Suspension adjusting method, device and system and storage medium

Abstract

The invention discloses a suspension adjusting method, device and system and a storage medium. The method comprises the steps that obstacle information of a vehicle marginal area is obtained through a sensor deployed in the vehicle marginal area; performing parametric modeling on a vehicle edge motion track based on the motion state information of the vehicle edge area; predicting an interaction risk level between the vehicle edge and the obstacle according to the obstacle information and the vehicle edge motion trail; determining a corresponding suspension adjustment level according to the predicted interaction risk level; generating a corresponding suspension adjusting instruction according to the suspension adjusting level; and the suspension adjusting instruction is sent to an air suspension control system so as to drive an air spring at the corresponding position of the vehicle to execute the local height adjusting action. By adopting the scheme provided by the invention, the risk that the edge of the vehicle is scratched can be reduced.

Description

Technical Field

[0001] This application relates to the field of automotive intelligent chassis and active safety technology, and in particular to a suspension adjustment method, device, system and storage medium. Background Technology

[0002] With the acceleration of urbanization, vehicles face complex road conditions caused by protrusions such as curbs, speed bumps, and manhole covers, which can easily lead to scrapes on the edges of the vehicle chassis (such as the front and rear bumpers and under the wheel arches). This not only causes vehicle damage and repair costs but may also affect driving safety. Although some high-end models are equipped with air suspension systems that can automatically adjust the suspension height according to preset conditions, current suspension height adjustment technology lacks the ability to accurately monitor vehicle edge areas (such as bumpers and wheel arches). It cannot effectively cope with the scrape risks caused by common obstacles such as curbs, manhole covers, and speed bumps on urban roads. Moreover, it is mostly based on the current ground clearance threshold to trigger suspension adjustment, which cannot predict collision risks in advance, resulting in scrapes on the vehicle edges.

[0003] Therefore, how to provide a suspension adjustment method to reduce the risk of scraping the vehicle's edges has become an urgent technical problem to be solved. Summary of the Invention

[0004] This application provides a suspension adjustment method, device, system, and storage medium to reduce the risk of edge scraping of a vehicle.

[0005] This application provides a suspension adjustment method, including: Information about obstacles in the vehicle's edge area is obtained through sensors deployed at the vehicle's edge. The vehicle edge motion trajectory is parametrically modeled based on the motion state information of the vehicle edge region; Predict the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory; The corresponding suspension adjustment level is determined based on the predicted level of interaction risk. Based on the suspension adjustment level, generate the corresponding suspension adjustment command; The suspension adjustment command is sent to the air suspension control system to drive the air springs at the corresponding positions of the vehicle to perform local height adjustment.

[0006] The beneficial effects of this application are as follows: This application uses sensors deployed in the vehicle edge area to acquire obstacle information, improving the monitoring capability of obstacles in the vehicle edge area; and it performs parametric modeling of the vehicle edge motion trajectory based on the motion state information of the vehicle edge area to predict the motion trajectory of the vehicle edge in advance; then, based on the obstacle information and the vehicle edge motion trajectory, it predicts the interaction risk level between the vehicle edge and the obstacle, determines the corresponding suspension adjustment level based on the predicted interaction risk level, and generates a corresponding suspension adjustment command; the suspension adjustment command is sent to the air suspension control system to drive the air springs at the corresponding positions of the vehicle to perform local height adjustment. Because the monitoring capability of the vehicle edge area is improved, the vehicle edge motion trajectory is predicted in advance, the risk of interaction between the vehicle edge and the obstacle is judged, and the suspension height is adjusted in advance, collisions caused by suspension adjustment lag are avoided, reducing the risk of scraping the vehicle edge.

[0007] In one embodiment, acquiring obstacle information in the vehicle edge region via sensors deployed in the vehicle edge region includes: Perception data about obstacles around the vehicle is collected by a multi-sensor array deployed at the edge of the vehicle, the multi-sensor array including at least millimeter-wave radar and vision sensors; Hardware-level time synchronization and data fusion processing are performed on the data collected by the millimeter-wave radar and visual sensor to generate obstacle information in the edge region.

[0008] In one embodiment, the parametric modeling of the vehicle edge motion trajectory based on the motion state information of the vehicle edge region includes: Obtain the coordinates of four control points at the edge of the vehicle; The coordinates of the four control points at the vehicle edge are substituted into the Bézier curve equation to dynamically simulate and update the motion trajectory of the vehicle edge.

[0009] In one embodiment, predicting the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory includes: The spatial features of point cloud acquired by the millimeter-wave radar and the temporal series features of roadside height acquired by the visual sensor are fused using a dual-channel spatiotemporal attention mechanism. Determine the nearest distance between the vehicle's edge and the obstacle; Based on the fused features and the nearest point distance between the vehicle edge and the obstacle, the interaction risk level is assessed in real time.

[0010] In one embodiment, the real-time assessment of the interaction risk level based on the fused features and the nearest point distance between the vehicle edge and the obstacle includes: When the distance to the obstacle is less than a first threshold and the vehicle speed is less than a first vehicle speed threshold, the interaction risk level is determined to be a warning level. When the distance to the obstacle is less than the second threshold and the vehicle speed is less than the second vehicle speed threshold, the interaction risk level is determined to be the mild adjustment level, wherein the second threshold is less than the first threshold; When the distance to the obstacle is less than the third threshold and the vehicle speed is greater than the third speed threshold, the interaction risk level is determined to be an emergency adjustment level, wherein the third threshold is less than the second threshold.

