System and method for damping adjustment of two-wheeled electric vehicle based on visual feedback

Abstract

The invention relates to the technical field of intelligent control of two-wheeled electric vehicles, in particular to a two-wheeled electric vehicle damping adjustment system and method based on visual feedback. The forward camera collects road images, and the image processing unit identifies feature road conditions through a lightweight model to obtain first image information. The state detection module obtains vehicle attitude and operation data through a Hall sensor and an accelerometer. The main control processor fuses information to calculate damping parameters and issues instructions, the damping adjustment executing mechanism dynamically adjusts the parameters of the damper according to the instructions, fine adjustment of damping force is achieved, the damping performance of the vehicle is improved, and the requirements of different road conditions are met. According to the technical scheme, the adaptability of the damping system to complex road conditions can be improved, the stability and comfort in the riding process are improved, and meanwhile reliable technical support is provided for intelligent development of the two-wheeled electric vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of two-wheeled electric vehicles, and in particular to a system and method for shock absorption adjustment of a two-wheeled electric vehicle based on visual feedback. Background Art

[0002] In the current field of two-wheeled electric vehicles, the shock absorption system mostly uses passive mechanical shock absorbers, such as hydraulic or spring structures. Their damping characteristics and stiffness parameters are usually set at the factory and cannot be adjusted in real time according to different road conditions. This traditional structure often fails to ensure both riding comfort and safety when encountering complex or changeable road conditions (such as potholes, speed bumps, and gravel roads).

[0003] To address these issues, some high-end electric vehicles are beginning to be equipped with active damping systems, such as those based on inertial measurement units (IMUs) like accelerometers and gyroscopes. These systems monitor vehicle posture and vibration intensity in real time and adjust damping parameters after vibration occurs, employing a feedback control mechanism. However, these solutions suffer from limitations such as feedback lag, blind spots, limited control accuracy, high costs, and low user engagement.

[0004] In summary, the prior art has not yet provided a method applicable to two-wheeled electric vehicles that can identify the road conditions ahead through a visual system and actively pre-adjust the shock absorption parameters. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a system and method for shock absorption adjustment of a two-wheeled electric vehicle based on visual feedback, so as to solve the problems of limitations in the prior art such as feedback lag, recognition blind spots, limited control accuracy, high cost and low user participation.

[0006] According to a first aspect of an embodiment of the present invention, a system for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback is provided, wherein the system comprises:

[0007] A forward-facing camera module, used to capture images of the road ahead;

[0008] an image processing unit, configured to run a preset lightweight image recognition model on the front road image to identify characteristic road conditions and obtain first image information;

[0009] A state detection module is used to obtain the posture information and operation data information of the target vehicle using preset sensors; the sensors include: Hall sensors and accelerometers;

[0010] a main control processor, configured to fuse the first image information with the posture information and operating data information of the target vehicle, calculate the currently required shock absorption parameters, and issue a first instruction;

[0011] The shock absorption adjustment actuator is used to dynamically adjust the parameters of the shock absorber according to the first instruction sent by the main control processor to achieve fine adjustment of the damping force.

[0012] Furthermore, it also includes:

[0013] The main control processor communicates with the front camera module via a CAN bus, and the front camera module is connected to the image processing unit using an AHD streaming interface.

[0014] Furthermore, it also includes:

[0015] The Hall sensor in the state detection module is integrated on the motor, the accelerometer is integrated in the electronic control, and the Hall sensor and the accelerometer are respectively connected to the main control processor via a CAN bus.

[0016] Furthermore, the shock absorption adjustment actuator adopts an adjustable hydraulic shock absorption device driven by a servo motor. The response time of the adjustable hydraulic shock absorption device driven by the servo motor is less than 150ms, and the adjustment accuracy can reach ±5%.

[0017] According to a second aspect of an embodiment of the present invention, a method for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback is provided, which is applied to any of the above-mentioned systems for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback, and is characterized in that the method comprises:

[0018] Collect images of the road ahead;

[0019] For the front road image, running a preset lightweight image recognition model to identify characteristic road conditions to obtain first image information;

[0020] Using preset sensors to obtain the target vehicle's posture information and operating data information; the sensors include: Hall sensors, accelerometers;

[0021] fusing the first image information with the posture information and operating data information of the target vehicle, calculating the currently required shock absorption parameters and issuing a first instruction;

[0022] The parameters of the shock absorber are dynamically adjusted according to the first instruction sent by the main control processor to achieve fine adjustment of the damping force.

[0023] Furthermore, the step of running a preset lightweight image recognition model on the road ahead image to identify characteristic road conditions and obtain first image information includes:

[0024] The YOLO-Tiny model is used to process the road ahead image in real time to obtain first image information; the first image information includes: speed bumps, potholes, and undulating road surface feature road condition information.

