Suspension control method, system, vehicle and device for a vehicle

Abstract

The application discloses a suspension control method and system of a vehicle, a vehicle and equipment. The suspension control method of the vehicle comprises the following steps: obtaining multi-source sensing data, wherein the multi-source sensing data comprises suspension height, suspension air bag pressure, road surface information and vehicle dynamic state information; generating an optimal control target of suspension parameters by using an NSGA-II multi-objective optimization algorithm according to the multi-source sensing data, wherein the optimization target of the NSGA-II multi-objective optimization algorithm comprises comfort, stability and vehicle energy consumption; generating a control sequence of an electromagnetic valve by using a DDPG deep reinforcement learning algorithm; and controlling the electromagnetic valve of the suspension according to the optimal control target and the control sequence of the electromagnetic valve, so as to realize adaptive adjustment of the suspension. The application provides an intelligent suspension control strategy with high precision, high reliability and low energy consumption for the vehicle, and has high application value.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more particularly to a suspension control method, system, vehicle, and equipment for a vehicle. Background Technology

[0002] Current electronically controlled air suspension systems mainly consist of a height sensor, solenoid valve assembly, electronic control unit, and air supply system, communicating via a CAN bus. These systems typically employ simple open-loop control strategies (such as solenoid valve control with fixed-time parameters), have a limited sensing dimension, and rely primarily on basic sensor data for reactive control. While existing technologies offer some capability in suspension height adjustment, load distribution, and basic fault diagnosis, they suffer from significant shortcomings in control accuracy and adaptability, failing to adequately meet the demands of vehicles, especially commercial vehicles, for high-precision, high-reliability suspension control. Summary of the Invention

[0003] Therefore, it is necessary to provide a vehicle suspension control method, system, vehicle, and equipment to address the aforementioned technical problems. This provides a high-precision, high-reliability, and low-energy-consumption intelligent suspension control strategy for vehicles such as new energy commercial vehicles, and has high application value.

[0004] Firstly, a suspension control method for a vehicle is provided, comprising: Obtain multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic status information; Based on the multi-source sensor data, the optimal control objectives of the suspension parameters are generated using the NSGA-II multi-objective optimization algorithm. The optimization objectives of the NSGA-II multi-objective optimization algorithm include comfort, stability, and vehicle energy consumption. The control sequence for the solenoid valve is generated using the DDPG deep reinforcement learning algorithm. Based on the optimal control objective and the control sequence of the solenoid valves, the solenoid valves of the suspension are controlled to achieve adaptive adjustment of the suspension.

[0005] In some examples, obtaining multi-source sensing data includes: The suspension height is obtained from the height sensors, wherein each axle is equipped with multiple independent height sensors, and the multiple height sensors are arranged in a spatially separated manner; The airbag pressure detected by the air pressure sensor is obtained, wherein the air pressure sensor is configured at the inlet and outlet of each airbag; Obtain road surface information detected by the front-facing camera and millimeter-wave radar; It acquires vehicle dynamic status information detected by IMU, wheel speed sensor and steering angle sensor.

[0006] In some examples, before generating the optimal control objective for the suspension parameters using the NSGA-II multi-objective optimization algorithm based on the multi-source sensor data, the method further includes: The multi-source sensor data is spatiotemporally aligned and fused; A road-vehicle coupling model is constructed based on the fused data, wherein the road-vehicle coupling model is used to extract the frequency domain / time domain features of road excitation; Create an LSTM-based road prediction module to predict road information ahead based on the output of the road-vehicle coupling model.

[0007] In some examples, generating the optimal control objective for the suspension parameters based on the multi-source sensor data using the NSGA-II multi-objective optimization algorithm includes: Based on the NSGA-II multi-objective optimization algorithm, combining comfort, stability and energy consumption objectives, the optimal suspension parameters are generated using the multi-source sensor data, and the optimal control objective of the suspension parameters is obtained based on the optimal suspension parameters.

[0008] In some examples, the generation of the control sequence for the solenoid valve using the DDPG deep reinforcement learning algorithm includes: The control sequence of the solenoid valve is obtained by using the DDPG deep reinforcement learning algorithm combined with the precise hysteresis model of the solenoid valve. The precise hysteresis model of the solenoid valve represents the mapping relationship between the opening / closing delay time and the air pressure.

[0009] In some examples, it also includes: Obtain the vehicle's current operating conditions; The optimal control target is adjusted based on the current driving conditions.