[0011] In one embodiment, determining the corresponding suspension adjustment level based on the predicted interaction risk level includes: When the interaction risk level is a warning level, the corresponding suspension adjustment level is determined to be the first adjustment level, which triggers a slight increase in the local suspension. When the interaction risk level is the mild adjustment level, the corresponding suspension adjustment level is determined to be the second adjustment level, which triggers a moderate increase in local suspension. When the interaction risk level is the emergency adjustment level, the corresponding suspension adjustment level is determined to be the third adjustment level, which triggers a rapid increase in the local suspension.

[0012] In one embodiment, the method further includes: The time window for predicting the interactive risk level is dynamically adjusted based on real-time vehicle speed; the higher the vehicle speed, the shorter the time window.

[0013] In one embodiment, the air suspension control system uses four independently controlled air springs and a parallel solenoid valve array as actuators, and controls the inflation and deflation rates of each air spring through pulse width modulation technology.

[0014] In one embodiment, generating a corresponding suspension adjustment command based on the suspension adjustment level includes: An improved fuzzy PID controller is used, which takes the obstacle information in the edge area, vehicle speed, and steering angle as inputs, and dynamically adjusts the PID control parameters according to the preset roadside safety distance threshold to generate the suspension adjustment command.

[0015] In one embodiment, the method operates in a distributed edge computing architecture, which includes a sensing unit, a decision-making unit, and an execution unit. The units communicate with each other via a CAN FD bus. The sensing unit performs the sensing data acquisition step and part of the data synchronization and processing step. The decision-making unit performs the remaining data synchronization and processing step, the trajectory modeling and risk prediction step, and the graded trigger judgment step. The execution unit performs the suspension adjustment execution step.

[0016] This application also provides a suspension adjustment device, including: The acquisition module is used to acquire obstacle information in the vehicle edge area through sensors deployed in the vehicle edge area; The modeling module is used to parametrically model the vehicle edge motion trajectory based on the motion state information of the vehicle edge region; The prediction module is used to predict the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory; The determination module is used to determine the corresponding suspension adjustment level based on the predicted level of interaction risk. The generation module is used to generate corresponding suspension adjustment commands based on the suspension adjustment level. The sending module is used to send the suspension adjustment command to the air suspension control system to drive the air spring at the corresponding position of the vehicle to perform a local height adjustment action.

[0017] In one embodiment, the acquisition module includes: The acquisition submodule is used to acquire perception data about obstacles around the vehicle through a multi-sensor array deployed in the edge area of ​​the vehicle, the multi-sensor array including at least millimeter-wave radar and vision sensors; The generation submodule is used to perform hardware-level time synchronization and data fusion processing on the data collected by the millimeter-wave radar and visual sensor to generate obstacle information in the edge region.

[0018] In one embodiment, the modeling module includes: The acquisition submodule is used to acquire the coordinates of four control points on the edge of the vehicle; The simulation submodule is used to substitute the coordinates of the four control points at the edge of the vehicle into the Bézier curve equation to dynamically simulate and update the motion trajectory of the vehicle edge.

[0019] In one embodiment, the prediction module includes: The fusion submodule is used to fuse the spatial features of point cloud acquired by the millimeter-wave radar and the temporal series features of roadside height acquired by the visual sensor using a dual-channel spatiotemporal attention mechanism. The determination submodule is used to determine the nearest distance between the vehicle edge and the obstacle; The evaluation submodule is used to assess the interaction risk level in real time based on the fused features and the nearest point distance between the vehicle edge and the obstacle.

[0020] In one embodiment, the evaluation submodule is further configured to: When the distance to the obstacle is less than a first threshold and the vehicle speed is less than a first vehicle speed threshold, the interaction risk level is determined to be a warning level. When the distance to the obstacle is less than the second threshold and the vehicle speed is less than the second vehicle speed threshold, the interaction risk level is determined to be the mild adjustment level, wherein the second threshold is less than the first threshold; When the distance to the obstacle is less than the third threshold and the vehicle speed is greater than the third speed threshold, the interaction risk level is determined to be an emergency adjustment level, wherein the third threshold is less than the second threshold.

[0021] In one embodiment, the determining module is further configured to: When the interaction risk level is a warning level, the corresponding suspension adjustment level is determined to be the first adjustment level, which triggers a slight increase in the local suspension. When the interaction risk level is the mild adjustment level, the corresponding suspension adjustment level is determined to be the second adjustment level, which triggers a moderate increase in local suspension. When the interaction risk level is the emergency adjustment level, the corresponding suspension adjustment level is determined to be the third adjustment level, which triggers a rapid increase in the local suspension.

[0022] In one embodiment, the apparatus further includes: The adjustment module is used to dynamically adjust the time window for predicting the interaction risk level based on the real-time vehicle speed; the higher the vehicle speed, the shorter the time window.

