Vehicle behavior estimation method and vehicle behavior estimation device

Abstract

The controller is provided with a vehicle behavior estimation unit that estimates the sprung and unsprung state quantities of the wheels. The vehicle behavior estimation unit estimates the sprung and unsprung state quantities of the front part and the sprung state quantity of the quantity of one rear wheel by means of a first AI estimation unit that is artificial intelligence, and the remaining sprung and unsprung state quantities of the rear part are calculated on the basis of the sprung and unsprung state quantities of the front part and the sprung state quantity of the quantity of one rear wheel. The vehicle behavior estimation unit estimates the sprung and unsprung state quantities of the rear portion as needed by a second AI estimation unit that is artificial intelligence.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a vehicle behavior estimation method and a vehicle behavior estimation device that estimate, for example, sprung and unsprung state amounts of wheels. BACKGROUND

[0002] Patent Literature 1 discloses a damper control system that is capable of using a machine learning algorithm and performing control of characteristics of a damper with high responsiveness and robustness. The damper control system has a damper control unit that controls characteristics of a damper used in a suspension of a vehicle, and a processing unit that accepts feedback data measured in the vehicle in relation to behavior of the vehicle, applies an arithmetic processing determined by execution of a machine learning algorithm to the feedback data, and outputs a control variable obtained by the arithmetic processing to the damper control unit. Further, the damper control unit controls the characteristics of the damper based on the control variable used in the inside of the damper control unit, and replaces the control variable used in the inside with a new control variable output by the processing unit.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2021-17168 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] However, according to the above-described prior art, a control variable is acquired by an arithmetic processing determined by execution of a machine learning algorithm (artificial intelligence), and the characteristics of the damper are controlled based on the control variable. However, such an update of the control variable based on artificial intelligence is performed at a high frequency, and the arithmetic load becomes excessively high, so there is a problem that it is not realistic from the viewpoint of computational resources.

[0008] An object of the present application is to provide a vehicle behavior estimation method and a vehicle behavior estimation device that are capable of reducing an arithmetic load of artificial intelligence.

[0009] MEANS FOR SOLVING THE PROBLEMS

[0010] One embodiment of the present application is a vehicle behavior estimation method that estimates sprung and unsprung state amounts of each wheel by artificial intelligence, estimates sprung and unsprung state amounts of a front portion and a state amount of a sprung portion of one wheel of a rear portion by artificial intelligence, calculates remaining sprung and unsprung state amounts of the rear portion from the sprung and unsprung state amounts of the front portion and the state amount of the sprung portion of the one wheel of the rear portion, and estimates the sprung and unsprung state amounts of the rear portion by artificial intelligence as necessary.

[0011] An embodiment of the present application is a vehicle behavior estimation device that estimates state amounts of each wheel on and off the spring by artificial intelligence, estimates state amounts of the front on and off the spring and the amount of the rear one wheel on the spring by artificial intelligence, calculates the remaining rear on and off the spring state amounts from the front on and off the spring state amounts and the amount of the rear one wheel on the spring, and estimates the rear on and off the spring state amounts by artificial intelligence as needed.

[0012] According to an embodiment of the present application, it is possible to reduce the computational load of artificial intelligence. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 is a whole structure diagram of a four-wheel vehicle that shows a controller to which an embodiment of the present application is applied.

[0014] Figure 2 is a control block diagram that shows the controller in Figure 1

[0015] Figure 3 is a block diagram that shows a vehicle behavior estimation section of the first and third embodiments.

[0016] Figure 4 is an explanatory diagram that shows an example of a neural network that constitutes an FL state estimation section.

[0017] Figure 5 is a block diagram that shows a vehicle behavior estimation section of the second embodiment.

[0018] Figure 6 is an explanatory diagram that shows a trajectory of a wheel at the time of a steady turn (steady cornering). DETAILED DESCRIPTION

[0019] Hereinafter, a case in which a vehicle behavior estimation method and a vehicle behavior estimation device of an embodiment of the present application are applied to a four-wheel vehicle will be exemplified, and detailed description will be made in accordance with the accompanying drawings.

[0020] Figure 1 and Figure 2 shows a vehicle behavior control device 1. The vehicle behavior control device 1 is constituted of a suspension device 5 that constitutes a damping force generation device and a controller 11 that constitutes a vehicle control device. Herein, in the vehicle body 2 that constitutes a vehicle body of a vehicle, left and right front wheels and left and right rear wheels (hereinafter, collectively referred to as wheels 3) are provided, for example, on the lower side. The wheels 3 are constituted of tires 4 that function as a spring that absorbs fine unevenness of a road surface. Figure 1

[0021] ​​The suspension device 5 is sandwichingly provided between the vehicle body 2 and the vehicle wheel 3. The suspension device 5 is composed of a suspension spring 6 (hereinafter, referred to as a spring 6) and a damping force adjustable shock absorber (hereinafter, referred to as a variable damper 7) which is provided in parallel with the spring 6 between the vehicle body 2 and the vehicle wheel 3.

[0022] Further, in Figure 2 one set of the suspension device 5 is provided between the vehicle body 2 and the vehicle wheel 3. However, the suspension device 5 is, for example, independently provided in total four sets between the four vehicle wheels 3 and the vehicle body 2, and Figure 2 one set of them is schematically shown in

[0023] Here, the variable damper 7 of the suspension device 5 is composed of a damping force adjustable hydraulic shock absorber which is sandwichingly provided between the vehicle body 2 and the vehicle wheel 3. In order to continuously adjust the characteristic of the generated damping force (i.e., the damping force characteristic) from a hard characteristic to a soft characteristic, a damping force variable actuator 8 composed of a damping force adjusting valve or the like is attached to the variable damper 7. Further, the damping force variable actuator 8 can not necessarily be a structure which continuously adjusts the damping force characteristic, and can be, for example, a structure which can adjust the damping force in a plurality of stages of two or more stages. In addition, the variable damper 7 can be of a pressure control type or a flow control type. The variable damper 7 can also be of a type which controls viscosity like a magnetorheological fluid or an electrorheological fluid.

