Suspension control device

Abstract

According to the present invention, a controller (suspension control device) comprises: a first sprung speed estimation unit that receives a signal from an IMU (sensor) provided in a vehicle body and estimates a sprung speed of the vehicle body on the basis of a learning result obtained through machine learning in advance; a second sprung speed estimation unit that receives a signal from the IMU and an FB current as an electric signal generated on the basis of unsprung vibration of a vehicle and estimates a sprung speed of the vehicle body on the basis of a learning result obtained through machine learning in advance; and an estimated value abnormality diagnosis unit (abnormality diagnosis unit) that compares the sprung speed output from the first sprung speed estimation unit with the sprung speed output from the second sprung speed estimation unit and determines whether or not the sprung speed is normal.

Description

Suspension control device

[0001] The present disclosure relates to a suspension control device that controls a suspension apparatus that suppresses vibration between a vehicle body and wheels.

[0002] Suspension control devices that estimate vehicle behavior using artificial intelligence (AI) such as neural networks are known. Abnormality determination for estimated values by AI can only be made based on, for example, whether the estimated value has reached an upper limit value or a lower limit value, or whether the estimated value is fixed at a constant value (value sticking). When normal values cannot be estimated using AI, there is a possibility that the performance of suspension control will deteriorate.

[0003] In contrast, Patent Document 1 discloses a configuration for detecting an abnormality in a suspension system that adjusts damping force using a trained model. In the system disclosed in Patent Document 1, abnormality determination is performed by comparing the sprung acceleration estimated from the stroke displacement and speed estimated by the trained model with the actual sprung acceleration acquired by a sensor.

[0004] Japanese Unexamined Patent Application Publication No. 2024-136422

[0005] Meanwhile, in the system disclosed in Patent Document 1, abnormality determination is performed by comparing the sprung acceleration estimated using the trained model with the actual sprung acceleration acquired by a sensor. This necessitates a sensor for acquiring the actual sprung acceleration, which poses the problem of increased manufacturing costs.

[0006] An object of one embodiment of the present invention is to provide a suspension control device that can perform abnormality determination of estimation information by AI with a simple configuration.

[0007] One embodiment of the present invention is a suspension control device for controlling a suspension device provided between a vehicle body and wheels in a vehicle, which suppresses vibrations between the vehicle body and the wheels, and comprises: a first sprung speed estimation unit that receives signals from a sensor provided on the vehicle body and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; a second sprung speed estimation unit that receives signals from the sensor and an electrical signal generated based on the unsprung vibration of the vehicle and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; and an abnormality diagnosis unit that compares the sprung speed output from the first sprung speed estimation unit and the sprung speed output from the second sprung speed estimation unit and determines whether it is normal or not.

[0008] One embodiment of the present invention is a suspension control device for controlling a suspension device provided between the vehicle body and the wheels in a vehicle, which suppresses vibrations between the vehicle body and the wheels, comprising: a sprung speed estimation unit that receives signals from a sensor provided on the vehicle body and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; a first rear wheel sprung speed estimation unit that estimates the sprung speed of the rear wheel in the direction of travel of the vehicle from the sprung speed; and a unit that receives signals from the sensor and electrical signals generated based on the unsprung vibration of the vehicle and performs prior machine learning The system includes: a relative speed estimation unit that estimates the relative speed between the vehicle body and the wheels based on the obtained learning results; a front wheel unsprung speed estimation unit that estimates the front wheel unsprung speed, which is the unsprung speed of the front wheels in the direction of travel of the vehicle, from the sprung speed and the relative speed; a second rear wheel sprung speed estimation unit that estimates the second rear wheel sprung speed, which is the sprung speed of the rear wheels in the direction of travel of the vehicle, from the front wheel unsprung speed, the relative speed and the vehicle speed of the vehicle; and an abnormality diagnosis unit that compares the first rear wheel sprung speed output from the first rear wheel sprung speed estimation unit with the second rear wheel sprung speed output from the second rear wheel sprung speed estimation unit and determines whether it is normal or not.

[0009] According to one embodiment of the present invention, anomaly detection of estimated information by AI can be performed with a simple configuration.

[0010] This is an explanatory diagram showing a vehicle to which a controller according to an embodiment of the present invention is applied. This is a block diagram of the vehicle state quantity calculation unit in Figure 1. This is a block diagram of the control command calculation unit in Figure 1. This is an explanatory diagram showing an example of a neural network for calculating sprung speed DNN. This is a flowchart showing the control process of the suspension device by the controller of the first embodiment. This is a flowchart showing the control process of the suspension device by the controller of the second embodiment. This is a flowchart showing the control process of the suspension device by the controller of the third embodiment.

[0011] The following will describe in detail, with reference to the attached drawings, an example of how the suspension control device according to an embodiment of the present invention can be applied to a four-wheeled vehicle.

[0012] Figure 1 shows a vehicle 1 to which a controller 21 as a suspension control device is applied. Vehicle 1 includes, for example, a suspension device 5, an IMU 11, and a controller 21 (ECU). In Figure 1, the lower side of the vehicle body 2 that constitutes the body of vehicle 1 is provided with, for example, left and right front wheels and left and right rear wheels (hereinafter collectively referred to as wheels 3). These wheels 3 are composed of tires 4, and the tires 4 act as springs that absorb fine irregularities in the road surface.

[0013] The suspension device 5 is a shock absorber provided by the vehicle 1. The suspension device 5 is installed interposed between the vehicle body 2 and the wheels 3. This suspension device 5 consists of a suspension spring 6 (hereinafter referred to as the spring 6) and a damping force adjustable shock absorber (hereinafter referred to as the variable damper 7) installed in parallel with the spring 6 between the vehicle body 2 and the wheels 3.

[0014] Note that Figure 1 shows a case where one set of suspension devices 5 is installed between the vehicle body 2 and the wheels 3. However, the suspension devices 5 can be installed in a total of four sets, for example, individually and independently between the four wheels 3 and the vehicle body 2, and only one of these sets is schematically shown in Figure 1.

[0015] Here, the variable damper 7 of the suspension system 5 is configured using a damping force adjustable hydraulic shock absorber interposed between the vehicle body 2 and the wheel 3. The variable damper 7 is the force generating mechanism of the suspension system 5. The variable damper 7 is provided between the vehicle body 2 and the wheel 3 and constitutes a relative displacement suppression device that changes the force that suppresses the relative displacement between the vehicle body 2 and the wheel 3.

[0016] The variable damper 7 is equipped with a damping force variable actuator 8, which consists of a damping force adjustment valve or the like, to continuously adjust the damping force characteristics (i.e., damping force characteristics) from hard characteristics to soft characteristics. The damping force variable actuator 8 does not necessarily have to be configured to continuously adjust the damping force characteristics; for example, it may be capable of adjusting the damping force in two or more stages. The variable damper 7 may also be a pressure-controlled type or a flow-control type. The variable damper 7 may also be a type that controls viscosity, such as a magnetorheological fluid or an electroviscous fluid.

[0017] The inertial measurement device 11 (hereinafter referred to as IMU 11) is a vehicle motion detection means that detects the momentum of the vehicle 1 in six axes. The IMU 11 is installed on the vehicle body 2, which is the so-called upper spring side. The IMU 11 includes, for example, three angular velocity sensors and three acceleration sensors. The IMU 11 is a multi-purpose sensor of the vehicle 1. The detection signals from the IMU 11 are used, for example, for traction control, brake control, anti-skid control, and automatic driving control of the vehicle 1.

[0018] The IMU 11 is mounted at any sensor position on the vehicle body 2. The IMU 11 detects the vertical sprung mass acceleration, the roll rate (angular velocity in the roll direction), and the pitch rate (angular velocity in the pitch direction) at the sensor position, and outputs the detected signals to the controller 21 described later. By knowing the mounting position (sensor position) of the IMU 11 in advance, the controller 21 can understand the vehicle motion. That is, if the vehicle body 2, which is the sprung mass, is considered a rigid body, the absolute vertical sprung mass velocity (sprung mass velocity) of each wheel can be geometrically calculated based on the sprung mass acceleration, roll rate, and pitch rate detected at any one point on the vehicle body 2. The sensor position of the IMU 11 may be, for example, the center of gravity of the vehicle body 2, or it may be a position other than the center of gravity.

