Vehicle magnetorheological damper control method and system with fusion vision detection

Abstract

The application discloses a vehicle magnetorheological damper control method and system with fusion visual detection, and the system comprises a controllable power supply, a plurality of magnetorheological dampers, an attitude sensor, a vehicle speed sensor, a vibration sensor, a camera, a controller and a holder. The method steps are as follows: 1) judging the generated road condition parameter type, if the road condition parameters are different, step 2) is entered, otherwise step 3) is entered; 2) control signals are sequentially generated according to the generation order of the road condition parameters, and are sequentially transmitted to the controllable power supply; 3) control signals are sequentially generated according to the order of the artificial driving road condition parameters, the short-term road condition parameters, the medium-term road condition parameters and the long-term road condition parameters, and are transmitted to the controllable power supply; 4) the controllable power supply respectively transmits currents to the magnetorheological dampers, so as to control the magnetorheological dampers to output damping forces to corresponding wheels. The application can significantly improve the comfort, stability and safety of the vehicle during driving.

Description

Technical Field

[0001] This invention relates to the field of vehicle control, specifically to a vehicle magnetorheological damper control method and system that integrates visual detection. Background Technology

[0002] Magnetorheological dampers can be used to construct intelligent semi-active suspensions with adjustable damping for vehicles. By controlling the magnetic field, the magnetorheological fluid inside the damper undergoes a phase transition between a highly fluid Newtonian fluid and a solid-like state, thereby changing the damping force of the damper. This transient controllability makes it highly valuable in the field of vehicle vibration reduction. As a semi-active suspension control system, in addition to the performance of the magnetorheological damping device, the detection and control of road condition parameters, which provide the control parameter decisions, are also crucial. Currently, vehicle magnetorheological systems generally obtain motion parameters such as acceleration, velocity, and displacement using traditional accelerometers based on infrared, inductive, and piezoelectric principles, thus achieving open-loop or closed-loop control. As an inductive load, the magnetorheological system suffers from significant response delay. Furthermore, when the vehicle travels at high speeds, the response time of traditional sensors indicates that a certain road condition has already occurred. At this point, the lag response of the magnetorheological system cannot handle the road condition in time, making it difficult to achieve optimal results. For example, when a vehicle travels at a high speed over a speed bump, traditional solutions require the accelerometer to detect an anomaly in the signal generated by the speed bump before controlling the magnetorheological damper. Furthermore, since magnetorheological systems typically have a response time on the order of tens of milliseconds, by the time the system responds, the vehicle may have already moved away from the speed bump. Clearly, if abnormal road conditions ahead of the vehicle can be detected in advance (i.e., pre-sensing), sufficient preprocessing and pre-set control strategies can be provided for the magnetorheological system, thereby accurately and promptly controlling the output damping force of the magnetorheological damper and achieving a more ideal vehicle vibration reduction effect.

[0003] Visual inspection technology primarily utilizes artificial intelligence-based computer vision to accurately and rapidly extract features from complex images. Traditional computer vision mainly employs manually designed feature extraction methods, which are only effective at extracting single features in specific environments. Their robustness and generalization ability are poor, making it difficult to achieve accurate recognition results in real-world applications. Summary of the Invention

[0004] The purpose of this invention is to provide a vehicle magnetorheological damper control system that integrates visual detection, including a controllable power supply, several magnetorheological dampers, an attitude sensor, a vehicle speed sensor, a vibration sensor, a camera, a controller, and a gimbal.

[0005] The controllable power supply transmits current to each magnetorheological damper, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

[0006] The magnetorheological damper is placed between the wheel and the frame;

[0007] The attitude sensor acquires the vehicle's tilt angle and the centripetal acceleration when the vehicle turns, and transmits them to the controller.

[0008] The vehicle speed sensor acquires the vehicle's speed and transmits it to the controller;

[0009] The vibration sensor acquires vibration parameters transmitted from the ground to the vehicle body and transmits them to the controller;

[0010] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0011] The gimbal is mounted on the vehicle body and is used to stabilize the camera;

[0012] The controller is equipped with an image processing network framework;

[0013] The image processing framework processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the following: long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0014] After receiving data from the attitude sensor, vehicle speed sensor, and vibration sensor, the controller processes the data to obtain human driving road condition parameters.

[0015] The controller stores a set of current intensity test values ​​for vehicle model and magnetorheological damper model under different long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters.

[0016] The controller searches the current intensity test value set, obtains the current intensity values ​​under the current long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters, and generates corresponding control signals, which are then transmitted to the controllable power supply.

[0017] Furthermore, the long-term road condition parameters refer to the road surface type on which the vehicle will travel at the future time t1, including but not limited to high-grade asphalt roads, general asphalt roads, cement roads, gravel roads, muddy roads, grasslands, snow, and ice.

[0018] The short-time-history road condition parameters include short-time-history road condition classifications and corresponding depth information;

[0019] The short-term road condition classification is based on the type of road condition target that the vehicle needs to cross in the future time t2 and the distance between the road condition target and the camera; t2 < t1; the road condition target types include, but are not limited to, potholes, raised road surfaces, manhole covers, speed bumps, and steps;

[0020] The mid-time history road condition parameters are mid-time history road condition types, including but not limited to uphill, downhill, and curves;

[0021] The human-generated driving road condition parameters include, but are not limited to, acceleration, deceleration, and centripetal acceleration corresponding to emergency start, emergency braking, and emergency cornering driving conditions.