[0024] The controller 11 constitutes a control device. The controller 11 as a control device which controls the damping characteristic of the variable damper 7 is composed of, for example, a microcomputer. The controller 11 is connected to, for example, a CAN 10 (Controller Area Network) which is a line network required for data communication. The controller 11 acquires data related to the behavior of the vehicle (hereinafter, referred to as behavior information) through the CAN 10. At this time, the behavior information includes, for example, front and rear accelerations (front and rear G), lateral acceleration (lateral G), steering angle, yaw rate, command current (FB current), wheel speed, and the like. Therefore, these behavior information are input to the controller 11. The output side of the controller 11 is connected to the damping force variable actuator 8 of the variable damper 7.

[0025] The controller 11 also includes a storage unit comprised of ROM, RAM, and nonvolatile memory. The storage unit of the controller 11 stores various programs, information (vehicle information), and data used to control the variable damper 7. Based on the behavior information, the controller 11 estimates the sprung velocity and relative velocity (piston velocity) as vehicle state quantities. Based on the estimated vehicle state quantities, the controller 11 calculates the force to be generated by the variable damper 7 (force generating mechanism) of the suspension system 5 and outputs a control signal (command current) to the variable damping force actuator 8 of the suspension system 5.

[0026] like Figure 2 As shown, the controller 11 includes a control unit 12, a weight parameter storage unit 13, and a data reading unit 14. The control unit 12 includes a vehicle behavior estimation unit 15 and a control value calculation unit 20.

[0027] The controller 11 obtains data related to vehicle behavior during driving (behavior information) from communication data transmitted within the vehicle via the CAN 10, calculates suspension control values ​​based on the behavior information and weight parameters in the storage unit 13, and controls the variable damper 7 of the suspension device 5.

[0028] In this embodiment, CAN is used as an example of an in-vehicle network, but other in-vehicle networks can also be used. Examples of in-vehicle networks include CAN FD (CAN with Flexible Data Rate), FlexRay, and in-vehicle Ethernet.

[0029] Next, the damping force control of the suspension device 5 performed by the controller 11 will be described. The vehicle behavior estimation unit 15 within the control unit 12 serves as the vehicle behavior estimation device. The vehicle behavior estimation unit 15 reads the weight parameters stored in the weight parameter storage unit 13 via the data readout unit 14. Furthermore, based on the behavior information and weight parameters received from the CAN 10, the vehicle behavior estimation unit 15 estimates the data related to suspension control required by the control value calculation unit 20. Specifically, it estimates the instantaneous values ​​related to suspension control. Based on the instantaneous values ​​input from the vehicle behavior estimation unit 15, the control value calculation unit 20 calculates a suspension control value (e.g., a target damping force) for controlling the damping force of the variable damper 7 of the suspension device 5. Based on the suspension control value, the control unit 20 outputs a command current as a control signal to the damping force variable actuator 8 of the variable damper 7.

[0030] In this embodiment, the vehicle behavior estimation unit 15 estimates data related to suspension control for each of the four wheels (sprung velocity, relative velocity), referring to behavior information in the data continuously transmitted on the CAN 10, such as wheel speed, longitudinal acceleration, lateral acceleration, and yaw rate. This estimation result is then transmitted to the subsequent control value calculation unit 20. The vehicle behavior estimation unit 15 estimates sensor data (vehicle state variables) related to suspension control for each of the four wheels. The vehicle behavior estimation unit 15 is comprised of, for example, a neural network. The weight parameters used in the neural network are pre-determined for the vehicle through machine learning. Therefore, the neural network also reflects the overall vehicle rigidity characteristics, improving the accuracy of suspension control that is tailored to the vehicle's behavior.

[0031] The weight parameters and behavior information are input to the vehicle behavior estimation unit 15. The vehicle behavior estimation unit 15 estimates instantaneous values, such as sensor data such as sprung velocity and relative velocity, based on the weight parameters and behavior information, and outputs the estimated values.

[0032] like Figure 3 As shown, the vehicle behavior estimation unit 15 includes a first AI estimation unit 16, a second AI estimation unit 17, a non-AI estimation unit 18, and an AI estimation selection unit 19. The vehicle behavior estimation unit 15 estimates the sprung and unsprung velocities of the four wheels using the first AI estimation unit 16, the second AI estimation unit 17, and the non-AI estimation unit 18. Based on the estimated sprung and unsprung velocities of the four wheels, the vehicle behavior estimation unit 15 obtains the sprung and relative velocities of the four wheels. The relative velocities are calculated as the difference between the sprung and unsprung velocities. The vehicle behavior estimation unit 15 outputs the sprung and relative velocities of the four wheels as vehicle state quantities to the subsequent control value calculation unit 20.

[0033] The first AI estimation unit 16 is an artificial intelligence (AI). It includes a FL state estimation unit 16A, a FR state estimation unit 16B, and an RL state estimation unit 16C. The FL state estimation unit 16A estimates the sprung and unsprung speeds of the left front wheel based on the behavior information. The FR state estimation unit 16B estimates the sprung and unsprung speeds of the right front wheel based on the behavior information. Thus, the first AI estimation unit 16 estimates the sprung and unsprung speeds of the left and right front wheels as the sprung and unsprung state quantities of the front vehicle (hereinafter referred to as the front sprung and unsprung state quantities). Furthermore, the RL state estimation unit 16C estimates the sprung speed of the left rear wheel based on the behavior information. Thus, the first AI estimation unit 16 estimates the sprung and unsprung state quantities of the front vehicle and the rear vehicle, as well as the sprung state quantity of the left rear vehicle, among the eight state quantities of the front sprung and unsprung vehicle and the rear sprung and unsprung vehicle.