[0010] In some examples, it also includes: In battery swapping mode, it responds to the driver's input commands to control the lifting of one side of the wheels; When a malfunction occurs, control the vehicle to switch to safe mode.

[0011] Secondly, a vehicle suspension control system is provided, comprising: The acquisition module is used to acquire multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information and vehicle dynamic status information; The control target determination module is used to generate the optimal control target of the suspension parameters based on the multi-source sensor data and using the NSGA-II multi-objective optimization algorithm, wherein the optimization targets of the NSGA-II multi-objective optimization algorithm include comfort, stability and vehicle energy consumption. The control timing determination module is used to generate the control sequence of the solenoid valve using the DDPG deep reinforcement learning algorithm. The control module is used to control the solenoid valves of the suspension according to the optimal control target and the control sequence of the solenoid valves, so as to realize the adaptive adjustment of the suspension.

[0012] Thirdly, a vehicle is provided, comprising: a suspension control system for the vehicle according to the second aspect described above.

[0013] Fourthly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the steps of the suspension control method for a vehicle according to the first aspect and any possible implementation thereof.

[0014] Fifthly, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the vehicle suspension control method of the first aspect and any possible implementation thereof.

[0015] Sixthly, a computer program product is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the vehicle suspension control method of the first aspect and any possible implementation thereof.

[0016] The embodiments of this application solve the problems of control accuracy, safety and reliability, environmental adaptability and energy efficiency of traditional electronically controlled suspension systems, and provide a high-precision, high-reliability and low-energy-consumption intelligent suspension control strategy for vehicles such as new energy commercial vehicles, which has high application value. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A flowchart of a vehicle suspension control method provided in an embodiment of this application; Figure 2 This is a structural block diagram of a vehicle suspension control system provided in an embodiment of this application; Figure 3 This is a structural block diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0018] The present application will now be described in further detail with reference to the embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the application. Furthermore, it should be noted that, for ease of description, only the parts relevant to the application are shown in the accompanying drawings.

[0019] It should be noted that, unless otherwise specified, the embodiments and features of the embodiments in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] The following describes in detail, with reference to the accompanying drawings, a suspension control method, system, vehicle, and device for a vehicle according to embodiments of this application.

[0021] Figure 1 This is a flowchart of a vehicle suspension control method according to one embodiment of this application. Figure 1 As shown, the vehicle suspension control method according to an embodiment of this application includes the following steps: S101: Obtain multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic status information.

[0022] In one embodiment of this application, obtaining multi-source sensor data includes: obtaining suspension height detected by a height sensor, wherein each axle is equipped with multiple independent height sensors, and the multiple height sensors are arranged spatially separately; obtaining suspension airbag pressure detected by a pressure sensor, wherein the pressure sensor is configured at the inlet and outlet of each suspension airbag; obtaining road surface information detected by a front-facing camera and millimeter-wave radar; and obtaining vehicle dynamic state information detected by an IMU, wheel speed sensor, and steering angle sensor.

[0023] Specifically, a triple-redundant height sensor can be used, namely: three independent height sensors are equipped on each axle, arranged in a spatially separated manner, and the impact of single-point failure is eliminated through a voting algorithm; high-precision air pressure sensors are configured at each airbag inlet and outlet, with a sampling frequency of up to 100Hz; a front-facing camera and millimeter-wave radar are integrated to identify road features 200m in advance; and an IMU, wheel speed sensor, and steering angle sensor are integrated to monitor the vehicle's dynamic status in real time.

[0024] S102: Based on the multi-source sensor data, the optimal control target of the suspension parameters is generated using the NSGA-II multi-objective optimization algorithm, wherein the optimization targets of the NSGA-II multi-objective optimization algorithm include comfort, stability and vehicle energy consumption.

[0025] It should be noted that before generating the optimal control target for suspension parameters using the NSGA-II multi-objective optimization algorithm based on the multi-source sensor data, the process also includes: spatiotemporal alignment and fusion of the multi-source sensor data; constructing a road-vehicle coupling model based on the fused data, wherein the road-vehicle coupling model is used to extract the frequency domain / time domain features of road excitation; and creating an LSTM-based road prediction module to predict the road surface information ahead based on the output of the road-vehicle coupling model.

