Dog clutch engagement control system

Abstract

In the present invention, a clutch actuator (5) causes a first clutch member (11) and a second clutch member (12) of a dog clutch (10) to move in a relative fashion in the axial direction. A phase difference sensor (6) detects the phase difference between the first clutch member (11) and the second clutch member (12). When a predetermined prediction execution condition is satisfied during an ongoing execution of a synchronization operation in which the rotational speed of an input shaft (3) is boosted toward the rotational speed of an output shaft (4) so as to synchronize the rotation of the input shaft (3) and the output shaft (4) due to an engagement instruction for causing the dog clutch (10) in a disengaged state to become engaged, a control device (7) performs the following processing. The control device (7): detects an engagement time when the dog clutch (10) can be engaged on the basis of the output of the phase difference sensor (6); predicts a future engagement time in the light of one past engagement time and an estimated rotational speed of the input shaft; and drives the clutch actuator (5) in accordance with the predicted future engagement time.

Description

Dog clutch engagement control system CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on Japanese Application No. 2023-121475 filed on July 26, 2023, the contents of which are incorporated herein by reference.

[0002] The present disclosure relates to a dog clutch engagement control system.

[0003] Conventionally, there has been known a dog clutch engagement control system that predicts a future engagement time when a first clutch member and a second clutch member, which rotate at different rotational speeds in a released state, can be engaged. For example, Patent Document 1 discloses a technology that predicts a future engagement time based on multiple past engagement times (engagement timings) detected from the output of a phase difference sensor, and drives a clutch actuator accordingly.

[0004] Japanese Patent Application Laid-Open No. 2021-025561

[0005] In the technology of Patent Document 1, the control device stores the timings of multiple past engagements and calculates an approximation line passing through each point to predict future engagement times. Because multiple past engagement times are used, predicting future engagement times takes time, which may impair responsiveness.

[0006] It is an object of the present disclosure to provide a dog clutch engagement control system that improves clutch engagement responsiveness.

[0007] The dog clutch engagement control system of the present disclosure includes a dog clutch, a clutch actuator, a phase difference sensor, and a controller.

[0008] The dog clutch has a first clutch member and a second clutch member, and the first clutch member and the second clutch member are switched between an engaged state and a disengaged state. The first clutch member is connected to the input shaft and has a plurality of first engagement teeth arranged in the circumferential direction. The second clutch member is connected to the output shaft and has a plurality of second engagement teeth arranged in the circumferential direction, which can mesh with the first engagement teeth directly or via an intermediate member.

[0009] In a dog clutch engagement control system mounted on a vehicle, the input shaft is connected to a motor or an internal combustion engine via a reducer, for example, and the output shaft is connected to drive wheels via an axle.

[0010] The clutch actuator moves the first clutch member and the second clutch member relative to each other in the axial direction, or, if an intermediate member is used, moves the intermediate member relative to the first clutch member and the second clutch member in the axial direction. The phase difference sensor detects the phase difference between the first clutch member and the second clutch member. The control device outputs a drive command to the clutch actuator in accordance with the engagement timing at which the dog clutch can be engaged.

[0011] The control device performs the following processing when a predetermined predicted execution condition is met while a synchronization operation is currently being performed in which the rotation speed of the input shaft is increased toward the rotation speed of the output shaft so as to synchronize the rotation of the input shaft and the output shaft by issuing an engagement command to engage a dog clutch that is in a released state.

[0012] The control device of the first aspect detects the engagement timing at which the dog clutch can be engaged based on the output of the phase difference sensor.

[0013] The control device of the second aspect calculates an estimated phase difference between the input shaft and the output shaft based on an initial phase difference between the input shaft and the output shaft at the start of a synchronizing operation and a cumulative phase difference converted from a time integral value of the difference in rotation speed between the input shaft and the output shaft during execution of the synchronizing operation. Depending on whether the phase difference sensor can detect a phase difference, the control device switches between detecting an engagement time at which the dog clutch can be engaged based on the output of the phase difference sensor or estimating the engagement time from the estimated phase difference.

[0014] In both the first and second aspects, the control device predicts a future engagement time from the timing of one previous engagement and the estimated rotation speed of the input shaft, and drives the clutch actuator according to the predicted future engagement time.

[0015] Here, the "last engagement time" includes information recently acquired by the control device. Information used when making current and future predictions based on information recently acquired by the control device is considered to be "past" information. Compared to the prior art of Patent Document 1, the present disclosure can improve clutch engagement responsiveness by predicting future engagement time from the last engagement time and the estimated rotation speed of the input shaft.

[0016] The above and other objects, features, and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which Fig. 1 is a diagram of an example of the configuration of a vehicle to which a dog clutch engagement control system is applied, Fig. 2 is a schematic diagram showing switching between 2WD and 4WD, Fig. 3 is a time chart showing the operation of switching from 2WD to 4WD, Fig. 4 is a schematic diagram showing state transitions of a dog clutch, Fig. 5 is a configuration diagram of the dog clutch engagement control system of this embodiment, Fig. 6 is a diagram explaining the detection principle of a phase difference sensor, and Fig. 7 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 8 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 9 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 10 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 11 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 12 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 13 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 14 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 15 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 16 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 17 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 18 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of a dog clutch, and Fig. 19 is a diagram illustrating a problem in a comparative example in predicting the future engagement timing of 8 is a main flowchart of clutch engagement control according to this embodiment, FIG. 9 is a sub-flowchart of S8 (engagement operation start determination) in FIG. 8, FIG. 10 is an estimated rotation speed map showing the estimated rotation speed of the input shaft during synchronization operation, FIG. 11 is a time chart of clutch engagement control according to the first embodiment (when estimated rotation speed>actual rotation speed), FIG. 12 is a time chart of clutch engagement control according to the first embodiment (when estimated rotation speed<actual rotation speed), and FIG. 13 is a time chart of clutch engagement control according to the second embodiment. FIG. 14 is a diagram illustrating the phase difference information acquisition means according to the comparative example and the second embodiment; FIG. 15 is a diagram illustrating the calculation of the cumulative phase difference; FIG. 16 is a flowchart illustrating the time base correction of the estimated rotation speed map according to the third embodiment; FIG. 17 is a time chart illustrating the time base correction of the estimated rotation speed map; FIG. 18 is a time chart of the clutch engagement control after the estimated rotation speed is corrected; and FIG. 19 is a flow chart of the input shaft actual rotation speed learning according to the fourth embodiment. FIG. 20 is a diagram explaining updating of the estimated rotation speed map by learning the actual input shaft rotation speed, FIG. 21 is a flowchart of target rotation speed increase gradient switching according to the fifth embodiment, FIG. 22 is a time chart of the synchronization operation according to the fifth embodiment, FIG. 23 is a time chart of the synchronization operation according to a modified example of the fifth embodiment, FIG. 24 is a diagram explaining a phase difference detection method based on a phase difference sensor signal, FIG. 25 is a phase difference detection pattern (1) when the second gradient is small (a1>a2), and FIG. 26 is a diagram explaining a method of detecting a phase difference based on a phase difference sensor signal,FIG. 27 is a phase difference detection pattern (1) when the second gradient is large (a1≒a2), FIG. 28 is a phase difference detection pattern (2) when the second gradient is large (a1≒a2), FIG. 29 is a flowchart of setting the idle rotation speed according to the sixth embodiment, FIG. 30 is a time chart of the synchronization operation of the sixth embodiment, FIG. 31 is a diagram explaining an example of setting the target idle rotation speed depending on the vehicle speed, FIG. 32 is a diagram explaining an example of setting the target idle rotation speed depending on the operating state other than the vehicle speed, and FIG. 33 is a diagram explaining an example of setting the target idle rotation speed depending on the operating state other than the vehicle speed according to the seventh embodiment. FIG. 34 is a time chart of clutch engagement control in the seventh embodiment, FIG. 35 is a diagram showing an example of setting a precedence operation time based on a delay factor, FIG. 36 is a flowchart of switching between execution and non-execution of engagement timing prediction in the eighth embodiment, FIG. 37 is a diagram comparing the number of binary pulses of a beat waveform based on vehicle speed, FIG. 38 is a diagram showing the relationship between vehicle speed and detectable rotation speed difference and phase difference detection time, FIG. 39 is a diagram showing switching between execution and non-execution of engagement timing prediction based on vehicle speed, and FIG. 40 is a diagram showing switching between execution and non-execution of engagement timing prediction based on other parameters.

[0017] A dog clutch engagement control system according to multiple embodiments of the present disclosure will be described with reference to the drawings. The basic system configurations of the first to eighth embodiments are the same, but the control configurations of the control devices are different. The first to eighth embodiments are collectively referred to as "the present embodiment." The dog clutch engagement control system of the present embodiment is a system that predicts the future engagement time of a dog clutch provided in a vehicle powertrain system when the clutch is in a released state, and drives a clutch actuator according to the predicted future engagement time to engage the dog clutch.