[0022] Furthermore, after the controller obtains the current intensity value, it also provides the driver with the recommended current intensity level under the current long-term road condition parameters and medium-term road condition parameters;

[0023] At this time, the controllable power supply provides a current range switch;

[0024] The driver can manually select the current intensity level recommended by the controller or choose their own.

[0025] Furthermore, the steps for obtaining control signals under short-term road condition parameters include:

[0026] a1) The controller calculates the lead distance between the road condition target and each magnetorheological damper; the lead distance is equal to the sum of the distance between the road condition target and the camera and the distance from the camera to each magnetorheological damper;

[0027] a2) By dividing the lead distance by the current vehicle speed, the lead time from the road condition target to each magnetorheological damper at the current vehicle speed can be calculated.

[0028] a3) The delay start-up time of each magnetorheological damper is calculated by subtracting the response time of the magnetorheological system from the lead time.

[0029] a4) Obtain the span dimension of the road condition target in the lateral direction of the road surface and the vehicle's driving deviation angle, and calculate the wheels that will cross the road condition target;

[0030] a5) By dividing the forward span dimension of the road condition target by the current vehicle speed, the holding time of each magnetorheological damper after delayed start is calculated.

[0031] a6) The controller searches for the current intensity test value set, obtains the current intensity value under the current short-term road condition parameters, and generates a control signal based on the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheel that will cross the road condition target.

[0032] The controller transmits control signals to a controllable power source, causing the magnetorheological damper corresponding to the wheel that will cross the road condition target to respond.

[0033] Furthermore, when long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated in an overlapping manner, the corresponding control signals are generated in an overlapping manner, thereby causing the controllable power supply to execute different control signals in an overlapping manner.

[0034] Furthermore, when long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated simultaneously, corresponding control signals are generated according to the priority order of the road condition parameters and transmitted to the controllable power supply; the priority order of the road condition parameters is: human-driven road condition parameters, short-term road condition parameters, medium-term road condition parameters, and long-term road condition parameters.

[0035] Furthermore, each magnetorheological damper has a corresponding vibration sensor.

[0036] Furthermore, methods for processing road condition images to obtain the distance between road targets and the camera include, but are not limited to, the following two:

[0037] b1) Use a binocular camera to acquire two-dimensional and three-dimensional road condition images, and perform image processing on the two-dimensional and three-dimensional road condition images to obtain the distance between the road condition target and the camera.

[0038] b2) Use fused depth information from a depth camera or ranging radar to obtain the distance between road targets and the camera.

[0039] Furthermore, the image processing framework's steps for processing the road condition image in front of the vehicle and the distance between the road condition target and the camera include:

[0040] c1) Perform image preprocessing on the road conditions in front of the vehicle, including adjusting the frame rate, resolution, and sampling rate;

[0041] c2) Based on the weight file, the video is further processed to achieve road surface feature discrimination and geometric position annotation; the methods for further processing the video include, but are not limited to, deep convolutional network models.

[0042] The method for a vehicle magnetorheological damper control system using the aforementioned fused visual detection includes the following steps:

[0043] 1) The tilt angle of the vehicle body and the centripetal acceleration of the vehicle when turning are obtained by using attitude sensors and transmitted to the controller;

[0044] The vehicle speed is obtained using a vehicle speed sensor and transmitted to the controller;

[0045] Vibration parameters transmitted from the ground to the vehicle body are obtained using vibration sensors and then transmitted to the controller.

[0046] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0047] 2) The image processing network framework of the controller processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0048] The controller processes data from the attitude sensor, vehicle speed sensor, and vibration sensor to obtain road condition parameters for human driving.

[0049] 3) The controller determines the type of generated road condition parameters. If long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are not generated simultaneously, then proceed to step 4); otherwise proceed to step 5.

[0050] 4) In the order of generating long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human driving road condition parameters, control signals are generated in sequence and transmitted to the controllable power supply in sequence, then jump to step 6).

[0051] 5) Generate corresponding control signals according to the priority order of road condition parameters and transmit them to the controllable power supply, then proceed to step 6); Specific steps include:

[0052] 5.1) The controller determines the current intensity value, the holding time after the magnetorheological damper is started, and the wheel to be controlled based on the current human driving road condition parameters, and generates a control signal, which is transmitted to the controllable power supply.

[0053] 5.2) The controller determines the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheels that will cross the target of the road condition based on the current short-term road condition parameters, and generates a control signal and transmits it to the controllable power supply.

[0054] 5.3) The controller determines the current intensity value based on the current mid-time road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0055] 5.4) The controller determines the current intensity value based on the current long-term road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0056] 6) The controllable power supply transmits current to each magnetorheological damper according to the order in which the control signals are received, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

[0057] The technical effects of this invention are undeniable. This invention utilizes advanced computer vision to achieve vehicle road condition pre-perception, that is, to obtain the target type (such as obstacles, uneven road surfaces, curves, slopes, etc.) and depth information (i.e., the distance between the target and the magnetorheological damper) of the road conditions in advance. Combined with the motion parameters such as vehicle speed and attitude provided by existing vehicle sensors, it can provide sufficient preprocessing time for the magnetorheological system, achieve a more ideal vehicle vibration reduction effect, and significantly improve the comfort, stability and safety of the vehicle during driving. This is of great significance for the upgrading of the intelligent vehicle industry. Attached Figure Description

[0058] Figure 1 Structure diagram of a vehicle magnetorheological damper control system that integrates visual inspection;

[0059] Figure 2 This is a system control block diagram;

[0060] Figure 3 A flowchart for long-term road condition procedures;

[0061] Figure 4 This is a flowchart for short-term road condition control.