[0026] For example, by using the adaptive Kalman filter algorithm to perform spatiotemporal alignment and fusion of multi-source heterogeneous data, a road-vehicle coupling model is constructed to realize the extraction of frequency domain / time domain features of road excitation. A road feature prediction module based on LSTM is developed, with a prediction accuracy of over 85%, and road condition changes can be predicted 2-3 seconds in advance.

[0027] In one embodiment of this application, based on the multi-source sensor data, the optimal control objective of the suspension parameters is generated using the NSGA-II multi-objective optimization algorithm. This includes: generating optimal suspension parameters based on the multi-source sensor data using the NSGA-II multi-objective optimization algorithm, combining comfort, stability, and energy consumption objectives, and obtaining the optimal control objective of the suspension parameters based on these optimal suspension parameters. In other words, based on the NSGA-II multi-objective optimization algorithm, an optimal set of suspension parameters is generated by comprehensively considering three objectives: comfort (e.g., considering vehicle acceleration RMS value), stability (e.g., considering roll angle), and energy consumption (e.g., considering air pump operating time).

[0028] S103: The control sequence of the solenoid valve is generated using the DDPG deep reinforcement learning algorithm.

[0029] In a specific example, the DDPG deep reinforcement learning algorithm is used to generate the control sequence of the solenoid valve. This includes: using the DDPG deep reinforcement learning algorithm combined with a precise hysteresis model of the solenoid valve to obtain the control sequence of the solenoid valve, wherein the precise hysteresis model of the solenoid valve represents the mapping relationship between the opening / closing delay time and the air pressure. In other words, by using the DDPG deep reinforcement learning algorithm combined with the precise hysteresis model of the solenoid valve (establishing the mapping relationship between the opening / closing delay time and the air pressure), precise control of the inflation / deflation timing is achieved.

[0030] S104: Based on the optimal control objective and the control sequence of the solenoid valve, the solenoid valve of the suspension is controlled to achieve adaptive adjustment of the suspension. PWM precise control can be executed to achieve precise control of the electronic valve, thereby realizing adaptive adjustment of the suspension.

[0031] In one embodiment of this application, the vehicle suspension control method further includes: obtaining the current driving conditions of the vehicle; and adjusting the optimal control target according to the current driving conditions.

[0032] In one embodiment of this application, the vehicle suspension control method further includes: in battery swapping mode, responding to the driver's input command to control the lifting of one side wheel; and in the event of a fault, controlling the vehicle to switch to a safety mode.

[0033] Specifically, it can incorporate software security mechanisms, such as: real-time health monitoring: performing a system self-check every 50ms, including sensor consistency checks and actuator response verification; fault diagnosis and prediction: providing a 2-hour advance warning of airbag malfunctions based on pressure-displacement mapping deviation analysis; predicting valve body wear status through solenoid valve current ripple analysis; and switching control strategies within 50ms when a fault occurs to ensure basic suspension functionality. It also includes a multi-level safety boundary design, including axle load distribution boundaries, vehicle tilt boundaries, and height adjustment boundaries.

[0034] Dedicated control for battery swapping heavy trucks, namely: bridge lifting mode: one-click lifting of one side of the wheel, supporting precise docking during the battery swapping process.

[0035] Based on driving route prediction, the air pump operation strategy is optimized to reduce energy consumption. During braking, the auxiliary air pressure is generated by the compression energy of the suspension, reducing the working time of the air pump. After stopping, it automatically switches to mechanical locking mode to maintain the suspension height without consuming energy.

[0036] The specific workflow is as follows: The system performs a power-on self-test, sensor calibration and parameter loading, establishes an initial vehicle model, acquires all sensor data, fuses multi-source data, updates the vehicle status, determines the current driving conditions, generates the optimal control target based on the NSGA-II algorithm, generates a precise solenoid valve control sequence based on the DDPG algorithm, checks whether the control command is within the safety boundary, and if so, outputs a PWM signal to control the solenoid valve. During the control process, the control effect can be recorded for online algorithm optimization.

[0037] In battery swapping mode: responds to driver commands, precisely controls the lifting of one side of the wheel, and quickly switches to safety mode in the event of a malfunction.

[0038] The vehicle suspension control method according to the embodiments of this application solves the problems of control accuracy, safety and reliability, environmental adaptability and energy efficiency of traditional electronically controlled suspension systems, and provides a high-precision, high-reliability and low-energy-consumption intelligent suspension control strategy for vehicles such as new energy commercial vehicles, which has high application value.

