JP7715102B2 - Vehicle control device - Google Patents

Description

The present invention relates to a vehicle control device.

Technology is known that limits the driving force of a vehicle when resonance occurs in the vehicle (see, for example, Patent Document 1).

Japanese Patent Application Laid-Open No. 2018-024278

The above technology limits the driving force only after resonance occurs in the vehicle, which may result in a delay in the start of driving force limitation.

The present invention therefore aims to provide a vehicle control device that can quickly initiate countermeasure control against vehicle resonance.

The above objective can be achieved by a vehicle control device that includes a calculation unit that calculates a calculated value based on sensor detection values related to vehicle behavior; a prediction unit that predicts whether the vehicle will resonate based on the calculated value using a machine-learned prediction model through supervised learning that inputs the calculated value before the vehicle resonates and outputs whether the vehicle will resonate; and an execution unit that, when it is predicted that the vehicle will resonate, executes countermeasure processing against the vehicle's resonance.

The detected value may include at least one of the longitudinal acceleration of the vehicle, the rotational speed of the vehicle's wheels, and the rotational speed of a rotating member that rotates in conjunction with a wheel that receives power from the vehicle's running power source, and the calculated value may include at least one of the average, maximum, minimum, variance, and standard deviation within a predetermined time period of at least one of the longitudinal acceleration of the vehicle, the difference between the rotational speeds of the vehicle's two wheels, the integrated value of the rotational fluctuation of the rotating member, and the difference between the vehicle's body speed and the vehicle's rotational speed.

The calculated value may include the total number of peak values within a predetermined time period of the waveform after bandpass filtering of the detected value.

The motor that serves as the vehicle's driving power source may be directly connected to one of the vehicle's wheels.

When it is predicted that the vehicle will resonate, the execution unit may, as the countermeasure processing, reduce the torque of the vehicle's driving power source compared to when it is not predicted that the vehicle will resonate.

The present invention provides a vehicle control device that can quickly initiate countermeasure control against vehicle resonance.

FIG. 1 is a schematic diagram of a vehicle. FIG. 2 shows an example of a waveform of the rotational speed of the left rear wheel after band-pass filtering. FIG. 3 is a flowchart showing an example of the resonance countermeasure control. FIG. 4 is a timing chart showing an example of the countermeasure processing. FIG. 5 is a timing chart showing another example of the countermeasure processing.

[General configuration of the vehicle]
1 is a schematic diagram of a vehicle 1. The vehicle 1 is an electric vehicle equipped with a motor 2 as a power source for traveling. The vehicle 1 is equipped with the motor 2, a propeller shaft 3, a differential gear 4, drive shafts 51 and 52, a left rear wheel 61, a right rear wheel 62, a left front wheel 63, a right front wheel 64, a PCU (Power Control Unit) 7, a battery 8, and an ECU (Electric Control Unit) 10.

Motor 2 is, for example, a permanent magnet synchronous motor or induction motor. Motor 2 functions as a prime mover that outputs torque when supplied with power, and also functions as a generator that generates electricity when vehicle 1 is braked. The power generated by motor 2 is supplied to battery 8 via PCU 7.

The motor 2 is directly connected to the left rear wheel 61 and right rear wheel 62 via the propeller shaft 3, differential gear 4, and drive shafts 51 and 52. Here, "the motor 2 is directly connected to the left rear wheel 61 and right rear wheel 62" means that no transmission, clutch, torque converter, etc. is provided between the motor 2 and the left rear wheel 61 and right rear wheel 62. The vehicle 1 moves when the torque of the motor 2 is transmitted to the left rear wheel 61 and right rear wheel 62. In the example shown in FIG. 1, the vehicle 1 is a rear-wheel drive vehicle in which the torque of the motor 2 is transmitted to the left rear wheel 61 and right rear wheel 62, but is not limited to this and may be a front-wheel drive vehicle or a four-wheel drive vehicle.

In vehicle 1, drive shafts 51 and 52 may be connected to the rotor of motor 2 via gears or directly.

Battery 8 is an electricity storage device composed of a secondary battery such as a nickel-metal hydride battery or a lithium-ion battery. Battery 8 is charged by power generated by motor 2, and can also be charged by power supplied from an external power source. Note that battery 8 is not limited to a secondary battery; it can be any electricity storage device that can generate DC voltage and be charged, such as a capacitor.