[0062] Figure 5 This is a flowchart for medium-term road condition control.

[0063] Figure 6 Flowchart for road condition control by human drivers;

[0064] Figure 7 This is a flowchart for priority control.

[0065] Figure 8 This is a schematic diagram of a short-term road condition implementation example.

[0066] In the diagram, 1-lane line (curve / slope), 2-pothole, 3-vehicle, 4-camera, 5-magnetorheological damper. Detailed Implementation

[0067] The present invention will be further described below with reference to embodiments, but it should not be construed that the scope of the present invention is limited to the following embodiments. Various substitutions and modifications made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention should be included within the scope of protection of the present invention.

[0068] Example 1:

[0069] See Figures 1 to 8 The vehicle magnetorheological damper control system integrating visual detection includes a controllable power supply, several magnetorheological dampers 5, attitude sensor, vehicle speed sensor, vibration sensor, camera 4, controller, and gimbal.

[0070] The controllable power supply transmits current to each magnetorheological damper, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

[0071] The magnetorheological damper is placed between the wheel and the frame;

[0072] The attitude sensor acquires the vehicle's tilt angle and the centripetal acceleration when the vehicle turns, and transmits them to the controller.

[0073] The vehicle speed sensor acquires the vehicle's speed and transmits it to the controller;

[0074] The vibration sensor acquires vibration parameters transmitted from the ground to the vehicle body and transmits them to the controller;

[0075] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0076] The gimbal is mounted on the vehicle body and is used to stabilize the camera;

[0077] The controller is equipped with an image processing network framework; the image processing network framework may be a deep neural network framework.

[0078] The image processing framework processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the following: long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0079] After receiving data from the attitude sensor, vehicle speed sensor, and vibration sensor, the controller processes the data to obtain human driving road condition parameters.

[0080] The controller stores a set of current intensity test values ​​for vehicle model and magnetorheological damper model under different long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters.

[0081] The controller searches the current intensity test value set, obtains the current intensity values ​​under the current long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters, and generates corresponding control signals, which are then transmitted to the controllable power supply.

[0082] The long-term road condition parameters refer to the road surface type on which the vehicle will travel at the future time t1, including but not limited to high-grade asphalt roads, general asphalt roads, cement roads, gravel roads, muddy roads, grasslands, snow, and ice.

[0083] The short-time-history road condition parameters include short-time-history road condition classifications and corresponding depth information;

[0084] The short-term road condition classification is based on the type of road condition target that the vehicle needs to cross in the future time t2 and the distance between the road condition target and the camera; t2 < t1; the road condition target types include, but are not limited to, potholes, raised road surfaces, manhole covers, speed bumps, and steps;

[0085] The mid-time history road condition parameters are mid-time history road condition types, including but not limited to uphill, downhill, and curves;

[0086] The human-generated driving road condition parameters include, but are not limited to, acceleration, deceleration, and centripetal acceleration corresponding to emergency start, emergency braking, and emergency cornering driving conditions.

[0087] After the controller obtains the current intensity value, it also provides the driver with the recommended current intensity level under the current long-term road condition parameters and medium-term road condition parameters.

[0088] At this time, the controllable power supply provides a current range switch;

[0089] The driver can manually select the current intensity level recommended by the controller or choose their own.

[0090] The steps to obtain control signals under short-term road condition parameters include:

[0091] a1) The controller calculates the lead distance between the road condition target and each magnetorheological damper; the lead distance is equal to the sum of the distance between the road condition target and the camera and the distance from the camera to each magnetorheological damper;

[0092] a2) By dividing the lead distance by the current vehicle speed, the lead time from the road condition target to each magnetorheological damper at the current vehicle speed can be calculated.

[0093] a3) The delay start-up time of each magnetorheological damper is calculated by subtracting the response time of the magnetorheological system from the lead time.

[0094] a4) Obtain the span dimension of the road condition target in the lateral direction of the road surface and the vehicle's driving deviation angle, and calculate the wheels that will cross the road condition target;

[0095] a5) By dividing the forward span dimension of the road condition target by the current vehicle speed, the holding time of each magnetorheological damper after delayed start is calculated.

[0096] a6) The controller searches for the current intensity test value set, obtains the current intensity value under the current short-term road condition parameters, and generates a control signal based on the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheel that will cross the road condition target.

[0097] The controller transmits control signals to a controllable power source, causing the magnetorheological damper corresponding to the wheel that will cross the road condition target to respond.

[0098] When long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated in an overlapping manner, the corresponding control signals are generated in an overlapping manner, thereby causing the controllable power supply to execute different control signals in an overlapping manner.

[0099] When long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated simultaneously, corresponding control signals are generated according to the priority order of the road condition parameters and transmitted to the controllable power supply. The priority order of the road condition parameters is: human-driven road condition parameters, short-term road condition parameters, medium-term road condition parameters, and long-term road condition parameters.

[0100] Each magnetorheological damper has a corresponding vibration sensor.

[0101] Methods for processing road condition images to obtain the distance between road objects and the camera include, but are not limited to, the following two:

[0102] b1) Use a binocular camera to acquire two-dimensional and three-dimensional road condition images, and perform image processing on the two-dimensional and three-dimensional road condition images to obtain the distance between the road condition target and the camera.