[0019] The CAN 12 (Controller Area Network) is connected to the controller 21. The CAN 12 is also connected to various multi-purpose sensors (not shown) of the vehicle, such as a vehicle speed sensor, wheel speed sensor, and steering angle sensor. The CAN 12 transmits various vehicle information, including the vehicle speed, which is the speed of the vehicle 1. As a result, the controller 21 can obtain information such as the vehicle speed through the CAN 12.

[0020] In this embodiment, CAN is used as an example of an in-vehicle network, but other in-vehicle networks may be used. Examples of in-vehicle networks include CAN FD (CAN with Flexible Data rate), FlexRay, and in-vehicle Ethernet.

[0021] The controller 21 is a suspension control device that controls a suspension device 5 installed between the vehicle body 2 and the wheels 3 in the vehicle 1. The controller 21 controls the suspension device 5 that suppresses vibrations between the vehicle body 2 and the wheels 3. The controller 21 has a first sprung speed estimation unit 28, a second sprung speed estimation unit 29, and an estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit). In addition, the controller 21 has a control command calculation unit 32 (control unit) that, when it receives a signal from the estimated value abnormality diagnosis unit 31 indicating that the rear wheel sprung speed is normal, controls the suspension device 5 to adjust the damping force according to the vibrations of the vehicle 1.

[0022] The controller 21 is configured as a control device for controlling the damping characteristics of the variable damper 7, and is, for example, a microcomputer. The controller 21 has a storage unit 22 consisting of ROM, RAM, non-volatile memory, etc. The storage unit 22 of the controller 21 stores various programs, information (sensor position, vehicle specifications), data, etc. for controlling the variable damper 7. The input side of the controller 21 is connected to the IMU 11 and also to the CAN 12, which is, for example, a network necessary for data communication. The output side of the controller 21 is connected to the damping force variable actuator 8 of the variable damper 7.

[0023] The controller 21 calculates (estimates) the sprung speed and relative speed as vehicle state variables based on data acquired from the IMU 11. At this time, the relative speed is the relative speed (stroke speed) between the sprung and unsprung mass, and is the piston speed of the variable damper 7. Based on the estimated vehicle state variables, the controller 21 determines the force to be generated by the variable damper 7 (force generation mechanism) of the suspension device 5 and outputs a control signal (command current) to the damping force variable actuator 8 of the suspension device 5.

[0024] As shown in Figures 1 and 2, the controller 21 includes a damper control unit 23, a weight parameter storage unit 24, and a data reading unit 25. The damper control unit 23 includes a receiving unit 26, a vehicle state quantity calculation unit 27, and a control command calculation unit 32. The weight parameter storage unit 24 may be part of the storage unit 22, or it may be a separate unit from the storage unit 22.

[0025] The receiving unit 26 acquires data from the IMU 11 regarding the sprung mass acceleration (floor vertical acceleration), roll rate, and pitch rate at the sensor position. The receiving unit 26 also acquires data related to the vehicle's behavior (hereinafter referred to as behavior information) through the CAN 12. This behavior information includes, for example, longitudinal acceleration (longitudinal G), lateral acceleration (lateral G), steering angle, yaw rate, vehicle speed, and wheel speed of each wheel. The controller 21 does not need to acquire data directly from the IMU 11; it may acquire it through the CAN 12.

[0026] The vehicle state quantity calculation unit 27 calculates (estimates) the sprung speed and relative speed as vehicle state quantities based on data acquired from, for example, the IMU 11 and CAN 12. The vehicle state quantity calculation unit 27 includes a first sprung speed estimation unit 28, a second sprung speed estimation unit 29, and an estimated value anomaly diagnosis unit 31. Based on data acquired from the IMU 11 and CAN 12, the vehicle state quantity calculation unit 27 estimates the data related to suspension control required by the control command calculation unit 32, specifically, instantaneous values ​​related to suspension control (sprung speed, relative speed).

[0027] The first sprung mass velocity estimation unit 28 receives signals from sensors (such as the IMU 11) on the vehicle body 2 and estimates the sprung mass velocity of the vehicle body 2 based on learning results obtained in advance through machine learning. Based on the signals from the IMU 11, the first sprung mass velocity estimation unit 28 acquires the sprung mass acceleration (floor vertical acceleration), roll rate, and pitch rate at the sensor positions. Based on these sprung mass acceleration, roll rate, and pitch rate at the sensor positions, the first sprung mass velocity estimation unit 28 estimates the first rear wheel sprung mass velocity. The first sprung mass velocity estimation unit 28 includes a sprung mass velocity calculation DNN 28A and a first rear wheel sprung mass velocity calculation unit 28B.

[0028] The DNN28A for calculating sprung speed is, for example, artificial intelligence (AI). The DNN28A for calculating sprung speed is a sprung speed estimation unit that receives signals from the IMU11, which is a sensor of the vehicle body 2, and estimates the sprung speed of the vehicle body 2 based on learning results obtained by machine learning in advance. Specifically, the DNN28A for calculating sprung speed estimates the sprung speed for all four wheels as vehicle behavior based on sprung speed (floor vertical acceleration), roll rate, and pitch rate. In other words, the DNN28A for calculating sprung speed estimates the vertical speed of the vehicle body 2 at the mounting position of the suspension device 5.

[0029] As an example, the specific configuration of the DNN28A for calculating sprung velocity will be explained with reference to Figure 4. As shown in Figure 4, the DNN28A for calculating sprung velocity is composed of, for example, a deep neural network. The neural network is composed of a three-layer hierarchical neural network in which elements of an input layer (number of elements i) 101, a hidden layer (number of elements j) 102, and an output layer (number of elements k) 103 are hierarchically connected. Each element of the input layer 101 and each element of the hidden layer 102 are connected by weights W1ij (i=1 to I, j=1 to J), and each element of the hidden layer 102 and each element of the output layer 103 are connected by weights W2jk (j=1 to J, k=1 to K). This weight information (hereinafter referred to as weight parameters) is represented by a matrix of weights W1ij and weights W2jk. The weight parameters are determined in advance by machine learning and are stored in the weight parameter storage unit 24. In this example, we have shown a fully connected neural network with a single hidden layer 102, but we are not limited to this. For example, a neural network with two or more hidden layers 102 is also possible.

[0030] The neural network's machine learning is performed based on data of the vehicle's sprung velocity, acquired in advance by a data acquisition vehicle, and behavioral information data including sprung acceleration (floor vertical acceleration), roll rate, and pitch rate. The data acquisition vehicle has the same specifications as the vehicle on which the DNN28A for calculating sprung velocity is installed, and is equipped with various sensors (accelerometers, etc.) for acquiring the vehicle's state variables. In the neural network's machine learning, the weight parameters are adjusted to learn the correlation between the vehicle's state variable (sprung velocity) data and the behavioral information data acquired by the data acquisition vehicle. The weight parameters obtained as a result of the learning are stored in the weight parameter storage unit 24.

[0031] In this embodiment, the DNN28A for calculating sprung speed uses machine learning to correlate vehicle state data and behavioral information data acquired by a data acquisition vehicle, but the present invention is not limited to this. For example, a vehicle model can be constructed according to the vehicle on which the DNN28A for calculating sprung speed is installed, and the DNN28A for calculating sprung speed may perform machine learning based on data acquired by simulation using this vehicle model.

[0032] The input layer 101 of the neural network receives time-series data such as sprung mass acceleration (floor vertical acceleration), roll rate, and pitch rate. The output layer 103 outputs instantaneous values ​​of the sprung mass velocity of the suspension device 5, assumed to be mounted on each wheel 3 of the vehicle. The number of elements in the hidden layer 102 is generally determined from the number of elements in the input layer 101 and the output layer 103, but it is set to the number that maximizes the accuracy of the vehicle behavior estimation by the neural network. The number of elements in the output layer 103 is determined by the output specifications of the vehicle behavior estimation.