The ECU 10 includes a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), and a storage device, and performs various controls by executing programs stored in the ROM and storage device. The ECU 10 is an example of a control device for the vehicle 1, and functionally implements the prediction unit and execution unit, which will be described in more detail below.

The ECU 10 is electrically connected to an ignition switch 90, wheel rotation speed sensors 91-94, axle rotation speed sensor 95, and a longitudinal acceleration sensor 96. The ignition switch 90 detects the on/off state of the ignition. The wheel rotation speed sensors 91-94 detect the rotation speeds of the left rear wheel 61, right rear wheel 62, left front wheel 63, and right front wheel 64, respectively. The longitudinal acceleration sensor 96 detects the longitudinal acceleration of the vehicle 1. The wheel rotation speed sensors 91-94, axle rotation speed sensor 95, and longitudinal acceleration sensor 96 are examples of sensors related to the behavior of the vehicle 1.

In such a vehicle 1, resonance may occur while the vehicle 1 is traveling. Resonance in the vehicle 1 occurs primarily when the vehicle 1 is traveling on an undulating road. Resonance while traveling on an undulating road is thought to occur when the frequency at which the left rear wheel 61 and other wheels repeatedly spin and contact the ground due to forced fluctuations from the undulating road overlaps with the vehicle 1's unsprung resonance frequency and drivetrain resonance frequency. In particular, in this embodiment, the motor 2 is directly connected to the left rear wheel 61 and right rear wheel 62 without the intervention of a clutch or transmission. This makes it easy for vibrations to be transmitted from the road surface to the vehicle 1. In this embodiment, the ECU 10 predicts whether resonance will occur using a prediction model described below, and if it is predicted that the vehicle 1 will resonate, it executes countermeasures to deal with the vehicle 1's resonance. The prediction model is pre-stored in the ROM of the ECU 10.

[Prediction model]
The prediction model is a machine-learned prediction model using supervised learning, and inputs a calculated value (described below) before the vehicle 1 resonates, and outputs whether the vehicle 1 will resonate. The calculated value is calculated based on detection values of sensors related to the behavior of the vehicle 1. The sensors related to the behavior of the vehicle 1 are the wheel rotation speed sensors 91 to 94, the axle rotation speed sensor 95, and the longitudinal acceleration sensor 96 described above. For supervised learning, for example, any one or a combination of logistic regression, random forest, support vector machine, k-nearest neighbor method, neural network, and the like may be used.

The detected values are the longitudinal acceleration of the vehicle 1, and the rotational speeds of the left rear wheel 61, right rear wheel 62, left front wheel 63, right front wheel 64, and propeller shaft 3. Here, the propeller shaft 3 is an example of a rotating member that rotates in conjunction with the left rear wheel 61 and right rear wheel 62, which receive power from the motor 2, which is the power source for running the vehicle 1. Therefore, for example, instead of the rotational speed of the propeller shaft 3, the rotational speed of the rotor of the motor 2 may be used. Also, if a gear is provided between the motor 2 and the propeller shaft 3, the rotational speed of that gear may be used.

The calculated values are the average, maximum, minimum, variance, and standard deviation within a specified time period of the longitudinal acceleration of the vehicle 1, the rotational speed of the left rear wheel 61, the rotational speed of the right rear wheel 62, the longitudinal acceleration, the difference in rotational speed between the left rear wheel 61 and the right rear wheel 62, the difference in rotational speed between the left front wheel 63 and the right front wheel 64, the difference in rotational speed between the left rear wheel 61 and the left front wheel 63, and the difference in rotational speed between the right rear wheel 62 and the right front wheel 64.

Furthermore, the calculated values include the average value, maximum value, minimum value, variance value, and standard deviation value of the integrated value of the rotational fluctuation amount of the propeller shaft 3 within a specified time period. The integrated value of the rotational fluctuation amount is calculated as follows. The rotational fluctuation amount is calculated based on a signal obtained by removing noise from the rotational speed component of the propeller shaft 3 using bandpass filtering. The rotational fluctuation amount may be, for example, the difference between the target rotational speed and the actual rotational speed, or the difference between the maximum and minimum values of the rotational speed of the propeller shaft 3 while the propeller shaft 3 rotates a specified angle. The integrated value is calculated by integrating such rotational fluctuation amount within a specified time period. Such integrated value is calculated every specified time period, and the average value, maximum value, minimum value, variance value, and standard deviation value are calculated from these multiple integrated values.