[0103] b2) Use fused depth information from a depth camera or ranging radar to obtain the distance between road targets and the camera.

[0104] The image processing framework processes the road conditions in front of the vehicle and the distance between road targets and the camera in the following steps:

[0105] c1) Perform image preprocessing on the road conditions in front of the vehicle, including adjusting the frame rate, resolution, and sampling rate;

[0106] c2) Based on the weight file, the video is further processed to achieve road surface feature discrimination and geometric position annotation; the methods for further processing the video include, but are not limited to, deep convolutional network models.

[0107] The method for a vehicle magnetorheological damper control system using the aforementioned fused visual detection includes the following steps:

[0108] 1) The tilt angle of the vehicle body and the centripetal acceleration of the vehicle when turning are obtained by using attitude sensors and transmitted to the controller;

[0109] The vehicle speed is obtained using a vehicle speed sensor and transmitted to the controller;

[0110] Vibration parameters transmitted from the ground to the vehicle body are obtained using vibration sensors and then transmitted to the controller.

[0111] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0112] 2) The image processing network framework of the controller processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0113] The controller processes data from the attitude sensor, vehicle speed sensor, and vibration sensor to obtain road condition parameters for human driving.

[0114] 3) The controller determines the type of generated road condition parameters. If long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are not generated simultaneously, then proceed to step 4); otherwise proceed to step 5.

[0115] 4) In the order of generating long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human driving road condition parameters, control signals are generated in sequence and transmitted to the controllable power supply in sequence, then jump to step 6).

[0116] 5) Generate corresponding control signals according to the priority order of road condition parameters and transmit them to the controllable power supply, then proceed to step 6); Specific steps include:

[0117] 5.1) The controller determines the current intensity value, the holding time after the magnetorheological damper is started, and the wheel to be controlled based on the current human driving road condition parameters, and generates a control signal, which is transmitted to the controllable power supply.

[0118] 5.2) The controller determines the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheels that will cross the target of the road condition based on the current short-term road condition parameters, and generates a control signal and transmits it to the controllable power supply.

[0119] 5.3) The controller determines the current intensity value based on the current mid-time road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0120] 5.4) The controller determines the current intensity value based on the current long-term road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0121] 6) The controllable power supply transmits current to each magnetorheological damper according to the order in which the control signals are received, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

[0122] Example 2:

[0123] A vehicle magnetorheological damper control system based on fusion vision detection includes a controllable power supply, a magnetorheological damper, an attitude sensor, a vehicle speed sensor, a vibration sensor, a camera, a gimbal, a controller, and other sensors, such as... Figure 1 As shown.

[0124] The controllable power supply is used to generate a constant or waveform current to the magnetorheological damper; furthermore, the controller obtains the output current feedback of the controllable power supply to form a closed-loop current control.

[0125] The magnetorheological damper is placed between the wheel and the frame (along with a spring element), and can be installed on the front wheel, the rear wheel, or all wheels depending on actual needs.

[0126] The attitude sensor is used to obtain the vehicle body's tilt angle and the centripetal acceleration when the vehicle turns; the system can also directly read the attitude sensor information already available in the vehicle system.

[0127] The vehicle speed sensor is used to obtain the vehicle's speed; the system can also directly read the vehicle speed sensor information already present in the vehicle system.

[0128] The vibration sensor is used to acquire vibration parameters transmitted from the ground to the vehicle body, such as vibration accelerometers, displacement sensors, etc.; more preferably, each magnetorheological damper has a corresponding vibration sensor.

[0129] The camera is used to acquire road conditions in front of the vehicle. It can also acquire the distance between the target object in the road conditions and the camera in two ways: one is image processing from two-dimensional to three-dimensional images (e.g., binocular camera), and the other is to fuse depth information or ranging radar in the camera (e.g., depth camera).

[0130] The gimbal is mounted on the vehicle body and is used to stabilize the camera to obtain stable images.

[0131] The controller is equipped with a deep convolutional network framework for processing camera image data and acquiring road condition parameters; the controller also processes relevant data from attitude sensors, vehicle speed sensors, and vibration sensors; the controller processes control signals required for different operating conditions and sends them to a controllable power supply; furthermore, the controller that processes the camera and sensor signals can be separated from the controller that sends the control signals to the controllable power supply and processes the control signals required for different operating conditions.

[0132] The road condition type parameters and their corresponding control methods are as follows:

[0133] The first type of road condition parameter is the long-term road condition classification (i.e., the type of road surface the vehicle travels on over a long period of time), including but not limited to high-grade asphalt roads, general asphalt roads, cement roads, gravel roads, muddy roads, grasslands, snow, and ice. The long-term road condition classification is obtained from camera data or verified by vibration sensors. The controller autonomously sends corresponding control signals to the controllable power supply, which controls each magnetorheological damper to make the same response. Specifically, it is based on the current intensity corresponding to the specific vehicle model and magnetorheological damper model after testing on various types of road surfaces. The controller can also issue road condition classification prompts to the driver (voice or information display), and the controllable power supply provides a current level switch, allowing the driver to manually select the current intensity level suggested by the controller or selected by the driver. Figure 3 The diagram shows an example of a workflow.