[0025] Furthermore, the fusing of the first image information with the posture information and the operating data information of the target vehicle, calculating the currently required shock absorption parameters and issuing a first instruction includes:

[0026] Calculate the acceleration, peak acceleration, and FFT amplitude within a period of time using the first image information, the posture information, and the operating data information of the target vehicle;

[0027] Based on the acceleration, peak acceleration, and FFT amplitude during the period, the turbulence level is obtained according to the following formula:

[0028] Turbulence level = w1×RMS acceleration + w2×peak acceleration + w3×FFT amplitude (1)

[0029] Among them, w1, w2, and w3 are adjustable weights ranging from [0–1], which are set according to the vehicle model and mode;

[0030] Using the obtained turbulence level, a target damping coefficient is determined according to the following formula:

[0031] Target damping = K1×bump level + K2×weight + K3×speed + K4×road condition factor (2)

[0032] Among them, K1, K2, K3, K4: adjustment coefficient, representing the contribution of each factor to the final shock absorption;

[0033] The target damping coefficient is used to characterize the currently required shock absorption parameters.

[0034] Furthermore, the calculating of the acceleration within a period of time by using the first image information, the posture information and the operation data information of the target vehicle includes:

[0035] Calculate the acceleration within a period of time using the first image information, the posture information and the operating data information of the target vehicle according to the following formula;

[0036]

[0037] Where ai is the instantaneous acceleration value when the sensor is sampling, and N is the number of sampling points.

[0038] Furthermore, the calculating the peak acceleration by using the first image information, the posture information and the operating data information of the target vehicle includes:

[0039] Get the maximum value of acceleration in a period of time and get the peak acceleration, as follows:

[0040] Peak=max(∣a1∣,∣a2∣,...,∣aN∣)(4);

[0041] Peak: represents peak acceleration, which records the maximum acceleration in each period (after taking the absolute value) and reflects sudden turbulence;

[0042] a1, a2, ..., aN: acceleration values at each sampling moment;

[0043] |a1|,|a2|,...,|aN|: take the absolute value of each acceleration value respectively;

[0044] max(|a1|,|a2|,...,|aN|): The maximum value among these absolute values is the peak acceleration.

[0045] Furthermore, the calculating of the FFT amplitude using the first image information, the posture information and the operation data information of the target vehicle includes:

[0046] Collect original acceleration signals;

[0047] The DC component of the original acceleration signal is removed using the following formula to obtain a result after the DC component is removed;

[0048]

[0049] Among them, a centered (t): acceleration signal after removing DC component; a(t): original acceleration signal, which is a function of time t; The average value of the original acceleration signal a(t); N: the number of sampling points of the original acceleration signal; a(i): the value of the original acceleration signal at the i-th sampling point;

[0050] Converting the result after removing the DC component into a frequency domain signal through fast Fourier transform to obtain an amplitude spectrum corresponding to each frequency component;

[0051] Obtain the frequency component with the largest amplitude in the amplitude spectrum corresponding to the frequency component and the following formula to obtain the frequency component corresponding to the main frequency;

[0052]

[0053] Among them, the formula means that in the positive frequency part (k∈[1,N / 2], generally only positive frequency is considered), the amplitude of all frequency components |A(f k The maximum value among )| is taken as the FFT amplitude;

[0054] The FFT amplitude is determined using the frequency component corresponding to the main frequency.

[0055] The technical solutions provided by the embodiments of the present invention may have the following beneficial effects:

[0056] 1. Realize feedforward shock absorption control and improve response speed

[0057] The forward visual recognition module identifies the road conditions 5-10 meters ahead in advance. The central control processor predicts the time it will reach a specific section of road based on the vehicle's current speed, thereby completing the shock absorption parameter adjustments in advance, overcoming the delays and inefficiencies caused by the "after-the-fact response" of traditional shock absorption systems.

[0058] 2. Multi-source fusion perception significantly improves adjustment accuracy

[0059] This system integrates multiple sensing modules such as image recognition, Hall sensors, accelerometers, etc., and comprehensively analyzes road conditions and vehicle status, making the calculation of shock absorption parameters more accurate and reasonable, avoiding misjudgment caused by a single sensor.

[0060] 3. Adaptive strategy based on scene characteristics to enhance adaptability to complex road conditions

[0061] The system supports setting differentiated response strategies for different types of road conditions (such as speed bumps, gravel, undulations, etc.), and dynamically adjusts damping parameters based on real-time vibration spectrum analysis, so that the vehicle can maintain good stability and passability in various complex terrains.

[0062] 4. Significantly improve riding comfort and safety

[0063] Adjusting the shock absorber response hardness in advance can effectively reduce the direct force of impacts from potholes, speed bumps, etc. on the human body, reduce driving fatigue, improve handling stability and the safety level of the entire vehicle, and is especially suitable for urban commuting and tourist riding scenarios.