[0023] In one embodiment, the air suspension control system uses four independently controlled air springs and a parallel solenoid valve array as actuators, and controls the inflation and deflation rates of each air spring through pulse width modulation technology.

[0024] In one embodiment, the generation module includes: An improved fuzzy PID controller is used, which takes the obstacle information in the edge area, vehicle speed, and steering angle as inputs, and dynamically adjusts the PID control parameters according to the preset roadside safety distance threshold to generate the suspension adjustment command.

[0025] This application also provides a suspension adjustment system, including: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to implement the suspension adjustment method described in any of the above embodiments.

[0026] This application also provides a computer-readable storage medium, which, when the instructions in the storage medium are executed by a processor corresponding to the suspension adjustment system, enables the suspension adjustment system to implement the suspension adjustment method described in any of the above embodiments.

[0027] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0028] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0029] The accompanying drawings are provided to further illustrate the present application and form part of the specification. They are used together with the embodiments of the present application to explain the application and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a suspension adjustment method in one embodiment of this application; Figure 2 This is a schematic diagram of the structure of a suspension adjustment device according to an embodiment of this application; Figure 3 This is a schematic diagram of the hardware structure of a suspension adjustment system according to an embodiment of this application. Detailed Implementation

[0030] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application.

[0031] Figure 1 This is a flowchart of a suspension adjustment method according to an embodiment of this application, as follows: Figure 1 As shown, the method can be implemented as follows: S101-S106: In step S101, obstacle information in the vehicle edge area is obtained by sensors deployed in the vehicle edge area; In step S102, the motion trajectory of the vehicle edge is parametrically modeled based on the motion state information of the vehicle edge region; In step S103, the interaction risk level between the vehicle edge and the obstacle is predicted based on the obstacle information and the vehicle edge motion trajectory; In step S104, the corresponding suspension adjustment level is determined based on the predicted interaction risk level; In step S105, a corresponding suspension adjustment command is generated based on the suspension adjustment level; In step S106, the suspension adjustment command is sent to the air suspension control system to drive the air spring at the corresponding position of the vehicle to perform a local height adjustment action.

[0032] Obstacle information in the vehicle's edge area is acquired through sensors deployed at the vehicle's edge. A multi-sensor array deployed at the vehicle's edge area collects perception data about obstacles around the vehicle. This multi-sensor array includes at least millimeter-wave radar and a vision sensor. The obstacle information includes at least the distance and outline of obstacles in the vehicle's edge area. For example, short-range millimeter-wave radar arrays and vision sensors (such as wide-angle cameras or fisheye lenses) are distributed and deployed at the four corners of the front and rear bumpers and wheel arches to acquire obstacle distances and images in the vehicle's edge area. Then, the data collected by the millimeter-wave radar and vision sensors undergoes hardware-level time synchronization and data fusion processing to generate obstacle information for the edge area. Through hardware-level synchronization (such as FPGA) or high-precision timestamp alignment technology, the acquisition latency of all sensor data is controlled to within 10 milliseconds, providing a highly timely data foundation for subsequent predictive decisions.

[0033] For example, firstly, a short-range millimeter-wave radar array is deployed in the vehicle's edge areas (front and rear bumpers, and the four corners of the wheel arches) to form 360° coverage. Each radar unit uses 77GHz high-frequency millimeter-wave technology, which has high directionality and anti-interference capabilities, and can accurately detect obstacles at a distance of 0-2 meters. By arranging multiple radar units at specific angles to form a dense detection grid, the coverage blind spot problem of traditional single-point radar is solved. Simultaneously, wide-angle cameras and fisheye lenses are installed at the front and rear bumpers, and the Canny edge detection algorithm from OpenCV is used to extract the curb contour. To address the perspective distortion problem in low-angle areas such as below the wheel arches, multi-camera stitching technology and deep learning models (such as MobileNet-Canny fusion) are used for distortion compensation and feature extraction. This design can effectively identify key parameters such as the width and height of the curb, compensating for the shortcomings of the vision system in static obstacle detection. Furthermore, the data collected by multiple millimeter-wave radars and vision sensors is synchronized at the hardware level (such as FPGA) or through timestamp alignment technology to solve the multi-sensor data synchronization problem. By unifying the trigger timing of millimeter-wave radar, cameras, and vehicle height sensors, data latency is reduced to less than 10ms, ensuring system real-time performance. Simultaneously, using onboard edge computing chips (such as NVIDIA Jetson Orin) for real-time processing of multi-sensor data reduces system costs. The system also incorporates four-corner Bézier curve trajectory modeling. By deploying sensors on the front and rear bumpers and wheel arches, the coordinates of four control points are acquired, and cubic Bézier curves are used to dynamically model the edge trajectory. This method can accurately calculate the closest distance between the vehicle edge and obstacles, providing a mathematical basis for suspension adjustment.

[0034] Based on the motion state information of the vehicle edge region, the vehicle edge motion trajectory is parametrically modeled. Specifically, the coordinates of four control points at the vehicle edge are obtained, and these coordinates are substituted into the Bézier curve equation to dynamically simulate and update the vehicle edge motion trajectory. For example, the sensor positions deployed on the front and rear bumpers and wheel arches are used as control points, and the vehicle edge motion trajectory is calculated in real time using the Bézier curve formula. In one embodiment, the Bézier curve formula is as follows: P(t) = (1-t)³P0+ 3t(1-t)²P1+ 3t²(1-t)P2+ t³P3, (0≤t≤1); Where P(t) represents the coordinates of the point on the curve corresponding to time t; P i B represents the coordinates of the control point. i,3 (t) is a cubic Bernstein basis function.

