Semi-active suspension control system based on Transform architecture

Abstract

The invention belongs to the technical field of vehicle suspension system control, and discloses a semi-active suspension control system based on Transform architecture, which comprises the following steps: arranging a sensor network, and collecting vehicle dynamic data; using a preset algorithm to align asynchronous data from different sensors, and keeping time sequence consistency of heterogeneous data; the measured value of the sensor i at the time t is xi (t), in the invention, the weight distribution of the state quantity is reduced through a Transform mechanism, the error control quantity caused by the sensor error is avoided, the robustness of the suspension control system to the data exception is improved, and the reliability of the suspension control system is improved. According to the method, stable control can be kept under sensor faults, external attacks and false sensor data injection or extreme road conditions, through a DDPG reinforcement learning control algorithm, the adaptive capacity of the system in a complex and changeable environment is enhanced, different road conditions, loads and driving styles can be rapidly adapted, and the control effect is improved while the calculation efficiency is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle suspension system control, and in particular relates to a semi-active suspension control system based on a Transformer architecture. Background Art

[0002] As a core component in automotive engineering, the vehicle suspension system is responsible for buffering the vibration and impact of the vehicle during driving, and is crucial to improving ride comfort and driving stability. It is particularly important to adopt advanced strategies to control the controllable damping of the suspension to optimize the above performance. At present, the control strategy of the vehicle suspension system is undergoing a transformation from the traditional rule-based method to intelligent control. Passive, semi-active and active suspension systems have their own characteristics, but each has its own limitations. Although the passive suspension has a simple structure, it is difficult to adapt to complex road conditions and provide flexible vibration reduction effects; although the semi-active suspension can achieve a certain degree of regulation by adjusting the damping force of the shock absorber, it still lacks the ability to actively apply force; and although the active suspension has outstanding control performance, it is difficult to be widely promoted in ordinary vehicles due to its high cost and complex structure.

[0003] The integration of intelligent control technology, especially fuzzy control, adaptive control, neural network and genetic algorithm, has brought new vitality to suspension system control, improving the effect and flexibility. In the process of intelligent vehicle networking, the injection of false sensor data through the CAN bus poses a threat to the safety of the car. However, these methods still face challenges in scalability and real-time performance when dealing with complex actual working conditions, especially the lack of ability to handle abnormal sensor data. At the same time, although reinforcement learning has shown great potential in the field of control and can optimize control strategies through continuous interaction with the environment, its practical application in suspension systems is limited by its instability in the training process, insensitivity to FDI attacks, low sample efficiency and limited strategy generalization ability. The Transformer model has made remarkable achievements in fields such as natural language processing with its excellent sequence modeling ability. Its core self-attention mechanism can effectively capture long-term and short-term dependencies, which has great potential in processing time series sensor data. However, the application of the Transformer model to suspension system control is still an innovative attempt, and many technical difficulties need to be overcome, such as high computational complexity, optimization to meet real-time control requirements, and the need to design Transformer variants suitable for control tasks. Therefore, a semi-active suspension control system based on Transformer architecture is proposed to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide a semi-active suspension control system based on a Transformer architecture to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a semi-active suspension control system based on a Transformer architecture, comprising the following steps:

[0006] S1: deploy sensor network to collect vehicle dynamic data;

[0007] S2: Use a preset algorithm to align asynchronous data from different sensors to maintain the temporal consistency of heterogeneous data;

[0008] The measurement value of sensor i at time t is x i (t), align all sensor data to the same time point t k : where t j ≤t k <t j+1 ;

[0009] S3: Building a quarter vehicle two-degree-of-freedom model

[0010]

[0011] Where m s is the vehicle body mass, m u is the unsprung mass, F is the damping force of the magnetorheological damper, x s is the displacement of the vehicle body, x u is the unsprung displacement, x r is the wheel displacement, c is the uncontrollable damping of the damper, k s is the spring stiffness, k t is the wheel stiffness;

[0012] S4: Modeling of Magnetorheological Damper via Neural Networks:

[0013]

[0014] The damper experimental data is obtained through the preset equipment, and the data is fitted through the neural network to obtain the weight matrix w I , w H and w O ;

[0015] The damper neural network model is combined with the vehicle suspension dynamics model to obtain an integrated model;

[0016] Interact the reinforcement learning structure with the integrated model to train the neural network parameters;