[0064] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0066] Figure 1 is a diagram showing a system composition of a two-wheeled electric vehicle shock absorption adjustment system based on visual feedback according to an exemplary embodiment;

[0067] Figure 2 is a schematic diagram of the architecture of an intelligent shock absorption control system according to an exemplary embodiment;

[0068] Figure 3 The present invention is a flow chart of a method for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback according to an exemplary embodiment. DETAILED DESCRIPTION

[0069] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0070] Example 1

[0071] Figure 1 1 is a diagram showing a system composition of a two-wheeled electric vehicle shock absorption adjustment system based on visual feedback according to an exemplary embodiment, the system comprising:

[0072] 1. Forward Camera Module 01: Installed at the front of the vehicle, it captures images of the road ahead with a frame rate exceeding 30 fps and a coverage distance of 5–10 meters.

[0073] 2. Image Processing Unit 02: Runs a lightweight image recognition model (such as YOLO-Tiny) to process images in real time and identify characteristic road conditions such as potholes, speed bumps, and undulations.

[0074] 3. Status detection module 03: includes Hall sensor and accelerometer, used to provide vehicle posture and operation data.

[0075] 4. Main control processor 04: integrates image recognition results and status information, calculates the current required vibration reduction parameters, and issues control instructions.

[0076] 5. Shock absorption adjustment actuator 05: The shock absorption actuator adopts an adjustable hydraulic shock absorption device driven by a servo motor. The servo motor can quickly change the valve opening or the hydraulic oil channel to achieve fine adjustment of the damping force. The response time is less than 150ms and the adjustment accuracy can reach ±5%.

[0077] In practice, the image processing module identifies the road type and characteristics ahead. Based on this image information and vehicle speed data, the main control unit calculates the type of road section the vehicle will reach in a few seconds. The system then sets a damping response strategy based on different road conditions. For example, for speed bumps and potholes, the system increases damping to mitigate vibration; for flat roads, the system reduces damping stiffness to conserve energy; and for undulating roads, the system adjusts the response frequency based on the spectral characteristics. Control commands are then transmitted via PWM to the damping control module, which adjusts the actual damping.

[0078] During implementation, the system collects key data in real time through multi-sensor modules: the visual recognition module provides visual recognition results, combined with road surface information such as road condition level (flat, slightly bumpy, severely potholed) and road condition type (speed bumps, gravel, continuous undulating road); at the same time, it obtains vehicle status information, including vehicle speed that determines response time and adjustment frequency, and instantaneous vibration intensity (acceleration) collected by the three-axis accelerometer, laying a data foundation for subsequent parameter calculations.

[0079] The main control unit first calculates the vibration intensity index based on the sensor data. Using a root mean square (RMS) algorithm, it compiles statistics on acceleration values over a period of time to reflect the overall vibration intensity. It also records the peak acceleration—the maximum absolute value of acceleration during that period—to indicate the severity of sudden vibrations. In spectral feature analysis, the triaxial accelerometer first collects the vehicle's vertical (Z-axis) acceleration signal. After removing the DC component, the time-domain signal is converted to the frequency domain using a fast Fourier transform (FFT), and the amplitude corresponding to the dominant frequency component (FFT amplitude) is extracted. Different amplitude and frequency combinations can distinguish road vibration characteristics. For example, high-frequency, high-amplitude vibrations correspond to the fine vibrations of a gravel road, low-frequency, high-amplitude vibrations correspond to the undulations of a bridge deck, and sudden, high-amplitude vibrations indicate speed bumps or potholes.

[0080] The system combines RMS acceleration, peak acceleration, and FFT amplitude to quantify the level of bumpiness using a weighted formula. The weighting coefficients can be flexibly adjusted based on the vehicle model and mode. Ultimately, a linear combination of the bumpiness level, vehicle speed, load (weight), and road condition factor is used to calculate the target damping coefficient. For example, a speed bump corresponds to a road condition factor of +30%, while a flat road corresponds to -50%. This allows the damping system's damping parameters to be dynamically adjusted to suit the vehicle's driving needs under different road conditions.

[0081] Furthermore, the shock absorber device uses an electronic hydraulic shock absorber to control the shock absorber hardness by controlling the voltage / current of the proportional valve and adjusting the damping aperture or the flow resistance of the hydraulic circuit; the central control processor generates a PWM duty cycle control signal according to the target damping value, and the format is, for example: PWM duty cycle range: 0–100%, corresponding to minimum to maximum damping.

[0082] For specific implementation, please refer to Figure 2 , Figure 2The architecture of the intelligent damping control system is demonstrated. A front-facing camera captures images of the road ahead, while edge computing devices detect road conditions. Accelerometers detect road bumps, and both data are transmitted to the intelligent central control system. The intelligent central control system interacts with the cloud platform via a 4G network. Users can select modes and set parameters using an app or through the instrument cluster. Based on this information, the intelligent central control system adjusts the damping strength and, in turn, controls the electronically controlled damping system, achieving intelligent control of vehicle damping, effectively improving driving comfort and safety.

[0083] Furthermore, users can view real-time traffic condition recognition results and system response status through the APP, and make personalized settings or update algorithm models.