[0035] The risk level of interaction between the vehicle edge and the obstacle is predicted based on the obstacle information and the vehicle edge motion trajectory.

[0036] This application utilizes a dual-channel spatiotemporal attention mechanism to fuse the spatial features of point clouds acquired by the millimeter-wave radar and the temporal series features of roadside height acquired by the visual sensor. Specifically, it includes two parallel attention channels: one analyzes the 3D point cloud data generated by the millimeter-wave radar, focusing on areas with high point cloud density (i.e., strong obstacle reflection); the other analyzes the continuous frame image sequence captured by the visual sensor, extracting the temporal features of roadside or obstacle height changes. The outputs of the two channels are weighted and fused to dynamically adjust the contribution of different features, forming a more comprehensive environmental perception. For example, when driving at high speeds, spatial features have a higher weight (due to potentially insufficient visual frame rate); during sharp turns, temporal features have a higher weight (due to more significant relative movement of obstacles).

[0037] Furthermore, the closest distance between the vehicle edge and the obstacle is determined. Based on the Bézier curve trajectory model, the theoretical distance from each point on the vehicle edge trajectory to the obstacle is calculated. Through spatial alignment technology, radar point clouds and visual features are mapped to a unified trajectory coordinate system to determine the point on the trajectory closest to the obstacle, which serves as the basis for key risk assessment.

[0038] Furthermore, based on the fused features and the nearest point distance between the vehicle edge and the obstacle, the interaction risk level is assessed in real time. For example, when the obstacle distance is less than a first threshold and the vehicle speed is less than a first speed threshold, the interaction risk level is determined to be a warning level; when the obstacle distance is less than a second threshold and the vehicle speed is less than a second speed threshold, the interaction risk level is determined to be a gradual adjustment level, where the second threshold is less than the first threshold; when the obstacle distance is less than a third threshold and the vehicle speed is greater than a third speed threshold, the interaction risk level is determined to be a rapid adjustment level, where the third threshold is less than the second threshold.

[0039] Of course, it can also obtain real-time vehicle parameters, such as vehicle speed, acceleration, and steering status. By integrating and fusing features, the nearest point distance, and real-time vehicle parameters, multi-dimensional information can be used to output the interaction risk probability using a probabilistic model (such as logistic regression or neural network) to reflect the possibility of a collision. Based on the interaction risk probability, it can be mapped to the corresponding interaction risk level (such as low, medium, and high), thus realizing intelligent decision-making from single distance judgment to multi-factor comprehensive evaluation.

[0040] The corresponding suspension adjustment level is determined based on the predicted interaction risk level. In one embodiment, a pre-determined correspondence between the interaction risk level and the suspension adjustment level is established. When the interaction risk level is a warning level, the corresponding suspension adjustment level is determined as a first adjustment level, which triggers a slight local suspension rise (e.g., 5mm); when the interaction risk level is a moderate adjustment level, the corresponding suspension adjustment level is determined as a second adjustment level, which triggers a moderate local suspension rise (e.g., 10mm); when the interaction risk level is a rapid adjustment level, the corresponding suspension adjustment level is determined as a third adjustment level, which triggers a rapid local suspension rise (e.g., 15mm).

[0041] In another embodiment, when generating suspension adjustment commands, the suspension adjustment height is determined based on the suspension adjustment level, the distance between the vehicle and the obstacle, and the obstacle height, and a corresponding adjustment command is generated. Specifically, the obstacle height is obtained; if the obstacle height is higher than a first preset height, an obstacle warning is issued; if the obstacle height is lower than the first preset height, the corresponding basic adjustment height is determined based on the suspension adjustment level; for example, each adjustment level corresponds to a basic adjustment height range, such as 5mm for warning level, 10mm for gradual adjustment level, and 15mm for rapid adjustment level. The distance between the vehicle and the obstacle is obtained, and a distance compensation factor is calculated based on the distance between the vehicle and the obstacle, and a height correction factor is calculated based on the obstacle height; the distance compensation factor is calculated based on the ratio of the actual obstacle distance to a preset threshold, with a larger compensation factor for closer distances. For example, a compensation factor of 1.5 corresponds to a distance less than 10m; a compensation factor of 1.2 corresponds to a distance between 10-20m; and a compensation factor of 1.0 corresponds to a distance greater than 20m. Based on the obstacle height, a corresponding height correction factor is determined: 1.5 for obstacles greater than 30cm, 1.2 for obstacles between 20-30cm, and 1.0 for obstacles less than 20cm. Finally, the final suspension adjustment height is calculated based on the base height, distance compensation factor, and height correction factor. Final suspension adjustment height = Base height × Distance compensation factor × Height correction factor.