[0017] S5: Apply DDPG reinforcement learning algorithm to calculate the control quantity for the state quantity. The DDPG network structure is Actor-Crit ic structure, where the Actor network generates actions: a t =μ(s|θ μ ), the Critic network evaluates the state-action value Q(s,a|θ Q );

[0018] S6: The state quantity is passed into the Transformer architecture to replace the traditional Actor network, analyze the correlation of the state quantity, distinguish between valid information and state quantity errors, and convert the data x i (t k ) is passed into the DDPG Transformer architecture;

[0019] S7: Combine the encoded high-dimensional data X with the weights Point multiplication to obtain a weighted matrix Each row of V represents a high-dimensional representation of sensor data;

[0020] S8: Transpose the Q and K matrices to obtain QK T matrix;

[0021] S9: Divide the K matrix dimension by the obtained attention weight result Scaling is performed to calculate the attention weight matrix Perform softmax activation function transformation and convert the result into percentage;

[0022] S10: The transformed attention weights Multiplying with matrix V gives Each element in the result is the sum of the attention weights of each sample data corresponding to all other sample data in the 64-dimensional Q, K matrix. Adjust V matrix weights Get the result Attention(Q,K,V);

[0023] S11: self-attention matrix head i =Attention(Q,K,V) to perform multiple groups of parallel training and obtain the multi-head attention mechanism result Concat(head1,…,head h ), and the multi-head attention mechanism result is combined with the multi-head weight matrix W O MultiHead(Q,K,V)=Concat(head1,…,head h )W O , each head iIt is the result of the attention mechanism in different spaces;

[0024] S12: Replace the traditional reinforcement learning Actor network architecture with the Transformer architecture and generate a parameterized strategy μ(s|θ μ ) to output the optimal action a under a given state t =μ(s|θ μ ), whose goal is to maximize the expected cumulative reward: The Critic network is used to evaluate the state-action value function Q(s,a|θ Q ), and optimize its parameters by minimizing the following loss function: L(θ Q )=E (s,a,r,s′)～D [(r+γQ(s′,μ(s′|θ μ )|θ Q )-Q(s,a|θ Q )) 2 ] Where r is the immediate reward, γ is the discount factor, and D is the data in the experience replay buffer;

[0025] S13: Design reinforcement learning reward function;

[0026] S14: Dual network architecture, including two independent Actor networks and With two independent Critic networks and The minimum action amount and state-action value are calculated by the following formula: The loss function is defined as:

[0027]

[0028] S15: Store the data obtained during the interaction with the environment in the playback buffer D, randomly sample batches of data from the buffer for training, and improve sample utilization. For the dual Actor and Cr it ic network update mechanism, the target network parameters θ are maintained and updated separately. μ ' and θ Q ', each time the network parameters are updated, the target network uses the soft update rule: Actor network update rule: θ μ '←τθ μ +(1-τ)θ μ ', Cr it ic network update rule: θ Q '←τθ Q +(1-τ)θ Q ', where τ is the soft update parameter, which balances the current network parameters with the parameters to be updated, adjusts the network parameters, and obtains the control signal I.

[0029] Preferably, in S1, the sensor network includes an accelerometer, a gyroscope, a displacement sensor and a pressure sensor.

[0030] Preferably, in S1, the vehicle dynamic data includes vehicle body acceleration, suspension travel, and tire deformation.

[0031] Preferably, in S4, the preset equipment includes a suspension test bench.

[0032] Preferably, in S4, the output value of the neural network is the controllable damping force F of the magnetorheological damper.

[0033] Preferably, the state quantity is passed into the Transformer architecture to replace the traditional Actor network, analyze the correlation of the state quantity, distinguish valid information and state quantity error, and convert the data x i (t k ) is passed into the DDPG Transformer architecture and includes the following steps:

[0034] First, the data x i (t k ) Upgraded to 64-dimensional S t , using learnable Fourier features S t Position code: PE (pos,2i) = sin(pos·w i ),PE (pos,2i+1) =cos(pos·w i ), where w i is a learnable parameter

[0035] Preferably, in S4, the integrated model is:

[0036]

[0037] In the formula, m s is the vehicle body mass, m u is the unsprung mass, F is the damping force of the magnetorheological damper, x s is the displacement of the vehicle body, x u is the unsprung displacement, x r is the wheel displacement, c is the uncontrollable damping of the damper, k s is the spring stiffness, k t is the wheel stiffness.

[0038] Preferably, in S2, the preset algorithm includes an interpolation algorithm.