[0018] 1 and 2, a configuration example of a vehicle 90 to which a dog clutch engagement control system is applied will be described. The vehicle 90 is an electric vehicle powered by two MGs (motor generators) 81, 82, one for the front wheels and one for the rear wheels. The MGs 81, 82 function as electric motors during power running and as generators during regenerative running. The vehicle 90 can switch between two-wheel drive (2WD) using only the front wheels 91 and four-wheel drive (4WD) using the front wheels 91 and the rear wheels 92.

[0019] The front-wheel MG 81 is always connected to the front wheels 91 via a differential gear 93 and a connecting shaft 95. Therefore, the front wheels 91 are always driven wheels. Meanwhile, a dog clutch 10 is provided in the power transmission path from the rear-wheel MG 82 to the rear wheels 92. In the example shown in FIG. 1 , the dog clutch 10 is provided between the rear-wheel MG 82 and the differential gear 94, but the dog clutch 10 may be provided closer to the rear wheels 92 than the differential gear 94.

[0020] When the dog clutch 10 is in a disengaged state, the rear wheel MG 82 is not connected to the rear wheels 92, and the rear wheels 92 rotate as driven wheels of the front wheels 91. When the dog clutch 10 is in an engaged state, the rear wheel MG 82 is connected to the rear wheels 92 via a differential gear 94 and a connecting shaft 96. At this time, the rear wheels 92 become drive wheels in addition to the front wheels 91. Note that speed reducers 87, 88 may be provided on the output shafts of the front wheel MG 81 and the rear wheel MG 82, respectively.

[0021] In this way, the vehicle 90 switches between 2WD and 4WD by switching the dog clutch 10 between the released state and the engaged state. For example, 2WD with 1MG driving is selected on a flat road with a low load, and 4WD with 2MG driving is selected on a slope requiring high driving force. When switching from 4WD to 2WD, it is necessary to reduce drag loss by disengaging the clutch and improve power consumption. When switching from 2WD to 4WD, it is necessary to achieve responsive acceleration performance by engaging a highly responsive clutch. If clutch engagement is delayed, the torque response of the disengaged rear wheel MG 82 is delayed, which delays the torque response of the entire vehicle. As a result, torque response to the driver's accelerator operation is delayed, affecting drivability.

[0022] When a higher-level vehicle control device (not shown) determines that a switch from 2WD to 4WD is necessary based on the vehicle's driving state, external environment such as the road surface, or a driver's instruction, it sends an engagement command to dog clutch engagement control system 100. When dog clutch engagement control system 100 receives the engagement command when dog clutch 10 is released, it engages dog clutch 10. Dog clutch engagement control system 100 includes dog clutch 10, clutch actuator 5, phase difference sensor 6, and control device 7. In the following drawings, clutch actuator 5 may be referred to as "ACT" as appropriate.

[0023] The dog clutch 10 has a first clutch member 11 connected to the input shaft 3 and a second clutch member 12 connected to the output shaft 4. The first clutch member 11 has a plurality of first engagement teeth 13 arranged in the circumferential direction and rotates about its axis. The second clutch member 12 has second engagement teeth 14 formed in the circumferential direction that can directly mesh with the first engagement teeth 13 and rotates coaxially with the first clutch member 11 in the same direction. In the configuration example of FIG. 1 , a reducer 88 is provided between the rear wheel MG 82 and the first clutch member 11. The rotation of the motor shaft 83 of the rear wheel MG 82 is reduced in speed by the reducer 88 and transmitted to the input shaft 3.

[0024] The clutch actuator 5 moves the first clutch member 11 and the second clutch member 12 relative to each other in the axial direction. The clutch actuator 5 is not limited to being provided on the first clutch member 11 side, but may also be provided on the second clutch member 12 side. When the first clutch member 11 and the second clutch member 12 move toward each other, they enter an engaged state in which the first engagement teeth 13 and the second engagement teeth 14 mesh with each other. When the first clutch member 11 and the second clutch member 12 move away from each other, they enter a released state in which they are no longer meshed with each other. In other words, the relative axial movement of the first clutch member 11 and the second clutch member 12 switches between the engaged state and the released state of the first clutch member 11 and the second clutch member 12. Note that a sleeve-type clutch other than this type of clutch will be described in the "Other Embodiments" section.

[0025] When the input shaft 3 and the output shaft 4 are rotating at different rotation speeds with the dog clutch 10 in a disengaged state, the phase difference sensor 6 detects the phase difference between the first clutch member 11 and the second clutch member 12, i.e., the phase difference between the input shaft 3 and the output shaft 4, and outputs a phase difference sensor signal to the control device 7. The dog clutch engagement control system 100 of this embodiment is mounted on a vehicle 90 in which the output shaft 4 is connected to rear wheels 92, which are drive wheels.

[0026] Hereinafter, the "time when the dog clutch 10 is engageable" will be referred to as the "engagement time." The "engagement time" includes not only a single time when an engagement operation is actually performed, but also multiple times when engagement is possible but the clutch is not engaged. The control device 7 detects past engagement times based on the phase difference sensor signal, predicts future engagement times using a method described below, and drives the clutch actuator 5 according to the predicted future engagement time. More specifically, the control device 7 outputs a drive command to the clutch actuator 5 prior to the future engagement time by an advance operation time T_ad required for the operation of the clutch actuator 5. The control device 7 also controls the drive of the rear wheel MG 82, thereby controlling the rotation speed of the input shaft 3.

[0027] A typical engagement operation when switching from 2WD to 4WD will be described with reference to Figures 3 and 4. Figure 3 shows the changes in input and output shaft rotation speeds and clutch actuator stroke. The output shaft rotation speed corresponds to the rotation speed of the axle, which is proportional to the vehicle speed. During period I from time t0 to t1, the control device 7 receives a switching command and determines the start of the switching operation. The command to switch from 2WD to 4WD generates an "engagement command to engage a dog clutch that is in a released state."

[0028] In response to the engagement command, the control device 7 drives the rear wheel MG 82 during period II from time t1 to t2, and increases the input shaft rotation speed Nin toward the output shaft rotation speed Nout so as to synchronize the rotation of the input shaft 3 with the rotation of the output shaft 4. This operation of matching the rotation speeds of the input and output shafts is called a "synchronization operation." During periods I and II, the dog clutch 10 is in a fully released state as shown in FIG. 4. Note that the symbols for times t0, t1, etc. on the time axis of each of the following time charts will be used independently for each diagram.

[0029] A target rotational speed difference ΔN_tgt between the input shaft rotational speed and the output shaft rotational speed is set to reflect the allowable limit of engagement shock when the dog clutch 10 is engaged. When the input / output shaft rotational speed difference reaches or falls below the target rotational speed difference ΔN_tgt at time t2, a drive command is output to the clutch actuator 5. During period III from time t2 to t3, the clutch stroke changes from 0 to Sto, and the dog clutch 10 transitions from the fully released state in FIG. 4 to the boundary engagement state.

[0030] Phase alignment of the clutch members 11, 12 is performed during a period IV from time t3 to t4. The time required for phase alignment may be extended depending on the initial phase of the dog clutch 10. Once phase alignment is complete, the clutch stroke changes from Sto to Stf during a period V from time t4 to t5, and the dog clutch 10 transitions from the boundary state of engagement shown in FIG. 4 to the fully engaged state.

[0031] Of the total switching time, which is the sum of periods I to V, the time required for speed matching during period II and phase matching during period IV are dominant. In the conventional technology of Patent Document 1 (JP 2021-025561 A), the phase matching time is shortened by predicting meshing timing based on a phase difference sensor signal. In this embodiment, the objective is to further shorten the phase matching time compared to the conventional technology of Patent Document 1 and achieve highly responsive switching from 2WD to 4WD by clutch engagement.

[0032] The configuration of the dog clutch engagement control system of this embodiment will be described with reference to Figures 5 and 6. From Figure 5 onwards, we will focus only on clutch engagement on the rear wheel 92 side of the vehicle 90 and will not refer to the regenerative operation of the MG, so the rear wheel MG 82 will be simply referred to as the "motor 82." The rotation speed of the motor 82 is detected by a rotation speed sensor 23 such as a resolver and converted to the rotation speed of the input shaft 3 by multiplying it by the reduction ratio of the reducer 88. In Figure 5, the converted input shaft rotation speed is input to the engagement timing prediction unit 72 of the control device 7. Meanwhile, the rotation speed of the output shaft 4 is detected by a rotation speed sensor 24 such as a wheel speed sensor, and the converted value is input to the engagement timing prediction unit 72. Hereinafter, the reference numerals for "input shaft 3" and "output shaft 4" relating to the rotation speeds will be omitted as appropriate.