[0042] Based on the suspension adjustment level, a corresponding suspension adjustment command is generated. For example, an improved fuzzy PID controller can be used, taking the obstacle information in the edge area, vehicle speed, and steering angle as inputs, and dynamically adjusting the PID control parameters according to a preset roadside safety distance threshold to generate the suspension adjustment command.

[0043] The suspension adjustment command is sent to the air suspension control system to drive the air springs at the corresponding positions on the vehicle to perform local height adjustment. The air suspension control system can use four independently controlled air springs and a parallel solenoid valve array as actuators, and control the inflation and deflation rates of each air spring using pulse width modulation technology. Alternatively, the air suspension control system can use hydraulic actuators instead of solenoid valves, achieving precise suspension adjustment through hydraulic pumps and cylinders.

[0044] When adjusting suspension height, a local suspension adjustment priority strategy can be adopted, raising the suspension height only on the side with interactive risks to avoid stability issues caused by changes in overall vehicle height. Specifically, an actuator design using four independent air springs at each corner and a parallel solenoid valve array, combined with a high-pressure air tank (above 750kPa), achieves millisecond-level predictive suspension adjustment. Each air spring is equipped with two sets of parallel solenoid valves, and the inflation and deflation rates are controlled through PWM pulse width modulation technology, reducing the single-wheel height adjustment time to within 1 second, far superior to the 5-10 second adjustment time of traditional systems.

[0045] In addition, a second preset height can be set. When the final suspension adjustment height is less than the second preset height, partial suspension adjustment is used. When the final suspension adjustment height is greater than the second preset height, overall suspension adjustment is used to avoid poor user experience caused by vehicle imbalance.

[0046] When the suspension condition is triggered, the interaction time between the vehicle edge and the obstacle can be determined based on the vehicle edge trajectory prediction, and the suspension can be raised 0.5-1 seconds in advance according to the interaction time to cover the blind spot of traditional threshold triggering.

[0047] In this application, the system can also employ a tiered triggering logic, setting a three-level adjustment strategy based on obstacle distance and vehicle speed: at low speeds (V<30km / h), priority is given to triggering adjustment using vision + radar fusion data; at medium speeds (30km / h≤V<50km / h), the system relies on a trajectory prediction model; and at high speeds (V≥50km / h), the prediction window is shortened to 0.5 seconds to ensure the rationality of the triggering timing. This design balances real-time performance with energy consumption, avoiding energy loss caused by frequent adjustments.

[0048] In this application, the time window for predicting the interaction risk level is dynamically adjusted based on real-time vehicle speed; the higher the vehicle speed, the shorter the time window. A predictive adjustment mechanism is employed, which calculates the prediction time window to anticipate the timing of the risk. The prediction time window is the estimated time from the current moment until the vehicle's edge interacts with the obstacle (i.e., the closest distance decreases to zero). The length of this window directly affects the lead time and response timing of the suspension adjustment, and is a key parameter for achieving millisecond-level predictive adjustment. For example, the prediction time window can be dynamically adjusted according to the following formula:

[0049] Where W is the prediction time window; Dmin is the nearest point distance; vrel is the relative speed between the vehicle and the obstacle; α(V, θ) is a correction factor, which is related to vehicle speed and steering angle. For example, at high speeds (V>60km / h), α is reduced to shorten the prediction time due to time constraints; at low speeds (V<20km / h), α is increased to extend the prediction time to reduce misjudgments; and during sharp turns (θ change rate is greater than the preset value), α is dynamically adjusted according to the steering angle change rate.

[0050] In one embodiment, the method operates in a distributed edge computing architecture, which includes a sensing unit, a decision-making unit, and an execution unit. The units communicate with each other via a CAN FD bus. The sensing unit performs the sensing data acquisition step and part of the data synchronization and processing step. The decision-making unit performs the remaining data synchronization and processing step, the trajectory modeling and risk prediction step, and the graded trigger judgment step. The execution unit performs the suspension adjustment execution step.

[0051] In one specific embodiment, the system hardware comprises: 1. Distributed short-range millimeter-wave radar array (8 elements, 77 GHz); 2. Wide-angle camera + fisheye lens (2 units, 120° field of view); 3. Edge computing chip (NVIDIA Jetson Orin, 16GB memory); 4. High-pressure gas storage tank (750kPa, capacity 5L); 5. Four independent air springs at the corners + parallel solenoid valve array; 6. Vehicle height sensors (4 units, installed near the shock absorbers).

[0052] The software architecture adopts a layered design, including: 1. Data Acquisition Layer: Responsible for real-time acquisition and preprocessing of data from multiple sensors; 2. Feature Extraction Layer: Extracts key features using a spatiotemporal attention mechanism; 3. Risk Prediction Layer: Calculates interaction risk based on the Bézier curve model; 4. Control Decision Layer: An improved fuzzy PID algorithm is used to generate control commands; 5. Execution control layer: Controls the air suspension system via CAN FD bus.

[0053] By employing distributed multi-sensor fusion detection technology and an edge computing-driven intelligent air suspension adjustment mechanism, this system achieves precise monitoring and real-time control of vehicle ground clearance, overcoming the shortcomings of traditional ground clearance detection technologies in terms of edge coverage, real-time performance, environmental adaptability, and predictability. The core innovation of this system lies in combining edge trajectory modeling with a spatiotemporal attention mechanism to achieve millisecond-level predictive suspension adjustment, providing a comprehensive chassis protection solution for passenger vehicles.