[0039] Preferably, in S13, the reinforcement learning reward function is The reward function is composed of four parts: body acceleration, suspension travel, tire deformation, and suspension dissipative constraint. The body acceleration and suspension travel are multiplied by weight coefficients a1 and a2 respectively for weight distribution. In addition, obstacle functions P, Q, and N are added. The obstacle function P constrains the suspension travel and keeps the suspension travel within the limit range, so that (x s -x u ) at the maximum suspension travel x max Within the range, the obstacle function Q constrains the suspension wheel deformation, and the obstacle function N constrains the magnetorheological semi-active suspension input.

[0040] Preferably, the barrier function The barrier function The barrier function

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] In the present invention, the Transformer mechanism is used to reduce the weight distribution of state quantities, avoid the erroneous control caused by sensor errors, improve the robustness of the suspension control system to data anomalies, and maintain stable control under sensor failure, external attack, injection of false sensor data or extreme road conditions.

[0043] Through the DDPG reinforcement learning control algorithm, the system's adaptability in complex and changing environments is enhanced. It can quickly adapt to different road conditions, loads and driving styles, while ensuring computing efficiency and improving control effects.

[0044] Compared with traditional methods, the present invention improves ride comfort by 36.7% and handling stability by 30.45% on average under various road conditions. Through lightweight design and optimization algorithm, the real-time performance of the control system is ensured and the interpretability of the system is improved. The basis of control decisions can be understood by analyzing the attention weight of the Transformer, which reduces the system's dependence on precise physical models and enhances the generalization ability in practical applications.

[0045] The present invention is different from existing traditional control methods and traditional intelligent algorithms. Traditional algorithms cannot accurately characterize the current state when sensors have errors, which in turn affects the precise and smooth control of the vehicle body. The method described in the present invention is a deep deterministic policy gradient (DDPG) reinforcement learning control method based on a data-driven Transformer architecture. The Transformer architecture is used to perform correlation analysis on abnormal data to obtain the importance weights between the data. A smaller weight coefficient is used to reduce the impact of abnormal data on the control amount. The weighted data is subjected to reinforcement learning training to obtain a control method that meets the control requirements and is robust to sensor failures, injection of false sensor data by external attacks, or extreme road conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is the overall control flow chart of the control method of the present invention;

[0047] Figure 2 This is a diagram showing the internal structure of a specific controller of the present invention;

[0048] Figure 3 A structural diagram of a two-degree-of-freedom suspension in an environment that interacts with a control algorithm in the present invention;

[0049] Figure 4 A schematic diagram of vehicle body acceleration when a convex road simulation experiment is performed using the suspension control system of the present invention;

[0050] Figure 5 A schematic diagram of the suspension dynamic travel when the suspension is controlled by the present invention to perform a convex road simulation experiment;

[0051] Figure 6 The figure is a schematic diagram of the dynamic and static load ratio of tires when a vehicle convex road simulation experiment is carried out using the control method of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1-Figure 6 As shown, the present invention provides the following technical solutions:

[0054] The present invention provides a system for processing data anomalies and optimizing suspension control by using a Transformer architecture and reinforcement learning, characterized in that the system comprises the following steps:

[0055] S1: Design a comprehensive sensor network, using accelerometers, gyroscopes, displacement sensors and pressure sensors to collect vehicle dynamic data including body acceleration, suspension travel, tire deformation, etc. There are two data anomalies in the state quantity collected by the sensor, namely, point error caused by electromagnetic interference and interval error caused by unstable power supply. The present invention can be used to reduce the possible impact.

[0056] S2: Use interpolation algorithms to align asynchronous data from different sensors to ensure the timing consistency of heterogeneous data. The measurement value of sensor i at time t is x i (t), the present invention uses an interpolation method to align all sensor data to a unified time point t k : where t j ≤t k <t j+1 .

[0057] S3: For the establishment of the reinforcement learning interactive environment, the suspension system is modeled using the vehicle dynamics model. Establish a quarter vehicle two-degree-of-freedom model Where m s is the vehicle body mass, m u is the unsprung mass, F is the damping force of the magnetorheological damper, x s is the displacement of the vehicle body, x u is the unsprung displacement, x r is the wheel displacement, c is the uncontrollable damping of the damper, k s is the spring stiffness, k t Wheel stiffness.