[0134] The second type of road condition parameter is the short-term road condition classification (i.e., the type of road condition target that the vehicle needs to cross in a very short time) and its depth information (i.e., the distance between the target and the camera), including but not limited to potholes, raised road surfaces, manhole covers, speed bumps, steps, and other obstacles; the short-term road condition classification is obtained from camera data; the controller calculates the lead distance between the road condition target and each magnetorheological damper, specifically the distance of the road condition target obtained by the camera plus the known distance between the camera and each magnetorheological damper, and then divides this lead distance by the current vehicle speed obtained by the vehicle speed sensor to obtain the lead time of the road condition target to each magnetorheological damper at the current vehicle speed; the controller sends corresponding control signals to the controllable power supply, and the controllable power supply controls each magnetorheological damper to make different responses, specifically the current intensity corresponding to various road surfaces and vehicle speeds after testing according to the specific vehicle model and magnetorheological damper model; the aforementioned The delay start-up time of each magnetorheological damper is obtained by subtracting the response time of the magnetorheological system from the lead time. Furthermore, the controller acquires the forward span of the road condition target in front of the road surface. This forward span is divided by the current vehicle speed obtained by the vehicle speed sensor, which gives the holding time of each magnetorheological damper after the delay start-up. Furthermore, the controller acquires the lateral span of the road condition target on the road surface. Using the vehicle's yaw angle obtained by the attitude sensor, the controller calculates the wheels that will cross the road condition target, thereby controlling the magnetorheological dampers corresponding to those wheels to respond accordingly. Magnetorheological dampers corresponding to wheels that will not cross the road condition target do not respond. Furthermore, if the vehicle is undergoing a gear change, the controller considers the effect of driving acceleration when calculating the lead time, holding time, and wheel crossing time. Furthermore, if the delay start-up time is less than 0, no response is made. Figure 4 The diagram shows an example of a workflow.

[0135] The third type of road condition parameter is a medium-time road condition classification, including uphill, downhill, and curves. The medium-time road condition classification is obtained from camera data or verified by attitude sensors. The controller autonomously sends corresponding control signals to the controllable power supply, which controls each magnetorheological damper to make the same response. Specifically, it is based on the current intensity corresponding to the specific vehicle model and magnetorheological damper model after testing on various types of road surfaces. The controller can also issue road condition classification prompts to the driver (voice or information display), and the controllable power supply provides a current level switch, allowing the driver to manually select the current intensity level recommended by the controller or selected by the driver. Figure 5 The diagram illustrates one example of a workflow. (Note: This type of control method is essentially the same as the first type.)

[0136] The fourth type of road condition parameters are for human-driven road conditions such as emergency start, emergency braking, and emergency cornering. The parameters for this type of condition (acceleration, deceleration, and centripetal acceleration) are obtained by attitude sensors and vehicle speed sensors. The controller autonomously sends corresponding control signals to the controllable power supply, which controls each magnetorheological damper to make the same response. Specifically, the current intensity is determined based on the specific vehicle model and magnetorheological damper model after testing on various types of road surfaces. Figure 6 The diagram shows an example of a workflow.

[0137] When the above-mentioned road condition parameters occur simultaneously, the corresponding control methods are implemented concurrently; when the above four types of road condition parameters occur simultaneously, the corresponding control methods are implemented according to the priority of "fourth type > second type > third type > first type". Figure 7 The diagram shows an example of a workflow.

[0138] When applying the above current intensity to the road condition parameters: in addition to applying a step current, a gradually increasing waveform current can also be applied; when removing the applied current, a gradually decreasing waveform current can also be applied.

[0139] The control system is a strongly coupled system with multiple inputs and multiple outputs. The input parameters and output parameters influence each other. The multiple inputs include feature information and distance information of the road surface collected by the camera, speed information collected by the speed sensor, information collected by the vibration sensor, and vehicle status information collected by the attitude sensor. The multiple outputs include multiple independent magnetorheological dampers that can be controlled separately to output corresponding damping forces.

[0140] After identifying the road surface type parameters, the core control concept is classification control, employing different control strategies for different road conditions. For example, in short-term road conditions, different control strategies are used for different road condition objectives; for instance, when a vehicle goes over a speed bump, the magnetorheological dampers on the front wheels can be activated first, followed by the magnetorheological dampers on the rear wheels, to achieve better comfort, rather than activating them simultaneously; similarly, when a vehicle goes over potholes, the magnetorheological dampers on the wheels that have passed over the potholes are activated to achieve better comfort, rather than activating them all.

[0141] When using deep convolutional network algorithms to perform feature recognition on road condition information, this control method mainly includes the following steps: 1. Pre-select an image set containing the target to be detected for annotation and training to generate corresponding weight files; 2. Read the video information transmitted from the camera and perform image preprocessing, including adjusting the frame rate, resolution, sampling rate, etc.

[0142] 3. Based on the weight file, the video is further processed using a deep convolutional network model to achieve road feature discrimination and geometric location annotation.

[0143] Example 3:

[0144] For the second type of short-term road condition classification, a vehicle magnetorheological damper control method and system diagram integrating visual detection are provided, as shown in the figure below. Figure 8 As shown, when a vehicle 3 equipped with four magnetorheological dampers is traveling on a road surface with lane markings 1, a pothole 2 appears ahead. The vehicle uses road condition visual detection technology to identify the type, lead distance, and span of the pothole, and calculates in advance the wheel that will pass over it, obtaining the lead time for that wheel to pass over the pothole. This lead time is then subtracted from the response time of the magnetorheological system to obtain the delayed start time. When the delayed start time has elapsed, the controller sends a control signal to the controllable power supply. The controllable power supply adjusts the damping force of the corresponding magnetorheological damper 5. When the wheel passes over the pothole, the magnetorheological damper provides better vibration reduction, thus achieving a better riding experience.

