Magnetorheological semi-active suspension switching control system and method based on image recognition

Abstract

The invention discloses a magnetorheological semi-active suspension switching control system and method based on image recognition. The method comprises the steps that a monocular camera is adopted to shoot a road surface image in front of a vehicle in real time, and obstacle recognition is carried out through a pre-trained deep learning model; according to real-time vehicle speed data collected by the vehicle speed sensor, the controller matches control parameters adapting to the current vehicle speed and the obstacle type by using a control parameter switching rule table obtained through simulation in advance in combination with an obstacle recognition result; according to the control parameters and the control algorithm, the controller calculates expected current needed by the magnetorheological damper on the basis of vehicle dynamic data collected in real time; the current source outputs actual current to the magnetorheological damper according to the calculated expected current, and the damping force of the magnetorheological damper is changed by adjusting current input of the magnetorheological damper. According to the invention, real-time identification of front obstacles can be realized, the comfort and the stability of the suspension system under complex road conditions are improved through an optimized control strategy, and particularly, the vibration reduction performance is obviously improved when the suspension system passes through typical obstacles.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle control, and in particular to a magnetorheological semi-active suspension switching control system and method based on image recognition. Background Art

[0002] With the rapid development of the automobile industry, vehicle suspension systems have gradually evolved from traditional passive suspensions to semi-active and active suspension systems. The passive suspension system only absorbs vibrations from the road surface through springs and dampers, lacking adjustability, resulting in poor comfort and stability of the vehicle under complex road conditions. The semi-active suspension system adjusts the damping force of the damper to adapt to different road conditions, thereby improving comfort and stability. Compared with passive suspension, semi-active suspension can respond to different road conditions more flexibly and improve the dynamic performance of the vehicle. Active suspension further adjusts the suspension mechanical characteristics of the vehicle through an active control system to achieve a more precise control effect. Despite this, semi-active suspension is still the most common choice in the market today due to its low implementation cost and control complexity, especially in mid-to-high-end models.

[0003] Magnetorheological dampers (MR dampers) have been widely used in recent years as the core component of semi-active suspension systems. Magnetorheological dampers use the characteristics of magnetorheological fluid to adjust the viscosity of the fluid by changing the strength of the external magnetic field, thereby controlling the damping force. This principle enables the MR damper to quickly adjust the damping force within milliseconds and achieve rapid response to vehicle vibrations. Compared with traditional hydraulic dampers, MR dampers have better adjustability and response speed, and can provide different damping effects under different working conditions, adapting to various complex road conditions. Due to its excellent performance and controllability, MR dampers are widely used in suspension systems of automobiles, railways and other means of transportation.

[0004] The study of suspension control algorithms is one of the key technologies in semi-active suspension systems. The control algorithm is used to adjust the working state of the damper according to the signal input (such as vehicle speed, acceleration, etc.) and the dynamic response of the vehicle to optimize the comfort and handling of the vehicle under different road conditions. Common suspension control algorithms include methods based on fuzzy logic control (Fuzzy Logic), adaptive control, robust control, and optimal control. In addition, in recent years, model predictive control (MPC) has been gradually applied to suspension systems as a forward-looking control method. MPC can adjust the parameters of the damper in advance based on future road conditions, thereby achieving more precise control. However, the implementation of suspension control algorithms still faces certain challenges, especially in the balance between multiple objectives (such as comfort and handling stability). For example, a comfort control method for unmanned vehicles that integrates vehicle speed planning and preview semi-active suspension is disclosed in a Chinese patent (application number: CN202010080594.7), which only controls and optimizes comfort and ignores the handling stability of the vehicle.