[0039] The vehicle suspension control method of this application adopts the NSGA-II multi-objective optimization algorithm and hierarchical control architecture, combined with the DDPG deep reinforcement learning algorithm and the solenoid valve precise hysteresis model for collaborative optimization.

[0040] Differences from existing technologies: Existing technology: The simple open-loop control strategy, such as fixed time parameter control such as "output right rear airbag solenoid valve drive 'ON', central solenoid valve drive 'OFF' for 0.2 seconds, wait for 0.2 seconds", cannot adapt to complex working conditions.

[0041] The embodiments of this application construct a three-layer control architecture (upper layer NSGA-II multi-objective optimization, middle layer DDPG reinforcement learning, and lower layer PWM precise control). The upper layer simultaneously optimizes three major objectives: comfort (vehicle acceleration RMS value), stability (roll angle), and energy consumption (air pump working time). The middle layer dynamically adjusts control parameters by establishing a precise mapping relationship between the solenoid valve opening / closing delay time and air pressure and temperature. The lower layer implements 10kHz PWM control. The three layers work together to achieve real-time adaptive adjustment of control parameters.

[0042] It adopts a triple sensor redundancy configuration, a dual solenoid valve parallel redundancy design, and a safety degradation strategy with a response time of less than 50ms.

[0043] Differences from existing technologies: Existing technology mainly relies on a single condition such as "vehicle speed = 0" or "EPB parking status = Applied" as a safety premise. Abnormal height status recognition requires more than 20 seconds, and suspension function is usually directly disabled when there is a malfunction.

[0044] The embodiments of this application implement triple sensor redundancy (using spatial separation arrangement and a three-out-of-two voting mechanism) and dual-path solenoid valve redundancy in hardware; construct multi-dimensional safety verification (vehicle speed, attitude, load, road surface, etc.) in software, and cover advanced diagnostic methods such as pressure-displacement mapping deviation and current ripple analysis for fault diagnosis; the safety degradation strategy is completed within 50ms, and can still maintain some basic functions even in the case of dual-point failure, which meets the ASIL-B functional safety level requirements.

[0045] By integrating multi-source environmental perception data to construct a road-vehicle coupled model, we can achieve LSTM-based road feature prediction and multi-axis load-height coordinated control.

[0046] Differences from existing technologies: Existing technology relies solely on basic altitude and barometric pressure sensor data, lacks environmental awareness, and makes control decisions based on reactive control of the current state.

[0047] The embodiments of this application combine multi-source sensing devices such as front-facing cameras, millimeter-wave radar, IMU, wheel speed sensors, and steering angle sensors, and perform spatiotemporal alignment and fusion through an adaptive Kalman filter algorithm; develop a road surface prediction module based on LSTM, which can predict road condition changes 2-3 seconds in advance; and design a multi-axis load-height collaborative control algorithm to ensure that the height deviation of each axis does not exceed the preset range when the load changes, thus achieving a qualitative change from "reactive control" to "predictive control".

[0048] The "bridge-raising mode" designed for battery-swapping heavy-duty trucks, low-energy consumption optimization strategies, and regulatory compliance guarantee mechanisms.

[0049] Differences from existing technologies: Existing technology: It does not take into account the special needs of new energy commercial vehicles, lacks support for battery swapping mode, and does not adequately consider energy consumption optimization and regulatory compliance.

[0050] Embodiments of this application include: developing a one-click "bridge-raising mode" to support precise docking during battery swapping; designing an intelligent air source management strategy to optimize air pump operation based on driving route prediction, combined with regenerative braking energy recovery technology; and incorporating real-time axle load monitoring and overload safety control to adapt to the development trend of new energy commercial vehicles.

[0051] Through multi-dimensional technological innovation, the problems of control precision, safety and reliability, environmental adaptability and energy efficiency of traditional electronically controlled suspension systems have been solved, providing a high-precision, high-reliability and low-energy-consumption intelligent suspension control solution for new energy commercial vehicles, which has significant technological advancement and application value.

[0052] Overall, it achieves precise adaptive control of suspension height and stiffness, significantly improving the vehicle's driving performance under various operating conditions.

[0053] The NSGA-II multi-objective optimization algorithm is adopted to achieve the optimal balance between comfort, stability and energy consumption in the system.