[0145] Example 4:

[0146] A deep convolutional network in a vehicle magnetorheological damper control method that integrates road condition visual detection aims to identify and detect road condition types. It adopts a one-stage target detection algorithm based on bounding box regression, such as SSD and YOLO. These algorithms no longer require a region proposal stage and directly generate the target's category probability and location coordinates from the network. The final detection result can be obtained directly after a single detection, which has good detection speed and accuracy.

[0147] Example 5:

[0148] The vehicle magnetorheological damper control system integrating visual detection includes a controllable power supply, several magnetorheological dampers, attitude sensors, vehicle speed sensors, vibration sensors, cameras, controllers, and gimbals.

[0149] The controllable power supply transmits current to each magnetorheological damper, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

[0150] The magnetorheological damper is placed between the wheel and the frame;

[0151] The attitude sensor acquires the vehicle's tilt angle and the centripetal acceleration when the vehicle turns, and transmits them to the controller.

[0152] The vehicle speed sensor acquires the vehicle's speed and transmits it to the controller;

[0153] The vibration sensor acquires vibration parameters transmitted from the ground to the vehicle body and transmits them to the controller;

[0154] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0155] The gimbal is mounted on the vehicle body and is used to stabilize the camera;

[0156] The controller is equipped with an image processing network framework;

[0157] The image processing framework processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the following: long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0158] After receiving data from the attitude sensor, vehicle speed sensor, and vibration sensor, the controller processes the data to obtain human driving road condition parameters.

[0159] The controller stores a set of current intensity test values ​​for vehicle model and magnetorheological damper model under different long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters.

[0160] The controller searches the current intensity test value set, obtains the current intensity values ​​under the current long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters, and generates corresponding control signals, which are then transmitted to the controllable power supply.

[0161] Example 6:

[0162] The main contents of the vehicle magnetorheological damper control system integrating visual detection are shown in Example 5. The long-term road condition parameters are the road surface types on which the vehicle will travel at the future time t1, including but not limited to high-grade asphalt road, general asphalt road, cement road, gravel road, mud road, grassland, snow, and ice.

[0163] The short-time-history road condition parameters include short-time-history road condition classifications and corresponding depth information;

[0164] The short-term road condition classification is based on the type of road condition target that the vehicle needs to cross in the future time t2 and the distance between the road condition target and the camera; t2 < t1; the road condition target types include, but are not limited to, potholes, raised road surfaces, manhole covers, speed bumps, and steps;

[0165] The mid-time history road condition parameters are mid-time history road condition types, including but not limited to uphill, downhill, and curves;

[0166] The human-generated driving road condition parameters include, but are not limited to, acceleration, deceleration, and centripetal acceleration corresponding to emergency start, emergency braking, and emergency cornering driving conditions.

[0167] Example 7:

[0168] The main contents of the vehicle magnetorheological damper control system integrating visual detection are shown in Example 5. In this system, after the controller obtains the current intensity value, it also provides the driver with the recommended current intensity level under the current long-term road condition parameters and medium-term road condition parameters.

[0169] At this time, the controllable power supply provides a current range switch;

[0170] The driver can manually select the current intensity level recommended by the controller or choose their own.

[0171] Example 8:

[0172] The vehicle magnetorheological damper control system integrating visual detection is described in Example 5. The steps for obtaining control signals under short-term road condition parameters include:

[0173] a1) The controller calculates the lead distance between the road condition target and each magnetorheological damper; the lead distance is equal to the sum of the distance between the road condition target and the camera and the distance from the camera to each magnetorheological damper;

[0174] a2) By dividing the lead distance by the current vehicle speed, the lead time from the road condition target to each magnetorheological damper at the current vehicle speed can be calculated.

[0175] a3) The delay start-up time of each magnetorheological damper is calculated by subtracting the response time of the magnetorheological system from the lead time.

[0176] a4) Obtain the span dimension of the road condition target in the lateral direction of the road surface and the vehicle's driving deviation angle, and calculate the wheels that will cross the road condition target;

[0177] a5) By dividing the forward span dimension of the road condition target by the current vehicle speed, the holding time of each magnetorheological damper after delayed start is calculated.

[0178] a6) The controller searches for the current intensity test value set, obtains the current intensity value under the current short-term road condition parameters, and generates a control signal based on the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheel that will cross the road condition target.

[0179] The controller transmits control signals to a controllable power source, causing the magnetorheological damper corresponding to the wheel that will cross the road condition target to respond.

[0180] Example 9:

[0181] The main contents of the vehicle magnetorheological damper control system integrating visual detection are shown in Example 5. In this system, when long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated in an overlapping manner, the corresponding control signals are generated in an overlapping manner, thereby enabling the controllable power supply to execute different control signals in an overlapping manner.

[0182] Example 10:

[0183] The main contents of the vehicle magnetorheological damper control system integrating visual detection are shown in Example 5. When long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated simultaneously, corresponding control signals are generated according to the priority order of the road condition parameters and transmitted to the controllable power supply. The priority order of the road condition parameters is: human-driven road condition parameters, short-term road condition parameters, medium-term road condition parameters, and long-term road condition parameters.