[0005] Although significant progress has been made in suspension control algorithms, the use of road surface information in suspension systems is still relatively limited in practical applications. Most current suspension systems rely on dynamic signals such as real-time vehicle speed and acceleration for control, while ignoring the impact of the road conditions ahead. With the development of sensor technologies such as lidar and cameras, the acquisition of road surface information ahead has become more feasible. Nevertheless, effectively integrating this information into the suspension control system, especially dynamically adjusting suspension parameters under various road conditions, remains a challenging research problem. Most existing solutions rely only on the current dynamic response of the vehicle and lack forward-looking information processing capabilities, resulting in the suspension system failing to fully exert its due performance. For example, a control method and system for a semi-active suspension disclosed in a Chinese patent (application number: CN202310079638.8) can determine a fuzzy adjustment strategy and adjust control parameters based on the performance parameters of the semi-active suspension and random disturbance information of the road surface, but it still requires the vehicle to drive on the corresponding road surface before the control parameters can be adjusted, and it is impossible to achieve pre-adjustment of the control algorithm, resulting in poor results. Summary of the invention

[0006] The purpose of the present invention is to provide a magnetorheological semi-active suspension switching control system based on image recognition in order to solve the problems existing in the above-mentioned prior art.

[0007] The technical solution for achieving the purpose of the present invention is as follows: on the one hand, a magnetorheological semi-active suspension switching control system based on image recognition is provided, the system comprising a monocular camera, a vehicle speed sensor, an acceleration sensor, a controller, a current source and a plurality of magnetorheological dampers;

[0008] The monocular camera is used to collect the road surface image in front of the vehicle in real time;

[0009] The vehicle speed sensor is used to collect vehicle speed data in real time;

[0010] The sensor module is used to collect vehicle dynamic data in real time;

[0011] The controller is used to identify obstacles based on the deep learning image recognition model and the road surface image; it is also used to match the control parameters adapted to the current vehicle speed and obstacle type based on the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result, through the control parameter switching rule table obtained in advance through simulation; it is also used to calculate the expected current based on the control algorithm, control parameters and vehicle dynamic data collected by the sensor module, and the current reflects the damping force required by the magnetorheological damper;

[0012] The current source is used to output actual current to the magnetorheological damper according to the calculation result output by the controller;

[0013] The magnetorheological damper adjusts its current input through a current source to change its damping force so as to mitigate or enhance the vibration of the vehicle.

[0014] Furthermore, the process of establishing the deep learning image recognition model includes:

[0015] Step 1.1, data collection and preprocessing: A large number of road images are collected through a monocular camera and image enhancement is performed; the collected road images include road images under different lighting, weather and road conditions, and road images containing a variety of different types of typical obstacles, and all images are labeled, including the category of the obstacle;

[0016] Step 1.2, constructing a data set from the road surface images collected in step 1.1, and dividing the data set into a training set, a validation set, and a test set;

[0017] Step 1.3, model training and optimization: the convolutional neural network model is trained using the training set through a back propagation algorithm; during the training process, each type of obstacle is classified using a cross entropy loss function, and the model weights are continuously adjusted through an optimization algorithm to minimize errors and improve recognition accuracy;

[0018] Step 1.4, model verification and tuning: After the model training is completed, performance evaluation is performed on the validation set and test set, and the model is further tuned based on the evaluation results.

[0019] Furthermore, step 1.2 constructs a data set from the road surface image collected in step 1.1, specifically by performing transformation processing including at least rotation, scaling, and translation on the road surface image collected in step 1.1 to increase the amount of image data, and constructs a data set from the original image and the image after the transformation processing.

[0020] Furthermore, during the training process of step 1.3, data augmentation techniques and regularization methods are used to prevent overfitting.

[0021] Furthermore, in step 1.4, the recognition capability of the model is evaluated by at least calculating the accuracy, recall and F1 score indicators of the model under different obstacle types.

[0022] Furthermore, the control parameters include but are not limited to ceiling-hook damping coefficients of different obstacles at different vehicle speeds, and the control algorithm includes but is not limited to ceiling-hook control algorithm.

[0023] Furthermore, the process of establishing the control parameter switching rule table specifically includes:

[0024] Step 2.1, determine the current control target according to different driving conditions;

[0025] Step 2.2, using a simulation model based on vehicle dynamic response, and determining the optimal ceiling damping coefficient under different control objectives through an optimization algorithm at different vehicle speeds and obstacle types;

[0026] Step 2.3, based on the results of step 2.1 and step 2.2, a rule table is formed, which lists the optimal skyhook damping coefficients under different driving conditions.