[0033] The phase difference sensor 6 is similar to that disclosed in Patent Document 1 and is composed of a magnetic detection element such as a Hall element and a magnet. The phase difference sensor 6 has a detection range SA (see FIG. 6 ) in the axial direction that straddles the first engagement teeth 13 and the second engagement teeth 14, and is disposed so as to face the clutch axis Z from the radial outside so as not to interfere with the dog clutch 10. The phase difference sensor 6 detects the total area of ​​the first engagement teeth 13 and the second engagement teeth 14 that pass through the detection range SA as the dog clutch 10 rotates, based on changes in magnetic flux intensity.

[0034] The pitch angle P is the angle obtained by dividing the angle of one rotation of the first clutch member 11 and the second clutch member 12 (i.e., 360 degrees) by the number of teeth. For example, the number of teeth of the first engagement teeth 13 and the second engagement teeth 14 is 36, and the pitch angle P is 10 degrees. In Figure 6 and the following figures, an example configuration with 36 teeth is assumed, and the range of the phase difference is shown as ±5 degrees.

[0035] The top diagram in Figure 6 shows the general phase difference Δθ during rotation. The middle diagram shows a state where the rotational phases of the first engagement teeth 13 and the second engagement teeth 14 are the same, that is, "phase difference Δθ = 0." In this state, the first clutch member 11 and the second clutch member 12 cannot engage with each other. The detection range SA includes both the tooth portions of the first engagement teeth 13 and the second engagement teeth 14, and the sensor output is maximized when the total area is maximized. The detection range SA includes both the gap portions of the first engagement teeth 13 and the second engagement teeth 14, and the sensor output is minimized when the total area is minimized.

[0036] The lower diagram in Figure 6 shows a state in which the rotational phase between the first engagement teeth 13 and the second engagement teeth 14 is shifted by half the pitch angle P, resulting in a "phase difference Δθ = ±(½)P (= ±5 deg)." In this state, the first clutch member 11 and the second clutch member 12 are engageable. The detection range SA includes one tooth portion of the first engagement teeth 13 and the second engagement teeth 14 and the gap portion of the other, and the sensor output becomes an intermediate value at the timing when the total area becomes an intermediate value between the maximum and minimum values.

[0037] Returning to FIG. 5 , the control device 7 includes a filter unit 71, an engagement timing prediction unit 72, an actuator drive unit 73, and a motor control unit 74. The filter unit 71 filters the output of the phase difference sensor 6 to remove fluctuation components in a predetermined frequency range. As a result, the filtered phase difference sensor output signal disclosed in, for example, FIG. 4(b) of Patent Document 1 is input to the engagement timing prediction unit 72. The node of the beat waveform in the phase difference sensor output signal corresponds to the engagement timing. The filter unit 71 may be configured as a circuit using filter elements, or may be configured as software. The relationship between the time constant of the filtering process and the anticipatory operation time T_ad will be described later in a seventh embodiment.

[0038] The engagement time prediction unit 72 acquires the filtered phase difference sensor signal, the input shaft rotation speed, and the output shaft rotation speed. When the engagement time prediction unit 72 receives an engagement instruction when the dog clutch 10 is released, it causes the motor control unit 74 to rotate the motor 82, and starts a synchronization operation to increase the input shaft rotation speed to a target rotation speed. The engagement time prediction unit 72 also internally stores an estimated rotation speed map, which will be described later. The engagement time prediction unit 72 predicts future engagement times based on the phase difference sensor signal and the estimated rotation speed, and outputs a drive instruction to the actuator drive unit 73.

[0039] Furthermore, in the sixth embodiment, for example, the motor control unit 74 sets the target idle rotation speed based on the vehicle speed, navigation information, camera information in front of the vehicle, weather information, etc. Details will be described later.

[0040] The actuator driving unit 73 drives the clutch actuator 5 in the forward direction at the timing instructed by the engagement timing prediction unit 72, thereby engaging the dog clutch 10. Furthermore, upon receiving a release instruction from a higher-level vehicle control device, the actuator driving unit 73 drives the clutch actuator 5 in the reverse direction, thereby disengaging the dog clutch 10.

[0041] For convenience, in Figure 5, the motor control unit 74 is shown as part of the control unit 7, but in reality, the control unit 7 is made up of a control unit of the dog clutch engagement control system 100 and a separate MG control unit that are functionally linked together. The MG control unit controls the power supply from the power source to the motor 82 by operating the inverter, with the aim of mainly controlling drive during regenerative powering of the rear wheels 92 after clutch engagement. In contrast, the motor control unit 74 in this embodiment exclusively controls the input shaft rotation speed during synchronization before clutch engagement. In other words, it is considered that at least the part of the MG control unit that performs synchronization operation corresponds to the motor control unit 74.

[0042] Next, referring to Figure 7, the problems of the conventional technology disclosed in Patent Document 1 will be described as a comparative example. As shown in Figures 3 to 5 of Patent Document 1, in the technology of the comparative example, while the rotation speed of the motor is increasing toward the rotation speed of the axle, the control device stores two or more past engagement times (meshing timings Tp) from the output of the phase difference sensor signal and calculates an approximate line passing through each point. The control device predicts the future engagement time (meshing timing Tf) along the approximate line and adds a correction value Cv corresponding to the rotation speed difference to the predicted value.

[0043] As the input shaft rotation speed Nin increases toward the output shaft rotation speed Nout during synchronization, the input / output shaft rotation speed difference ΔNio decreases. When the input shaft rotation speed Nin reaches the minimum detectable rotation speed Nmin_sen, it becomes possible to detect the phase difference at the sampling interval of the phase difference sensor 6. The input / output shaft rotation speed difference ΔNio at this time is represented as the detectable rotation speed difference ΔN_sen. Furthermore, when the input shaft rotation speed Nin reaches the minimum engageable rotation speed Nmin_eng, engagement within the allowable limit of engagement shock becomes possible. The input / output shaft rotation speed difference ΔNio at this time is represented as the target rotation speed difference ΔN_tgt. The "rotation speed difference" is defined as 0 or a positive value.

[0044] In the following description, "present" refers to the timing at which future engagement times are predicted during synchronization. When a prediction is made currently using recently acquired information, the information used is considered to be "past" information. The current input shaft rotation speed must be equal to or greater than the minimum engageable rotation speed Nmin_eng.

[0045] In Figure 7 and subsequent figures, the past engagement times used for prediction are represented by the symbol "Eng_p," and the predicted future engagement times are represented by the symbol "Eng_f." The upper diagram in Figure 7 shows a comparative example of predicted times when future engagement times are predicted from the past two engagement times. The first and second past engagement times are represented by "Eng_p1" and "Eng_p2," respectively.

[0046] After two previous engagement times Eng_p1 and Eng_p2 have been detected, the future engagement time Eng_f is predicted from the approximate line calculated by the calculation device. In the comparative example, the prediction takes time because the history of two or more previous engagement times is required. As a result, it is not possible to achieve highly responsive clutch operation, and drivability may be reduced.

[0047] Therefore, the aim of this embodiment to solve the problem is to predict future engagement timing Eng_f from one previous engagement timing Eng_p, thereby shortening the prediction time, as shown in the diagram below. To achieve this, instead of calculating an approximation line from multiple previous engagement timings, the estimated rotation speed of the input shaft is used. The control device of this embodiment predicts future engagement timing from one previous engagement timing and the estimated rotation speed of the input shaft and drives the clutch actuator, thereby achieving highly responsive clutch engagement and improving drivability.

[0048] However, the control device 7 that receives the engagement command does not always predict the future engagement time. The control device 7 of this embodiment predicts the future engagement time using the above method when predetermined prediction execution conditions are met. Examples of when the prediction execution conditions are met and when they are not met will be described later in the eighth embodiment.

[0049] (First Embodiment) Next, specific configurations of multiple embodiments of the dog clutch engagement control system 100 will be described. The first and second embodiments represent two basic technologies that form the backbone of the system, while the third to eighth embodiments represent applied technologies that are appropriately added to the first and second embodiments. First, with reference to the flowcharts in Figures 8 and 9, the basic steps of clutch engagement control that are common to each embodiment will be described. In the following explanation of the flowcharts, the symbol "S" denotes a step. S5, indicated by a dashed line, is not performed in the first embodiment, but is performed in the second embodiment.

[0050] In conjunction with the explanation of the flowchart, the estimated rotation speed map in Figure 10 and the time charts in Figures 11 and 12 will be referred to. Of the four diagrams in each of Figures 11 and 12, the top diagram shows changes in the input shaft rotation speed. In this embodiment, the output shaft rotation speed (i.e., vehicle speed) is basically constant, and the input shaft rotation speed increases toward the "target rotation speed," which is a rotation speed equivalent to the output shaft rotation speed. In the time chart of the synchronization operation, the "rotation speed" that changes with time t is the input shaft rotation speed. The actual input shaft rotation speed is represented as "real rotation speed f_real(t)," and the input shaft rotation speed estimated from the output characteristics of the motor 82 is represented as "estimated rotation speed f_n(t)."