[0033] The first rear wheel sprung speed calculation unit 28B is a first rear wheel sprung speed estimation unit that estimates the sprung speed of the rear wheels in the direction of vehicle travel from the sprung speed. The first rear wheel sprung speed calculation unit 28B receives the sprung speeds for all four wheels output from the sprung speed calculation DNN 28A. The first rear wheel sprung speed calculation unit 28B selects the sprung speeds for two rear wheels from the sprung speeds for all four wheels output from the sprung speed calculation DNN 28A. The first rear wheel sprung speed calculation unit 28B outputs the selected sprung speeds for the two rear wheels as the first rear wheel sprung speed.

[0034] The second sprung speed estimation unit 29 receives signals from sensors (such as the IMU 11) and electrical signals (e.g., FB current) generated based on the unsprung vibration of the vehicle 1, and estimates the sprung speed of the vehicle body 2 based on learning results obtained in advance through machine learning. Specifically, the second sprung speed estimation unit 29 acquires the sprung acceleration (floor vertical acceleration), roll rate, and pitch rate at the sensor position based on the signals received from the IMU 11. In addition, the second sprung speed estimation unit 29 acquires command current values ​​for each variable damper 7 (variable damping force actuator 8) based on the FB current signal received from the control command calculation unit 32. At this time, the FB current signal is the command current value output from the control command calculation unit 32, and is a value corresponding to the damping force generated by the variable damper 7. The second sprung mass velocity estimation unit 29 estimates the second rear wheel sprung mass velocity based on the sprung mass acceleration, roll rate, pitch rate at these sensor positions, and the FB current for each variable damper 7. The second sprung mass velocity estimation unit 29 includes a relative speed calculation DNN 29A, a front wheel sprung mass velocity calculation unit 29B, a front wheel relative speed calculation unit 29C, a rear wheel relative speed calculation unit 29D, a front wheel unsprung mass velocity calculation unit 29E, and a second rear wheel sprung mass velocity calculation unit 29F.

[0035] The relative speed calculation DNN29A is a relative speed estimation unit that estimates the relative speed between the vehicle body 2 and the wheels 3. At this time, the relative speed calculation DNN29A receives signals from sensors (IMU11, etc.) and electrical signals (FB current, etc.) generated based on the unsprung vibration of the vehicle 1, and estimates the relative speed between the vehicle body 2 and the wheels 3 based on the learning results obtained in advance by machine learning. Specifically, the relative speed calculation DNN29A estimates the relative speed for the four wheels as vehicle behavior based on the sprung acceleration (floor vertical acceleration), roll rate, pitch rate at the sensor position and the FB current for all four wheels. The relative speed calculation DNN29A uses a mathematical model of the vehicle 1, which takes values ​​indicating sprung acceleration (floor vertical acceleration), roll rate, pitch rate, and FB current as inputs, to determine the relative speed (stroke speed) of the suspension device 5 for each wheel. At this time, the relative speed calculation DNN29A is, for example, an artificial intelligence model constructed by machine learning. In other words, the mathematical model of vehicle 1 is an artificial intelligence (AI) model.

[0036] The relative velocity calculation DNN 29A receives the sprung mass acceleration (floor vertical acceleration), roll rate, pitch rate, and FB current for all four wheels as input, and estimates the stroke velocity of each wheel using a neural network that has been trained to estimate the stroke velocity of each wheel. The relative velocity calculation DNN 29A reads the weight parameters stored in the weight parameter storage unit 24 via the data reading unit 25. The relative velocity calculation DNN 29A estimates the stroke velocity based on the sprung mass acceleration (floor vertical acceleration), roll rate, pitch rate, FB current, and weight parameters at the sensor position.

[0037] The neural network of DNN29A for relative velocity calculation is configured similarly to the neural network of DNN28A for sprung speed calculation. The machine learning of the neural network is performed based on data of vehicle state variables (relative velocity) acquired in advance by a data acquisition vehicle, and data of sprung speed acceleration (floor vertical acceleration), roll rate, pitch rate, and FB current at the sensor position. However, this is not limited to this, for example, a vehicle model corresponding to the vehicle on which DNN29A for relative velocity calculation is installed may be constructed, and DNN29A for relative velocity calculation may perform machine learning of the neural network based on data acquired by simulation using this vehicle model.

[0038] The weight parameters of the trained neural network are stored in the weight parameter storage unit 24. Therefore, the neural network of the relative velocity calculation DNN 29A is constructed using the weight parameters stored in the weight parameter storage unit 24.

[0039] The front wheel sprung speed calculation unit 29B selects the sprung speed for the two front wheels from the sprung speeds for all four wheels output from the sprung speed calculation DNN 28A. The front wheel sprung speed calculation unit 29B outputs the selected sprung speed for the two front wheels.

[0040] The front wheel relative speed calculation unit 29C selects the relative speed of the two front wheels from the relative speeds (stroke speeds) of all four wheels output from the relative speed calculation DNN 29A. The front wheel relative speed calculation unit 29C outputs the selected relative speed of the two front wheels as the front wheel relative speed.

[0041] The rear wheel relative speed calculation unit 29D selects the relative speed of the two rear wheels from the relative speeds (stroke speeds) of all four wheels output from the relative speed calculation DNN 29A. The rear wheel relative speed calculation unit 29D outputs the selected relative speed of the two rear wheels as the rear wheel relative speed.

[0042] The front wheel unsprung speed calculation unit 29E is a front wheel unsprung speed estimation unit that estimates the unsprung speed of the front wheels in the direction of travel of the vehicle 1 from the sprung speed and relative speed. The front wheel unsprung speed calculation unit 29E receives the front wheel sprung speed output from the front wheel sprung speed calculation unit 29B and the front wheel relative speed output from the front wheel relative speed calculation unit 29C as inputs. Based on the front wheel sprung speed output from the front wheel sprung speed calculation unit 29B and the front wheel relative speed output from the front wheel relative speed calculation unit 29C, the front wheel unsprung speed calculation unit 29E calculates the unsprung speed for the two front wheels. The front wheel unsprung speed calculation unit 29E outputs the calculated unsprung speed for the two front wheels as the front wheel unsprung speed.

[0043] The second rear wheel sprung speed calculation unit 29F is a second rear wheel sprung speed estimation unit that estimates the sprung speed of the rear wheels in the direction of travel of the vehicle 1 from the front wheel unsprung speed, relative speed, and vehicle speed of the vehicle 1. The second rear wheel sprung speed calculation unit 29F receives the rear wheel relative speed output from the rear wheel relative speed calculation unit 29D, the front wheel unsprung speed output from the front wheel unsprung speed calculation unit 29E, and the delay time calculated by the delay time calculation unit 30 from the vehicle speed. At this time, the delay time is the time it takes for the rear wheels to reach the position of the front wheels, and is calculated, for example, based on the vehicle speed and wheelbase. The second rear wheel sprung speed calculation unit 29F outputs the sprung speed for the two rear wheels as the second rear wheel sprung speed, based on the rear wheel relative speed output from the rear wheel relative speed calculation unit 29D, the front wheel unsprung speed output from the front wheel unsprung speed calculation unit 29E, and the delay time calculated from the vehicle speed by the delay time calculation unit 30. Specifically, the second rear wheel sprung speed calculation unit 29F estimates the rear wheel unsprung speed based on the front wheel unsprung speed output from the front wheel unsprung speed calculation unit 29E and the delay time calculated from the vehicle speed by the delay time calculation unit 30. The second rear wheel sprung speed calculation unit 29F calculates the sprung speed for the two rear wheels based on the rear wheel unsprung speed estimated from the front wheel unsprung speed and delay time, and the rear wheel relative speed output from the rear wheel relative speed calculation unit 29D. The second rear wheel sprung speed calculation unit 29F outputs the calculated unsprung speed for the two rear wheels as the second rear wheel unsprung speed.