[0027] Furthermore, the control objectives described in step 2.1 include: when driving at low speed, giving priority to optimizing comfort; when driving at high speed, giving priority to optimizing handling stability; the low-speed and high-speed customized settings.

[0028] Furthermore, the control target in step 2.1 is dynamically adjusted according to the driver's demand or the automatic detection system.

[0029] On the other hand, a magnetorheological semi-active suspension switching control method based on image recognition is provided, the method comprising:

[0030] Step 1: Use a monocular camera to capture the road image in front of the vehicle in real time, and identify obstacles through a pre-trained deep learning model;

[0031] Step 2: The controller uses the control parameter switching rule table obtained by simulation in advance to match the control parameters suitable for the current vehicle speed and obstacle type according to the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result;

[0032] Step 3, the controller transmits the matched control parameters to the control algorithm, and the control algorithm calculates the expected current required by the magnetorheological damper based on the real-time collected vehicle dynamic data;

[0033] Step 4: the current source outputs actual current to the magnetorheological damper according to the calculated expected current, and the damping force of the magnetorheological damper is changed by adjusting the current input of the magnetorheological damper to reduce or enhance the vibration of the vehicle.

[0034] Compared with the prior art, the present invention has the following significant advantages:

[0035] (1) The present invention can identify obstacles ahead in real time through a monocular camera, and dynamically switch control parameters based on the identification results to optimize the performance of the suspension system, especially significantly improving comfort and stability when passing through typical obstacles.

[0036] (2) Using deep learning algorithms for obstacle recognition can improve the accuracy and real-time performance of recognition at a lower cost.

[0037] The present invention is further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the system composition of the present invention.

[0039] Figure 2 Flowchart for obstacle recognition and control parameter switching. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0041] Combination Figure 1 , the present invention provides a magnetorheological semi-active suspension switching control system based on image recognition, the system comprising a monocular camera, a vehicle speed sensor, an acceleration sensor, a controller, a current source and a plurality of magnetorheological dampers;

[0042] The monocular camera is used to collect the road surface image in front of the vehicle in real time;

[0043] The vehicle speed sensor is used to collect vehicle speed data in real time;

[0044] The sensor module (including but not limited to the vehicle acceleration sensor) is used to collect vehicle dynamic data in real time;

[0045] The controller is used to identify obstacles based on the deep learning image recognition model and the road surface image; it is also used to match the control parameters adapted to the current vehicle speed and obstacle type based on the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result, through the control parameter switching rule table obtained in advance through simulation; it is also used to calculate the expected current based on the control algorithm, control parameters and vehicle dynamic data collected by the sensor module, and the current reflects the damping force required by the magnetorheological damper, thereby affecting the response of the suspension system (whenever the system identifies a specific obstacle and combines it with the current vehicle speed, the controller will match the most suitable ceiling damping coefficient from the rule table. The matched control parameters will be passed to the control algorithm, and the response of the suspension system will be adjusted to meet the current comfort or handling stability target);

[0046] The current source is used to output actual current to the magnetorheological damper according to the calculation result output by the controller;

[0047] The magnetorheological damper adjusts its current input through a current source to change its damping force, thereby affecting the vibration response of the vehicle suspension system and providing optimal comfort and stability.

[0048] Furthermore, in one embodiment, the process of establishing the deep learning image recognition model includes:

[0049] Step 1.1, data collection and preprocessing: A large number of road images are collected through a monocular camera and image enhancement is performed; the collected road images include road images under different lighting, weather and road conditions, and road images containing a variety of different types of typical obstacles, and all images are labeled, including the category of the obstacle;

[0050] Here, typical obstacles include speed bumps, puddles, potholes, etc. Categories of obstacles include speed bumps, puddles, etc.