[0058] S4: Magnetorheological damper modeling using neural network: The damper experimental data is obtained through the suspension test bench, and the data is fitted through the neural network to obtain the weight matrix w I， w H , w O The output value of the neural network is the controllable damping force F of the magnetorheological damper. The damper neural network model is combined with the vehicle suspension dynamics model to obtain an integrated model: The reinforcement learning structure interacts with the environment model and trains the neural network parameters to achieve the best control effect.

[0059] S5: Apply the DDPG reinforcement learning algorithm to the state quantity to calculate the control quantity. The DDPG algorithm is a model-free reinforcement learning algorithm suitable for continuous action space. The DDPG network structure is an Actor-Critic structure. The Actor network generates actions: a t =μ(s|θ μ), the Critic network evaluates the state-action value Q(s,a|θ Q ).

[0060] S6: The state quantity is transferred to the Actor network improved by this patent - Transformer architecture to replace the traditional Actor network, analyze the correlation of the state quantity, and distinguish between valid information and state quantity errors. i (t k ) is passed into the DDPG Transformer architecture. First, the data x i (t k ) Upgraded to 64-dimensional S t , using learnable Fourier features S t Position code: PE (pos,2i) = sin(pos·w i ),PE (pos,2i+1) =cos(pos·w i ), where w i are learnable parameters.

[0061] S7: Combine the encoded high-dimensional data X with the weights Point multiplication to obtain a weighted matrix Each row of V represents a high-dimensional representation of sensor data.

[0062] S8: Transpose the Q and K matrices to obtain QK T The matrix is ​​used to obtain the relative relationship between data at different positions. The size of the numerical result reflects the degree of relationship between data, which is the attention weight. The larger the number, the closer the relationship between the data, and the smaller the number, the more distant the relationship between the data. The attention weight can reflect the relationship between the attention data and other data. For abnormal data measured by sensors. Since the deviation between abnormal data and real data is large, the relative position in high-dimensional space is far, that is, the relationship with other data is relatively loose, and the attention weight obtained is small, which reduces the impact of abnormal data on control effect.

[0063] S9: Divide the K matrix dimension by the obtained attention weight result Scale the calculated attention weight matrix Perform softmax activation function transformation and convert the result into percentage. The row and column represent the correlation between each high-dimensional data and other data, and the sum of each row is 1.

[0064] S10: The transformed attention weights Multiplying with matrix V gives Each element in the result is the sum of the attention weights of each sample data corresponding to all other sample data in the 64-dimensional Q, K matrix. Adjust V matrix weights Get the accurate result Attention(Q,K,V).

[0065] S11: self-attention matrix head i =Attention(Q,K,V) to perform multiple groups of parallel training and obtain the multi-head attention mechanism result Concat(head1,…,head h ). Combine the multi-head attention mechanism result with the multi-head weight matrix W O Multiply them together to get the multi-head attention result MultiHead(Q,K,V)=Concat(head1,…,head h )W O Each head i It is the result of the attention mechanism in different spaces. It can capture multiple characteristics of the input data in parallel, analyze abnormal data by comparing weight results from multiple spatial angles, and minimize the impact of abnormal data by reducing weights.

[0066] S12: The present invention adopts the Actor-critical architecture of the DDPG algorithm. It mainly includes two parts: the Actor network and the critical network. In the present invention, the Transformer architecture replaces the traditional reinforcement learning Actor network architecture, and the parameterized strategy μ(s|θ μ ) to output the optimal action a under a given state t =μ(s|θ μ ). Its goal is to maximize the expected cumulative reward: The Critic network is used to evaluate the state-action value function Q(s,a|θ Q ), and optimize its parameters by minimizing the following loss function: L(θ Q )=E (s,a,r,s′)～D [(r+γQ(s′,μ(s′|θ μ )|θ Q )-Q(s,a|θ Q )) 2 ]Where r is the immediate reward, γ is the discount factor, and D is the data in the experience replay buffer.

[0067] S13: The present invention designs an optimal reinforcement learning reward function for the control objective of the magnetorheological semi-active suspension system The suspension system has three purposes: reducing vehicle acceleration, maintaining handling stability, and maintaining safe driving conditions. In response to the above three requirements, the reward function is composed of four parts: vehicle acceleration, suspension travel, tire deformation, and suspension dissipative constraint. The vehicle acceleration and suspension travel are multiplied by weight coefficients a1 and a2 respectively to distribute weights. In addition, obstacle functions P, Q, and N are added. Among them, the obstacle function Constrain the suspension travel and keep the suspension travel within the limit range, so that (x s -x u ) at the maximum suspension travel x max range, to prevent damage to the suspension system and comfort. Obstacle function Q = The suspension wheel deformation is constrained to maintain a firm contact between the wheel and the road surface, so that it remains within the allowable range and the vehicle reaches a safe driving condition. The input of magnetorheological semi-active suspension is constrained to meet the dissipative constraint requirements.