[0044] The estimated value abnormality diagnosis unit 31 is an abnormality diagnosis unit that compares the sprung velocities output from the first sprung velocity estimation unit 28 and the second sprung velocity estimation unit 29, and determines whether they are normal or not. The first rear wheel sprung velocity output from the first rear wheel sprung velocity calculation unit 28B and the second rear wheel sprung velocity output from the second rear wheel sprung velocity calculation unit 29F are input to the estimated value abnormality diagnosis unit 31. The estimated value abnormality diagnosis unit 31 compares the first rear wheel sprung velocity estimated by the first sprung velocity estimation unit 28 and the second rear wheel sprung velocity estimated by the second sprung velocity estimation unit 29, and determines whether the estimated values including the first rear wheel sprung velocity and the second rear wheel sprung velocity are normal or not.

[0045] The estimated value abnormality diagnosis unit 31 calculates, for example, the absolute value of the difference (velocity difference) between the two rear wheel sprung velocities as the difference between the first rear wheel sprung velocity and the second rear wheel sprung velocity of the left rear wheel. The estimated value abnormality diagnosis unit 31 determines whether the difference between the two rear wheel sprung velocities is equal to or greater than a predetermined threshold. When the difference between the two rear wheel sprung velocities is smaller than the threshold, the estimated value abnormality diagnosis unit 31 determines that the estimated value by AI is normal, and outputs a signal (normal signal) indicating the determination result of "normal". On the other hand, when the difference between the two rear wheel sprung velocities is equal to or greater than the threshold, the estimated value abnormality diagnosis unit 31 determines that the estimated value by AI is abnormal, and outputs a signal (abnormal signal) indicating the determination result of "abnormal".

[0046] As with the left rear wheel, the estimated value abnormality diagnosis unit 31 also determines, for the right rear wheel, whether the estimated value by AI is normal based on the difference between the two rear wheel sprung velocities. Specifically, the estimated value abnormality diagnosis unit 31 calculates, for example, the absolute value of the difference (velocity difference) between the two rear wheel sprung velocities as the difference between the first rear wheel sprung velocity and the second rear wheel sprung velocity of the right rear wheel. The estimated value abnormality diagnosis unit 31 determines whether the difference between the two rear wheel sprung velocities is equal to or greater than a predetermined threshold. When the difference between the two rear wheel sprung velocities is smaller than the threshold, the estimated value abnormality diagnosis unit 31 determines that the estimated value by AI is normal, and outputs a normal signal indicating the determination result of "normal". On the other hand, when the difference between the two rear wheel sprung velocities is equal to or greater than the threshold, the estimated value abnormality diagnosis unit 31 determines that the estimated value by AI is abnormal, and outputs an abnormal signal indicating the determination result of "abnormal".

[0047] Note that the threshold value is the maximum value at which the estimated value can be determined to be normal based on the difference between the two rear-wheel sprung velocities, and is appropriately set in consideration of detected values, calculated values and the like obtained from actual vehicles, simulations, and the like.

[0048] The control command calculation unit 32 is a control unit that controls the damping force of the suspension device 5. As shown in FIG. 3, the control command calculation unit 32 receives a signal indicating whether the rear-wheel sprung velocity is normal from the estimated value abnormality diagnosis unit 31. In addition to this, the control command calculation unit 32 receives the sprung velocities for four wheels estimated by the first sprung velocity estimation unit 28, the relative velocities for four wheels estimated by the second sprung velocity estimation unit 29, and the vehicle CAN signal including vehicle information transmitted to the CAN 12. When a normal signal (normal signal) is received from the estimated value abnormality diagnosis unit 31, the control command calculation unit 32 controls the suspension device 5 to adjust the damping force according to the vibration of the vehicle 1. Specifically, the control command calculation unit 32 calculates a control command value (command current value) for controlling the damping force of the variable damper 7 of the suspension device 5, based on, for example, the sprung velocities for four wheels estimated by the first sprung velocity estimation unit 28 and the relative velocities for four wheels estimated by the second sprung velocity estimation unit 29. At this time, the control amount for controlling the variable damper 7 is a control command value for controlling the damping force of the variable damper 7. Specifically, the control command calculation unit 32 calculates the command current value as a control command value for improving the riding comfort of the vehicle based on, for example, bilinear optimal control, skyhook control, H∞ control, or the like. The control command calculation unit 32 outputs a command current as a control signal to the damping force variable actuator 8 of the variable damper 7 based on the command current value. At this time, the command current output by the control command calculation unit 32 is input to the second sprung velocity estimation unit 29 (the DNN 29A for relative velocity calculation) as an FB current (feedback current). The control command calculation unit 32 may calculate a control command value for improving steering stability, without being limited to the riding comfort of the vehicle.

[0049] If an abnormal signal (abnormal signal) is received, the control command calculation unit 32 stops controlling the suspension device 5. At this time, the control command calculation unit 32 switches to fail-safe mode. For this reason, the control command calculation unit 32 outputs a current command, for example, so that the variable damper 7 generates a hard damping force. However, the control command calculation unit 32 may also output a current command equivalent to a conventional passive damper (hereinafter referred to as a conventional equivalent current command) when it receives an abnormal signal from the estimated value abnormal diagnosis unit 31.

[0050] Next, with reference to Figure 5, the control process of the suspension device 5 by the controller 21 (damper control unit 23) will be explained. Each step in the flowchart shown in Figure 5 will be denoted as "S" (for example, step 1 will be "S1"). This also applies to Figures 6 and 7.

[0051] In S1, the sprung speed calculation DNN 28A of the first sprung speed estimation unit 28 estimates the sprung speed of each wheel based on the sprung speed acceleration, roll rate, and pitch rate at the sensor position. At this time, the sprung speed calculation DNN 28A reads weight parameters, which have been pre-learned to show the correlation between the sprung speed acceleration, roll rate, pitch rate, and the sprung speed of each wheel, from the weight parameter storage unit 24. The sprung speed calculation DNN 28A estimates the sprung speed for all four wheels based on the read weight parameters and the sprung speed acceleration, roll rate, and pitch rate at the sensor position.

[0052] In S2, the relative velocity calculation DNN 29A of the second sprung mass velocity estimation unit 29 estimates the relative velocity (stroke velocity) of each wheel based on the sprung mass acceleration, roll rate, pitch rate, and FB current at the sensor position. At this time, the relative velocity calculation DNN 29A reads weight parameters, which have been pre-learned to represent the correlation between the sprung mass acceleration, roll rate, pitch rate, FB current, and the relative velocity of each wheel, from the weight parameter storage unit 24. The relative velocity calculation DNN 29A estimates the relative velocity for all four wheels based on the read weight parameters and the sprung mass acceleration, roll rate, pitch rate, and FB current at the sensor position.

[0053] In S3, the first sprung speed estimation unit 28 calculates the rear wheel sprung speed from the sprung speed estimated by the sprung speed calculation DNN 28A (estimated sprung speed). Specifically, the first sprung speed estimation unit 28 uses the first rear wheel sprung speed calculation unit 28B to select the sprung speed for two rear wheels from the sprung speed for all four wheels (estimated sprung speed) estimated by the sprung speed calculation DNN 28A. The first rear wheel sprung speed calculation unit 28B outputs the selected sprung speed for two rear wheels as the first rear wheel sprung speed.

[0054] In S4, the second sprung speed estimation unit 29 calculates the front wheel unsprung speed from the sprung speed estimated by the sprung speed calculation DNN 28A (estimated sprung speed) and the relative speed estimated by the relative speed calculation DNN 29A (estimated relative speed). Specifically, the front wheel sprung speed calculation unit 29B selects the sprung speed for the two front wheels from the sprung speed for all four wheels output from the sprung speed calculation DNN 28A. The front wheel relative speed calculation unit 29C selects the relative speed for the two front wheels from the relative speed for all four wheels (stroke speed) output from the relative speed calculation DNN 29A. The front wheel relative speed calculation unit 29C calculates the selected relative speed for the two front wheels as the front wheel relative speed. The front wheel unsprung speed calculation unit 29E calculates the unsprung speed for the two front wheels based on the front wheel sprung speed output from the front wheel sprung speed calculation unit 29B and the front wheel relative speed output from the front wheel relative speed calculation unit 29C. The second sprung speed estimation unit 29 outputs the calculated unsprung speed for the two front wheels as the front wheel unsprung speed.