[0051] Ideally, the estimated rotation speed f_n(t) should match the actual rotation speed f_real(t), but in reality, they may not match. Fig. 11 shows a case where the estimated rotation speed f_n(t) exceeds the actual rotation speed f_real(t), and Fig. 12 shows a case where the estimated rotation speed f_n(t) is lower than the actual rotation speed f_real(t).

[0052] The difference between the target rotation speed and the estimated rotation speed is represented as the “estimated rotation speed difference ΔN_n,” and the difference between the target rotation speed and the actual rotation speed is represented as the “actual rotation speed difference ΔN_real.” The definitions of the detectable minimum rotation speed Nmin_sen, the engageable minimum rotation speed Nmin_eng, the detectable rotation speed difference ΔN_sen, and the target rotation speed difference ΔN_tgt are as shown in FIG.

[0053] The second diagram from the top shows the change in phase difference. As described above with reference to FIG. 6 , when the number of teeth is 36, the phase difference Δθ varies within a range of ±5° centered around 0°. Now that the rotation speed difference has fallen below the minimum detectable rotation speed Nmin_sen, the timing at which "Δθ = ±5°" is detected as the "past engagement timing Eng_p used for prediction." In addition, an estimated phase difference, indicated by the dashed line, is calculated based on the estimated rotation speed. After the present, the timing at which the estimated phase difference will be ±5° is predicted as the "future engagement timing Eng_f."

[0054] The third diagram from the top shows the ON / OFF of the drive command by the actuator drive unit 73. The bottom diagram shows the stroke of the clutch actuator 5. The clutch actuator 5 starts to be driven at a timing that is earlier than the future engagement timing Eng_f by the advance operation time T_ad.

[0055] The main flowchart in FIG. 8 starts with an engagement command input at time t0. At this time, the target rotation speed of the input shaft rises in a stepwise manner from 0. In S1, synchronization of the input and output shafts is started. In S2, an estimated rotation speed f_n(t) of the input shaft is calculated. As shown in FIG. 10, the estimated rotation speed f_n(t) is stored in an estimated rotation speed map that represents the step response to the target rotation speed as a function of time t. The estimated rotation speed of the input shaft at time tn from engagement command time t0 is represented as f_n(tn).

[0056] The estimated rotation speed may be calculated using a formula, or the nominal actual response value of the motor 82 may be applied. Since the target rotation speed correlates with the vehicle speed, multiple estimated rotation speed maps are stored according to the vehicle speed. Also, multiple maps may be stored for the temperature and drive voltage of the motor 82. By using an estimated rotation speed map according to the conditions, the error between the map value and the rotation speed of the actual behavior is reduced, improving the accuracy of the engagement timing prediction in S7.

[0057] In S3, the engagement timing prediction unit 72 acquires information on one or more engagement timings. In S4, it is determined whether the difference between the current actual rotation speed and the target rotation speed is equal to or less than the detectable rotation speed difference ΔN_sen and whether one previous engagement timing to be used in predicting future engagement timings has been acquired. If the result in S4 is NO, the process returns to before S3, and S3 is repeated until the result in S4 is YES.

[0058] S5, indicated by a dashed line, will be described later in a second embodiment. In S6, the engagement timing prediction unit 72 predicts a future phase difference from an estimated rotation speed difference, which is the difference between the estimated rotation speed and the target rotation speed. The phase difference at each future time is calculated as the sum of the current phase difference and a cumulative phase difference calculated from the integral of the estimated rotation speed difference from the present to the future. The formula for calculating the cumulative phase difference is the same as the formula for calculating the phase difference in a phase difference undetectable region, which will be described later in a third embodiment.

[0059] In S7, the engagement timing prediction unit 72 predicts, as the future engagement timing, the engageable timing at which the predicted phase difference is "Δθ = ±P / 2" (see FIG. 6). In the example shown in FIG. 11, the engageable timing closest to the present is predicted as the future engagement timing. However, the engageable timing predicted is not limited to the engageable timing closest to the present, and the engageable timing that is the second or subsequent time from the present may also be predicted as the future engagement timing.

[0060] In S8, it is determined whether the engagement operation of the clutch actuator 5 can be started. Figure 9 shows a sub-flowchart for determining whether to start the engagement operation. In S81, a target rotation speed difference ΔN_tgt is set according to the allowable limit of the engagement shock. For simplicity, this embodiment assumes that the vehicle speed, i.e., the output shaft rotation speed, is constant during the synchronization operation. However, in reality, the vehicle may accelerate or decelerate during the synchronization operation. The concept of the target rotation speed difference ΔN_tgt in this case will be described in the "Other Embodiments" section.

[0061] In S82, it is determined whether one of the following two conditions A and B is met. The rotation speed differences are compared in absolute value. [Condition A]: Estimated rotation speed difference ΔN_n at future engagement time≦target rotation speed difference ΔN_tgt [Condition B]: Current actual rotation speed difference ΔN_real≦target rotation speed difference ΔN_tgt

[0062] In the example of Fig. 11 where the estimated rotation speed is higher than the actual rotation speed, Condition B is not satisfied but Condition A is satisfied. In the example of Fig. 12 where the estimated rotation speed is lower than the actual rotation speed, Condition A is not satisfied but Condition B is satisfied.

[0063] The determination in S82 is repeated until a YES determination is made. If a YES determination is made in S82, a determination is made in S83 that the engaging operation can be started. Then, the process returns to the main flowchart of FIG. 8, and a YES determination is made in S8.

[0064] In S9, the engagement timing prediction unit 72 sets a pre-emptive operation time T_ad according to the delay time required for the clutch actuator 5 to transition from the released state to the engagement boundary state (see FIG. 4) after receiving a drive command. The standard pre-emptive operation time T_ad is stored as a default value. In the seventh embodiment, the pre-emptive operation time T_ad is calculated each time using parameters. In S10, the actuator drive unit 73 outputs a drive command to the clutch actuator 5 at a timing that is earlier than the predicted future engagement time by the pre-emptive operation time T_ad.

[0065] Second Embodiment Switching of the phase difference information acquisition means according to the second embodiment will be described with reference to Figures 13 to 15. As described above with reference to Figure 11 and other figures, the phase difference cannot be detected by the phase difference sensor 6 unless the input shaft rotation speed Nin is equal to or greater than the minimum detectable rotation speed Nmin_sen and the input / output shaft rotation speed difference is equal to or less than the detectable rotation speed difference ΔN_sen. In other words, because the start of prediction of the engagement time depends on the detectable rotation speed difference ΔN_sen, prediction of the engagement time is delayed. Furthermore, if the input / output shaft rotation speed difference exceeds the detectable rotation speed difference ΔN_sen during engagement time prediction, prediction cannot continue.

[0066] Therefore, in the second embodiment, when a predetermined prediction execution condition is met while a synchronization operation is currently being performed, the control device 7 switches the means for acquiring phase difference information depending on whether the phase difference can be detected by the phase difference sensor, which is determined from the current input / output shaft rotational speed difference. Specifically, the control device 7 calculates an estimated phase difference between the input shaft and the output shaft based on the initial phase difference between the input shaft and the output shaft at the start of the synchronization operation and the cumulative phase difference converted from the time integral of the rotational speed difference between the input shaft and the output shaft during the synchronization operation. Depending on whether the phase difference can be detected by the phase difference sensor, the control device 7 switches between detecting the engagement timing based on the output of the phase difference sensor and estimating the engagement timing from the estimated phase difference.

[0067] The flowchart of Fig. 13 is related to S3 "Obtain engagement timing information" in the main flowchart of Fig. 8, and includes steps S31 to S33. In S31, it is determined whether the input / output shaft rotational speed difference ΔNio is equal to or less than the detectable rotational speed difference ΔN_sen. If the answer is YES in S31, in S32, the engagement timing prediction unit 72 obtains phase difference information based on the phase difference sensor signal. If the answer is NO in S31, in S33, the engagement timing prediction unit 72 obtains phase difference information from the initial phase difference and the accumulated phase difference obtained by integrating the input / output shaft rotational speed difference.

[0068] 14, the transition of the phase difference information acquisition means when the input / output shaft rotational speed difference ΔNio changes from a state equal to or less than the detectable rotational speed difference ΔN_sen to a state where it exceeds the detectable rotational speed difference ΔN_sen, and then returns to a state equal to or less than the detectable rotational speed difference ΔN_sen will be described. The middle diagram shows the control of a comparative example in which the phase difference information acquisition means is the phase difference sensor signal alone. The bottom diagram shows the control of a second embodiment in which the phase difference information acquisition means switches between the phase difference sensor signal and integral calculation of the input / output shaft rotational speed difference.