[0034] Like the first AI estimation unit 16, the second AI estimation unit 17 is also an artificial intelligence. It includes an RL state estimation unit 17A and an RR state estimation unit 17B. The RL state estimation unit 17A estimates the unsprung velocity of the left rear wheel based on the behavior information. The RR state estimation unit 17B estimates the sprung and unsprung velocities of the right rear wheel based on the behavior information. Thus, the second AI estimation unit 17 estimates the unsprung velocities of the left and right rear wheels, as well as the sprung velocity of the right rear wheel, as sprung and unsprung state quantities of the rear portion (hereinafter referred to as rear sprung state quantities).

[0035] The FL state estimation unit 16A, the FR state estimation unit 16B, the RL state estimation unit 16C, the RL state estimation unit 17A, and the RR state estimation unit 17B are configured similarly to the neural network of the vehicle behavior estimation unit disclosed in Japanese Patent Application Laid-Open No. 2022-191913, for example.

[0036] As an example, refer to Figure 4 The specific structure of the FL state estimation unit 16A will be described. Figure 4 As shown, the FL state estimation unit 16A is configured, for example, by a neural network. The neural network comprises a three-layer hierarchical neural network, hierarchically connecting elements of the input layer (number of elements i) 101, the hidden layer (number of elements j) 102, and the output layer (number of elements k) 103. Each element of the input layer 101 is connected to each element of the hidden layer 102 by weights W1ij (i = 1 to I, j = 1 to J), and each element of the hidden layer 102 is connected to each element of the output layer 103 by weights W2jk (j = 1 to J, k = 1 to K). Information on these weights (hereinafter referred to as weight parameters) is represented by a matrix of weights W1ij and W2jk. The weight parameters are pre-calculated through machine learning and stored in the weight parameter storage unit 13. While this example shows a fully connected neural network with a single hidden layer 102, the present invention is not limited to this. For example, a neural network having two or more hidden layers 102 may also be employed.

[0037] Machine learning in the neural network is performed based on vehicle state quantity data (sprung and unsprung speeds) and behavior information data previously acquired by the data acquisition vehicle. The data acquisition vehicle is a vehicle of the same specifications as the vehicle equipped with the first AI estimation unit 16 and the second AI estimation unit 17 of the vehicle behavior estimation unit 15, and is equipped with various sensors (such as speed sensors and acceleration sensors) that acquire vehicle state quantities. During machine learning in the neural network, weight parameters are adjusted to learn the correlation between the vehicle state quantity data (sprung and unsprung speeds) acquired by the data acquisition vehicle and the behavior information data. The weight parameters obtained as a result of the learning are stored in the weight parameter storage unit 13.

[0038] Furthermore, in this embodiment, the first AI estimation unit 16 and the second AI estimation unit 17 perform machine learning based on the correlation between vehicle state quantity data and behavior information data acquired by the data acquisition vehicle. However, the present invention is not limited to this. For example, a vehicle model corresponding to the vehicle equipped with the first AI estimation unit 16 and the second AI estimation unit 17 may be constructed, and the first AI estimation unit 16 and the second AI estimation unit 17 may perform machine learning based on data acquired through simulation using this vehicle model.

[0039] like Figure 4 As shown, the neural network's input layer 101 consists of a first input layer 101A, which receives time-series data on the left front wheel's wheel speed, and a second input layer 101B, which receives time-series data on the vehicle's longitudinal acceleration. The output layer 103 outputs, for example, the instantaneous values ​​of the sprung and unsprung velocities of the suspension system 5, which is assumed to be mounted on the vehicle's left front wheel. The number of elements in the hidden layer 102 is generally determined based on the number of elements in the input layer 101 and the output layer 103, and is set to maximize the accuracy of the vehicle behavior estimation using the neural network. The number of elements in the output layer 103 is determined by the output specifications of the vehicle behavior estimation.

[0040] exist Figure 4 For simplicity of explanation, the neural network of the FL state estimation unit 16A used to control the suspension system 5 mounted on the left front wheel of the vehicle is shown as an example. The FR state estimation unit 16B, RL state estimation unit 16C, RL state estimation unit 17A, and RR state estimation unit 17B are also composed of the same neural network as the FL state estimation unit 16A. Therefore, the vehicle behavior estimation unit 15 includes the FL state estimation unit 16A, FR state estimation unit 16B, RL state estimation unit 16C, RL state estimation unit 17A, and RR state estimation unit 17B, which are composed of neural networks.

[0041] In addition, Figure 4 In FIG, the FL state estimating unit 16A is configured to output the sprung and unsprung velocities of the left front wheel based on the wheel speed and longitudinal acceleration as behavior information. The present invention is not limited to this. The FL state estimating unit 16A may also output the sprung and unsprung velocities of the left front wheel based on the wheel speed, longitudinal acceleration, lateral acceleration, and yaw rate as behavior information. Furthermore, the FL state estimating unit 16A may also output the sprung and unsprung velocities of the left front wheel based on the wheel speed, longitudinal acceleration, lateral acceleration, yaw rate, steering angle, and the command current supplied to the variable damping force actuator 8 of the left front wheel as behavior information. This also applies to the FR state estimating unit 16B, the RL state estimating unit 16C, the RL state estimating unit 17A, and the RR state estimating unit 17B.