[0084] This invention fully integrates visual perception and intelligent control technologies, and has the following intelligent functions, significantly improving the system's automation level, user interaction experience, and adaptive adjustment capabilities:

[0085] 1. Automatic and manual dual mode switching

[0086] The system supports automatic adjustment mode. In this mode, after the user starts the electric vehicle, no manual intervention is required. The system will automatically adjust the shock absorption parameters in real time based on the camera recognition results and the vehicle's operating status (such as speed, vibration intensity, etc.) to achieve dynamic adaptive adjustment.

[0087] At the same time, the system provides a manual control mode, where users can independently set the hardness and response strategy of the shock absorber through the APP interface or vehicle-mounted buttons to meet personalized control under specific needs.

[0088] 2. Personalized parameter configuration

[0089] The system provides a variety of riding preference settings. Users can select or create different adjustment style templates through the APP, such as:

[0090] "Comfort First": Increases the sensitivity to small vibrations;

[0091] "Sports Mode": appropriately increases the damping hardness to enhance the sense of control;

[0092] "Energy Saving Mode": Automatically reduces damping on flat roads to reduce energy consumption.

[0093] Users can also customize the weight parameters in the control algorithm (such as bump level coefficient, speed factor, etc.) to match different users' weight, riding habits or terrain preferences.

[0094] 3. Remote monitoring and configuration function

[0095] The system's accompanying mobile phone APP has a remote interaction function. Users can connect to the vehicle's main control unit via a BLE (Bluetooth Low Energy) or 4G communication module to achieve:

[0096] View the road condition identification information ahead and the current status of the shock absorber in real time;

[0097] Remote adjustment of shock absorption parameter configuration;

[0098] View historical road condition recognition records and shock absorption response data;

[0099] Perform online firmware upgrades and algorithm model updates to support continuous iterative optimization.

[0100] More specifically, the main control unit and the visual perception module communicate through the CAN bus to ensure that the image recognition results and sensor data can be transmitted to the control core within millisecond-level delays to meet high-frequency control requirements.

[0101] Furthermore, the forward camera module 01 is connected to the image processing unit 02 using an AHD streaming interface to ensure the stability and clarity of the video stream.

[0102] Furthermore, the shock absorption adjustment actuator 05 adopts an adjustable hydraulic shock absorption device driven by a servo motor. The servo motor can quickly change the valve opening or the hydraulic oil channel to achieve fine adjustment of the damping force. The response time is less than 150ms and the adjustment accuracy can reach ±5%.

[0103] Furthermore, the Hall sensor in the status detection module 03 is integrated into the motor, and the accelerometer is integrated into the electronic control, which is connected to the intelligent central control via the CAN bus, and has high stability and anti-interference capabilities.

[0104] At the software implementation level, the embedded software system developed in the main control unit encompasses multiple functional modules: the sensor driver layer is responsible for real-time acquisition and caching of multi-source data such as vehicle speed, acceleration, and image recognition results; the data fusion and judgment logic module integrates and calculates sensor information to accurately identify the upcoming road condition type and level of bumpiness; the damping parameter calculation algorithm module calculates the current required damping target value in real time based on predefined formulas and strategies; and the control signal output module converts the damping parameters into PWM duty cycles or servo motor control signals to drive the shock absorber adjustment mechanism. Meanwhile, a mobile app communicates with the main control unit via a Bluetooth Low Energy (BLE) or 4G module, displaying real-time images of the road ahead, classification information, damping status, and vehicle operating status. It also provides a user-friendly interface for personalized settings, allowing users to select damping styles such as Comfort, Standard, and Sport, or manually adjust the default damping strategy. It also features remote fault diagnosis and firmware updates.

[0105] Furthermore, the system integration adopts a modular design, and each hardware module (forward-facing camera, image processing unit 02, etc.) is connected to a standardized interface through a unified CAN communication protocol to ensure compatibility and scalability. The integration process includes: the hardware integration stage installs the module to the electric vehicle platform and arranges the wiring in a regular manner; the software integration stage embeds the image recognition algorithm and control drive logic; the system joint debugging stage verifies the consistency and stability through multi-scenario real-vehicle testing; the APP platform docking stage completes the connection test and data interaction function development. According to the test, the system power consumption has an impact of less than 3% on battery life, and the communication delay is less than 80ms, meeting the real-time requirements. The implementation effect verification shows that the system can stably identify urban road conditions and perform feedforward damping adjustment during continuous riding. The average delay in the 20-minute commuting test is less than 180ms. The shock absorption comfort score is 27% higher than that of traditional mechanical preset shock absorbers, and user satisfaction has been significantly improved.

[0106] Compared with the existing technology, the present invention has the following significant advantages and positive effects:

[0107] 1. Realize feedforward shock absorption control and improve response speed

[0108] The forward visual recognition module identifies the road conditions 5-10 meters ahead in advance. The central control processor predicts the time it will reach a specific section of road based on the vehicle's current speed, thereby completing the shock absorption parameter adjustments in advance, overcoming the delays and inefficiencies caused by the "after-the-fact response" of traditional shock absorption systems.