[0069] As the input shaft rotation speed Nin decreases toward 0, the input / output shaft rotation speed difference ΔNio exceeds the detectable rotation speed difference ΔN_sen at time tz. As the input shaft rotation speed Nin increases again in response to an engagement command after the input shaft has stopped, the input / output shaft rotation speed difference ΔNio becomes equal to or less than the detectable rotation speed difference ΔN_sen at time ta. The period from time tz to time ta is an undetectable region. In the comparative example, after an engagement command is issued, it is only after the undetectable region has been passed that the engagement timing can be predicted. Thereafter, the future engagement timing Eng_f is predicted using the past engagement timing Eng_p.

[0070] In contrast, in the second embodiment, the phase difference sensor detection value immediately before time tz is stored as the initial phase difference. Furthermore, while the input shaft is rotating in the non-detection region, the cumulative phase difference is calculated by integrating the converted value of the input / output shaft rotation speed difference ΔNio, and the value obtained by adding the cumulative phase difference to the initial phase difference is estimated as the phase difference. Therefore, at time ta, it becomes possible to predict future engagement timing Eng_f using the most recent engagement timing Eng_p. This may make it possible to predict future engagement timing Eng_f earlier than in the comparative example.

[0071] An example of calculating the cumulative phase difference will be described with reference to FIG. 15 . The input / output shaft rotation speed difference is expressed as a function of time f(t). The rotation speed difference f(t) [rpm] is divided by the number of teeth using equation (1) in FIG. 15 to convert it into a phase change per hour [deg / sec]. The input / output shaft rotation speed difference is integrated over the interval from time α to time β using equation (2) in FIG. 15 . As shown in equation (3) in FIG. 15 , the integral value of the input / output shaft rotation speed difference is divided by the pitch angle P (= 360 / number of teeth), and the remainder (mod (360 / number of teeth)) is calculated as the cumulative phase difference. The input / output shaft rotation speed difference may also be subjected to a Hilbert transform.

[0072] In the second embodiment, in the undetectable region where the input / output shaft rotational speed difference ΔNio exceeds the detectable rotational speed difference ΔN_sen, the phase difference is estimated by integrating the input / output shaft rotational speed difference, thereby shortening the time required to predict the future clutch engagement timing, thereby shortening the total clutch engagement time from the start of synchronization.

[0073] Third Embodiment Time axis correction of the estimated rotation speed map according to a third embodiment will be described with reference to Figures 16 to 18. In the first embodiment, the engagement timing prediction unit 72 always uses a pre-stored estimated rotation speed map to predict future engagement timing Eng_f from the past engagement timing Eng_p. However, due to a discrepancy between the estimated rotation speed in the pre-stored map and the actual rotation speed, an error may occur in the future engagement timing Eng_f predicted using the map. Therefore, in the third embodiment, the estimated rotation speed map is corrected based on the actual rotation speed.

[0074] The flowchart of Fig. 16 shows the details of S5 "correct the estimated rotation speed", which is executed when the determination in S4 of the main flowchart of Fig. 8 is YES. In Fig. 17, the two-dot chain line indicates the actual rotation speed f_real(t), the thin dashed line indicates the estimated rotation speed f_n(t) before correction, and the thick dashed line indicates the estimated rotation speed f_n(t) after correction. ＊ The time from the engagement command to the previous engagement timing Eng_p used for prediction is referred to as an engagement timing appearance time t_sen.

[0075] After the previous engagement timing is obtained, in S51, the actual rotation speed f_real(t_sen) at the engagement timing appearance time t_sen is obtained. In S52, an equivalent time t_comp such that "f_n(t_comp) = f_real(t_sen)" is calculated using a map of the pre-correction estimated rotation speed f_n(t). In Figure 17, the point of f_real(t_sen) is moved horizontally, and the time value at the point where it intersects with the map of the estimated rotation speed f_n(t) is the equivalent time t_comp.

[0076] In S53, a time difference Δt_comp between the corresponding time t_comp and the engagement timing appearance time t_sen is calculated. In S54, it is determined whether the time difference Δt_comp is equal to or greater than a correction execution threshold, and if NO, the process ends without performing correction. If YES in S54, in S55, the ratio of the engagement timing appearance time t_sen to the corresponding time t_comp is calculated as a correction magnification k. In S56, the time axis of the estimated rotation speed map is corrected by a factor of k. In other words, the estimated rotation speeds f_n(t) and f_n before and after correction are ＊ The relationship with (t) is "f_n ＊ (t)=f_n(k×t)" and the time axis is corrected by k times.

[0077] As shown in FIG. 18, the corrected estimated rotation speed f_n ＊ (t) closely matches the actual rotation speed f_real(t). In this way, the control device 7 stores an estimated rotation speed map that defines changes in the estimated rotation speed with respect to the drive time of the input shaft during synchronization. During synchronization, the control device 7 obtains the actual rotation speed f_real(t_sen) of the input shaft at the previous engagement time, and calculates a corrected estimated rotation speed by multiplying the time axis of the estimated rotation speed map by a correction factor k so that the estimated rotation speed f_n(t_comp) approaches the actual rotation speed.

[0078] In the third embodiment, the prediction of future engagement timing is improved by using an estimated rotation speed map whose time axis is corrected in accordance with actual behavior. Note that if the rotation speed axis (vertical axis) of the estimated rotation speed map is corrected instead of the time axis (horizontal axis), the convergence of the rotation speed difference with respect to the target rotation speed changes, and the effect of reducing future estimation errors due to the smaller rotation speed difference is small. In contrast, in the third embodiment, the time axis of the map is corrected, so the convergence of the rotation speed difference is maintained, and the effect of reducing estimation errors is large.

[0079] 19 and 20, learning of the actual input shaft rotation speed according to a fourth embodiment will be described. The estimated rotation speed in a map stored in advance under standard conditions may not match the actual input shaft rotation speed due to individual differences in the motor 82, variations in temperature, power supply voltage, etc. Therefore, in the fourth embodiment, the relationship between the drive time and the actual rotation speed when increasing the input shaft rotation speed relative to the target rotation speed with the dog clutch 10 in a disengaged state is learned, and the estimated rotation speed map is updated based on the learning result.

[0080] In S00 of the flowchart in Figure 19, it is determined whether the dog clutch 10 is in a released state, and if the answer is YES, the process proceeds to S061, where learning begins. The time chart in Figure 20 shows the change in the actual input shaft rotation speed (0, X1n to Xnn) from learning start time t0 to learning completion time tn. t1 to tn represent time as well as the time from time t0.

[0081] At time t0, when the target rotation speed is stepped up from 0 rpm to the set value in S061, the engagement timing prediction unit 72 starts learning. In S062, the engagement timing prediction unit 72 acquires a profile of the time from time t0 and the actual input shaft rotation speeds X1n to X3n from time t1 to t3. In S063, the engagement timing prediction unit 72 saves the learned estimated rotation speed map. In other words, the previous estimated rotation speed map is updated. In S064, the target rotation speed is returned to 0 rpm at time tn.

[0082] In the fourth embodiment, variations in the accuracy of estimating the input shaft rotation speed due to factors such as the motor temperature and the power supply voltage can be absorbed through learning, thereby improving the estimation accuracy.

[0083] Fifth Embodiment Switching of the increase gradient of the target rotation speed of the input shaft according to a fifth embodiment will be described with reference to Figures 21 to 28. "Target rotation speed" refers to a target value of the input shaft rotation speed, and "gradient" refers to the rate of change over time. During synchronization, the control device 7 increases the input shaft rotation speed to a final target rotation speed that corresponds to the "driving state of the vehicle," such as vehicle speed. The final target rotation speed in the fifth embodiment is the minimum engageable rotation speed at which the clutch can be engaged within the allowable limit of engagement shock, and is determined according to the output shaft rotation speed converted from vehicle speed.

[0084] During synchronization, it is preferable to define a linear profile of the target rotation speed as a function of time from the start and increase the input shaft rotation speed along the linear profile in order to improve estimation accuracy. However, if the maximum gradient corresponding to the maximum torque of the motor 82 is uniformly applied over the entire period to increase the rotation speed along a single linear profile, the phase difference detection time becomes long, and there are cases where highly responsive clutch engagement is not possible. The reason for this will be explained in detail later.

[0085] Therefore, in the fifth embodiment, the synchronization operation is divided into a plurality of periods, including a first period and a second period, and the target rotation speed is increased along a polygonal profile formed by connecting the linear gradients of the respective periods. The first period is a period from the start of the synchronization operation during which the input shaft rotation speed increases to the minimum rotation speed at which the phase difference can be detected by the phase difference sensor 6, i.e., the minimum detectable rotation speed Nmin_sen (see FIG. 7 , etc.). The second period follows the first period and is a period during which the input shaft rotation speed further increases from the minimum detectable rotation speed Nmin_sen. The second gradient in the second period is set smaller than the first gradient in the first period.