[0042] like Figure 3 As shown, the non-AI estimation unit 18 calculates three state quantities on the rear wheel side based on the estimation results of the first AI estimation unit 16. This reduces the computational load of the vehicle behavior estimation unit 15. The non-AI estimation unit 18 includes an RL state estimation unit 18A and an RR state estimation unit 18B. The RL state estimation unit 18A estimates the unsprung speed of the left rear wheel based on the estimation results and behavior information of the FL state estimation unit 16A. The RR state estimation unit 18B estimates the unsprung speed of the right rear wheel based on the estimation results and behavior information of the FR state estimation unit 16B. The RR state estimation unit 18B estimates the sprung speed of the right rear wheel based on the estimation results and behavior information of the FL state estimation unit 16A, the FR state estimation unit 16B, and the RL state estimation unit 16C.

[0043] Specifically, the RL state estimator 18A calculates the unsprung speed of the left rear wheel by delaying the unsprung speed of the left front wheel estimated by the FL state estimator 16A of the first AI estimator 16 by the wheelbase. The RR state estimator 18B calculates the unsprung speed of the right rear wheel by delaying the unsprung speed of the right front wheel estimated by the FR state estimator 16B of the first AI estimator 16 by the wheelbase. Furthermore, the RR state estimator 18B calculates the sprung speed of the right rear wheel, representing the remaining wheel, based on the sprung speeds of the three wheels estimated by the FL state estimator 16A, FR state estimator 16B, and RL state estimator 16C, treating the vehicle body 2 as a rigid body. Thus, the non-AI estimator 18 calculates the unsprung speeds of the left and right rear wheels, as well as the sprung speed of the right rear wheel, as the rear sprung vertical state quantities.

[0044] The AI ​​estimation selection unit 19 selects whether to calculate the unsprung speed of the rear wheels, etc., by the non-AI estimation unit 18 or to estimate the unsprung speed of the rear wheels, etc., by the second AI estimation unit 17. For example, the AI ​​estimation selection unit 19 selects either the non-AI estimation unit 18 or the second AI estimation unit 17 based on the steering angle included in the behavior information. For example, when the steering angle is smaller than a predetermined angle, the road surface trajectory of the front wheels and the road surface trajectory of the rear wheels are approximately the same. Therefore, the AI ​​estimation selection unit 19 selects the non-AI estimation unit 18 when the steering angle is smaller than the predetermined angle. In this case, the non-AI estimation unit 18 calculates the sprung and unsprung speeds of the rear wheels based on the sprung and unsprung speeds of the front wheels. On the other hand, when the steering angle is larger than the predetermined angle, the AI ​​estimation selection unit 19 selects the second AI estimation unit 17. In this case, the second AI estimation unit 17 estimates the sprung and unsprung speeds of the rear wheels based on the behavior information.

[0045] Thus, the vehicle behavior estimating unit 15 estimates the state amounts of the front and the rear suspensions and the state amount of the spring of the one wheel of the rear by the first AI estimating unit 16 as artificial intelligence, and the remaining state amounts of the front and the rear suspensions are calculated from the state amounts of the front and the state amount of the spring of the one wheel of the rear, and the state amounts of the front and the rear suspensions are estimated by the second AI estimating unit 17 as artificial intelligence as necessary. In this way, the vehicle behavior estimating unit 15 estimates the vehicle behavior of the rear (the state amounts of the rear suspensions) by the mathematical expression based on the estimation result of the vehicle behavior of the front (the state amounts of the front suspensions) by the artificial intelligence (the first AI estimating unit 16). Thus, in the present embodiment, the computational load of the estimation by the artificial intelligence can be reduced as compared with the case where the state amounts of the front and the rear suspensions are estimated by the artificial intelligence.

[0046] Further, the vehicle behavior estimating unit 15 estimates the state amounts of the front and the rear suspensions and the state amount of the spring of the one wheel of the rear by the first AI estimating unit 16 and the second AI estimating unit 17 as artificial intelligence in the case where it is determined that it is difficult to estimate the state amounts of the rear suspensions from the state amounts of the front suspensions due to the steering or the like. Thus, with respect to the state amounts of the front and the rear suspensions of the four wheels, the prescribed estimation accuracy can be ensured, and the performance of the suspension control can be ensured.

[0047] The vehicle behavior estimating unit 15 estimates the state amounts of the front and the rear suspensions and the state amount of the spring of the one wheel of the rear by the artificial intelligence at ordinary times, and estimates the state amounts of the front and the rear suspensions by the artificial intelligence only in the case where the trajectories of the wheels of the front and the rear are not coincident. Thus, at ordinary times, the computational load of the estimation by the artificial intelligence can be reduced. On the other hand, in the case where the trajectories of the wheels of the front and the rear are not coincident, the state amounts of the front and the rear suspensions are estimated by the first AI estimating unit 16 and the second AI estimating unit 17 as artificial intelligence, and the performance of the suspension control can be ensured.

[0048] The vehicle behavior estimating unit 15 estimates the state amounts of the front and the rear suspensions and the state amount of the spring of the one wheel of the rear by the first AI estimating unit 16 as artificial intelligence, and the remaining state amounts are calculated from the state amounts of the front and the state amount of the spring of the one wheel of the rear. Thus, the non-AI estimating unit 18 of the vehicle behavior estimating unit 15 can calculate the state amount of the rear suspension (the suspension velocity) based on the state amount of the front suspension (the suspension velocity). In addition, for the non-AI estimating unit 18 of the vehicle behavior estimating unit 15, the vehicle body 2 is regarded as a rigid body from the state amounts of the springs of the three wheels (the spring velocities), and thus the state amount of the spring of the remaining one wheel (the right rear wheel) (the spring velocity) can be calculated.