[0184] Example 11:

[0185] The main contents of the vehicle magnetorheological damper control system integrating visual detection are shown in Example 5, wherein each magnetorheological damper has a corresponding vibration sensor.

[0186] Example 12:

[0187] The vehicle magnetorheological damper control system integrating visual detection is described in Example 5. The methods for processing road condition images to obtain the distance between road condition targets and the camera include, but are not limited to, the following two:

[0188] b1) Use a binocular camera to acquire two-dimensional and three-dimensional road condition images, and perform image processing on the two-dimensional and three-dimensional road condition images to obtain the distance between the road condition target and the camera.

[0189] b2) Use fused depth information from a depth camera or ranging radar to obtain the distance between road targets and the camera.

[0190] Example 13:

[0191] The vehicle magnetorheological damper control system integrating visual detection is described in Example 5. The image processing framework processes the road condition image in front of the vehicle and the distance between the road condition target and the camera, including the following steps:

[0192] c1) Perform image preprocessing on the road conditions in front of the vehicle, including adjusting the frame rate, resolution, and sampling rate;

[0193] c2) Based on the weight file, the video is further processed to achieve road surface feature discrimination and geometric position annotation; the methods for further processing the video include, but are not limited to, deep convolutional network models.

[0194] Example 14:

[0195] The method for a vehicle magnetorheological damper control system using the aforementioned fused visual detection includes the following steps:

[0196] 1) The tilt angle of the vehicle body and the centripetal acceleration of the vehicle when turning are obtained by using attitude sensors and transmitted to the controller;

[0197] The vehicle speed is obtained using a vehicle speed sensor and transmitted to the controller;

[0198] Vibration parameters transmitted from the ground to the vehicle body are obtained using vibration sensors and then transmitted to the controller.

[0199] The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera.

[0200] 2) The image processing network framework of the controller processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters.

[0201] The controller processes data from the attitude sensor, vehicle speed sensor, and vibration sensor to obtain road condition parameters for human driving.

[0202] 3) The controller determines the type of generated road condition parameters. If long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are not generated simultaneously, then proceed to step 4); otherwise proceed to step 5.

[0203] 4) In the order of generating long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human driving road condition parameters, control signals are generated in sequence and transmitted to the controllable power supply in sequence, then jump to step 6).

[0204] 5) Generate corresponding control signals according to the priority order of road condition parameters and transmit them to the controllable power supply, then proceed to step 6); Specific steps include:

[0205] 5.1) The controller determines the current intensity value, the holding time after the magnetorheological damper is started, and the wheel to be controlled based on the current human driving road condition parameters, and generates a control signal, which is transmitted to the controllable power supply.

[0206] 5.2) The controller determines the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheels that will cross the target of the road condition based on the current short-term road condition parameters, and generates a control signal and transmits it to the controllable power supply.

[0207] 5.3) The controller determines the current intensity value based on the current mid-time road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0208] 5.4) The controller determines the current intensity value based on the current long-term road condition parameters and generates a control signal, which is then transmitted to the controllable power supply;

[0209] The order in which the corresponding control signals are generated is determined by the priority of the road condition parameters. The priority is: human-driven road conditions > short-term road conditions > medium-term road conditions > long-term road conditions. If two, three, or four road condition parameters are generated at the same time, then two, three, or four of the above steps are executed in order of priority of the road condition parameters.

[0210] For example, when generating both human-driven road condition parameters and long-term road condition parameters simultaneously, first execute step 5.1), then step 5.4), and finally jump to step 6).

[0211] 6) The controllable power supply transmits current to each magnetorheological damper according to the order in which the control signals are received, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

Claims

1. A vehicle magnetorheological damper control system integrating visual inspection, characterized in that, It includes a controllable power supply, several magnetorheological dampers, attitude sensors, vehicle speed sensors, vibration sensors, cameras, controllers, and a gimbal. The controllable power supply transmits current to each magnetorheological damper, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel. The magnetorheological damper is placed between the wheel and the frame; The attitude sensor acquires the vehicle's tilt angle and the centripetal acceleration when the vehicle turns, and transmits them to the controller. The vehicle speed sensor acquires the vehicle's speed and transmits it to the controller; The vibration sensor acquires vibration parameters transmitted from the ground to the vehicle body and transmits them to the controller; The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera. The gimbal is mounted on the vehicle body and is used to stabilize the camera; The controller is equipped with an image processing network framework; The image processing network framework processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the following: long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters. After receiving data from the attitude sensor, vehicle speed sensor, and vibration sensor, the controller processes the data to obtain human driving road condition parameters. The controller stores a set of current intensity test values ​​for vehicle model and magnetorheological damper model under different long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters. The controller searches the current intensity test value set, obtains the current intensity values ​​under the current long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters, and generates the corresponding control signal, which is then transmitted to the controllable power supply. The steps to obtain control signals under short-term road condition parameters include: S1) The controller calculates the lead distance between the road condition target and each magnetorheological damper; the lead distance is equal to the sum of the distance between the road condition target and the camera and the distance from the camera to each magnetorheological damper; S2) By dividing the lead distance by the current vehicle speed, the lead time from the road condition target to each magnetorheological damper at the current vehicle speed is calculated. S3) The delay start-up time of each magnetorheological damper is calculated by subtracting the response time of the magnetorheological system from the lead time. S4) Obtain the span dimension of the road condition target in the lateral direction of the road surface and the vehicle's driving deviation angle, and calculate the wheels that will cross the road condition target; S5) By dividing the forward span dimension of the road condition target by the current vehicle speed, the holding time of each magnetorheological damper after delayed start is calculated. S6) The controller searches for the current intensity test value set, obtains the current intensity value under the current short-term road condition parameters, and generates a control signal based on the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheel that will cross the road condition target. The controller transmits control signals to a controllable power source, causing the magnetorheological damper corresponding to the wheel that will cross the road condition target to respond. When long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated in an overlapping manner, the corresponding control signals are generated in an overlapping manner, thereby causing the controllable power supply to execute different control signals in an overlapping manner. When long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are generated simultaneously, corresponding control signals are generated according to the priority order of the road condition parameters and transmitted to the controllable power supply. The priority order of the road condition parameters is: human-driven road condition parameters, short-term road condition parameters, medium-term road condition parameters, and long-term road condition parameters.

