of all tires enters the target value ±0.1psi range. The compressor is then turned off and the valve is locked. In addition, when the vehicle speed is >60km / h, the curve radius is <50m, or the road surface adhesion coefficient is too low, the inflation and deflation functions are disabled.

[0013] On the other hand, a vehicle tire pressure adaptive control method based on multimodal road condition recognition includes: Road condition recognition: Multiple sensors simultaneously collect road condition-related data, make preliminary judgments according to their respective judgment logics, and fuse the judgment results of each sensor to determine the current road condition type; Driving style acquisition: The driving style acquisition module reads throttle, brake and steering operation data in real time through the vehicle CAN bus, counts the number of rapid acceleration and hard braking according to preset rules, and generates driving style labels and corresponding correction parameters. Tire pressure adjustment: Based on the identified road condition type, the corresponding road condition correction coefficient is retrieved from the preset correction coefficient table; the tire temperature is obtained in real time to calculate the temperature compensation value, and the driving style correction item generated by the driving style acquisition module is also obtained. By obtaining the reference tire pressure in the vehicle manual, and combining the road condition correction coefficient, temperature compensation value and driving style correction item, the target tire pressure is calculated through the road condition-tire pressure mapping model. Based on the difference between the target tire pressure and the current tire pressure, the inflation or deflation mode is selected for adjustment. Safety control: Real-time monitoring of tire temperature, deformation, and vehicle driving status. When the risk of over-inflation or deflation of the tires is detected, or when the vehicle is in high-speed, large steering angle, or low-adhesion conditions, a safety locking mechanism is triggered to disable the tire pressure regulation function.

[0014] Compared with existing technologies, this on-board tire pressure dynamic adjustment system and adaptive control method based on multimodal road condition recognition has the following advantages: I. This invention, by employing multi-sensor fusion for road condition recognition, eliminates reliance on driver experience and significantly improves recognition accuracy. At the same time, it enables dynamic inflation and deflation while driving without stopping. Through a dual-process design of zoned polling inflation and vibration-assisted deflation, it can respond to different road condition changes in real time and accurately, effectively improving the vehicle's adaptability to complex environments.

[0015] Second, this invention collects driving operation data in real time through a driving style acquisition module, generates driving style correction items, and performs personalized tire pressure adjustment based on driving habits, so that vehicle performance is more in line with the driver's needs and the overall adaptability and safety of the vehicle are enhanced.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 This is a block diagram of an on-board tire pressure dynamic adjustment system based on multimodal road condition recognition. Figure 2 This is a flowchart of an on-board tire pressure adaptive control method based on multimodal road condition recognition. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] Example 1 For off-road scenarios in snow, the multimodal road condition recognition module activates when the vehicle enters snow from an asphalt road. At this time, the outdoor temperature is -8℃, and the road surface has a complex situation with alternating layers of thin ice and compacted snow. The millimeter-wave radar first captures the ground echo scattering intensity of -22dB. Combined with the preset low scattering → snow / sand judgment logic, the road section is initially marked as a candidate for snow. The front camera collects road surface images within a 50-meter range ahead. After extracting texture features, it shows that the brightness value in the HSV color space is stable at around 0.85 (typical snow reflection characteristics), and the edge density is only 0.23 (sharp edges without rocks or other hard obstacles), with a matching degree of 94% with the built-in snow texture template. At the same time, the vehicle's accelerometer detects a peak value of 2.1Hz in the Z-axis vibration frequency spectrum. This is due to the low-frequency vibration caused by the buffering effect of snow on the vehicle, which is significantly different from the dry asphalt road surface (vibration peak 8-10Hz). Combining the real-time data of the three types of sensors, the formula is used to... After completing the fusion judgment, the final output of the snow road condition recognition result is given.

[0021] The driving style acquisition module operates synchronously with road condition recognition, retrieving nearly 10km of driving operation records in real time via the vehicle's CAN bus. Data shows that the driver's accelerator pedal opening change rate averaged 18% / s during this journey (far below the rapid acceleration threshold of 50% / s), the maximum brake pedal travel was only 45%, and the average steering wheel angular velocity was 25° / s. Furthermore, there were no instances of rapid acceleration with accelerator pedal opening >80% and a change rate >50% / s, nor were there any instances of emergency braking with brake pedal travel >70% and an action time <0.5s. Based on these characteristics, the module defines the driving style as smooth and outputs the corresponding correction parameter of -1.0kPa according to preset rules, providing a personalized adjustment basis for subsequent tire pressure calculation.