[0054] High-precision control is achieved through collaborative optimization of DDPG deep reinforcement learning and solenoid valve hysteresis model.

[0055] The parameter self-tuning mechanism based on operating condition identification enables the system to automatically adapt to 12 typical driving conditions, reducing the need for manual intervention.

[0056] A highly reliable suspension control system was built, which significantly reduced the failure rate and safety risks.

[0057] Specific technical effects: The triple sensor redundancy configuration reduces the failure rate of critical sensors, and the parallel redundancy design of dual solenoid valves improves the reliability of the actuator.

[0058] Advanced fault diagnosis technologies (pressure-displacement mapping deviation analysis, current ripple analysis) improve fault diagnosis coverage and can provide early warning of airbag aging issues.

[0059] Figure 2 This is a structural block diagram of a vehicle suspension control system according to one embodiment of this application. Figure 2 As shown, a vehicle suspension control system according to an embodiment of this application includes: an acquisition module 210, a control target determination module 220, a control timing determination module 230, and a control module 240, wherein: The acquisition module 210 is used to acquire multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information and vehicle dynamic status information; The control target determination module 220 is used to generate the optimal control target of the suspension parameters based on the multi-source sensor data and using the NSGA-II multi-objective optimization algorithm, wherein the optimization targets of the NSGA-II multi-objective optimization algorithm include comfort, stability and vehicle energy consumption. The control timing determination module 230 is used to generate the control sequence of the solenoid valve using the DDPG deep reinforcement learning algorithm; The control module 240 is used to control the solenoid valve of the suspension according to the optimal control target and the control sequence of the solenoid valve, so as to realize the adaptive adjustment of the suspension.

[0060] The vehicle suspension control method according to the embodiments of this application solves the problems of control accuracy, safety and reliability, environmental adaptability and energy efficiency of traditional electronically controlled suspension systems, and provides a high-precision, high-reliability and low-energy-consumption intelligent suspension control strategy for vehicles such as new energy commercial vehicles, which has high application value.

[0061] Specific limitations regarding the vehicle's suspension control system can be found in the above description of the vehicle's suspension control methods, and will not be repeated here. The various modules of the aforementioned vehicle suspension control system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the computer device's memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0062] Furthermore, a vehicle is provided, comprising: a suspension control system for the vehicle according to any of the above embodiments. This vehicle solves the problems of control accuracy, safety and reliability, environmental adaptability and energy efficiency of traditional electronically controlled suspension systems, and provides a high-precision, high-reliability, and low-energy-consumption intelligent suspension control strategy for vehicles such as new energy commercial vehicles, which has high application value.

[0063] Furthermore, other components and functions of the vehicle according to the embodiments of this application are known to those skilled in the art and will not be described in detail here.

[0064] In one embodiment, a computer device is provided. Figure 3 This is a structural block diagram of the computer device provided in the embodiments of this application, with reference to... Figure 3 The computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned embodiment of the vehicle suspension control method. For example, it executes: obtaining multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic state information; Based on the multi-source sensor data, the optimal control objectives of the suspension parameters are generated using the NSGA-II multi-objective optimization algorithm. The optimization objectives of the NSGA-II multi-objective optimization algorithm include comfort, stability, and vehicle energy consumption. The control sequence for the solenoid valve is generated using the DDPG deep reinforcement learning algorithm. Based on the optimal control objective and the control sequence of the solenoid valves, the solenoid valves of the suspension are controlled to achieve adaptive adjustment of the suspension.

[0065] This application also provides a computer-readable storage medium storing a computer program. When a processor executes the computer program, it implements the aforementioned vehicle suspension control method embodiment. For example, it executes: obtaining multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic state information; Based on the multi-source sensor data, the optimal control objectives of the suspension parameters are generated using the NSGA-II multi-objective optimization algorithm. The optimization objectives of the NSGA-II multi-objective optimization algorithm include comfort, stability, and vehicle energy consumption. The control sequence for the solenoid valve is generated using the DDPG deep reinforcement learning algorithm. Based on the optimal control objective and the control sequence of the solenoid valves, the solenoid valves of the suspension are controlled to achieve adaptive adjustment of the suspension.