[0055] In S5, the second sprung mass velocity estimation unit 29 calculates the rear wheel relative speed from the relative speed estimated by the relative speed calculation DNN 29A (estimated relative speed). Specifically, the second sprung mass velocity estimation unit 29 uses the rear wheel relative speed calculation unit 29D to select the relative speed of two rear wheels from the relative speeds (stroke speeds) of all four wheels output from the relative speed calculation DNN 29A. The rear wheel relative speed calculation unit 29D outputs the selected relative speed of two rear wheels as the rear wheel relative speed.

[0056] In S6, the delay time calculation unit 30 calculates the rear wheel delay time from the vehicle speed. Specifically, the delay time calculation unit 30 calculates the rear wheel delay time relative to the front wheel based on the vehicle speed and wheelbase.

[0057] In S7, the second sprung mass velocity estimation unit 29 calculates the second rear wheel sprung mass velocity based on the rear wheel relative speed output from the rear wheel relative speed calculation unit 29D, the front wheel unsprung mass velocity output from the front wheel unsprung mass velocity calculation unit 29E, and the delay time (rear wheel delay time) calculated from the vehicle speed by the delay time calculation unit 30. Specifically, the second rear wheel sprung mass velocity calculation unit 29F estimates the rear wheel unsprung mass velocity based on the front wheel unsprung mass velocity output from the front wheel unsprung mass velocity calculation unit 29E and the delay time calculated from the vehicle speed by the delay time calculation unit 30. The second rear wheel sprung mass velocity calculation unit 29F calculates the sprung mass velocity for the two rear wheels based on the rear wheel unsprung mass velocity estimated from the front wheel unsprung mass velocity and delay time, and the rear wheel relative speed output from the rear wheel relative speed calculation unit 29D. The second rear wheel sprung speed calculation unit 29F outputs the calculated unsprung speed for the two rear wheels as the second rear wheel unsprung speed.

[0058] In S8, the estimated value anomaly diagnosis unit 31 calculates the difference between the first rear wheel sprung speed output from the first rear wheel sprung speed calculation unit 28B and the second rear wheel sprung speed output from the second rear wheel sprung speed calculation unit 29F. At this time, the difference between the two rear wheel sprung speeds is the absolute value of the speed difference. The estimated value anomaly diagnosis unit 31 determines whether the difference between the two rear wheel sprung speeds is greater than or equal to a predetermined threshold. If the difference between the two rear wheel sprung speeds is less than the threshold, the estimated values ​​of the first sprung speed estimation unit 28 and the second sprung speed estimation unit 29 are considered normal. For this reason, "NO" is determined in S8, and steps S1 onwards are repeated.

[0059] At this time, the estimated value anomaly diagnosis unit 31 outputs a signal (normal signal) indicating that the estimated value is normal. The normal signal from the estimated value anomaly diagnosis unit 31 is input to the control command calculation unit 32. Therefore, the control command calculation unit 32 controls the damping force of the variable damper 7 based on the sprung speed estimated by the first sprung speed estimation unit 28 and the relative speed estimated by the second sprung speed estimation unit 29. Specifically, the control command calculation unit 32 calculates a control command value to adjust the damping force of the variable damper 7 based on the sprung speed estimated by the first sprung speed estimation unit 28 and the relative speed estimated by the second sprung speed estimation unit 29. As a result, the damper control unit 23 outputs a command current based on the control command value to the variable damping force actuator 8 of the variable damper 7, thereby controlling the damping force of the variable damper 7.

[0060] On the other hand, when the difference between the two rear wheel sprung mass velocities is greater than or equal to a threshold, the estimated values ​​of the first sprung mass velocity estimation unit 28 and the second sprung mass velocity estimation unit 29 are considered abnormal. For this reason, "YES" is determined in S8, and the process proceeds to S9. At this time, the estimated value abnormality diagnosis unit 31 outputs a signal (abnormality signal) indicating that the estimated value is abnormal.

[0061] In S9, the control command calculation unit 32 stops controlling the damping force of the variable damper 7 based on the abnormal signal from the estimated value abnormality diagnosis unit 31. At this time, the damper control unit 23 (control command calculation unit 32) outputs a command current to hard fix the damping force of the variable damper 7, and these control processes are terminated.

[0062] Thus, the controller 21 (suspension control device) according to this embodiment includes: a first sprung speed estimation unit 28 that receives a signal from the IMU 11 (sensor) of the vehicle body 2 and estimates the sprung speed of the vehicle body 2 based on learning results obtained in advance by machine learning; a second sprung speed estimation unit 29 that receives the FB current as an electrical signal generated based on the signal from the IMU 11 and the unsprung vibration of the vehicle 1 and estimates the sprung speed of the vehicle body 2 based on learning results obtained in advance by machine learning; and an estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit) that compares the sprung speeds output from the first sprung speed estimation unit 28 and the second sprung speed estimation unit 29 and determines whether or not it is normal.

[0063] Conventional technology can only determine abnormalities in the relative velocity between the sprung mass and unsprung mass by checking whether it is within the upper or lower limits using AI. Therefore, if a normal value cannot be estimated, there is a possibility that performance such as ride comfort will deteriorate.

[0064] In contrast, in this embodiment, abnormalities in the first sprung speed estimation unit 28 and the second sprung speed estimation unit 29 are determined by comparing the sprung speed (rear wheel sprung speed) based on the outputs of two DNNs (DNN 28A for sprung speed calculation and DNN 29A for relative speed calculation). As a result, the controller 21 detects if an abnormality occurs in either or both of the AI ​​estimated values ​​of the first sprung speed estimation unit 28 and the second sprung speed estimation unit 29. This makes it possible to suppress the control of the damping force of the suspension device 5 (variable damper 7) based on abnormal estimated values, thereby ensuring system performance. In addition, in this embodiment, there is no need to provide a separate sensor to determine abnormalities in the AI ​​estimated values, and abnormality determination of estimated information by AI can be performed with a simple configuration.

[0065] Furthermore, as a method for detecting abnormalities, it is conceivable to estimate the sprung speed, etc., based on a Kalman filter or wheel speed sensor, and compare this estimated value with the AI ​​estimated value. However, in this case, in addition to calibrating the AI ​​estimated value, calibration of the estimated values ​​by the Kalman filter or wheel speed sensor is required, which presents the challenge of increased calibration man-hours. In contrast, in this embodiment, calibration of the AI ​​estimated value can be performed using two DNNs (DNN28A for sprung speed calculation and DNN29A for relative speed calculation), thereby suppressing calibration man-hours and increasing productivity.

[0066] In this embodiment, the first sprung speed estimation unit 28 has a first rear wheel sprung speed calculation unit 28B (first rear wheel sprung speed estimation unit) that estimates the first rear wheel sprung speed, which is the sprung speed of the rear wheels in the direction of travel of the vehicle 1, from the sprung speed. The second sprung speed estimation unit 29 has a relative speed calculation DNN 29A (relative speed estimation unit) that estimates the relative speed between the vehicle body 2 and the wheels 3. The front wheel unsprung speed calculation unit 29E (front wheel unsprung speed estimation unit) estimates the front wheel unsprung speed, which is the unsprung speed of the front wheels in the direction of travel of the vehicle 1, from the sprung speed and relative speed. The second rear wheel sprung speed calculation unit 29F (second rear wheel sprung speed estimation unit) estimates the second rear wheel sprung speed, which is the sprung speed of the rear wheels in the direction of travel of the vehicle 1, from the front wheel unsprung speed, relative speed and vehicle speed of the vehicle 1.