[0051] Step 1.2, constructing a data set from the road surface images collected in step 1.1, and dividing the data set into a training set, a validation set, and a test set;

[0052] Here, the training set is used to learn the model, the validation set is used to adjust the model's hyperparameters, and the test set is used to evaluate the model's accuracy;

[0053] Step 1.3, model training and optimization: the convolutional neural network model is trained using the training set through a back propagation algorithm; during the training process, each type of obstacle is classified using a cross entropy loss function, and the model weights are continuously adjusted through an optimization algorithm to minimize errors and improve recognition accuracy;

[0054] Step 1.4, model verification and tuning: After the model training is completed, performance evaluation is performed on the validation set and test set. Based on the evaluation results, the model is further tuned to improve its recognition accuracy and real-time performance in practical applications.

[0055] Preferably, in some embodiments, step 1.2 constructs a data set from the road surface image collected in step 1.1, specifically: the road surface image collected in step 1.1 is subjected to transformation processing including at least rotation, scaling, and translation to increase the amount of image data, and the data set is jointly constructed by the original image and the image after the transformation processing.

[0056] Preferably, in some embodiments, during the training process of step 1.3, data enhancement techniques and regularization methods are used to prevent overfitting and ensure the generalization ability of the model in complex environments.

[0057] Preferably, in some embodiments, in step 1.4, the recognition capability of the model is evaluated by at least calculating the accuracy, recall and F1 score indicators of the model under different obstacle types.

[0058] Furthermore, in one of the embodiments, the control parameters include but are not limited to ceiling-hook damping coefficients of different obstacles at different vehicle speeds, and the control algorithm includes but is not limited to ceiling-hook control algorithm.

[0059] Furthermore, in one of the embodiments, the process of establishing the control parameter switching rule table specifically includes:

[0060] Step 2.1, determine the current control target according to different driving conditions;

[0061] Step 2.2, using a simulation model based on vehicle dynamic response, and determining the optimal ceiling damping coefficient under different control objectives through an optimization algorithm at different vehicle speeds and obstacle types;

[0062] Step 2.3, based on the results of step 2.1 and step 2.2, a rule table is formed, which lists the optimal skyhook damping coefficients under different driving conditions.

[0063] Preferably, in some embodiments, the control objectives in step 2.1 include: when driving at low speed, giving priority to optimizing comfort (reducing passengers' sense of vibration); when driving at high speed, giving priority to optimizing handling stability (improving vehicle stability); the low-speed and high-speed customized settings.

[0064] Preferably, in some embodiments, the control target in step 2.1 is dynamically adjusted according to the driver's demand or an automatic detection system.

[0065] Preferably, in some embodiments, the optimization algorithm in step 2.2 includes but is not limited to a genetic algorithm and a particle swarm optimization algorithm.

[0066] Combination Figure 2 , provides a magnetorheological semi-active suspension switching control method based on image recognition, specifically including:

[0067] Step 1: Use a monocular camera to capture the road image in front of the vehicle in real time, and identify obstacles through a pre-trained deep learning model;

[0068] Step 2: The controller uses the control parameter switching rule table obtained by simulation in advance to match the control parameters suitable for the current vehicle speed and obstacle type according to the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result;

[0069] Step 3, the controller transmits the matched control parameters to the control algorithm, and the control algorithm calculates the expected current required by the magnetorheological damper based on the real-time collected vehicle dynamic data;

[0070] Step 4: the current source outputs actual current to the magnetorheological damper according to the calculated expected current, and the damping force of the magnetorheological damper is changed by adjusting the current input of the magnetorheological damper to reduce or enhance the vibration of the vehicle.

[0071] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A magnetorheological semi-active suspension switching control system based on image recognition, characterized in that: The system includes a monocular camera, a vehicle speed sensor, an acceleration sensor, a controller, a current source and a plurality of magnetorheological dampers; The monocular camera is used to collect the road surface image in front of the vehicle in real time; The vehicle speed sensor is used to collect vehicle speed data in real time; The sensor module is used to collect vehicle dynamic data in real time; The controller is used to identify obstacles based on the deep learning image recognition model and the road surface image; It is also used to match the control parameters suitable for the current vehicle speed and obstacle type according to the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result through the control parameter switching rule table obtained in advance through simulation; It is also used to calculate the expected current based on the control algorithm, control parameters and vehicle dynamic data collected by the sensor module, and the current reflects the required damping force of the magnetorheological damper; The current source is used to output actual current to the magnetorheological damper according to the calculation result output by the controller; The magnetorheological damper adjusts its current input through a current source to change its damping force so as to mitigate or enhance the vibration of the vehicle.

2. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 1 is characterized in that: The process of establishing the deep learning image recognition model includes: Step 1.1, data collection and preprocessing: A large number of road images are collected through a monocular camera and image enhancement is performed; the collected road images include road images under different lighting, weather and road conditions, and road images containing a variety of different types of typical obstacles, and all images are labeled, including the category of the obstacle; Step 1.2, constructing a data set from the road surface images collected in step 1.1, and dividing the data set into a training set, a validation set, and a test set; Step 1.3, model training and optimization: the convolutional neural network model is trained using the training set through a back propagation algorithm; during the training process, each type of obstacle is classified using a cross entropy loss function, and the model weights are continuously adjusted through an optimization algorithm to minimize errors and improve recognition accuracy; Step 1.4, model verification and tuning: After the model training is completed, performance evaluation is performed on the validation set and test set, and the model is further tuned based on the evaluation results.

3. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 2 is characterized in that: Step 1.2 constructs a data set from the road surface image collected in step 1.1, specifically: the road surface image collected in step 1.1 is transformed by at least rotation, scaling, and translation to increase the image data volume, and the data set is jointly constructed by the original image and the transformed image.

4. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 2 is characterized in that: During the training process of step 1.3, data augmentation techniques and regularization methods are used to prevent overfitting.

5. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 2 is characterized in that: In step 1.4, the recognition ability of the model is evaluated by calculating at least the precision, recall and F1 score indicators of the model under different obstacle types.

6. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 1 is characterized in that: The control parameters include but are not limited to the ceiling damping coefficients of different obstacles at different vehicle speeds, and the control algorithm includes but is not limited to the ceiling control algorithm.

7. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 6 is characterized in that: The process of establishing the control parameter switching rule table specifically includes: Step 2.1, determining a current control target according to different driving conditions; the control target is dynamically adjusted according to the driver's needs or the automatic detection system; Step 2.2, using a simulation model based on vehicle dynamic response, and determining the optimal ceiling damping coefficient under different control objectives through an optimization algorithm at different vehicle speeds and obstacle types; Step 2.3, based on the results of step 2.1 and step 2.2, a rule table is formed, which lists the optimal skyhook damping coefficients under different driving conditions.

8. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 7 is characterized in that: The control objectives described in step 2.1 include: when driving at low speeds, giving priority to optimizing comfort; when driving at high speeds, giving priority to optimizing handling stability; and the low-speed and high-speed customized settings.

9. The magnetorheological semi-active suspension switching control system based on image recognition according to claim 7, characterized in that: The optimization algorithms in step 2.2 include genetic algorithm and particle swarm optimization algorithm.

10. A magnetorheological semi-active suspension switching control method based on image recognition based on the system according to any one of claims 1 to 9, characterized in that: The method comprises: Step 1: Use a monocular camera to capture the road image in front of the vehicle in real time, and identify obstacles through a pre-trained deep learning model; Step 2: The controller uses the control parameter switching rule table obtained by simulation in advance to match the control parameters suitable for the current vehicle speed and obstacle type according to the real-time vehicle speed data collected by the vehicle speed sensor and the obstacle recognition result; Step 3, the controller transmits the matched control parameters to the control algorithm, and the control algorithm calculates the expected current required by the magnetorheological damper based on the real-time collected vehicle dynamic data; Step 4: the current source outputs actual current to the magnetorheological damper according to the calculated expected current, and the damping force of the magnetorheological damper is changed by adjusting the current input of the magnetorheological damper to reduce or enhance the vibration of the vehicle.

Images