[0054] This application utilizes distributed multi-sensor fusion detection technology deployed in the vehicle's edge area to achieve comprehensive dynamic monitoring of key components such as bumpers and wheel arches. This makes the detection of ground clearance in the edge area more accurate and reliable, effectively solving the monitoring blind spot problem existing in traditional systems. Through four-corner Bezier curve trajectory modeling and a dual-channel spatiotemporal attention mechanism, the system can predict the interaction risks between the vehicle and obstacles in advance, making suspension adjustments more timely and avoiding the adjustment lag problem caused by traditional systems relying on the current ground clearance threshold. At the same time, the use of four-corner independent air springs and a parallel solenoid valve array design, combined with an environmental adaptive sensor combination, significantly improves the system's response speed and stability in complex road conditions such as rain, snow, and fog, providing safer and more intelligent chassis protection capabilities for family cars.

[0055] The beneficial effects of this application are as follows: This application uses sensors deployed in the vehicle edge area to acquire obstacle information, improving the monitoring capability of obstacles in the vehicle edge area; and it performs parametric modeling of the vehicle edge motion trajectory based on the motion state information of the vehicle edge area to predict the motion trajectory of the vehicle edge in advance; then, based on the obstacle information and the vehicle edge motion trajectory, it predicts the interaction risk level between the vehicle edge and the obstacle, determines the corresponding suspension adjustment level based on the predicted interaction risk level, and generates a corresponding suspension adjustment command; the suspension adjustment command is sent to the air suspension control system to drive the air springs at the corresponding positions of the vehicle to perform local height adjustment. Because the monitoring capability of the vehicle edge area is improved, the vehicle edge motion trajectory is predicted in advance, the risk of interaction between the vehicle edge and the obstacle is judged, and the suspension height is adjusted in advance, collisions caused by suspension adjustment lag are avoided, reducing the risk of scraping the vehicle edge.

[0056] In one embodiment, step S101 above can be implemented as steps A1-A2 as follows: In step A1, perception data about obstacles around the vehicle is collected by a multi-sensor array deployed at the edge of the vehicle. The multi-sensor array includes at least millimeter-wave radar and a vision sensor. In step A2, hardware-level time synchronization and data fusion processing are performed on the data collected by the millimeter-wave radar and the vision sensor to generate obstacle information in the edge region.

[0057] In one embodiment, step S102 above can be implemented as steps B1-B2 as follows: In step B1, the coordinates of four control points on the edge of the vehicle are obtained; In step B2, the coordinates of the four control points at the vehicle edge are substituted into the Bézier curve equation to dynamically simulate and update the motion trajectory of the vehicle edge.

[0058] In one embodiment, step S103 above can be implemented as steps C1-C3 as follows: In step C1, a dual-channel spatiotemporal attention mechanism is used to fuse the spatial features of the point cloud acquired by the millimeter-wave radar and the temporal series features of the roadside height acquired by the visual sensor. In step C2, the nearest distance between the vehicle edge and the obstacle is determined; In step C3, the interaction risk level is assessed in real time based on the fused features and the nearest point distance between the vehicle edge and the obstacle.

[0059] In one embodiment, step C3 above can be implemented as steps C31-C33: In step C31, when the obstacle distance is less than a first threshold and the vehicle speed is less than a first vehicle speed threshold, the interaction risk level is determined to be a warning level. In step C32, when the obstacle distance is less than the second threshold and the vehicle speed is less than the second vehicle speed threshold, the interaction risk level is determined to be a mild adjustment level, wherein the second threshold is less than the first threshold; In step C33, when the obstacle distance is less than the third threshold and the vehicle speed is greater than the third vehicle speed threshold, the interaction risk level is determined to be an emergency adjustment level, wherein the third threshold is less than the second threshold.

[0060] In one embodiment, step S104 above can be implemented as steps D1-D3 as follows: In step D1, when the interaction risk level is a warning level, the corresponding suspension adjustment level is determined to be the first adjustment level, which is to trigger a slight increase in the local suspension. In step D2, when the interaction risk level is the mild adjustment level, the corresponding suspension adjustment level is determined to be the second adjustment level, which is to trigger a moderate increase in local suspension. In step D3, when the interaction risk level is the emergency adjustment level, the corresponding suspension adjustment level is determined to be the third adjustment level, which triggers a rapid increase in the local suspension.

[0061] In one embodiment, the method may also be implemented as follows: The time window for predicting the interactive risk level is dynamically adjusted based on real-time vehicle speed; the higher the vehicle speed, the shorter the time window.

[0062] In one embodiment, the air suspension control system uses four independently controlled air springs and a parallel solenoid valve array as actuators, and controls the inflation and deflation rates of each air spring through pulse width modulation technology.

[0063] In one embodiment, step S105 above can be implemented as follows: An improved fuzzy PID controller is used, which takes the obstacle information in the edge area, vehicle speed, and steering angle as inputs, and dynamically adjusts the PID control parameters according to the preset roadside safety distance threshold to generate the suspension adjustment command.