[0068] S14: To improve system stability and control accuracy, a dual network architecture is used. In this architecture, there are two independent Actor networks. and With two independent Critic networks and The minimum action amount and state-action value are calculated by the following formula: The loss function is defined as: The dual Q network can effectively reduce the deviation in Q value estimation, thereby improving the stability of the training process and avoiding strategy oscillation and divergence.

[0069] S15: The present invention also uses the experience replay mechanism and the target network update mechanism to further improve the stability and training efficiency of the system. By storing the data obtained during the interaction with the environment in the replay buffer D, the Transformer mechanism can break the time correlation and randomly sample batches of data from the buffer for training, thereby improving sample utilization. For the dual actor and critical network update mechanisms, the target network parameters θ are maintained and updated with delays. μ ' and θ Q '. Each time the network parameters are updated, the target network uses the soft update rule: Actor network update rule: θ μ '←τθ μ +(1-τ)θ μ ', Crit ic network update rule: θ Q '←τθ Q +(1-τ)θ Q'. Where τ is a soft update parameter that balances the current network parameters with the parameters to be updated. The network parameters are adjusted to obtain the optimal control signal I.

[0070] In addition, in the present invention, in S12, a reinforcement learning control strategy based on a Transformer architecture is used to output a control quantity I in real time in response to road excitation applied to the vehicle, thereby achieving adaptive control of the vehicle suspension.

[0071] In summary, based on the above disclosed method, a suspension convex road simulation experiment is carried out in the present invention.

[0072] Assume that the height of the convex road surface in this experiment is h=0.065m, the length of the convex road surface is l=4m, and the vehicle travels through the road surface at a speed of v=36km / h.

[0073] It is expected that the vehicle body acceleration will be effectively reduced when driving on bumpy roads, the suspension travel will not exceed 0.07, and the dynamic and static load ratio will not exceed the rated limit of 1.

[0074] Based on the above parameters, a lane keeping simulation experiment is conducted and the following results are obtained: Figure 4-Figure 6 The simulation results shown are Figure 4 It can be seen that the above control method can effectively realize that the suspension system can effectively reduce the acceleration of the vehicle body when it is stimulated by the road surface. Figure 5 and Figure 6 It can be seen that the above control method can effectively achieve safe driving and operation of the vehicle.