2. The vehicle magnetorheological damper control system integrating visual detection according to claim 1, characterized in that, The long-term road condition parameters refer to the road surface type on which the vehicle will travel at the future time t1, including high-grade asphalt road, general asphalt road, cement road, gravel road, muddy road, grassland, snow, and ice. The short-time-history road condition parameters include short-time-history road condition classifications and corresponding depth information; The short-term road condition classification is based on the type of road condition target that the vehicle needs to cross in the future time t2 and the distance between the road condition target and the camera; t2 < t1; the road condition target types include potholes, raised road surfaces, manhole covers, speed bumps, and steps; The mid-time history road condition parameters are mid-time history road condition types, including uphill, downhill, and curves; The human-driven road condition parameters include acceleration, deceleration, and centripetal acceleration corresponding to emergency start, emergency braking, and emergency cornering driving conditions.

3. The vehicle magnetorheological damper control system integrating visual detection according to claim 1, characterized in that, After the controller obtains the current intensity value, it also provides the driver with the recommended current intensity level under the current long-term road condition parameters and medium-term road condition parameters. At this time, the controllable power supply provides a current range switch; The driver can manually select the current intensity level recommended by the controller or choose their own.

4. The vehicle magnetorheological damper control system integrating visual detection according to claim 1, characterized in that, Each magnetorheological damper has a corresponding vibration sensor.

5. The vehicle magnetorheological damper control system integrating visual detection according to claim 1, characterized in that, Methods for processing road condition images to obtain the distance between road objects and the camera include: A binocular camera is used to acquire two-dimensional and three-dimensional road condition images, and the images are processed to obtain the distance between the road condition target and the camera. The distance between road targets and the camera is obtained by using fused depth information from a depth camera or ranging radar.

6. The vehicle magnetorheological damper control system integrating visual detection according to claim 1, characterized in that, The image processing network framework processes the road conditions in front of the vehicle and the distance between road targets and the camera in the following steps: S1) Perform image preprocessing on the road conditions in front of the vehicle, including adjusting the frame rate, resolution, and sampling rate; S2) Based on the weight file, the video is further processed to achieve road surface feature discrimination and geometric position annotation; the methods for further processing the video include deep convolutional network models.

7. A method for a vehicle magnetorheological damper control system using the fusion vision detection as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step 1) Use the attitude sensor to obtain the vehicle's tilt angle and the centripetal acceleration when the vehicle is turning, and transmit it to the controller; The vehicle speed is obtained using a vehicle speed sensor and transmitted to the controller; Vibration parameters transmitted from the ground to the vehicle body are obtained using vibration sensors and then transmitted to the controller. The camera captures road conditions in front of the vehicle, and the controller processes the road conditions to obtain the road condition type and the distance between the road condition target and the camera. Step 2) The image processing network framework of the controller processes the road condition image in front of the vehicle and the distance between the road condition target and the camera to obtain one or more of the long-term road condition parameters, short-term road condition parameters, and medium-term road condition parameters. The controller processes data from the attitude sensor, vehicle speed sensor, and vibration sensor to obtain road condition parameters for human driving. Step 3) The controller determines the type of generated road condition parameters. If long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters are not generated simultaneously, proceed to step 4); otherwise, proceed to step 5. Step 4) Generate control signals in sequence according to the order of generating long-term road condition parameters, short-term road condition parameters, medium-term road condition parameters, and human-driven road condition parameters, and transmit them to the controllable power supply in sequence, then jump to step 6). Step 5) Generate corresponding control signals according to the priority order of road condition parameters and transmit them to the controllable power supply, then proceed to Step 6); Specific steps include: Step 5.1) The controller determines the current intensity value, the holding time after the magnetorheological damper is activated, and the wheel to be controlled based on the current human driving road condition parameters, and generates a control signal, which is then transmitted to the controllable power supply. Step 5.2) The controller determines the current intensity value, the lead time of the magnetorheological damper, the holding time of the magnetorheological damper after the delayed start, and the wheels that will cross the target of the road condition based on the current short-term road condition parameters, and generates a control signal and transmits it to the controllable power supply. Step 5.3) The controller determines the current intensity value based on the current mid-time road condition parameters and generates a control signal, which is then transmitted to the controllable power supply. Step 5.4) The controller determines the current intensity value based on the current long-term road condition parameters and generates a control signal, which is then transmitted to the controllable power supply. Step 6) The controllable power supply transmits current to each magnetorheological damper according to the order in which the control signals are received, thereby controlling each magnetorheological damper to respond and output damping force to the corresponding wheel.

Images