[0067] As a result, the first sprung speed estimation unit 28 estimates the sprung speed for all four wheels using the sprung speed calculation DNN 28A, and the first rear wheel sprung speed calculation unit 28B extracts the sprung speed for the two rear wheels as the first rear wheel sprung speed from the sprung speed for all four wheels. This allows the first sprung speed estimation unit 28 to estimate the first rear wheel sprung speed.

[0068] Furthermore, the second sprung speed estimation unit 29 estimates the relative speeds of all four wheels using the relative speed calculation DNN 29A, and estimates the front wheel unsprung speed from the sprung speed and relative speed using the front wheel unsprung speed calculation unit 29E. The second sprung speed estimation unit 29 estimates the rear wheel sprung speed of the vehicle 1 as the second rear wheel sprung speed using the second rear wheel sprung speed calculation unit 29F from the front wheel unsprung speed, relative speed and the vehicle speed of the vehicle 1. In this way, the second sprung speed estimation unit 29 can estimate the second rear wheel sprung speed separately from the first rear wheel sprung speed calculated by the first sprung speed estimation unit 28.

[0069] The controller 21 of this embodiment includes a control command calculation unit 32 (control unit) that receives a signal from the estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit) indicating whether the rear wheel sprung speed is normal or not, and when it receives a normal signal (normal signal), it controls the suspension device 5 to adjust the damping force according to the vibration of the vehicle, and when it receives an abnormal signal (abnormal signal), it stops controlling the suspension device 5.

[0070] Therefore, the controller 21 puts the system into fail-safe mode by stopping the operation of the suspension device 5 in the event of an abnormality. As a result, it is possible to suppress the deterioration of ride comfort beyond that of fail-safe mode.

[0071] The controller 21 (suspension control device) according to this embodiment includes a sprung speed calculation DNN 28A (sprung speed estimation unit) that receives a signal from the IMU 11 (sensor) of the vehicle body 2 and estimates the sprung speed of the vehicle body 2 based on learning results obtained by machine learning in advance, a first rear wheel sprung speed calculation unit 28B (first rear wheel sprung speed estimation unit) that estimates the first rear wheel sprung speed, which is the sprung speed of the rear wheel in the direction of travel of the vehicle 1, from the sprung speed, and the FB current, which is an electrical signal generated based on the signal from the IMU 11 (sensor) and the unsprung vibration of the vehicle 1, and estimates the relative speed between the vehicle body 2 and the wheel 3 based on learning results obtained by machine learning in advance The system includes a relative speed calculation DNN 29A (relative speed estimation unit) that estimates the sprung speed, a front wheel unsprung speed calculation unit 29E (front wheel unsprung speed estimation unit) that estimates the front wheel unsprung speed, which is the unsprung speed of the front wheels in the direction of travel of the vehicle 1, from the sprung speed and relative speed, a second rear wheel sprung speed calculation unit 29F (second rear wheel sprung speed estimation unit) that estimates the second rear wheel sprung speed, which is the sprung speed of the rear wheels in the direction of travel of the vehicle 1, from the front wheel unsprung speed, relative speed and vehicle speed of the vehicle 1, and an estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit) that compares the rear wheel sprung speeds (first rear wheel sprung speed and second rear wheel sprung speed) output from the first rear wheel sprung speed calculation unit 28B and the second rear wheel sprung speed calculation unit 29F to determine whether it is normal or not.

[0072] As a result, the controller 21 determines abnormalities in the rear wheel sprung speed by comparing the output of the two DNNs (DNN 28A for sprung speed calculation and DNN 29A for relative speed calculation). This allows the controller 21 to detect abnormalities in either or both of the AI ​​estimates of the sprung speed calculation DNN 28A and the relative speed calculation DNN 29A. Consequently, the controller 21 can suppress the control of the damping force of the suspension device 5 (variable damper 7) based on abnormal estimates, thereby ensuring system performance.

[0073] Next, Figures 1, 2, 3, and 6 show a second embodiment. The characteristic of the second embodiment is that, when the control unit receives an abnormal signal from the abnormality diagnosis unit, it stops controlling the suspension device for a predetermined time, then receives a signal from the abnormality diagnosis unit indicating whether the rear wheel sprung speed is normal or not, and controls the suspension device according to the received result. In the second embodiment, the same reference numerals are used for the same components as in the first embodiment described above, and their descriptions are omitted.

[0074] As shown in Figures 1 to 3, the controller 41 according to the second embodiment is configured in substantially the same way as the controller 21 according to the first embodiment. The controller 41 includes a damper control unit 42, a weight parameter storage unit 24, and a data reading unit 25. The damper control unit 42 includes a receiving unit 26, a vehicle state quantity calculation unit 27, and a control command calculation unit 43.

[0075] The control command calculation unit 43 is a control unit that controls the damping force of the suspension device 5. The control command calculation unit 43 is configured similarly to the control command calculation unit 32 in the first embodiment. Therefore, the control command calculation unit 43 receives a signal from the estimated value abnormality diagnosis unit 31 indicating whether the rear wheel sprung speed is normal or not. If a signal indicating normality (normal signal) is received, the control command calculation unit 43 controls the suspension device 5 to adjust the damping force according to the vibration of the vehicle 1. Specifically, the control command calculation unit 43 calculates a control command value (command current value) for controlling the damping force of the variable damper 7 of the suspension device 5 based on, for example, the sprung speed for all four wheels estimated by the first sprung speed estimation unit 28 and the relative speed for all four wheels estimated by the second sprung speed estimation unit 29. At this time, the control amount for controlling the variable damper 7 is the control command value for controlling the damping force of the variable damper 7. Based on the command current value, the control command calculation unit 43 outputs a command current as a control signal to the damping force variable actuator 8 of the variable damper 7. Furthermore, the control command calculation unit 43 may calculate control command values ​​not only for improving the ride comfort of the vehicle, but also for improving steering stability.

[0076] On the other hand, if an abnormal signal (abnormal signal) is received, the control command calculation unit 43 stops controlling the suspension device 5 for a predetermined time, then receives a signal from the estimated value abnormal diagnosis unit 31 (abnormal diagnosis unit) indicating whether the rear wheel sprung speed is normal or not, and controls the suspension device 5 according to the received result. While the control of the suspension device 5 is stopped, the control command calculation unit 43 outputs, for example, a current command or a conveyor-equivalent current command that generates a hard damping force in the variable damper 7. If an abnormal signal is received again after the predetermined time has elapsed, the control command calculation unit 43 also outputs, for example, a current command or a conveyor-equivalent current command that generates a hard damping force in the variable damper 7. Conversely, if a normal signal is received after the predetermined time has elapsed, the control command calculation unit 43 controls the suspension device 5 to adjust the damping force of the variable damper 7 based on the sprung speed for all four wheels estimated by the first sprung speed estimation unit 28 and the relative speed for all four wheels estimated by the second sprung speed estimation unit 29. The predetermined time for stopping control of the suspension device 5 is, for example, the time required for the road surface conditions to change and for the AI ​​estimated values ​​to return to normal. This time is set appropriately, taking into consideration actual vehicle detection values, simulation values, etc.

[0077] Next, with reference to Figure 6, the control process of the suspension device 5 by the controller 41 (damper control unit 42) will be described.

[0078] As shown in Figure 6, the control process of the suspension device 5 by the controller 41 in the second embodiment is substantially the same as the control process of the suspension device 5 by the controller 21 in the first embodiment.

[0079] However, if the difference in the speed of the two rear wheel sprung masses is greater than or equal to a threshold, and "YES" is determined in S8, the process proceeds to S11. At this time, the estimated value abnormality diagnosis unit 31 outputs a signal (abnormality signal) indicating that the estimated value is abnormal.