[0109] 2. Multi-source fusion perception significantly improves adjustment accuracy

[0110] This system integrates multiple sensing modules such as image recognition, Hall sensors, accelerometers, etc., and comprehensively analyzes road conditions and vehicle status, making the calculation of shock absorption parameters more accurate and reasonable, avoiding misjudgment caused by a single sensor.

[0111] 3. Adaptive strategy based on scene characteristics to enhance adaptability to complex road conditions

[0112] The system supports setting differentiated response strategies for different types of road conditions (such as speed bumps, gravel, undulations, etc.), and dynamically adjusts damping parameters based on real-time vibration spectrum analysis, so that the vehicle can maintain good stability and passability in various complex terrains.

[0113] 4. Significantly improve riding comfort and safety

[0114] Adjusting the shock absorber response hardness in advance can effectively reduce the direct force of impacts from potholes, speed bumps, etc. on the human body, reduce driving fatigue, improve handling stability and the safety level of the entire vehicle, and is especially suitable for urban commuting and tourist riding scenarios.

[0115] 5. Energy saving and personalization

[0116] In flat road scenarios, the system automatically reduces the shock absorption hardness to reduce unnecessary energy consumption; users can also personalize the shock absorption style (comfort / standard / sport mode) through the APP to achieve a dynamic balance between human-computer interaction and energy saving.

[0117] 6. Have good system scalability

[0118] This system adopts a modular design, and its image recognition algorithms, control strategies, and communication protocols all support remote upgrades and optimizations, facilitating the subsequent expansion of additional features (such as cloud-based sharing of road conditions and dynamic group riding linkage), providing a solid technical foundation for the intelligent development of two-wheeled electric vehicles. 2. Multi-source fusion perception significantly improves adjustment accuracy.

[0119] This system integrates multiple sensing modules such as image recognition, Hall sensors, accelerometers, etc., and comprehensively analyzes road conditions and vehicle status, making the calculation of shock absorption parameters more accurate and reasonable, avoiding misjudgment caused by a single sensor.

[0120] See also Figure 3 , Figure 3 The present invention is a flow chart of a method for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback according to an exemplary embodiment, the method comprising:

[0121] S1. Collecting images of the road ahead;

[0122] S2. For the road ahead image, run a preset lightweight image recognition model to identify characteristic road conditions to obtain first image information;

[0123] S3 uses a preset sensor to obtain the target vehicle's posture information, operating data information; the sensor includes: a Hall sensor, an accelerometer;

[0124] S4. Fusion of the first image information and the target vehicle's posture information and operating data information, calculates the current required damping parameters and issues a first instruction;

[0125] S5. Dynamically adjust the parameters of the shock absorber according to the first instruction sent by the main control processor to achieve fine adjustment of the damping force.

[0126] In a further specific implementation, the method of running a preset lightweight image recognition model on the front road image to identify characteristic road conditions and obtain first image information includes:

[0127] The YOLO-Tiny model is used to process the road ahead image in real time to obtain first image information; the first image information includes: speed bumps, potholes, and undulating road surface feature road condition information.

[0128] In a further specific implementation, the fusing of the first image information with the posture information and the operating data information of the target vehicle, calculating the currently required shock absorption parameters and issuing the first instruction includes:

[0129] Calculate the acceleration, peak acceleration, and FFT amplitude within a period of time using the first image information, the posture information, and the operating data information of the target vehicle;

[0130] Based on the acceleration, peak acceleration, and FFT amplitude during the period, the turbulence level is obtained according to the following formula:

[0131] Turbulence level = w1 × RMS acceleration + w2 × peak acceleration + w3 × FFT amplitude (1)

[0132] Among them, w1, w2, and w3 are adjustable weights ranging from [0–1], which are set according to the vehicle model and mode;

[0133] Using the obtained turbulence level, a target damping coefficient is determined according to the following formula:

[0134] Target damping = K1×bump level + K2×weight + K3×speed + K4×road condition factor (2)

[0135] Among them, K1, K2, K3, K4: adjustment coefficient, representing the contribution of each factor to the final shock absorption;

[0136] The target damping coefficient is used to characterize the currently required shock absorption parameters.

[0137] More specifically, the implementation process includes the following:

[0138] The forward-facing camera captures images, and the image processing module identifies the road type and features ahead.

[0139] The main control unit calculates the type of road section the vehicle will reach in a few seconds based on the image information and vehicle speed data.

[0140] The system sets the shock absorption response strategy according to different road conditions, for example:

[0141] Speed bumps / potholes: Enhance damping and reduce vibration

[0142] Flat road: Reduce shock absorption hardness to save energy

[0143] Undulating continuous road surface: Adjust the response frequency according to the spectrum characteristics

[0144] The control instructions are transmitted to the shock absorption adjustment module via PWM to adjust the actual damping.

[0145] Users can view real-time traffic condition recognition results and system response status through the APP, and make personalized settings or update algorithm models.