[0086] Please refer to Figures 21 and 22. The flowchart in Figure 21 is related to S1 "Start of synchronization operation" in the main flowchart in Figure 8, and includes steps numbered S12 to S17. The upper part of Figure 22 shows a polygonal line profile of the target rotation speed. This profile is switched depending on the "driving state of the vehicle" such as the vehicle speed. The symbols "t1" and "t2" in Figure 22 are used independently from the figures of other embodiments.

[0087] During a first period Tup1 from the start of synchronization t0 to time t1, the input / output shaft rotation speed Nin increases from 0 rpm to the minimum detectable rotation speed Nmin_sen at a first gradient a1. During a second period Tup2 from time t1 to time t2, the input / output shaft rotation speed Nin increases at a second gradient a2 from the minimum detectable rotation speed Nmin_sen to the minimum engageable rotation speed Nmin_eng, which is the final target rotation speed. After time t2, the input / output shaft rotation speed Nin is maintained at the minimum engageable rotation speed Nmin_eng.

[0088] The lower part of Figure 22 compares the first gradient a1 and the second gradient a2. The gradient values ​​are expressed in units of [rpm / sec], i.e., the rotational speed-time conversion rate. The first gradient a1 and the second gradient a2 are positive values, and the second gradient a2 is smaller than the first gradient a1 (0<a2<a1). The first gradient a1 corresponds to the maximum torque of the motor 82.

[0089] The processing of the flowchart is mainly performed by the "controller 7." In S12, the controller 7 acquires the first gradient a1, the second gradient a2, the minimum detectable rotation speed Nmin_sen, and the minimum engageable rotation speed Nmin_eng as parameters for the synchronization operation in accordance with the driving state such as the vehicle speed. In S13, after starting the synchronization operation, the controller 7 increases the input shaft rotation speed Nin at the first gradient a1. In S14, it is determined whether the input shaft rotation speed Nin is equal to or greater than the minimum detectable rotation speed Nmin_sen. S13 is repeated until a YES determination is made in S14.

[0090] In S15, the control device 7 increases the input shaft rotation speed Nin at a second gradient a2 during a second period Tup2. In S16, it is determined whether the input shaft rotation speed Nin is equal to the minimum engageable rotation speed Nmin_eng. S15 is repeated until a positive determination is made in S16. After a positive determination is made in S16, the control device 7 maintains the input shaft rotation speed Nin at a gradient of 0 in S17.

[0091] In this way, during the first period Tup1, the input shaft rotation speed Nin increases toward synchronization at the highest speed. During the second period Tup2, during which phase difference detection by the phase difference sensor 6 becomes possible, the second gradient a2 is set so as to minimize the phase difference detection time. Thus, in the fifth embodiment, the total engagement time is shortened by shortening the synchronization time and the phase difference detection time. Furthermore, the target rotation speed profile is set so as to achieve highly responsive clutch engagement in accordance with the vehicle operating conditions, such as vehicle speed.

[0092] FIG. 23 shows a time chart of the synchronization operation according to a modification of the fifth embodiment. The first gradient a1 in the first period Tup1 and the second gradient a2 at the beginning of the second period Tup2 are the same as those in the basic example of FIG. 22 . In this modification, when prediction of the future engagement time is completed at time ts during the second period Tup2, the second period Tup2 ends and the third period Tup3 begins. In the third period Tup3, the gradient is switched to a third gradient a3, which is larger than the second gradient a2, and the input shaft rotation speed Nin increases to the minimum engageable rotation speed Nmin_eng. For example, the third gradient a3 is set to be equal to the first gradient a1. After phase difference detection is completed, the synchronization time until the minimum engageable rotation speed Nmin_eng is shortened, thereby further shortening the total engagement time.

[0093] The reason why the phase difference detection time increases when the input shaft rotation speed increases at a high rate during phase difference detection will be explained with reference to FIGS. 24 to 28. In FIGS. 24 to 28, the phase difference sensor signal on the vertical axis represents the output voltage of the phase difference sensor 6, which is offset so that the output voltage is 0 V when the phase difference Δθ = 0 (see FIG. 6). A 20 msec scale is shown on the time axis for reference. When the phase difference Δθ between the first clutch member 11 and the second clutch member 12, which have 36 teeth, is ±5° (see FIG. 6), the output voltage of the phase difference sensor 6 is ±1 V. The dashed-dotted line in the figures represents the sensor output, which is a beat waveform with a frequency corresponding to the rotation speed difference, and the dashed line represents the envelope of the beat waveform. The solid-line pulse waveform represents the sensor output binarized so that the sensor output is 0 V when it is between 0 V and +1 V and (-1) V when it is between -1 V and less than 0 V.

[0094] The phase difference detection method will be described with reference to FIG. 24. "Detection start" indicates that phase difference detection begins when the input shaft rotation speed Nin reaches the minimum detectable rotation speed Nmin_sen, i.e., at time t1 in FIGS. 22 and 23. Here, phase difference detection refers to detecting the node of the envelope, i.e., the point where the envelope passes the output voltage (-1) V. This point is the "past engagement timing Eng_p used for prediction." After detection begins, two consecutive pulses rising from 0 V to (-1) V are detected. The first pulse is detected just before the envelope node, and the second pulse is detected across the envelope node. Phase difference detection is completed at the edge timing of the second pulse falling to 0 V.

[0095] The upper rows of Figures 25 to 28 show profiles of the target input shaft rotation speed. The first gradient a1 is the rotation speed increase gradient until phase difference detection begins and is common to all four figures. The second gradient a2 is the rotation speed increase gradient during phase difference detection. Figures 25 and 26 show two patterns of phase difference detection when the second gradient a2 is smaller than the first gradient a1 (a1 > a2). Figures 27 and 28 show two patterns of phase difference detection when the second gradient a2 is equal to the first gradient a1 (a1 ≒ a2) and relatively large. Each of the two patterns differs in the timing of the pulse that appears from the start of detection until the first envelope node.

[0096] 25 to 28, the interval from the previous engagement time Eng_p used for prediction to the future engagement time Eng_f is referred to as the "engagement time interval." As shown in FIGS. 25 and 26, when the second gradient a2 is smaller than the first gradient a1, the rotation speed difference at the time of phase difference detection is relatively large, so the engagement time interval is short. Therefore, the engagement operation at the future engagement time Eng_f is completed early.

[0097] In both Figures 25 and 26, after detection begins, the first pulse is detected by the time of the first envelope node. In pattern (1) of Figure 25, the first pulse falling edge is detected after engagement time Eng_p, and phase difference detection is completed successfully. In pattern (2) of Figure 26, the first pulse width after engagement time Eng_p is smaller than the resolution, so detection of the falling edge fails, and the next falling edge is detected. Therefore, although completion of phase difference detection is slightly delayed compared to pattern (1), the impact is small.

[0098] As shown in Figures 27 and 28, when the second gradient a2 is equal to and relatively large than the first gradient a1, the difference in rotation speed at the time of phase difference detection is small, resulting in a long engagement time interval. Therefore, it takes time for the engagement operation at the future engagement time Eng_f to occur. In pattern (1) of Figure 27, the first pulse is detected by the time of the first envelope node after detection begins, and phase difference detection is completed smoothly. However, because the engagement operation takes a long time, highly responsive clutch engagement is not achieved.

[0099] Furthermore, in pattern (2) of FIG. 28, a pulse is not generated by the time of the first envelope node after the start of detection, making it undetectable as shown by the dashed line. The first pulse is detected only before the second envelope node after the start of detection. This significantly delays the completion of phase difference detection compared to pattern (1) of FIG. 27. Furthermore, because it takes time until the engagement operation, the future engagement timing Eng_f is far in the future. This further reduces the responsiveness of the clutch engagement from the engagement command.

[0100] In this way, if the second gradient a2 during phase difference detection is relatively large, the engagement time interval becomes long, and depending on the timing of the pulse immediately after detection starts, the phase difference detection time may become even longer. Therefore, in the fifth embodiment, the increase gradient of the target rotation speed is switched in stages, and the second gradient a2 during phase difference detection is set to be smaller than the first gradient a1, thereby making it possible to shorten the engagement time interval and the phase difference detection time.

[0101] Sixth Embodiment An idle speed setting according to a sixth embodiment will be described with reference to FIGS. 29 to 32. In the synchronization operation of the above-described embodiments, the input shaft 3 is stopped in the disengaged state, and after the control device 7 receives an engagement command, the input shaft speed increases from 0 rpm toward a target speed corresponding to the output shaft speed. For convenience, the difference between the minimum engageable speed and the output shaft speed is ignored here. During high-speed driving, the difference between the input shaft speed and the target speed is large, so it takes time for the input shaft speed to reach the target speed and complete clutch engagement. Therefore, in the sixth embodiment, the input shaft speed is maintained at a predetermined idle speed in the disengaged state, thereby shortening the time from the engagement command until the input shaft speed reaches the target speed.