[0075] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A semi-active suspension control system based on Transformer architecture, characterized in that: The steps include: S1: deploy sensor network to collect vehicle dynamic data; S2: Use a preset algorithm to align asynchronous data from different sensors to maintain the temporal consistency of heterogeneous data; The measurement value of sensor i at time t is x i (t), align all sensor data to the same time point t k : where t j ≤t k <t j+1 ; S3: Building a quarter vehicle two-degree-of-freedom model Where m s is the vehicle body mass, m u is the unsprung mass, F is the damping force of the magnetorheological damper, x s is the displacement of the vehicle body, x u is the unsprung displacement, x r is the wheel displacement, c is the uncontrollable damping of the damper, k s is the spring stiffness, k t is the wheel stiffness; S4: Modeling of Magnetorheological Damper via Neural Networks: The damper experimental data is obtained through the preset equipment, and the data is fitted through the neural network to obtain the weight matrix w I , w H and w O ; The damper neural network model is combined with the vehicle suspension dynamics model to obtain an integrated model; Interact the reinforcement learning structure with the integrated model to train the neural network parameters; S5: Apply DDPG reinforcement learning algorithm to calculate the control quantity of the state quantity. The DDPG network structure is Actor-Critic structure, where the Actor network generates actions: a t =μ(s|θ μ ), the Critic network evaluates the state-action value Q(s,a|θ Q ); S6: The state quantity is passed into the Transformer architecture to replace the traditional Actor network, analyze the correlation of the state quantity, distinguish between valid information and state quantity errors, and convert the data x i (t k ) is passed into the DDPG Transformer architecture; S7: Combine the encoded high-dimensional data X with the weights Point multiplication to obtain a weighted matrix Each row of V represents a high-dimensional representation of sensor data; S8: Transpose the Q and K matrices to obtain QK T matrix; S9: Divide the K matrix dimension by the obtained attention weight result Scaling is performed to calculate the attention weight matrix Perform softmax activation function transformation and convert the result into percentage; S10: The transformed attention weights Multiplying with matrix V gives Each element in the result is the sum of the attention weights of each sample data corresponding to all other sample data in the 64-dimensional Q, K matrix. Adjust V matrix weights Get the result Attention(Q,K,V); S11: self-attention matrix head i =Attention(Q,K,V) to perform multiple groups of parallel training and obtain the multi-head attention mechanism result Concat(head1,…,head h ), and the multi-head attention mechanism result is combined with the multi-head weight matrix W O Multiply them together to get the multi-head attention result MultiHead(Q,K,V)=Concat(head1,…,head h )W O , each head i It is the result of the attention mechanism in different spaces; S12: Replace the traditional reinforcement learning Actor network architecture with the Transformer architecture and generate a parameterized strategy μ(s|θ μ ) to output the optimal action a under a given state t =μ(s|θ μ ), whose goal is to maximize the expected cumulative reward: The Critic network is used to evaluate the state-action value function Q(s,a|θ Q ), and optimize its parameters by minimizing the following loss function: L(θ Q )=E (s,a,r,s′)～D [(r+γQ(s′,μ(s′|θ μ )|θ Q )-Q(s,a|θ Q )) 2 ] Where r is the immediate reward, γ is the discount factor, and D is the data in the experience replay buffer; S13: Design reinforcement learning reward function; S14: Dual network architecture, including two independent Actor networks and With two independent critic networks and The minimum action amount and state-action value are calculated by the following formula: The loss function is defined as: S15: Store the data obtained during the interaction with the environment in the playback buffer D, randomly sample batches of data from the buffer for training, and improve sample utilization. For the dual Actor and Critic network update mechanism, the target network parameters θ are maintained and updated separately. μ ' and θ Q ', each time the network parameters are updated, the target network uses the soft update rule: Actor network update rule: θ μ '←τθ μ +(1-τ)θ μ ', Critic network update rule: θ Q '←τθ Q +(1-τ)θ Q ', where τ is the soft update parameter, which balances the current network parameters with the parameters to be updated, adjusts the network parameters, and obtains the control signal I.

2. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S1, the sensor network includes an accelerometer, a gyroscope, a displacement sensor and a pressure sensor.

3. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S1, the vehicle dynamic data includes vehicle body acceleration, suspension travel, and tire deformation.

4. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S4, the preset equipment includes a suspension test bench.

5. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S4, the output value of the neural network is the controllable damping force F of the magnetorheological damper.

6. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: The state quantity is passed into the Transformer architecture to replace the traditional Actor network, analyze the correlation of the state quantity, distinguish between valid information and state quantity errors, and convert the data x i (t k ) is passed into the DDPG Transformer architecture and includes the following steps: First, the data x i (t k ) Upgraded to 64-dimensional S t , using learnable Fourier features S t Position code: PE (pos,2i) = sin(pos·w i ),PE (pos,2i+1) =cos(pos·w i ), where w i are learnable parameters.

7. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S4, the integrated model is: In the formula, m s is the vehicle body mass, m u is the unsprung mass, F is the damping force of the magnetorheological damper, x s is the displacement of the vehicle body, x u is the unsprung displacement, x r is the wheel displacement, c is the uncontrollable damping of the damper, k s is the spring stiffness, k t is the wheel stiffness.

8. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S2, the preset algorithm includes an interpolation algorithm.

9. The semi-active suspension control system based on Transformer architecture according to claim 1, characterized in that: In S13, the reinforcement learning reward function is The reward function is composed of four parts: body acceleration, suspension travel, tire deformation, and suspension dissipative constraint. The body acceleration and suspension travel are multiplied by weight coefficients a1 and a2 respectively for weight distribution. In addition, obstacle functions P, Q, and N are added. The obstacle function P constrains the suspension travel and keeps the suspension travel within the limit range, so that (x s -x u ) at the maximum suspension travel x max Within the range, the obstacle function Q constrains the suspension wheel deformation, and the obstacle function N constrains the magnetorheological semi-active suspension input.

10. The semi-active suspension control system based on Transformer architecture according to claim 9, characterized in that: The barrier function The barrier function The barrier function

Images