[0146] In specific implementation, the working principle is as follows:

[0147] State perception stage: The system collects data in real time through the sensor module. The visual recognition module provides road condition levels (such as flat, slightly bumpy, and severely potholed) and road condition types (such as speed bumps, gravel roads, and continuously undulating roads). Vehicle state information is obtained by the speed sensor and three-axis accelerometer. The speed (Speed) determines the response time and adjustment frequency, and the acceleration (Accel) is captured by the three-axis accelerometer to capture the instantaneous vibration intensity.

[0148] Parameter calculation stage: The main control unit integrates the above data and dynamically calculates the damping parameters through a series of algorithms. First, the bump index is calculated, and the root mean square (RMS) value of acceleration is calculated using formula (3), which reflects the vibration intensity. Where ai is the instantaneous acceleration value when the sensor is sampling, and N is the number of sampling points.

[0149]

[0150] Furthermore, the peak acceleration is recorded using Equation 4 to capture sudden jolts.

[0151] Peak=max(∣a1∣,∣a2∣,...,∣aN∣) (4).

[0152] Next, perform FFT amplitude calculation: The vehicle's acceleration data in the vertical direction (usually the Z axis) is collected by a triaxial accelerometer. A 1- to 2-second sequence (e.g., 100-1000 data points) is used as the analysis window. The original signal is assumed to be: a(t), t = 0, Δt, 2Δt, ..., (N-1)Δt (Δt is the sampling interval, N is the number of sampling points, and the total sampling time is T = N·Δt).

[0153] In specific implementation, the original acceleration data is processed to remove the DC offset (remove the average value) through Equation 5 to ensure that the spectrum is not concentrated at 0 Hz.

[0154]

[0155] Among them, a centered (t): acceleration signal after removing DC component; a(t): original acceleration signal, which is a function of time t; The average value of the original acceleration signal a(t); N: the number of sampling points of the original acceleration signal; a(i): the value of the original acceleration signal at the i-th sampling point.

[0156] Convert the time domain acceleration data into frequency domain signal through FFT transformation;

[0157] The amplitude spectrum corresponding to each frequency component is obtained by the following formula.

[0158]

[0159] The frequency domain amplitude spectrum is calculated (modulo) using the following formula:

[0160]

[0161] In one embodiment, the amplitude of the main frequency component (i.e., FFT amplitude) is extracted:

[0162] In the transformation results, find the frequency component corresponding to the main frequency (maximum amplitude point);

[0163] The corresponding amplitude is the FFT amplitude, which indicates the intensity of the main frequency component of the current vibration;

[0164] More specifically, the frequency component with the largest amplitude in the amplitude spectrum corresponding to the frequency component is obtained by the following formula and the frequency component corresponding to the main frequency is obtained by the following formula;

[0165]

[0166] Among them, the formula means that in the positive frequency part (k∈[1,N / 2], generally only positive frequency is considered), the amplitude of all frequency components |A(f k The maximum value among )| is taken as the FFT amplitude;

[0167] The FFT amplitude is determined using the frequency component corresponding to the main frequency.

[0168] In specific implementation, different types of road surfaces can be effectively distinguished by analyzing vibration characteristics: when high-frequency and large-amplitude vibrations occur, it usually indicates a road surface with fine and continuous vibrations, such as a gravel road; if it is a low-frequency but large-amplitude vibration, it often corresponds to an undulating road surface, such as a bridge surface or a continuous gentle slope; and once a sudden high-amplitude vibration occurs, it is highly likely that a speed bump or pothole has been encountered.

[0169] Furthermore, the system combines RMS (root mean square) acceleration, peak acceleration, and FFT amplitude (the amplitude of the dominant frequency in the vibration spectrum, which measures the frequency of fluctuations) to quantify the vibration intensity of the current road conditions. This is achieved using the formula "Bump Level = w1 × RMS acceleration + w2 × peak acceleration + w3 × FFT amplitude," where w1, w2, and w3 are adjustable weights (ranging from 0 to 1) that can be flexibly set based on the vehicle model and driving mode.

[0170] Furthermore, the target damping coefficient is determined using the formula "Target Damping = K1 × Bump Level + K2 × Weight + K3 × Speed + K4 × Road Condition Factor." K1, K2, K3, and K4 are adjustment coefficients, reflecting their contribution to the final damping. The road condition factor is a preset value. For example, when encountering a speed bump, damping increases by 30%; on a flat road, damping decreases by 50%; and on a gravel road, damping increases by 20%. This calculation method allows the system to integrate multi-dimensional information to precisely adjust damping, adapting to different road conditions and vehicle states, improving driving comfort and stability.

[0171] More specifically, the process by which the central control processor dynamically adjusts the hardness of the hydraulic shock absorber is as follows:

[0172] More specifically, the process by which the central control processor dynamically adjusts the hardness of the hydraulic shock absorber is as follows:

[0173] 1. Real-time data acquisition

[0174] The central control processor receives real-time data input from multiple sources at a periodic frequency (e.g., every 50ms): image recognition results from the forward-facing camera preprocessed by the image processing unit, road condition type (such as speed bumps, potholes, gravel, flat), road location and distance (in meters), vehicle speed information obtained from the Hall effect sensor, and three-axis acceleration data collected by the accelerometer. This data forms the basis for the system to analyze road conditions and vehicle status.