[0102] Before the synchronization operation begins, the control device 7 maintains the input shaft rotation speed at a "predetermined idle speed greater than 0 and less than or equal to the output shaft rotation speed" when the dog clutch 10 is in a disengaged state. This approach is also effective when it is desired to shorten the time until the clutch is engaged, even when driving at high speeds, for example, when switching from 2WD to 4WD quickly. However, because continuing to idle in a disengaged state results in a decrease in power consumption, it is not desirable to set the idle speed higher than necessary. Therefore, it is necessary to set an appropriate target idle speed according to the vehicle's operating (or running) conditions.

[0103] The motor control unit 74 acquires the state of the dog clutch 10, for example, from the actuator drive unit 73. The motor control unit 74 also acquires information indicating the vehicle's running state, such as vehicle speed, navigation information, camera information, and weather information, from the outside directly or via other blocks.

[0104] In S00 of the flowchart in Figure 29, it is determined whether the dog clutch 10 is in a released state, and if the answer is YES, the process proceeds to S01. In S01, a target idle speed of the input shaft is set in accordance with the operating state (or running state) of the vehicle. In S02, the motor control unit 74 converges and maintains the input shaft speed at the target idle speed. Figure 30 shows the synchronization operation according to the sixth embodiment. In contrast to the comparative example in which the idle speed is not set, in the sixth embodiment in which the idle speed is set, the rotation speed difference ΔN between the input shaft rotation speed Nin and the target rotation speed at the start of the synchronization operation is smaller, and the time to reach the target rotation speed is shortened.

[0105] An example of setting the target idle speed based on vehicle speed will be described with reference to FIG. 31 . The control device 7 switches the idle speed depending on the vehicle speed. As shown in the upper diagram, if the idle speed is not set, the target speed arrival time becomes longer as the vehicle speed increases. If a vehicle speed (e.g., 40 km / h) corresponding to the upper limit of the target speed arrival time is set as the vehicle speed threshold, the target idle speed is set to 0 rpm below the vehicle speed threshold, and the target idle speed is set higher as the vehicle speed increases above the vehicle speed threshold, as shown in the lower diagram. This shortens the target speed arrival time above the vehicle speed threshold.

[0106] An example of setting the target idle speed based on driving conditions other than vehicle speed will be described with reference to Figure 32. Here, we will assume application to a 4WD vehicle with front and rear driving force distribution control. Since driving force distribution does not function when the clutch is released, in situations where driving force distribution is required, the clutch must be engaged with high response from the released state to enable driving force distribution.

[0107] Example 1 is when it is raining, snowing, or the road is frozen. The target idle speed is set to 0 rpm when the outside air temperature is equal to or higher than a temperature threshold (for example, 0°C), and is set to a level according to the weather when the outside air temperature is below the temperature threshold. For example, the level is set to 0 for weather other than snow or rain, level 1 for rain, and level 2 for snow, and the higher the level, the higher the target idle speed is set.

[0108] Example 2 is when turning a sharp curve. In a four-wheel independently drive vehicle in which the left and right wheels are driven independently in addition to the front and rear wheels, it is possible to distribute driving force between the wheels on the inside and outside of the turn when turning. The motor control unit 74 obtains the turning radius of the curve the vehicle is about to travel from, for example, navigation information. The smaller the turning radius of the sharper the curve, the greater the need for driving force distribution, so the target idle speed is set higher. The target idle speed may also be changed based on the vehicle yaw rate, which is result information during the turn.

[0109] Example 3 shows braking at a red light ahead or at the end of a traffic jam. The motor control unit 74 obtains the distance to the stop line of the red light ahead or the end of the traffic jam, i.e., the required braking distance, from, for example, an on-board camera. The shorter the required braking distance, the greater the need to brake in 4WD, so the target idle speed is set higher. The target idle speed may be switched based on the deceleration (negative acceleration) or other information resulting from braking. In the graphs of Examples 2 and 3, the target idle speed changes continuously with respect to the parameters of the turning radius and required braking distance, but the target idle speed may also change in steps with respect to the parameters.

[0110] In the sixth embodiment, the input shaft is allowed to idle while the dog clutch 10 is in the released state, thereby preventing the clutch engagement time from becoming longer depending on the vehicle operating conditions such as the vehicle speed. Alternatively, the clutch engagement time can be further shortened. For example, when the idle speed is set depending on the vehicle speed, the higher the vehicle speed, the smaller the difference in input and output shaft speeds in the released state is set, thereby preventing the synchronization operation time from becoming longer.

[0111] Seventh Embodiment The setting of the anticipatory operation time according to a seventh embodiment will be described with reference to FIGS. 33 to 35 . Various delay factors cause time variations from the time when the engagement time prediction unit 72 starts predicting the engagement time until the clutch actuator 5 is fully engaged based on a drive command from the actuator drive unit 73. Depending on the degree of variation, it may not be possible to engage the clutch actuator 5 at the predicted future engagement time. Therefore, in the seventh embodiment, the control device 7 estimates the delay time of the operation of the clutch actuator 5 from information on the delay factors related to the drive of the clutch actuator 5. This estimation includes detection. Based on the estimation result, the control device 7 starts driving the clutch actuator 5 earlier than the predicted future engagement time by an anticipatory operation time T_ad adjusted according to the delay time. The anticipatory operation time T_ad may be set by either the engagement time prediction unit 72 or the actuator drive unit 73.

[0112] In S91 of the flowchart in Figure 33, the engagement timing prediction unit 72 estimates the delay time of the clutch actuator 5 from the delay factors. In S92, the engagement timing prediction unit 72 adjusts the anticipatory operation time T_ad in accordance with the estimated delay time. In the third and bottom diagrams in Figure 34, the anticipatory operation time T_ad is longer and the actuator drive timing is earlier than in Figure 11.

[0113] 35, an example of setting the anticipatory operation time depending on delay factors will be described. Example 1 of the delay factor is the time constant of the filtering process in the filter unit 71 that filters the output signal of the phase difference sensor 6. The larger the time constant of the filtering process, the larger the delay of the phase difference sensor signal acquired by the engagement timing prediction unit 72 relative to the true phase difference, and therefore the anticipatory operation time T_ad is set to be longer.

[0114] A second example of a delay factor is the computational load of the CPU constituting the control device 7. The greater the computational load of the CPU, the longer the delay time from when a drive command is given to the clutch actuator 5 to when actual drive is completed due to processing delays and communication delays between CPUs via the CAN or the like, and therefore the advance operation time T_ad is set to be longer.

[0115] A third example of a delay factor is the temperature of the clutch actuator 5. For example, if the clutch actuator 5 is configured as a permanent magnet synchronous motor, the torque relative to the passing current changes depending on the actuator temperature. At high temperatures, the torque decreases due to demagnetization, and at low temperatures, the torque increases due to magnetization, so the operating speed of the clutch actuator 5 changes. The higher the actuator temperature, the slower the operating speed becomes, and the longer the time until engagement is complete, so the advance operating time T_ad is set to be.

[0116] In the seventh embodiment, the advance operation time T_ad is adjusted so as to reduce time variations due to delay factors related to the driving of the clutch actuator 5 and the prediction of the engagement time, thereby making it possible to engage the clutch at the predicted future engagement time.

[0117] Eighth Embodiment Switching between execution and non-execution of engagement time prediction according to an eighth embodiment will be described with reference to Figures 36 to 40. Predicting future engagement times is not always optimal for achieving highly responsive clutch engagement, and in some situations, predicting future engagement times may actually hinder high response. Therefore, in the eighth embodiment, when it is determined that a prediction execution condition is met based on the vehicle's operating state, the control device 7 predicts future engagement times and drives the clutch actuator 5. The first to seventh embodiments described above can be considered to execute each process on the assumption that a prediction execution condition is met.

[0118] Furthermore, when it is determined that the prediction execution condition is not satisfied depending on the operating state of the vehicle, the control device 7 does not predict the future engagement time. In this case, the control device 7 drives the clutch actuator 5 based on the difference in rotation speed between the input shaft 3 and the output shaft 4. More specifically, the actuator drive unit 73 outputs a drive command for the clutch actuator 5 when the input / output shaft rotation speed difference ΔNio becomes equal to or less than the target rotation speed difference ΔN_tgt and the input shaft 3 and the output shaft 4 are synchronized in rotation.