Furthermore, the calculated values include the average, maximum, minimum, variance, and standard deviation within a predetermined time period of the difference between the vehicle speed of the vehicle 1 and the rotational speed of any one of the left rear wheel 61, right rear wheel 62, left front wheel 63, and right front wheel 64. The vehicle speed is estimated based on the rotational speed of at least one of the left rear wheel 61, right rear wheel 62, left front wheel 63, and right front wheel 64. For example, the vehicle speed may be the maximum rotational speed among the rotational speeds of the left rear wheel 61, right rear wheel 62, left front wheel 63, and right front wheel 64. Note that the vehicle speed may be estimated based on the rotational speed of the wheels, or may be estimated using a conventionally known method, such as an acceleration sensor.

Furthermore, the calculated values include the total number of peak values within a specified time period of the waveform after band-pass filtering of the rotational speed of the left rear wheel 61, the total number of peak values within a specified time period of the waveform after band-pass filtering of the rotational speed of the left front wheel 63, and the total number of peak values within a specified time period of the waveform after band-pass filtering of the longitudinal acceleration.

Figure 2 shows an example of a waveform after bandpass filtering of the rotation speed of the left rear wheel 61. Bandpass filtering is a process for removing noise from the detected value indicating the rotation speed of the left rear wheel 61, passing a predetermined frequency band including resonant frequency components and removing other noise components. As shown in Figure 2, the ECU 10 calculates the total number of peak values within a predetermined reading range within a specified time period as the calculated value.

When learning the prediction model, the vehicle 1 is driven on flat roads, uneven roads, cobblestone roads, and undulating roads. As described above, resonance is likely to occur when driving on undulating roads. For this reason, while driving on an undulating road and before resonance occurs, the detected values of the rotational speed of the left rear wheel 61, etc., and the longitudinal acceleration fluctuate slightly, and these fluctuations are reflected in the calculated values described above. In this way, the prediction model can be generated. As described below, the ECU 10 uses the prediction model to predict whether resonance will occur before it actually occurs, and executes countermeasures against resonance.

[Resonance countermeasure control]
3 is a flowchart showing an example of resonance countermeasure control. This control is repeatedly executed while the ignition is on. The ECU 10 acquires the detected values of the wheel rotation speed sensors 91 to 94, the axle rotation speed sensor 95, and the longitudinal acceleration sensor 96 (step S1), and calculates the above-mentioned calculation values based on these detected values (step S2).

Next, the ECU 10 uses the above-described prediction model to determine whether resonance is predicted to occur in the vehicle 1 (step S3). If it is determined that resonance will not occur (No in step S3), the ECU 10 turns off the resonance prediction flag (step S4) and terminates this control. If resonance is predicted to occur (Yes in step S3), the ECU 10 turns on the resonance prediction flag (step S5) and executes countermeasure processing against resonance (step S6). In this embodiment, the countermeasure processing is processing to limit the maximum torque of the motor 2, which will be described in detail later.

In this way, it is possible to predict the occurrence of resonance early on based on the prediction model, and if resonance is predicted to occur, it is possible to start countermeasure processing early.

[Countermeasures]
FIG. 4 is a timing chart showing an example of the countermeasure process. FIG. 4 shows the transition of the resonance prediction flag and the maximum torque of the motor 2. When resonance is predicted to occur and the resonance prediction flag is switched from off to on (time t1), the maximum torque of the motor 2 is immediately limited. In this way, the countermeasure process is executed immediately after the occurrence of resonance is predicted, making it possible to avoid resonance. The countermeasure process shown in FIG. 4 is suitable for vehicles with a relatively large maximum torque of the motor 2. This is because even if the maximum torque of the motor 2 is immediately limited in this way, the motor 2 has a large reserve capacity, so the impact on drivability is small.

Figure 5 is a timing chart showing another example of countermeasure processing. Figure 5 shows the progression of the resonance prediction flag, the integrated value of the rotational fluctuation amount described above, and the maximum torque of motor 2. As shown in Figure 5, when the resonance prediction flag is switched from off to on (time t1), the threshold value for the integrated value used to limit the maximum torque of motor 2 is switched from value α to value β, which is smaller than value α. As a result, when the integrated value becomes equal to or greater than value β (time t2), the maximum torque of motor 2 is limited. Value α is a threshold value used to limit the maximum torque of motor 2 based on the integrated value when it is predicted that resonance will not occur. Value β is a threshold value used to limit the maximum torque of motor 2 based on the integrated value when it is predicted that resonance will occur.