[0175] 2. Predict approach time

[0176] The central control processor uses the formula based on the current vehicle speed V and obstacle distance D. The time T for the vehicle to reach the target road section is predicted. When T is less than the system's preset response time threshold (e.g., 1.5 seconds), the system immediately enters the adjustment preparation state and triggers the shock absorption parameter adjustment process in advance to ensure that the damping pre-adjustment is completed before the vehicle arrives, avoiding the lag of traditional feedback control.

[0177] 3. Bump level assessment

[0178] Based on real-time acceleration data, the processor quantifies the current vibration intensity through comprehensive calculation, namely the "bump level": Bump level = w1 × RMS acceleration + w2 × peak acceleration + w3 × FFT amplitude;

[0179] Among them, w1, w2, and w3 are weight coefficients that can be optimized according to user settings or vehicle models to balance the influence of different vibration characteristics; RMS acceleration reflects the overall intensity of vibration, and FFT amplitude captures the dominant frequency characteristics of vibration (such as high-frequency fine vibrations or low-frequency fluctuations) through spectral analysis.

[0180] 4. Target damping calculation

[0181] The processor combines the bump level, vehicle status, and road condition type to determine the target damping value of the hydraulic shock absorber using the formula: Target Damping = K1 × Bump Level + K2 × Weight + K3 × Speed + K4 × Road Condition Factor. K1-K4 are adjustment coefficients, corresponding to the contribution weight of each factor to the shock absorption effect. The Road Condition Factor is a preset correction term, as shown in Table 1. For example:

[0182] Table 1

[0183]

[0184] Flat road (bump level 0-2): factor -30%, strategy is to reduce damping to reduce energy consumption and improve comfort;

[0185] Speed bumps (bump level 4-6): factor +50%, the strategy is to increase damping to suppress impact;

[0186] Pothole road (bump level 6-8): Factor +70%, the strategy is to increase cushioning to prevent chassis impact;

[0187] Undulating roads (bump levels 3-5): Factor +40% (periodic adjustment), the strategy is to dynamically match the rhythm of road surface fluctuations.

[0188] 5. Control output generation

[0189] The shock absorber uses an electronic hydraulic shock absorber, which adjusts the damping aperture or hydraulic circuit flow resistance by controlling the voltage / current of the proportional valve to achieve precise control of the shock absorber hardness. The central control processor generates a corresponding PWM (pulse width modulation) duty cycle signal based on the target damping value, where:

[0190] The PWM duty cycle range is 0–100%, corresponding to continuously adjustable damping from minimum to maximum;

[0191] The signal drives the actuator in real time through the electronic control system, ensuring that the response time matches the predicted road condition approach time, realizing "pre-perception-pre-adjustment" closed-loop control.

[0192] This process enables the hydraulic shock absorption system to adaptively adjust according to real-time road conditions and vehicle status through multi-dimensional data fusion and dynamic algorithm matching, significantly improving riding stability and comfort. The present invention uses forward visual recognition to perceive the road conditions ahead in advance, and implements feedforward shock absorption control in combination with vehicle speed prediction to achieve the arrival time, breaking through the delay limitation of the traditional "after-the-fact response" and significantly improving the response speed; the multi-source fusion perception module comprehensively processes multiple data such as images, Hall sensors, accelerometers, etc., avoiding the problem of misjudgment of a single sensor, making the calculation of shock absorption parameters more accurate and reasonable; the adaptive strategy based on scene characteristics dynamically adjusts the damping parameters for different road conditions (such as speed bumps, gravel roads, etc.) to enhance adaptability to complex terrain; the mobile terminal APP personalizes the shock absorption style (comfort / standard / sport) and automatic energy-saving adjustment on flat roads, taking into account both user experience and energy consumption optimization; the modular design supports remote upgrades and expansions, facilitating subsequent function extensions (such as road condition cloud sharing). This method system effectively improves riding comfort, handling stability and safety level, achieves comprehensive optimization of response speed, adjustment accuracy, scene adaptation, personalized experience and system scalability, and provides an advanced technical solution for the intelligent development of two-wheeled electric vehicles.

[0193] It's important to note that during continuous riding, the system consistently identifies common urban road conditions and makes appropriate feedforward damping adjustments based on vehicle speed and predicted time. In a 20-minute commute test, the system achieved an average latency of less than 180ms, and a 27% improvement in damping comfort (compared to traditional mechanical preset shock absorbers), resulting in high user satisfaction.

[0194] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0195] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" is at least two.

[0196] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0197] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0198] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0199] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0200] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0201] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0202] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A system for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback, characterized in that: The system comprises: A forward-facing camera module, used to capture images of the road ahead; an image processing unit, configured to run a preset lightweight image recognition model on the front road image to identify characteristic road conditions and obtain first image information; A state detection module is used to obtain the posture information and operation data information of the target vehicle using preset sensors; the sensors include: Hall sensors and accelerometers; a main control processor, configured to fuse the first image information with the posture information and operating data information of the target vehicle, calculate the currently required shock absorption parameters, and issue a first instruction; The shock absorption adjustment actuator is used to dynamically adjust the parameters of the shock absorber according to the first instruction sent by the main control processor to achieve fine adjustment of the damping force.