[0064] In one embodiment, the method operates in a distributed edge computing architecture, which includes a sensing unit, a decision-making unit, and an execution unit. The units communicate with each other via a CAN FD bus. The sensing unit performs the sensing data acquisition step and part of the data synchronization and processing step. The decision-making unit performs the remaining data synchronization and processing step, the trajectory modeling and risk prediction step, and the graded trigger judgment step. The execution unit performs the suspension adjustment execution step.

[0065] Figure 2 This is a schematic diagram of the structure of a suspension adjustment device according to one embodiment of this application, as shown below. Figure 2 As shown, the device includes: The acquisition module 201 is used to acquire obstacle information in the edge area of ​​the vehicle through sensors deployed in the edge area of ​​the vehicle; Modeling module 202 is used to perform parametric modeling of the vehicle edge motion trajectory based on the motion state information of the vehicle edge region; Prediction module 203 is used to predict the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory; The determination module 204 is used to determine the corresponding suspension adjustment level based on the predicted interaction risk level; The generation module 205 is used to generate a corresponding suspension adjustment command based on the suspension adjustment level; The sending module 206 is used to send the suspension adjustment command to the air suspension control system to drive the air spring at the corresponding position of the vehicle to perform a local height adjustment action.

[0066] In one embodiment, the acquisition module includes: The acquisition submodule is used to acquire perception data about obstacles around the vehicle through a multi-sensor array deployed in the edge area of ​​the vehicle, the multi-sensor array including at least millimeter-wave radar and vision sensors; The generation submodule is used to perform hardware-level time synchronization and data fusion processing on the data collected by the millimeter-wave radar and visual sensor to generate obstacle information in the edge region.

[0067] In one embodiment, the modeling module includes: The acquisition submodule is used to acquire the coordinates of four control points on the edge of the vehicle; The simulation submodule is used to substitute the coordinates of the four control points at the edge of the vehicle into the Bézier curve equation to dynamically simulate and update the motion trajectory of the vehicle edge.

[0068] In one embodiment, the prediction module includes: The fusion submodule is used to fuse the spatial features of point cloud acquired by the millimeter-wave radar and the temporal series features of roadside height acquired by the visual sensor using a dual-channel spatiotemporal attention mechanism. The determination submodule is used to determine the nearest distance between the vehicle edge and the obstacle; The evaluation submodule is used to assess the interaction risk level in real time based on the fused features and the nearest point distance between the vehicle edge and the obstacle.

[0069] In one embodiment, the evaluation submodule is further configured to: When the distance to the obstacle is less than a first threshold and the vehicle speed is less than a first vehicle speed threshold, the interaction risk level is determined to be a warning level. When the distance to the obstacle is less than the second threshold and the vehicle speed is less than the second vehicle speed threshold, the interaction risk level is determined to be the mild adjustment level, wherein the second threshold is less than the first threshold; When the distance to the obstacle is less than the third threshold and the vehicle speed is greater than the third speed threshold, the interaction risk level is determined to be an emergency adjustment level, wherein the third threshold is less than the second threshold.

[0070] In one embodiment, the determining module is further configured to: When the interaction risk level is a warning level, the corresponding suspension adjustment level is determined to be the first adjustment level, which triggers a slight increase in the local suspension. When the interaction risk level is the mild adjustment level, the corresponding suspension adjustment level is determined to be the second adjustment level, which triggers a moderate increase in local suspension. When the interaction risk level is the emergency adjustment level, the corresponding suspension adjustment level is determined to be the third adjustment level, which triggers a rapid increase in the local suspension.

[0071] In one embodiment, the apparatus further includes: The adjustment module is used to dynamically adjust the time window for predicting the interaction risk level based on the real-time vehicle speed; the higher the vehicle speed, the shorter the time window.

[0072] In one embodiment, the air suspension control system uses four independently controlled air springs and a parallel solenoid valve array as actuators, and controls the inflation and deflation rates of each air spring through pulse width modulation technology.

[0073] In one embodiment, the generation module includes: An improved fuzzy PID controller is used, which takes the obstacle information in the edge area, vehicle speed, and steering angle as inputs, and dynamically adjusts the PID control parameters according to the preset roadside safety distance threshold to generate the suspension adjustment command.

[0074] Figure 3 This is a schematic diagram of the hardware structure of a suspension adjustment system according to an embodiment of this application, as shown below. Figure 3 As shown, the suspension adjustment system includes: At least one processor 320; and, Memory 304 communicatively connected to the at least one processor 320; wherein, The memory 304 stores instructions that can be executed by the at least one processor 320 to implement the suspension adjustment method described in any of the above embodiments.

[0075] Reference Figure 3 The suspension adjustment system 300 may include one or more of the following components: a processing component 302, a memory 304, a power supply component 306, an input / output (I / O) interface 308, a sensor component 310, and a communication component 312.

[0076] Processing component 302 typically controls the overall operation of suspension adjustment system 300. Processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the method described above. Furthermore, processing component 302 may include one or more modules to facilitate interaction between processing component 302 and other components. The processor 320 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0077] Memory 304 is configured to store various types of data to support the operation of suspension adjustment system 300. Examples of this data include instructions for any application or method operating on suspension adjustment system 300. Memory 304 may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. Memory 304 may also be an external storage device of the terminal device, such as a plug-in hard disk equipped on the terminal device. Memory 304 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Memory 304 is used to store programs and data required by this application. Memory 304 may also be used to temporarily store data that has been output or will be output.