[0049] The vehicle behavior estimation unit 15 estimates the on-spring and off-spring state quantities at the front and the rear by artificial intelligence in a case where it is determined that the tracks of the wheels at the front and the rear are different. Therefore, for example, even in a situation where the off-spring state quantity at the rear cannot be calculated from the off-spring state quantity at the front, the on-spring and off-spring state quantities of all the wheels can be estimated by the first AI estimation unit 16 and the second AI estimation unit 17 as artificial intelligence.

[0050] Further, in the first embodiment, the first AI estimation unit 16 is configured to estimate the on-spring speed of the left rear wheel in addition to the on-spring speed and the off-spring speed of the left front wheel and the right front wheel, but the present application is not limited to this. The first AI estimation unit 16 can estimate the on-spring speed of the right rear wheel instead of the left rear wheel. In this case, the second AI estimation unit 17 and the non-AI estimation unit 18 estimate or calculate the on-spring speed of the left rear wheel, which is the remaining one wheel, on the basis of the on-spring speeds of the three wheels estimated by the first AI estimation unit 16.

[0051] Next, Figure 2 and Figure 5 A second embodiment will be described. The second embodiment is characterized in that a vehicle behavior estimation unit estimates or calculates on-spring accelerations and off-spring accelerations of four wheels and then acquires on-spring speeds and off-spring speeds of the four wheels. Further, in the second embodiment, the same reference numerals are attached to the same structural elements as those of the first embodiment described above, and the description thereof will be omitted.

[0052] The vehicle behavior estimation unit 21 of the second embodiment is a vehicle behavior estimation device. The vehicle behavior estimation unit 21 reads the weight parameters stored in the weight parameter storage 13 via the data readout unit 14. Further, the vehicle behavior estimation unit 21 estimates data required for suspension control in the control value calculation unit 20, specifically, instantaneous values involved in suspension control, on the basis of the behavior information from the CAN 10 and the weight parameters.

[0053] The vehicle behavior estimation unit 21 includes a first AI estimation unit 22, a second AI estimation unit 23, a non-AI estimation unit 24, an AI estimation selection unit 25, and an integrator 26. The vehicle behavior estimation unit 21 estimates on-spring accelerations and off-spring accelerations of the four wheels by the first AI estimation unit 22, the second AI estimation unit 23, and the non-AI estimation unit 24. The vehicle behavior estimation unit 21 integrates the estimated on-spring accelerations and off-spring accelerations of the four wheels to acquire on-spring speeds and off-spring speeds of the four wheels. The vehicle behavior estimation unit 21 acquires on-spring speeds and relative speeds of the four wheels on the basis of the on-spring speeds and the off-spring speeds of the four wheels. The vehicle behavior estimation unit 21 outputs the on-spring speeds and the relative speeds of the four wheels as vehicle state quantities to the control value calculation unit 20 at the rear stage.

[0054] The first AI estimation unit 22 is artificial intelligence. The first AI estimation unit 22 includes an FL state estimation unit 22A, an FR state estimation unit 22B, and an RL state estimation unit 22C. The FL state estimation unit 22A estimates the sprung and unsprung accelerations of the front left wheel on the basis of the behavior information. The FR state estimation unit 22B estimates the sprung and unsprung accelerations of the front right wheel on the basis of the behavior information. Thus, the first AI estimation unit 22 estimates the sprung and unsprung accelerations of the front left and right wheels as the front unsprung and sprung state quantities. In addition to this, the RL state estimation unit 22C estimates the unsprung acceleration of the rear left wheel on the basis of the behavior information. Thus, the first AI estimation unit 22 estimates the front unsprung and sprung state quantities and the rear left unsprung state quantity among the eight state quantities of the front unsprung and sprung state quantities and the rear unsprung and sprung state quantities.

[0055] The second AI estimation unit 23 is also artificial intelligence like the first AI estimation unit 22. The second AI estimation unit 23 includes an RL state estimation unit 23A and an RR state estimation unit 23B. The RL state estimation unit 23A estimates the unsprung acceleration of the rear left wheel on the basis of the behavior information. The RR state estimation unit 23B estimates the sprung and unsprung accelerations of the rear right wheel on the basis of the behavior information. Thus, the second AI estimation unit 23 estimates the unsprung accelerations of the rear left and right wheels and the sprung acceleration of the rear right wheel as the rear unsprung and sprung state quantities.

[0056] The FL state estimation unit 22A, the FR state estimation unit 22B, the RL state estimation unit 22C, the RL state estimation unit 23A, and the RR state estimation unit 23B of the second embodiment are configured like the FL state estimation unit 16A, the FR state estimation unit 16B, the RL state estimation unit 16C, the RL state estimation unit 17A, and the RR state estimation unit 17B of the first embodiment. That is, the FL state estimation unit 22A, the FR state estimation unit 22B, the RL state estimation unit 22C, the RL state estimation unit 23A, and the RR state estimation unit 23B of the second embodiment are configured by a neural network. The neural network is input with a weight parameter in which the correlation between the behavior information and the unsprung acceleration and the like is learned. The FL state estimation unit 22A, the FR state estimation unit 22B, the RL state estimation unit 22C, the RL state estimation unit 23A, and the RR state estimation unit 23B estimate the unsprung acceleration and the like on the basis of the weight parameter and the behavior information.

[0057] The non-AI estimation unit 24 calculates the three state quantities of the rear wheels based on the estimation results of the first AI estimation unit 22. Thereby, the non-AI estimation unit 24 realizes a reduction in the computational load of the vehicle behavior estimation unit 21. The non-AI estimation unit 24 includes an RL state estimation unit 24A and an RR state estimation unit 24B. The RL state estimation unit 24A estimates the unsprung acceleration of the left rear wheel based on the estimation result of the FL state estimation unit 22A and the behavior information. The RR state estimation unit 24B estimates the unsprung acceleration of the right rear wheel based on the estimation result of the FR state estimation unit 22B and the behavior information. The RR state estimation unit 24B estimates the sprung acceleration of the right rear wheel based on the estimation results of the FL state estimation unit 22A, the FR state estimation unit 22B, and the RL state estimation unit 22C and the behavior information.