[0080] In S11, the control command calculation unit 43, based on the abnormal signal from the estimated value abnormality diagnosis unit 31, stops controlling the suspension device 5 (variable damper 7) for a predetermined time, then receives a signal from the estimated value abnormality diagnosis unit 31 indicating whether the rear wheel sprung speed is normal or not, and controls the suspension device 5 according to the received result. If an abnormal signal is received again after the predetermined time has elapsed, the control command calculation unit 43 continues to stop controlling the suspension device 5 (variable damper 7). At this time, the control command calculation unit 43 outputs, for example, a current command or a conveyor-equivalent current command that generates a hard damping force for the variable damper 7. Conversely, if a normal signal is received after the predetermined time has elapsed, the control command calculation unit 43 resumes controlling the suspension device 5 (variable damper 7). Specifically, the control command calculation unit 43 controls the suspension device 5 to adjust the damping force of the variable damper 7 based on the sprung speed of the four wheels estimated by the first sprung speed estimation unit 28 and the relative speed of the four wheels estimated by the second sprung speed estimation unit 29.

[0081] Thus, the second embodiment can achieve substantially the same effects as the first embodiment. In the second embodiment, when the control command calculation unit 43 (control unit) receives an abnormal signal from the estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit), it stops controlling the suspension device for a predetermined time, then receives a signal from the estimated value abnormality diagnosis unit 31 indicating whether the rear wheel sprung speed is normal or not, and controls the suspension device 5 according to the reception result. In this way, even after stopping control due to abnormality detection, the control command calculation unit 43 in the second embodiment has the estimated value abnormality diagnosis unit 31 re-diagnose whether it is normal or not at regular intervals. As a result, for example, on a road surface where the AI ​​estimation is not problematic, the estimated value abnormality diagnosis unit 31 determines that it is normal, and the control command calculation unit 43 can resume controlling the suspension device 5. As a result, deterioration of ride comfort due to the suspension device 5 control stoppage can be suppressed.

[0082] Next, Figures 1, 2, 3, and 7 show a third embodiment. The characteristic of the third embodiment is that, when the control unit receives an abnormal signal from the abnormal diagnosis unit, it stops controlling the suspension device, performs the diagnosis by the abnormal diagnosis unit again, and stops controlling the suspension device until it receives a normal signal from the abnormal diagnosis unit. In the third embodiment, the same reference numerals are used for the same components as in the first embodiment described above, and their descriptions are omitted.

[0083] As shown in Figures 1 to 3, the controller 51 according to the third embodiment is configured in substantially the same way as the controller 21 according to the first embodiment. The controller 51 includes a damper control unit 52, a weight parameter storage unit 24, and a data reading unit 25. The damper control unit 52 includes a receiving unit 26, a vehicle state quantity calculation unit 27, and a control command calculation unit 53.

[0084] The control command calculation unit 53 is a control unit that controls the damping force of the suspension device 5. The control command calculation unit 53 is configured similarly to the control command calculation unit 32 in the first embodiment. Therefore, the control command calculation unit 53 receives a signal from the estimated value abnormality diagnosis unit 31 indicating whether the rear wheel sprung speed is normal or not. If a signal indicating normality (normal signal) is received, the control command calculation unit 53 controls the suspension device 5 to adjust the damping force according to the vibration of the vehicle 1. Specifically, the control command calculation unit 53 calculates a control command value (command current value) for controlling the damping force of the variable damper 7 of the suspension device 5 based on, for example, the sprung speed for all four wheels estimated by the first sprung speed estimation unit 28 and the relative speed for all four wheels estimated by the second sprung speed estimation unit 29. At this time, the control amount for controlling the variable damper 7 is the control command value for controlling the damping force of the variable damper 7. Based on the command current value, the control command calculation unit 53 outputs a command current as a control signal to the damping force variable actuator 8 of the variable damper 7. Furthermore, the control command calculation unit 53 may calculate control command values ​​not only for improving the ride comfort of the vehicle, but also for improving steering stability.

[0085] On the other hand, if an abnormal signal (abnormal signal) is received, the control command calculation unit 53 stops controlling the suspension device 5 and performs a diagnosis again by the estimated value abnormal diagnosis unit 31 (abnormal diagnosis unit). The control command calculation unit 53 stops controlling the suspension device 5 until it receives a normal signal from the estimated value abnormal diagnosis unit 31. While the control of the suspension device 5 is stopped, the control command calculation unit 53 outputs, for example, a current command or a conveyor-equivalent current command that causes the variable damper 7 to generate a hard damping force. On the other hand, if the estimated value abnormal diagnosis unit 31 determines that it is normal after the abnormal diagnosis again and receives a normal signal from the estimated value abnormal diagnosis unit 31, the control command calculation unit 53 controls the suspension device 5 to adjust the damping force of the variable damper 7 based on the sprung speed for all four wheels estimated by the first sprung speed estimation unit 28 and the relative speed for all four wheels estimated by the second sprung speed estimation unit 29.

[0086] Next, with reference to Figure 7, the control process of the suspension device 5 by the controller 51 (damper control unit 52) ​​will be explained.

[0087] As shown in Figure 7, the control process of the suspension device 5 by the controller 51 in the third embodiment is substantially the same as the control process of the suspension device 5 by the controller 21 in the first embodiment.

[0088] However, if the difference in the speed of the two rear wheel sprung masses is greater than or equal to a threshold, and "YES" is determined in S8, the process proceeds to S21. At this time, the estimated value abnormality diagnosis unit 31 outputs a signal (abnormality signal) indicating that the estimated value is abnormal.

[0089] In S21, the control command calculation unit 53 stops controlling the suspension device 5 (variable damper 7) based on the abnormal signal from the estimated value abnormal diagnosis unit 31, and re-executes the abnormal diagnosis by the estimated value abnormal diagnosis unit 31 (application reset). The control command calculation unit 53 stops controlling the suspension device 5 until it receives a normal signal from the estimated value abnormal diagnosis unit 31.

[0090] On the other hand, if the estimated value abnormal diagnosis unit 31 determines that the abnormality diagnosis is normal after re-executing the abnormality diagnosis, and receives a normal signal from the estimated value abnormality diagnosis unit 31, the control command calculation unit 53 resumes control of the suspension device 5 (variable damper 7). Specifically, the control command calculation unit 53 controls the suspension device 5 to adjust the damping force of the variable damper 7 based on the sprung speed for all four wheels estimated by the first sprung speed estimation unit 28 and the relative speed for all four wheels estimated by the second sprung speed estimation unit 29.

[0091] Thus, the third embodiment can obtain substantially the same effects as the first embodiment. In the third embodiment, when the control command calculation unit 53 (control unit) receives an abnormal signal from the estimated value abnormality diagnosis unit 31 (abnormality diagnosis unit), it stops controlling the suspension device 5, performs the diagnosis by the estimated value abnormality diagnosis unit 31 again, and stops controlling the suspension device 5 until it receives a normal signal from the estimated value abnormality diagnosis unit 31. On the other hand, when the estimated value abnormality diagnosis unit 31 determines that the abnormality diagnosis is normal after the abnormality diagnosis is performed again, and receives a normal signal from the estimated value abnormality diagnosis unit 31, the control command calculation unit 53 resumes controlling the suspension device 5 (variable damper 7). This allows for repeated abnormality diagnosis even after control has stopped due to an abnormality, and for example, on road surfaces where the AI ​​estimation indicates no problems, control of the suspension device 5 can be resumed. As a result, deterioration of ride comfort due to the suspension device 5 being stopped can be suppressed.

[0092] In the first embodiment, the second sprung speed estimation unit 29 (relative speed calculation DNN 29A) receives a signal of FB current to the variable damper 7 of each wheel from the control command calculation unit 32 as an electrical signal generated based on the unsprung vibration of the vehicle 1, but the present invention is not limited to this. The second sprung speed estimation unit 29 (relative speed calculation DNN 29A) may, for example, receive a signal of unsprung acceleration (acceleration signal) of each wheel from an unsprung acceleration sensor provided on each wheel, receive a relative signal (displacement sensor signal) of the variable damper 7 of each wheel from a vehicle height sensor provided on each wheel, or receive a wheel speed signal of each wheel from a wheel speed sensor provided on each wheel as an electrical signal generated based on the unsprung vibration of the vehicle 1.