[0022] Upon receiving the snow road condition recognition result and the -1.0 kPa driving style correction item, the tire pressure calculation module immediately starts the calculation. First, it retrieves the baseline tire pressure data from the vehicle control system. The standard road tire pressure specified in the vehicle manual is 36 psi. Then, it matches the correction coefficient corresponding to the snow road condition. According to the system's built-in road condition-tire pressure mapping model, the tire pressure needs to be reduced in snowy scenarios to increase the contact area. The correction coefficient is set to -15.0 kPa. At the same time, the tire's built-in temperature sensor reports that the current tire temperature is -5℃. According to the temperature compensation formula "0.2 × (tire temperature - 25)", it is calculated that: 0.2 × (-5 - 25) = -6.0 kPa. Since the tire temperature is lower than the ideal temperature of 25℃, the tire pressure needs to be further reduced to offset the tire pressure contraction caused by the low temperature. Finally, the driving style correction item of -1.0 kPa is substituted, and the target tire pressure is calculated to be 32.8 psi through the model's comprehensive calculation.

[0023] Once the target tire pressure is determined, the dynamic adjustment module quickly enters the deflation process. First, the tire pressure sensor detects that the current tire pressure of all four tires is 36 psi (standard for highway driving), which is significantly different from the target tire pressure of 32.8 psi, meeting the deflation conditions. The control center then sends an opening command to the high-speed solenoid valves of the four wheels and activates the tire sidewall vibrator. The vibration breaks the stable state of the gas inside the tire, accelerating the gas from the solenoid valve, improving the deflation efficiency by 40% compared to the non-vibration mode. During the adjustment process, the system provides real-time tire pressure feedback every 2 seconds through a high-precision pressure sensor. When the tire pressures of the left front, right front, left rear, and right rear tires are detected to be stable at 32.7 psi, 32.8 psi, 32.9 psi, and 32.8 psi respectively (all within ±0.1 psi error range), a closing command is immediately sent. The solenoid valves complete the closing action within 50 ms and trigger the mechanical self-locking device to prevent valve loosening and leakage due to road bumps. The entire adjustment process takes 18 seconds, far less than the 3-5 minutes required for traditional manual adjustment.

[0024] After adjustment, the system continuously monitored the driving status to verify the effect. Detection using the tire sidewall embedded strain gauge array revealed that the tire contact area increased by 32% compared to before adjustment, effectively improving snow grip. During the subsequent 5km driving on snowy roads, the vehicle stability system recorded zero skid warnings in real time, reducing the skid probability from 38% in the traditional system to 9%, with more stable vehicle posture, especially when cornering. At the same time, the tire temperature remained between -5℃ and -3℃, and the deformation rate remained stable at 0.02% (below the safety threshold of 0.05%), without triggering any safety locking mechanisms. Finally, the vehicle successfully passed through the snowy road section at a speed of 60km / h, verifying the system's reliability and adaptability in complex winter road conditions.

[0025] Example 2 For sections of road alternating between rocks and mud, when the multimodal road condition recognition module is working, if the vehicle encounters a rocky section immediately upon entering the off-road section, the vehicle's accelerometer detects a significant peak of 18Hz in the Z-axis vibration frequency spectrum. This is due to the road surface being covered with sharp protrusions, causing high-frequency vibrations from the tires colliding with them. Simultaneously, the tire deformation sensor detects a sudden change in the sideslip angle of 6 times per second, each change corresponding to the instantaneous deformation of the tire as it rolls over a protrusion. The millimeter-wave radar receives a ground echo scattering intensity of -10dB, exhibiting high scattering characteristics, indicating that the road surface is hard and uneven. The camera... The system captured textures with a density >0.8 at the road edge, further confirming the hardness of the rock. After fusion of data from multiple sensors, the system determined the current road surface to be rocky within 80ms. After driving 1.2km, the road condition changed to a muddy section. The vehicle's accelerometer showed that the peak value of the Z-axis vibration frequency spectrum dropped to 5Hz. This was because the soft mud buffered the vibration. The tire deformation sensor detected a continuous increase in the sideslip angle, indicating that the tire was slowly sinking. The road surface seen by the camera appeared as an irregular paste with an HSV saturation >0.6, which was consistent with the visual characteristics of a muddy road surface. The system updated the road condition to a muddy road surface.

[0026] The driving style acquisition module worked continuously throughout the process, acquiring driving operation data in real time via the vehicle's CAN bus. The records showed that within the past 5km, the driver repeatedly accelerated rapidly, with a throttle pedal opening change rate >60% / s. There were also 3 instances of emergency braking, each with a brake pedal travel of 80% and an action time of <0.4s. The average steering wheel angular velocity was >50° / s. These data indicate a relatively aggressive driving style. Based on this, the module generated a +2.0kPa driving style correction item to enhance tire support under aggressive driving conditions.