[0066] This application provides a computer program product including instructions that, when executed, cause the method described in this application embodiment to be performed. For example, it can execute... Figure 1 The various steps of the suspension control method for the vehicle shown include, for example, obtaining multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic state information. Based on the multi-source sensor data, the optimal control objectives of the suspension parameters are generated using the NSGA-II multi-objective optimization algorithm. The optimization objectives of the NSGA-II multi-objective optimization algorithm include comfort, stability, and vehicle energy consumption. The control sequence for the solenoid valve is generated using the DDPG deep reinforcement learning algorithm. Based on the optimal control objective and the control sequence of the solenoid valves, the solenoid valves of the suspension are controlled to achieve adaptive adjustment of the suspension.

[0067] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0068] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0069] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A suspension control method for a vehicle, characterized in that, include: Obtain multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information, and vehicle dynamic status information; Based on the multi-source sensor data, the optimal control objectives of the suspension parameters are generated using the NSGA-II multi-objective optimization algorithm. The optimization objectives of the NSGA-II multi-objective optimization algorithm include comfort, stability, and vehicle energy consumption. The control sequence for the solenoid valve is generated using the DDPG deep reinforcement learning algorithm. Based on the optimal control objective and the control sequence of the solenoid valves, the solenoid valves of the suspension are controlled to achieve adaptive adjustment of the suspension.

2. The vehicle suspension control method according to claim 1, characterized in that, The acquisition of multi-source sensing data includes: The suspension height is obtained from the height sensors, wherein each axle is equipped with multiple independent height sensors, and the multiple height sensors are arranged in a spatially separated manner; The airbag pressure detected by the air pressure sensor is obtained, wherein the air pressure sensor is configured at the inlet and outlet of each airbag; Obtain road surface information detected by the front-facing camera and millimeter-wave radar; It acquires vehicle dynamic status information detected by IMU, wheel speed sensor and steering angle sensor.

3. The vehicle suspension control method according to claim 1, characterized in that, Before generating the optimal control objective for the suspension parameters using the NSGA-II multi-objective optimization algorithm based on the multi-source sensor data, the process also includes: The multi-source sensor data is spatiotemporally aligned and fused; A road-vehicle coupling model is constructed based on the fused data, wherein the road-vehicle coupling model is used to extract the frequency domain / time domain features of road excitation; Create an LSTM-based road prediction module to predict road information ahead based on the output of the road-vehicle coupling model.

4. The vehicle suspension control method according to claim 1, characterized in that, The step of generating the optimal control objective for the suspension parameters based on the multi-source sensor data and using the NSGA-II multi-objective optimization algorithm includes: Based on the NSGA-II multi-objective optimization algorithm, combining comfort, stability and energy consumption objectives, the optimal suspension parameters are generated using the multi-source sensor data, and the optimal control objective of the suspension parameters is obtained based on the optimal suspension parameters.

5. The vehicle suspension control method according to claim 1, characterized in that, The generation of the control sequence for the solenoid valve using the DDPG deep reinforcement learning algorithm includes: The control sequence of the solenoid valve is obtained by using the DDPG deep reinforcement learning algorithm combined with the precise hysteresis model of the solenoid valve. The precise hysteresis model of the solenoid valve represents the mapping relationship between the opening / closing delay time and the air pressure.

6. The vehicle suspension control method according to claim 1, characterized in that, Also includes: Obtain the vehicle's current operating conditions; The optimal control target is adjusted based on the current driving conditions.

7. The suspension control method for a vehicle according to any one of claims 1-6, characterized in that, Also includes: In battery swapping mode, it responds to the driver's input commands to control the lifting of one side of the wheels; When a malfunction occurs, control the vehicle to switch to safe mode.

8. A vehicle suspension control system, characterized in that, include: The acquisition module is used to acquire multi-source sensor data, wherein the multi-source sensor data includes suspension height, suspension airbag pressure, road surface information and vehicle dynamic status information; The control target determination module is used to generate the optimal control target of the suspension parameters based on the multi-source sensor data and using the NSGA-II multi-objective optimization algorithm, wherein the optimization targets of the NSGA-II multi-objective optimization algorithm include comfort, stability and vehicle energy consumption. The control timing determination module is used to generate the control sequence of the solenoid valve using the DDPG deep reinforcement learning algorithm. The control module is used to control the solenoid valves of the suspension according to the optimal control target and the control sequence of the solenoid valves, so as to realize the adaptive adjustment of the suspension.

9. A vehicle, characterized in that, include: The vehicle suspension control system according to claim 8.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the vehicle suspension control method according to any one of claims 1-7.

Images