[0058] Specifically, the RL state estimation unit 24A performs a delay process of the amount of wheelbase based on the unsprung acceleration of the left front wheel estimated by the FL state estimation unit 22A of the first AI estimation unit 22, and thereby calculates the unsprung acceleration of the left rear wheel. The RR state estimation unit 24B performs a delay process of the amount of wheelbase based on the unsprung acceleration of the right front wheel estimated by the FR state estimation unit 22B of the first AI estimation unit 22, and thereby calculates the unsprung acceleration of the right rear wheel. In addition, the RR state estimation unit 24B regards the vehicle body 2 as a rigid body based on the three quantities of the sprung accelerations of the wheels estimated by the FL state estimation unit 22A, the FR state estimation unit 22B, and the RL state estimation unit 22C, and thereby calculates the sprung acceleration of the right rear wheel of the remaining one of the quantities. Thereby, the non-AI estimation unit 24 calculates the unsprung accelerations of the left and right rear wheels and the sprung acceleration of the right rear wheel as the rear sprung and unsprung state quantities.

[0059] The AI estimation selection unit 25 selects whether to calculate the unsprung acceleration of the rear wheels and the like by the non-AI estimation unit 24 or to estimate the unsprung acceleration of the rear wheels and the like by the second AI estimation unit 23. The AI estimation selection unit 25 of the second embodiment is configured substantially similarly to the AI estimation selection unit 19 of the first embodiment.

[0060] Accordingly, the AI estimation selection unit 25 selects either one of the non-AI estimation unit 24 and the second AI estimation unit 23, for example, based on the steering angle included in the behavior information. The AI estimation selection unit 19 selects the non-AI estimation unit 24 when the steering angle is smaller than a predetermined angle. At this time, the non-AI estimation unit 24 calculates the sprung and unsprung accelerations of the rear wheels based on the sprung and unsprung accelerations of the front wheels. On the other hand, the AI estimation selection unit 25 selects the second AI estimation unit 23 when the steering angle is larger than the predetermined angle. At this time, the second AI estimation unit 23 estimates the sprung and unsprung accelerations of the rear wheels based on the behavior information.

[0061] The integrator 26 integrates the sprung and unsprung accelerations of the four wheels estimated by the first AI estimation section 22, the second AI estimation section 23, and the non-AI estimation section 24. Thereby, the integrator 26 calculates and outputs the sprung and unsprung speeds of the four wheels. The vehicle behavior estimation section 21 acquires the sprung and relative speeds of the four wheels on the basis of the sprung and unsprung speeds of the four wheels.

[0062] Thus, in the second embodiment, substantially the same operational effects as those of the first embodiment can be obtained. In the second embodiment, the first AI estimation section 22, the second AI estimation section 23, and the non-AI estimation section 24 of the vehicle behavior estimation section 21 estimate or calculate the sprung and unsprung accelerations of the four wheels as the state quantities of the sprung and unsprung. Therefore, the vehicle behavior estimation section 21 can acquire the sprung and unsprung speeds of the four wheels by integrating the sprung and unsprung accelerations of the four wheels.

[0063] In addition, in the second embodiment, the first AI estimation section 22 is configured to estimate the sprung acceleration of the left rear wheel in addition to the sprung and unsprung accelerations of the left and right front wheels, but the present application is not limited to this. The first AI estimation section 22 can estimate the sprung acceleration of the right rear wheel instead of the left rear wheel. In this case, the second AI estimation section 23 and the non-AI estimation section 24 estimate or calculate the sprung acceleration of the left rear wheel, which is the remaining one of the four wheels, on the basis of the sprung accelerations of the three wheels estimated by the first AI estimation section 22.

[0064] Next, Figure 2 , Figure 3 and Figure 6 A third embodiment will be described. The third embodiment is characterized in that the vehicle behavior estimation section calculates the trajectories of the wheels of the front portion and the wheels of the rear portion on the basis of the steering angle or the lateral acceleration, and determines the deviation of the trajectories. Further, in the third embodiment, the same reference numerals are attached to the same structural elements as those of the above-described first embodiment, and the description thereof will be omitted.

[0065] The vehicle behavior estimation section 31 of the third embodiment is a vehicle behavior estimation device. The vehicle behavior estimation section 31 reads the weight parameters stored in the weight parameter storage section 13 via the data readout section 14. Further, the vehicle behavior estimation section 31 estimates the data related to the suspension control required in the control value calculation section 20, specifically, the instantaneous values related to the suspension control, on the basis of the behavior information from the CAN 10 and the weight parameters.

[0066] The vehicle behavior estimation unit 31 includes the first AI estimation unit 16, the second AI estimation unit 17, the non-AI estimation unit 18, and an AI estimation selection unit 32. The vehicle behavior estimation unit 31 estimates the sprung and unsprung speeds of the four wheels by the first AI estimation unit 16, the second AI estimation unit 17, and the non-AI estimation unit 18. The vehicle behavior estimation unit 31 acquires the sprung and relative speeds of the four wheels on the basis of the estimated sprung and unsprung speeds of the four wheels. The vehicle behavior estimation unit 31 outputs the sprung and relative speeds of the four wheels to the control value calculation unit 20 in the rear stage as vehicle state quantities.