2. The system according to claim 1, wherein: Also includes: The main control processor communicates with the front camera module via a CAN bus, and the front camera module is connected to the image processing unit using an AHD streaming interface.

3. The system according to claim 1, wherein: Also includes: The Hall sensor in the state detection module is integrated on the motor, the accelerometer is integrated in the electronic control, and the Hall sensor and the accelerometer are respectively connected to the main control processor via a CAN bus.

4. The system according to claim 1, wherein: The damping adjustment actuator adopts an adjustable hydraulic damping device driven by a servo motor. The response time of the adjustable hydraulic damping device driven by the servo motor is less than 150ms, and the adjustment accuracy can reach ±5%.

5. A method for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback, applied to a system for adjusting shock absorption of a two-wheeled electric vehicle based on visual feedback according to any one of claims 1 to 5, characterized in that: The method comprises: Collect images of the road ahead; For the front road image, running a preset lightweight image recognition model to identify characteristic road conditions to obtain first image information; Using preset sensors to obtain the target vehicle's posture information and operating data information; the sensors include: Hall sensors, accelerometers; fusing the first image information with the posture information and operating data information of the target vehicle, calculating the currently required shock absorption parameters and issuing a first instruction; The parameters of the shock absorber are dynamically adjusted according to the first instruction sent by the main control processor to achieve fine adjustment of the damping force.

6. The method according to claim 5, characterized in that The step of running a preset lightweight image recognition model on the front road image to identify characteristic road conditions and obtain first image information includes: The YOLO-Tiny model is used to process the road ahead image in real time to obtain first image information; the first image information includes: speed bumps, potholes, and undulating road surface feature road condition information.

7. The method according to claim 5, characterized in that The fusing of the first image information with the posture information and the operating data information of the target vehicle, calculating the currently required shock absorption parameters and issuing a first instruction includes: Calculate the acceleration, peak acceleration, and FFT amplitude within a period of time using the first image information, the posture information, and the operating data information of the target vehicle; Based on the acceleration, peak acceleration, and FFT amplitude during the period, the turbulence level is obtained according to the following formula: Bump level = w1 × RMS acceleration + w2 × peak acceleration + w3 × FFT amplitude (1) Among them, w1, w2, and w3 are adjustable weights ranging from [0–1], which are set according to the vehicle model and mode; Using the obtained turbulence level, a target damping coefficient is determined according to the following formula: Target damping = K1 × bump level + K2 × weight + K3 × speed + K4 × road condition factor (2) Among them, K1, K2, K3, K4: adjustment coefficient, representing the contribution of each factor to the final shock absorption; The target damping coefficient is used to characterize the currently required shock absorption parameters.

8. The method according to claim 7, characterized in that The calculating the acceleration within a period of time by using the first image information, the posture information and the operation data information of the target vehicle includes: Calculate the acceleration within a period of time using the first image information, the posture information and the operating data information of the target vehicle according to the following formula; Where ai is the instantaneous acceleration value when the sensor is sampling, and N is the number of sampling points.

9. The method according to claim 7, characterized in that The calculating the peak acceleration by using the first image information, the posture information and the operation data information of the target vehicle includes: Get the maximum value of acceleration in a period of time and get the peak acceleration, as follows: Peak=max(∣a1∣,∣a2∣,...,∣aN∣) (4); Peak: represents peak acceleration, which records the maximum acceleration in each period (after taking the absolute value) and reflects sudden turbulence; a1, a2, ..., aN: acceleration values at each sampling moment; |a1|,|a2|,...,|aN|: take the absolute value of each acceleration value respectively; max(|a1|,|a2|,...,|aN|): The maximum value among these absolute values is the peak acceleration.

10. The method according to claim 7, characterized in that The calculating the FFT amplitude using the first image information, the posture information and the operation data information of the target vehicle includes: Collect original acceleration signals; The DC component of the original acceleration signal is removed using the following formula to obtain a result after the DC component is removed; Among them, a centered (t): acceleration signal after removing DC component; a(t): original acceleration signal, which is a function of time t; The average value of the original acceleration signal a(t); N: the number of sampling points of the original acceleration signal; a(i): the value of the original acceleration signal at the i-th sampling point; Converting the result after removing the DC component into a frequency domain signal through fast Fourier transform to obtain an amplitude spectrum corresponding to each frequency component; Obtain the frequency component with the largest amplitude in the amplitude spectrum corresponding to the frequency component and the following formula to obtain the frequency component corresponding to the main frequency; Among them, the formula means that in the positive frequency part (k∈[1,N / 2], generally only positive frequency is considered), the amplitude of all frequency components |A(f k The maximum value among )| is taken as the FFT amplitude; The FFT amplitude is determined using the frequency component corresponding to the main frequency.

Images