Therefore, if it is predicted that resonance will not occur, the maximum torque of motor 2 is limited when the integrated value reaches or exceeds value α (time t3). In this way, if the prediction model predicts that resonance will occur, the threshold is switched to value β, allowing the maximum torque of motor 2 to be limited early.

The countermeasure processing shown in Figure 5 is suitable for vehicles with a relatively small maximum torque of motor 2. This is because vehicles with a relatively small maximum torque have a small driving force, and limiting the maximum torque of motor 2 has a significant impact on drivability. It is also possible to limit the maximum torque when the rotation fluctuation integrated value begins to increase.

In the above embodiment, the vehicle 1 is an electric vehicle, but this is not limiting. For example, the vehicle may be an engine vehicle equipped with an engine as a power source for running. The vehicle may also be a hybrid vehicle equipped with an engine and a motor as a power source for running. In these vehicles, too, the occurrence of resonance can be suppressed by reducing the torque of these power sources for running as a countermeasure when resonance is predicted.

In the above embodiment, the motor 2 is directly connected to the left rear wheel 61 and the right rear wheel 62, but this is not limited to this, and the contents of the above embodiment can also be applied to mechanisms that are connected via a clutch, transmission, etc.

In the above embodiment, the detected values are the longitudinal acceleration, the rotational speeds of the left rear wheel 61, right rear wheel 62, left front wheel 63, right front wheel 64, and propeller shaft 3, but are not limited to these and may be at least one of the above values.

The calculated values may include at least one of the average, maximum, minimum, variance, and standard deviation within a specified time period of at least one of the following: the longitudinal acceleration of vehicle 1, the difference between the rotational speeds of two wheels of vehicle 1, the integrated value of the rotational fluctuation amount of a rotating member, and the difference between the body speed of vehicle 1 and the rotational speed of one of the wheels of vehicle 1.

The calculated value may include the total number of peak values within a predetermined time period of the waveform after bandpass filtering for at least one of the detected values.

Although the present invention has been described in detail above with reference to specific embodiments, the present invention is not limited to these specific embodiments, and various modifications and variations are possible within the scope of the invention as set forth in the claims.

1. Vehicle 2. Motor (power source for driving)
3. Propeller shaft (rotating member)
10 ECU (controller, prediction unit, execution unit)
61 Left rear wheel 62 Right rear wheel 63 Left front wheel 64 Right front wheel 91 to 94 Wheel rotation speed sensors (sensors)
95 Shaft rotation speed sensor (sensor)
96 Front and rear acceleration sensor (sensor)

Claims (5)

a calculation unit that calculates a calculation value based on a detection value of a sensor relating to a behavior of the vehicle;
a prediction unit that predicts whether the vehicle will resonate based on the calculated value using a prediction model that has been machine-learned through supervised learning and that receives the calculated value before the vehicle resonates as an input and outputs whether the vehicle will resonate; and
an execution unit that, when it is predicted that the vehicle will resonate, executes a countermeasure process for the vehicle resonance. the detected value includes at least one of a longitudinal acceleration of the vehicle, a rotational speed of a wheel of the vehicle, and a rotational speed of a rotating member that rotates in conjunction with the wheel that receives power from a power source for running the vehicle,
2. The vehicle control device of claim 1, wherein the calculated value includes at least one of an average value, a maximum value, a minimum value, a variance value, and a standard deviation value within a predetermined time of at least one of the longitudinal acceleration of the vehicle, the difference between the rotational speeds of two wheels of the vehicle, the integrated value of the rotational fluctuation amount of the rotating member, and the difference between the body speed of the vehicle and the rotational speed of one of the wheels of the vehicle. The vehicle control device of claim 2, wherein the calculated value includes the total number of peak values within a predetermined time period of the waveform of the detected value after bandpass filtering. The vehicle control device of claim 3, wherein the motor that is the power source for driving the vehicle is directly connected to one of the wheels of the vehicle. 5. A vehicle control device according to claim 1, wherein the execution unit, when it is predicted that the vehicle will resonate, reduces the torque of the vehicle's driving power source as the countermeasure processing compared to when it is not predicted that the vehicle will resonate.