[0067] The AI estimation selection unit 32 selects whether to calculate the unsprung speed of the rear wheel and the like by the non-AI estimation unit 18 or to estimate the unsprung speed of the rear wheel and the like by the second AI estimation unit 17. The AI estimation selection unit 32 calculates the trajectories of the wheels in the front and rear portions on the basis of the steering angle and the lateral acceleration, and determines the shift of the trajectories. Specifically, the AI estimation selection unit 32 calculates the trajectories of the wheels in the front and rear portions on the basis of the steering angle, the lateral acceleration, and the vehicle speed.

[0068] The following shows a specific calculation method of the shift of the trajectories. For example, in the case where a turn is assumed to be a steady circular turn (steady-state circular turn), the yaw rate r is expressed by the formula of mathematical formula 1.

[0069] [mathematical formula 1]

[0070]

[0071] Here, δ denotes the steering angle, A denotes a stability coefficient (stability factor), V denotes the vehicle speed, Lf denotes the distance in the front-rear direction from the center of gravity to the front wheel, and Lr denotes the distance in the front-rear direction from the center of gravity to the rear wheel. At this time, the sum of Lf and Lr becomes the wheelbase Lw. The side slip angle (side slip angle) β at this time is expressed by the formula of mathematical formula 2.

[0072] [mathematical formula 2]

[0073]

[0074] Here, K is a proportional constant called cornering power. Further, in the case where the lateral acceleration a is used as an input, the yaw rate r can be calculated by the formula of mathematical formula 3.

[0075] [mathematical formula 3]

[0076]

[0077] The relationship between the center of gravity point position and the turn center (Xcen, Ycen) is expressed by the formula of mathematical formula 5 on the basis of the formula of mathematical formula 4.

[0078] [Math. 4]

[0079]

[0080] [Math. 5]

[0081]

[0082] The coordinates (Xfl, Yfl), (Xrl, Yrl) of the front and rear wheels, respectively, are based on the dimensions of the vehicle body 2 and are expressed by the formula of Math. 6.

[0083] [Math. 6]

[0084]

[0085] Based on these, the turning radius Rfl of the front wheel is expressed by the formula of Math. 7.

[0086] [Math. 7]

[0087]

[0088] Likewise, the turning radius Rrl of the rear wheel is expressed by the formula of Math. 8.

[0089] [Math. 8]

[0090]

[0091] The AI estimation selection section 32 determines that it is difficult to estimate the rear wheel information from the front wheel information when the difference between the turning radius Rfl of the front wheel and the turning radius Rrl of the rear wheel is larger than the tire width. At this time, the AI estimation selection section 32 switches from the non-AI estimation section 18 to the second AI estimation section 17 to acquire the sprung speed and the like of the rear wheel. That is, in the case where the difference between the turning radius Rfl of the front wheel and the turning radius Rrl of the rear wheel is larger than the tire width, the non-AI estimation section 18 does not calculate the sprung speed and the like of the rear wheel by the mathematical formula, but estimates the sprung speed and the like of the rear wheel as the second AI estimation section 17 of artificial intelligence.

[0092] Further, as shown in Math. 1 to Math. 8, the trajectories of the front and rear wheels of the vehicle can be calculated based on either one of the steering angle δ and the lateral acceleration a. Therefore, the vehicle behavior estimation section 31 calculates the trajectories of the front and rear wheels of the vehicle based on the steering angle δ or the lateral acceleration a. However, the vehicle behavior estimation section 31 can also calculate the trajectories of the front and rear wheels of the vehicle based on both the steering angle δ and the lateral acceleration a. For example, the vehicle behavior estimation section 31 can also compare the trajectories of the wheels calculated based on the steering angle δ and the trajectories of the wheels calculated based on the lateral acceleration a, and use the larger one (the maximum value) to calculate the final trajectories of the wheels.

[0093] Thus, in the third embodiment, substantially the same effect as the first embodiment can be obtained. In the third embodiment, the vehicle behavior estimation unit 31 calculates the trajectories of the front wheels and the rear wheels based on the steering angle δ or the lateral acceleration a, and determines the deviation of the trajectories. Therefore, the vehicle behavior estimation unit 31 can select the non-AI estimation unit 18 when the calculated deviation of the trajectories is smaller than the tire width, and calculate the sprung mass speed of the rear wheels and the like by the mathematical expression. On the other hand, the vehicle behavior estimation unit 31 can select the second AI estimation unit 17 as the artificial intelligence when the calculated deviation of the trajectories is larger than the tire width, and estimate the sprung mass speed of the rear wheels and the like.

[0094] In addition, in the third embodiment, the first AI estimation unit 16, the second AI estimation unit 17, and the non-AI estimation unit 18 of the first embodiment are assumed to be used, but the first AI estimation unit 22, the second AI estimation unit 23, and the non-AI estimation unit 24 of the second embodiment can also be used.

[0095] In the third embodiment, the AI estimation selection unit 32 is assumed to switch from the non-AI estimation unit 18 to the second AI estimation unit 17 when the deviation of the trajectories of the front wheels and the rear wheels is larger than a prescribed value, but the present application is not limited to this. The AI estimation selection unit 32 can also switch from the non-AI estimation unit 18 to the second AI estimation unit 17 when the vehicle is backing up, for example. That is, the AI estimation selection unit 32 can select either the second AI estimation unit 17 or the non-AI estimation unit 18 according to the traveling direction of the vehicle. This structure can be applied to the first and second embodiments.

[0096] In the first embodiment, the case where the controller 11 acquires the behavior information of the vehicle including the wheel speed through the CAN 10 is described as an example, but the present application is not limited to this. The controller 11 can also directly acquire the detection values from various sensors, such as the detection values of various sensors. In addition, the controller 11 can also acquire the behavior information from other controllers and the like. This structure can be applied to the second and third embodiments.