[0093] In the first embodiment, the estimated value anomaly diagnosis unit 31 compares the sprung speeds of the two rear wheels output from the first sprung speed estimation unit 28 and the second sprung speed estimation unit 29 to determine whether the estimated value is normal or not. However, the present invention is not limited to this. The estimated value anomaly diagnosis unit 31 may compare the sprung speed of one wheel, either the left or right rear wheel, to determine whether the estimated value is normal or not. The estimated value anomaly diagnosis unit 31 may compare the sprung speed of the front wheels to determine whether the estimated value is normal or not. That is, the estimated value anomaly diagnosis unit 31 may compare the sprung speeds of all four wheels to determine whether the estimated value is normal or not, or it may compare the sprung speed of any one of the four wheels to determine whether the estimated value is normal or not. This configuration can also be applied to the second and third embodiments.

[0094] In the first embodiment, the controller 21 acquires data such as vehicle speed through the CAN 12, but the present invention is not limited to this. For example, the controller may acquire vehicle speed data directly from the vehicle speed sensor. That is, the controller 21 may, for example, acquire detection values ​​from various sensors directly from those sensors. The controller 21 may also acquire behavioral information, including vehicle speed, from other controllers or the like. This configuration can also be applied to the second and third embodiments.

[0095] The first embodiment describes a case where the IMU 11 is installed outside the controller 21 and is received by the controller 21 for control. The present invention is not limited to this, and the IMU may be mounted inside the controller. This configuration can also be applied to the second and third embodiments.

[0096] In the first embodiment, the first sprung mass velocity estimation unit 28 and the second sprung mass velocity estimation unit 29 receive signals from the IMU 11 (Inertial Measurement Unit) which acts as a sensor of the vehicle body 2, but the present invention is not limited to this. The sensors of the vehicle body 2 may be vertical acceleration sensors, roll direction angular velocity sensors, pitch direction angular velocity sensors, etc., provided on the vehicle body 2. Therefore, the first sprung mass velocity estimation unit 28 and the second sprung mass velocity estimation unit 29 may receive signals from acceleration sensors, angular velocity sensors, etc. This configuration can also be applied to the second and third embodiments.

[0097] In the embodiments described above, the case in which a variable damper 7 consisting of a semi-active damper is used as the force generating mechanism was explained as an example. The present invention is not limited to this, and an active damper (either an electric actuator or a hydraulic actuator) may be used as the force generating mechanism. In the embodiments, the case in which a force generating mechanism that generates an adjustable force between the vehicle body 2 and the wheel 3 is configured with a variable damper 7 consisting of a damping force adjustable hydraulic shock absorber was explained as an example. The present invention is not limited to this, and for example, the force generating mechanism may be configured with an air suspension, stabilizer (Kinesus), electromagnetic suspension, etc., in addition to a hydraulic shock absorber.

[0098] In the embodiments described above, a control device for a suspension system used in a four-wheeled automobile was used as an example. However, the present invention is not limited to this, and may be applied to two-wheeled vehicles, three-wheeled vehicles, and also to work vehicles, transport vehicles such as trucks and buses.

[0099] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0100] This application claims priority under Japanese Patent Application No. 2025-067909, filed on 17 April 2025. The entire disclosure of Japanese Patent Application No. 2025-067909, filed on 17 April 2025, including the specification, claims, drawings, and abstract, is incorporated into this application by reference.

[0101] 1: Vehicle, 2: Body, 3: Wheels, 5: Suspension system, 7: Variable damper (force generation mechanism), 8: Variable damping force actuator, 11: IMU (sensor), 12: CAN, 21, 41, 51: Controller (suspension control device), 23, 42, 52: Damper control unit, 27: Vehicle state quantity calculation unit, 28: First sprung mass velocity estimation unit, 28A: DNN for sprung mass velocity calculation (sprung mass velocity estimation unit), 28B: First rear wheel sprung mass velocity calculation unit (first rear wheel sprung mass 29: Speed ​​estimation unit, 29: Second sprung mass speed estimation unit, 29A: DNN for relative speed calculation (relative speed estimation unit), 29B: Front wheel sprung mass speed calculation unit, 29C: Front wheel relative speed calculation unit, 29D: Rear wheel relative speed calculation unit, 29E: Front wheel unsprung mass speed calculation unit (front wheel unsprung mass speed estimation unit), 29F: Second rear wheel sprung mass speed calculation unit (second rear wheel sprung mass speed estimation unit), 30: Delay time calculation unit, 31: Estimated value anomaly diagnosis unit (anomaly diagnosis unit), 32, 43, 53: Control command calculation unit (control unit)

Claims

1. A suspension control device for controlling a suspension device provided between the vehicle body and the wheels in a vehicle, which suppresses vibrations between the vehicle body and the wheels, comprising: a first sprung speed estimation unit that receives signals from a sensor provided on the vehicle body and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; a second sprung speed estimation unit that receives signals from the sensor and an electrical signal generated based on the unsprung vibration of the vehicle and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; and an abnormality diagnosis unit that compares the sprung speed output from the first sprung speed estimation unit and the sprung speed output from the second sprung speed estimation unit and determines whether it is normal or not.

2. A suspension control device according to claim 1, wherein the first sprung speed estimation unit has a first rear wheel sprung speed estimation unit that estimates a first rear wheel sprung speed, which is the sprung speed of the rear wheel in the direction of travel of the vehicle, from the estimated sprung speed, and the second sprung speed estimation unit has a relative speed estimation unit that estimates the relative speed between the vehicle body and the wheel, a front wheel unsprung speed estimation unit that estimates a front wheel unsprung speed, which is the unsprung speed of the front wheel in the direction of travel of the vehicle, from the sprung speed and the relative speed, and a second rear wheel sprung speed estimation unit that estimates a second rear wheel sprung speed, which is the sprung speed of the rear wheel in the direction of travel of the vehicle, from the front wheel unsprung speed, the relative speed and the vehicle speed of the vehicle.

3. A suspension control device according to claim 2, further comprising a control unit that receives a judgment result signal output from the abnormality diagnosis unit, controls the suspension device to adjust the damping force according to the vibration of the vehicle if a normal signal is received, and stops controlling the suspension device if an abnormal signal is received.

4. A suspension control device according to claim 3, wherein the control unit, upon receiving a signal indicating an abnormality from the abnormality diagnosis unit, stops controlling the suspension device for a predetermined time, then receives a judgment result signal output from the abnormality diagnosis unit, and controls the suspension device according to the received result.

5. A suspension control device according to claim 3, wherein the control unit, upon receiving an abnormal signal from the abnormality diagnosis unit, stops controlling the suspension device, performs the diagnosis by the abnormality diagnosis unit again, and stops controlling the suspension device until it receives a normal signal from the abnormality diagnosis unit.

6. A suspension control device for controlling a suspension device provided between the vehicle body and the wheels in a vehicle, which suppresses vibrations between the vehicle body and the wheels, comprising: a sprung speed estimation unit that receives signals from a sensor provided on the vehicle body and estimates the sprung speed of the vehicle body based on learning results obtained by prior machine learning; a first rear wheel sprung speed estimation unit that estimates a first rear wheel sprung speed, which is the sprung speed of the rear wheel in the direction of travel of the vehicle, from the sprung speed; a relative speed estimation unit that receives signals from the sensor and an electrical signal generated based on the unsprung vibration of the vehicle and estimates the relative speed between the vehicle body and the wheels based on learning results obtained by prior machine learning; and a front wheel unsprung speed estimation unit that estimates a front wheel unsprung speed, which is the unsprung speed of the front wheel in the direction of travel of the vehicle, from the sprung speed and the relative speed. A suspension control device comprising: a second rear wheel sprung speed estimation unit that estimates a second rear wheel sprung speed, which is the sprung speed of the rear wheels in the direction of travel of the vehicle, from the front wheel unsprung speed, the relative speed, and the vehicle speed of the vehicle; and an abnormality diagnosis unit that compares the first rear wheel sprung speed output from the first rear wheel sprung speed estimation unit with the second rear wheel sprung speed output from the second rear wheel sprung speed estimation unit and determines whether it is normal or not.

Images