[0027] When driving on rocky terrain, the tire pressure calculation module uses 36 psi as the baseline tire pressure from the vehicle manual. Considering that higher tire pressure is needed on rocky surfaces to improve puncture resistance, a road condition correction factor of +20.0 kPa is selected. At this point, the tire temperature rises to 35°C due to driving, and appropriate compensation is made based on the effect of temperature on tire pressure. Combined with the +2.0 kPa correction item corresponding to aggressive driving, the target tire pressure is determined to be approximately 45 psi. After entering muddy terrain, in order to increase the tire contact area and reduce sinking, the road condition correction factor is adjusted to -28.0 kPa. The tire temperature rises to 38°C, and the temperature compensation value changes accordingly. The driving style correction item remains at +2.0 kPa, and the final calculated target tire pressure is approximately 22 psi.

[0028] The dynamic adjustment module performs inflation on rocky sections. Upon detecting that the current tire pressure (36 psi) is lower than the target value of 45 psi, it immediately starts a scroll compressor with a flow rate of 35 L / min, inflating the tires sequentially: left front, right front, left rear, and right rear. The first inflation cycle lasts 5 seconds, raising the tire pressure to 38 psi. The second cycle reduces the inflation time by 30% to 3.5 seconds, raising the tire pressure to 43 psi. The third cycle further reduces the inflation time by 30% to 2.45 seconds, ultimately stabilizing the tire pressure at 45.0 ± 0.1 psi. Within the specified range, the tire deformation rate remained below 0.03% throughout the entire process, with no abnormalities observed. After inflation was completed, the compressor was shut off and self-locked, taking 32 seconds. Upon entering the muddy section, deflation was performed. Since the current tire pressure of 45 psi was higher than the target value of 22 psi, the system activated the four-wheel high-speed solenoid valves and simultaneously activated the sidewall vibrator to accelerate deflation. After 12 seconds, the tire pressure dropped to 22 psi, and the valves were then closed and self-locked. At this point, the tire temperature remained at approximately 38°C, with a deformation rate of <0.02%. The adjustment process was safe and reliable.

[0029] Performance verification shows that on rocky sections, after tire pressure adjustment, the tire ground pressure distribution is uniform, successfully driving over multiple sharp protrusions without puncture, and reducing the risk of tire blowout from 12.7% to 1.3%. On muddy sections, the tire sinking depth is reduced from 12cm to 5cm, traction is increased by 180N, and the vehicle successfully passes through a 15cm deep muddy area, fully demonstrating the system's adaptive adjustment capability and safety in complex off-road scenarios.

[0030] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition, characterized in that, The system includes: a multimodal road condition recognition module, a driving style acquisition module, a tire pressure calculation module, and a dynamic adjustment execution module; The multimodal road condition recognition module identifies road condition types by fusing data from millimeter-wave radar, cameras, vehicle accelerometers, and tire deformation sensors. The driving style acquisition module obtains operation data through the vehicle CAN bus, and combines the number of rapid accelerations and brakings recorded by the vehicle control system to generate driving style labels. The tire pressure calculation module uses a reference tire pressure as a basis, combined with fixed correction coefficients for different road conditions, temperature compensation based on the difference between tire temperature and 25°C, and driving style correction items, to calculate the target tire pressure through a road condition-tire pressure mapping model. The dynamic adjustment execution module includes an on-board air pump, a high-speed solenoid valve, and a tire sidewall vibrator. It performs inflation and deflation operations according to the target tire pressure and has a safety locking function.

2. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, The multimodal road condition recognition module includes a millimeter-wave radar, a camera, a vehicle accelerometer, a tire deformation sensor, and a road condition change rate monitoring unit. The millimeter-wave radar determines road conditions by detecting the intensity of ground echo scattering; low scattering corresponds to sandy or snowy terrain. The camera determines road conditions by detecting texture features; high edge density corresponds to rocky surfaces. The vehicle accelerometer determines road conditions by detecting the Z-axis vibration frequency spectrum; a peak value of 3-8Hz corresponds to muddy surfaces. The tire deformation sensor determines road conditions by detecting the rate of change of the sideslip angle; a sudden increase in the rate of change of the sideslip angle corresponds to soft ground. Simultaneously, by collecting data on the vehicle's rapid acceleration / deceleration frequency, steering amplitude, and speed fluctuations, the module identifies the driver's driving style.

3. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 2, characterized in that, The multimodal road condition recognition module identifies road condition types by fusing data, and its recognition formula is as follows: ， in, It is a comprehensive road condition identification index, used to ultimately determine the road condition type. , , , These are the weighting coefficients of the recognition results from each sensor, summing to 1. They are set based on the recognition accuracy and reliability of the sensors. It is a millimeter-wave radar traffic identification index. It is the camera's visual road condition recognition index. It is the vehicle accelerometer road condition recognition index. It is the road condition recognition index of the tire deformation sensor.

4. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, The driving style acquisition module obtains driving operation data through the vehicle CAN bus and vehicle control system, specifically including: accelerator pedal opening change rate, brake pedal travel and application time, and steering wheel angular velocity; it also records the number of rapid accelerations per unit time, defined as accelerator pedal opening > 80% and change rate > 50% / s, and the number of emergency brakings, defined as brake pedal travel > 70% and application time < 0.5s. Based on this data, a driving style label is generated.

5. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, In the tire pressure calculation module, the fixed correction coefficients for different road conditions include 0.0 kPa for highways, -35.0 kPa for sandy areas, -28.0 kPa for muddy areas, -15.0 kPa for snowy areas, and +20.0 kPa for rocky areas.

6. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, The tire pressure calculation module calculates the target tire pressure using a road condition-tire pressure mapping model, and the calculation formula is as follows: ， in, This is the target tire pressure, which is the final tire pressure value that needs to be adjusted to. That is the reference tire pressure. It is the first Correction factors for different road conditions It is the road condition recognition state coefficient. When the first road condition is recognized... When dealing with various road conditions, ,otherwise , This is the temperature compensation coefficient, with a value of 0.

2. This is the current tire temperature. The ideal tire temperature is fixed at 25°C. This is the driving style correction factor, determined based on the driving style label. It is a quantitative value for driving style.

7. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, The dynamic adjustment execution module includes an on-board air pump, a high-speed solenoid valve, a tire deformation monitoring device, and an environmental sensing center. The on-board air pump is an oil-free silent scroll compressor with a flow rate ≥35L / min and noise <65dB. The high-speed solenoid valve is driven by piezoelectric ceramics with a response time <50ms. The tire deformation monitoring uses an embedded strain gauge array on the tire sidewall with an accuracy of ±0.05%. The environmental sensing center uses the NVIDIA Orin+ LiDAR point cloud classification algorithm with a processing latency <80ms.

8. The vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition according to claim 1, characterized in that, The dynamic adjustment execution module detects that the current tire pressure is higher than the target tire pressure, initiates the deflation program, opens the high-speed solenoid valve of the corresponding tire and activates the tire sidewall vibrator to accelerate pressure release, monitors the tire pressure in real time, and closes and locks the valve after reaching the target value. When the current tire pressure is detected to be lower than the target tire pressure, the scroll compressor is started, and the solenoid valves of each tire are opened in the order of left front-right front-left rear-right rear. After the first round of inflation lasts 5 seconds, the valves are closed, the current tire pressure is detected and the difference from the target value is calculated. If the difference is >0.5psi, the inflation time of the next round is reduced by 30%, and the adjustment is carried out in turn until the tire pressure of all tires is within the target value ±0.1psi range. The compressor is then turned off and the valves are locked. Furthermore, the inflation and deflation functions are disabled when the vehicle speed is >60km / h, the curve radius is <50m, or the road surface adhesion coefficient is too low.

9. A vehicle-mounted tire pressure adaptive control method based on multimodal road condition recognition, the method being used to control the vehicle-mounted tire pressure dynamic adjustment system based on multimodal road condition recognition as described in any one of claims 1-8, characterized in that, The method includes: Road condition recognition: Multiple sensors simultaneously collect road condition-related data, make preliminary judgments according to their respective judgment logics, and fuse the judgment results of each sensor to determine the current road condition type; Driving style acquisition: The driving style acquisition module reads throttle, brake and steering operation data in real time through the vehicle CAN bus, counts the number of rapid acceleration and hard braking according to preset rules, and generates driving style labels and corresponding correction parameters. Tire pressure adjustment: Based on the identified road condition type, the corresponding road condition correction coefficient is retrieved from the preset correction coefficient table; the tire temperature is obtained in real time to calculate the temperature compensation value, and the driving style correction item generated by the driving style acquisition module is also obtained. By obtaining the reference tire pressure in the vehicle manual, and combining the road condition correction coefficient, temperature compensation value and driving style correction item, the target tire pressure is calculated through the road condition-tire pressure mapping model. Based on the difference between the target tire pressure and the current tire pressure, the inflation or deflation mode is selected for adjustment. Safety control: Real-time monitoring of tire temperature, deformation, and vehicle driving status. When the risk of over-inflation or deflation of the tires is detected, or when the vehicle is in high-speed, large steering angle, or low-adhesion conditions, a safety locking mechanism is triggered to disable the tire pressure regulation function.

Images