[0078] The power supply assembly 306 provides power to the various components of the suspension adjustment system 300. The power supply assembly 306 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the suspension adjustment system 300.

[0079] I / O interface 308 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc.

[0080] The sensor assembly 310 includes one or more sensors for providing status assessments of various aspects of the suspension adjustment system 300. Additionally, the sensor assembly 310 can detect the on / off state of the suspension adjustment system 300, the relative positioning of components, and the operational status of the suspension adjustment system 300 or a component of the suspension adjustment system 300. In some embodiments, the sensor assembly 310 may include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor, etc.

[0081] Communication component 312 is configured to enable suspension adjustment system 300 to provide wired or wireless communication capabilities with other devices and cloud platforms. Suspension adjustment system 300 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 316 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 316 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0082] In an exemplary embodiment, the suspension adjustment system 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the suspension adjustment method described in any of the above embodiments.

[0083] This application also provides a computer-readable storage medium, which, when the instructions in the storage medium are executed by a processor corresponding to the suspension adjustment system, enables the suspension adjustment system to implement the suspension adjustment method described in any of the above embodiments.

[0084] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0085] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0086] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0088] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A suspension adjustment method, characterized in that, include: Information about obstacles in the vehicle's edge area is obtained through sensors deployed at the vehicle's edge. The vehicle edge motion trajectory is parametrically modeled based on the motion state information of the vehicle edge region; Predict the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory; The corresponding suspension adjustment level is determined based on the predicted level of interaction risk. Based on the suspension adjustment level, generate the corresponding suspension adjustment command; The suspension adjustment command is sent to the air suspension control system to drive the air springs at the corresponding positions of the vehicle to perform local height adjustment.

2. The method as described in claim 1, characterized in that, The acquisition of obstacle information in the vehicle edge area through sensors deployed in the vehicle edge area includes: Perception data about obstacles around the vehicle is collected by a multi-sensor array deployed at the edge of the vehicle, the multi-sensor array including at least millimeter-wave radar and vision sensors; Hardware-level time synchronization and data fusion processing are performed on the data collected by the millimeter-wave radar and visual sensor to generate obstacle information in the edge region.

3. The method as described in claim 1, characterized in that, The parametric modeling of the vehicle edge motion trajectory based on the motion state information of the vehicle edge region includes: Obtain the coordinates of four control points at the edge of the vehicle; The coordinates of the four control points at the vehicle edge are substituted into the Bézier curve equation to dynamically simulate and update the motion trajectory of the vehicle edge.

4. The method as described in claim 1, characterized in that, The step of predicting the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory includes: The spatial features of point cloud acquired by the millimeter-wave radar and the temporal series features of roadside height acquired by the visual sensor are fused using a dual-channel spatiotemporal attention mechanism. Determine the nearest distance between the vehicle's edge and the obstacle; Based on the fused features and the nearest point distance between the vehicle edge and the obstacle, the interaction risk level is assessed in real time.

5. The method as described in claim 4, characterized in that, The interaction risk level is assessed in real time based on the fused features and the nearest point distance between the vehicle edge and the obstacle, including: When the distance to the obstacle is less than a first threshold and the vehicle speed is less than a first vehicle speed threshold, the interaction risk level is determined to be a warning level. When the distance to the obstacle is less than the second threshold and the vehicle speed is less than the second vehicle speed threshold, the interaction risk level is determined to be the mild adjustment level, wherein the second threshold is less than the first threshold; When the distance to the obstacle is less than the third threshold and the vehicle speed is greater than the third speed threshold, the interaction risk level is determined to be an emergency adjustment level, wherein the third threshold is less than the second threshold.

6. The method as described in claim 1, characterized in that, The method further includes: The time window for predicting the interactive risk level is dynamically adjusted based on real-time vehicle speed; the higher the vehicle speed, the shorter the time window.

7. The method as described in claim 1, characterized in that, The air suspension control system uses four independently controlled air springs and a parallel solenoid valve array as actuators, and controls the inflation and deflation rates of each air spring through pulse width modulation technology.

8. A suspension adjustment device, characterized in that, include: The acquisition module is used to acquire obstacle information in the vehicle edge area through sensors deployed in the vehicle edge area; The modeling module is used to parametrically model the vehicle edge motion trajectory based on the motion state information of the vehicle edge region; The prediction module is used to predict the interaction risk level between the vehicle edge and the obstacle based on the obstacle information and the vehicle edge motion trajectory; The determination module is used to determine the corresponding suspension adjustment level based on the predicted level of interaction risk. The generation module is used to generate corresponding suspension adjustment commands based on the suspension adjustment level; The sending module is used to send the suspension adjustment command to the air suspension control system to drive the air spring at the corresponding position of the vehicle to perform a local height adjustment action.

9. A suspension adjustment system, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to implement the suspension adjustment method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor corresponding to the suspension adjustment system, the suspension adjustment system is able to implement the suspension adjustment method as described in any one of claims 1-7.

Images