[0097] In the above-described embodiments, the case where the variable damper 7 composed of a semi-active damper is used as the force generation mechanism is described as an example. The present application is not limited to this, and an active damper (either one of an electric actuator and a hydraulic actuator) can also be used as the force generation mechanism. In the above-described embodiments, the case where the force generation mechanism that generates an adjustable force between the vehicle body 2 side and the wheel 3 side is composed of the variable damper 7 composed of a damper force adjustment type hydraulic buffer is described as an example. The present application is not limited to this, and the force generation mechanism can also be composed of an air suspension, a stabilizer (kinetic), an electromagnetic suspension, and the like in addition to the hydraulic buffer.

[0098] In the above-described embodiments, the vehicle behavior device for a four-wheel vehicle is exemplified and described. However, the present application is not limited to this, and can be applied to, for example, a work vehicle, a truck as a carrier vehicle, a bus, and the like.

[0099] Further, the present application is not limited to the above-described embodiments, and includes various modifications. For example, the above-described embodiments are described in detail in order to easily understand the present application, and are not limited to necessarily having all the structures described. In addition, a part of the structure of an embodiment can be replaced with the structure of another embodiment, and in addition, the structure of another embodiment can be added to the structure of an embodiment. In addition, to a part of the structure of each embodiment, addition, deletion, and replacement of other structures can be performed.

[0100] This application claims priority based on Japanese Patent Application No. 2023-045993 filed on March 22, 2023. The entire disclosure of Japanese Patent Application No. 2023-045993 filed on March 22, 2023, including the specification, claims, drawings, and abstract, is incorporated herein by reference in its entirety.

[0101] Explanation of Reference Signs

[0102] 1: vehicle behavior control device; 2: vehicle body; 3: wheel; 5: suspension device; 7: damping force adjustment type shock absorber (variable damper); 8: damping force variable actuator; 11: controller; 15, 21, 31: vehicle behavior estimation section; 16, 22: first AI estimation section; 17, 23: second AI estimation section; 18, 24: non-AI estimation section; 19, 25, 32: AI estimation selection section.

Claims

1. A vehicle behavior estimation method that estimates the sprung and unsprung state quantities of each wheel using artificial intelligence. Using artificial intelligence to estimate the sprung and unsprung state quantities of the front and the sprung state quantity of one rear wheel; The remaining rear sprung and unsprung state quantities are calculated based on the front sprung and unsprung state quantities and the sprung state quantity for one rear wheel; and As needed, artificial intelligence is used to estimate the sprung and unsprung state quantities of the rear part.

2. The vehicle behavior estimation method according to claim 1, wherein: Normally, artificial intelligence is used to estimate the sprung and unsprung state quantities of the front and the sprung state quantity of one rear wheel. Only when the trajectories of the front wheels and the rear wheels do not coincide with each other, the sprung and unsprung state quantities of the rear are estimated using artificial intelligence.

3. The vehicle behavior estimation method according to claim 1, wherein: Only the front sprung and unsprung state quantities and the rear left or right sprung state quantities among the eight state quantities of the front sprung and unsprung and the rear sprung and unsprung are estimated by artificial intelligence, and the remaining state quantities are calculated based on the front sprung and unsprung state quantities and the sprung state quantity of one rear wheel.

4. The vehicle behavior estimation method according to claim 1, wherein: When it is determined that the tracks of the front wheels and the tracks of the rear wheels are different from each other, the sprung and unsprung state quantities are estimated by artificial intelligence at both the front and rear parts.

5. The vehicle behavior estimation method according to claim 2 or claim 4, wherein: The tracks of the front wheels and the rear wheels are calculated based on the steering angle, and the deviation of the tracks is determined.

6. The vehicle behavior estimation method according to claim 2 or claim 4, wherein: The tracks of the front wheels and the rear wheels are calculated based on the lateral acceleration, and the deviation of the tracks is determined.

7. A vehicle behavior estimation device that estimates the sprung and unsprung state quantities of each wheel using artificial intelligence. Using artificial intelligence to estimate the sprung and unsprung state quantities of the front and the sprung state quantity of one rear wheel; The remaining rear sprung and unsprung state quantities are calculated based on the front sprung and unsprung state quantities and the sprung state quantity for one rear wheel; and As needed, artificial intelligence is used to estimate the sprung and unsprung state quantities of the rear part.

8. The vehicle behavior estimation device according to claim 7, wherein: Normally, artificial intelligence is used to estimate the sprung and unsprung state quantities of the front and the sprung state quantity of one rear wheel. Only when the trajectories of the front wheels and the rear wheels do not coincide with each other, the sprung and unsprung state quantities of the rear are estimated using artificial intelligence.

9. The vehicle behavior estimation device according to claim 7, wherein: Only the front sprung and unsprung state quantities and the rear left or right sprung state quantities among the eight state quantities of the front sprung and unsprung and the rear sprung and unsprung are estimated by artificial intelligence, and the remaining state quantities are calculated based on the front sprung and unsprung state quantities and the sprung state quantity of one rear wheel.

10. The vehicle behavior estimation device according to claim 7, wherein: When it is determined that the tracks of the front wheels and the tracks of the rear wheels are different from each other, the sprung and unsprung state quantities are estimated by artificial intelligence at both the front and rear parts.

11. The vehicle behavior estimation device according to claim 8 or claim 10, wherein: The tracks of the front wheels and the rear wheels are calculated based on the steering angle, and the deviation of the tracks is determined.

12. The vehicle behavior estimation device according to claim 8 or claim 10, wherein: The tracks of the front wheels and the rear wheels are calculated based on the lateral acceleration, and the deviation of the tracks is determined.

Images