[0119] Here, "not predicting future engagement timing" does not necessarily mean that the entire process of obtaining the previous engagement timing based on the phase difference sensor signal after an engagement instruction is issued and predicting future engagement timing from the estimated rotation speed is stopped. For example, the engagement timing prediction unit 72 may always perform calculations up to predicting future engagement timing, and then switch whether or not to output a drive instruction to the actuator drive unit 73 by referring to the future engagement timing. In this way, switching between "referencing / not referring to future engagement timing" is also interpreted as one aspect of "executing / not executing engagement timing prediction."

[0120] In S11 of the flowchart in FIG. 36, a clutch engagement command is input. In S71, it is determined whether the prediction execution condition is met. If YES in S71, the control device 7 predicts the future engagement time in S72 and drives the clutch actuator 5 in accordance with the predicted future engagement time. If NO in S71, it is determined in S73 whether the current input / output shaft rotational speed difference ΔNio is equal to or less than the target rotational speed difference ΔN_tgt. At the timing when YES is determined in S73, the process proceeds to S74, where the actuator drive unit 73 outputs a drive command to the clutch actuator 5.

[0121] For example, if the vehicle speed is lower than a predetermined threshold, it is determined that the prediction execution condition is not met. As shown in Figure 37, a vehicle speed of 20 km / h is compared with a vehicle speed of 40 km / h as an example of "low speed" and "high speed." If the difference in input and output shaft rotation speeds is the same, the beat waveforms, i.e., the envelope waveforms, will match. Since the frequency of the phase difference sensor signal is proportional to the vehicle speed, the width of the binarized pulses is shorter at higher speeds and longer at lower speeds. The number of pulses appearing between nodes of the beat waveform is greater at higher speeds and fewer at lower speeds.

[0122] Therefore, to prevent erroneous detection when detecting nodes of the beat waveform, the detectable rotation speed difference ΔN_sen must be set smaller as the speed decreases, and the minimum detectable rotation speed Nmin_sen becomes higher. As a result, when the input shaft rotation speed increases, the phase difference detection time and engagement time become longer as the speed decreases. The relationship between vehicle speed, detectable rotation speed difference ΔN_sen, and phase difference detection time is shown in Figure 38.

[0123] Therefore, as shown in Figure 39, the control device 7 predicts future engagement times when the vehicle speed is equal to or greater than the vehicle speed threshold, but does not predict when the vehicle speed is below the vehicle speed threshold. When the vehicle speed is below the vehicle speed threshold, the fact that the input and output shafts have achieved rotational synchronization is used as a trigger to output a drive command for the clutch actuator 5. This allows for early clutch engagement without waiting for the next engagement time when the vehicle speed is below the vehicle speed threshold. For example, hysteresis may be provided for the vehicle speed threshold, and the ON threshold, which switches from non-execution to execution, may be set higher than the OFF threshold, which switches from execution to non-execution. This makes it possible to prevent control hunting due to vehicle speed fluctuations near the threshold.

[0124] With reference to Figure 40, an example will be described in which execution / non-execution of engagement timing prediction is switched using other parameters. The parameter in Example 1 is the initial stroke of the clutch actuator 5. For example, if the clutch actuator 5 is intentionally moved from the fully released position, the initial stroke at the time of the engagement command will be greater than the normal value (= 0). If an engagement operation is performed from an abnormal position, an error will occur in the actuator stroke from the engagement boundary position to the fully engaged position, which may prevent the clutch from engaging at the predicted future engagement time.

[0125] Therefore, the control device 7 predicts the future engagement time if the initial stroke at the time of the engagement command is between 0 and the allowable value δL, but does not predict if the initial stroke exceeds the allowable value δL. If the initial stroke exceeds the allowable value δL, the fact that the input and output shafts have achieved rotational synchronization is used as a trigger to output a drive command for the clutch actuator 5. This makes it possible to avoid the clutch not being able to engage at the predicted future engagement time and to achieve early clutch engagement.

[0126] The parameter in Example 2 is a change in the target driving force or output shaft rotation speed during the engagement timing prediction. The target driving force is the driving force expected for the vehicle due to user operation, such as the amount of accelerator operation, the amount of brake operation, or the amount of cruise control operation. The change may be evaluated either as the amount of change before and after the change or as the rate of change. If the change in the target driving force or the output shaft rotation speed is large, the estimation accuracy of the target rotation speed difference ΔN_tgt deteriorates, and the prediction accuracy of the future engagement timing deteriorates.

[0127] Therefore, the control device 7 predicts the future engagement time if the change in the target driving force or output shaft rotation speed during engagement time prediction is equal to or less than a threshold value, but does not perform the prediction if the change in the target driving force or output shaft rotation speed is greater than the threshold value.If the change in the target driving force or output shaft rotation speed during engagement time prediction is greater than the threshold value, the fact that the input and output shafts have achieved rotational synchronization is used as a trigger to output a drive command for the clutch actuator 5.This makes it possible to prevent a deterioration in the prediction accuracy of the future engagement time.

[0128] The parameter in Example 3 is whether or not torque limitation occurs on the motor 82 when an engagement command is issued or when the engagement timing is being predicted. For example, if the torque of the motor 82 is limited due to a drop in the battery SOC or abnormal overheating of the windings, it becomes difficult to follow the target rotation speed in the synchronization operation, the accuracy of the estimated rotation speed deteriorates, and the accuracy of the prediction of the future engagement timing deteriorates.

[0129] Therefore, if there is no torque limit on the motor 82 at the time of issuing an engagement command or during engagement timing prediction, the control device 7 predicts the future engagement time, but does not perform the prediction if there is a torque limit. If there is a torque limit on the motor 82, the fact that the input and output shafts have achieved rotational synchronization is used as a trigger to output a drive command for the clutch actuator 5. This makes it possible to prevent a deterioration in the prediction accuracy of the future engagement time.

[0130] In the eighth embodiment, in situations where the accuracy of the prediction cannot be guaranteed or where predicting the engagement time would actually hinder high response, the actuator is driven based only on the input / output shaft rotational speed difference without executing the engagement time prediction, thereby enabling appropriate operation according to the situation. For example, at low vehicle speeds, the phase difference detection time is long, so not executing the engagement time prediction allows the clutch to be engaged sooner.

[0131] 1, 5, etc. has a face-type gear configuration in which the clutch members 11, 12 themselves move relative to one another in the axial direction, allowing direct meshing between the first engagement teeth 13 of the first clutch member 11 and the second engagement teeth 14 of the second clutch member 12. In addition to this configuration, a sleeve-type clutch may be employed in which the clutch members are stationary and a movable sleeve is provided as a separate component as an intermediary member.

[0132] A sleeve-type clutch is disclosed, for example, in Japanese Patent Application Laid-Open No. 2010-96190. In a sleeve-type clutch, first engagement teeth of a first clutch member and second engagement teeth of a second clutch member can mesh with each other via a sleeve serving as an intermediary member. A clutch actuator axially moves the sleeve relative to the first clutch member and the second clutch member, thereby switching the engaged and disengaged states of the mesh clutch. The detection portion of the phase difference sensor is adjusted so that different sensor signals are output depending on whether the state is engaged or disengaged.

[0133] (b) The dog clutch 10 is not limited to being provided between the rear wheel motor 82 and the rear wheels 92 in the 2WD / 4WD switchable vehicle 90 illustrated in FIG. 1 , but may also be provided, for example, between the front wheel motor and the front wheels in a front-wheel drive (FF) vehicle.

[0134] (c) The phase difference sensor 6 is not limited to one that detects the total area of ​​the engaging teeth 13, 14 within the detection range SA as exemplified in Figures 5 and 6, but may be one that can detect information correlated with the difference in rotational position between the input shaft 3 and the output shaft 4 and convert it into a phase difference. In the "switching of means for acquiring phase difference information" according to the second embodiment, "acquisition of phase difference information based on a phase difference sensor" includes acquisition of information based on various phase difference sensors.

[0135] (d) The rotational drive source connected to the input shaft 3 is not limited to the motor 82, but may be an internal combustion engine or the like. In the fourth embodiment, instead of the drive time from the start of energization of the motor 82, the relationship between the drive time from the start of operation of the internal combustion engine or the like and the input shaft rotation speed may be learned. Furthermore, the dog clutch engagement control system may be applied to power transmission mechanisms of general machinery in addition to vehicle powertrains.

[0136] (e) In the above embodiment, the target rotation speed difference ΔN_tgt suppresses engagement shock when the vehicle accelerates or decelerates during synchronization. If the dog clutch 10 is engaged when the actual rotation speed of the input shaft is greater than the target rotation speed, a shock occurs on the side that accelerates the vehicle. If the dog clutch 10 is engaged when the actual rotation speed of the input shaft is smaller than the target rotation speed, a shock occurs on the side that decelerates the vehicle.

[0137] Therefore, the rotation speed difference may be evaluated not as an absolute value but as a positive or negative value, and for example, when the vehicle is accelerating, the case where "actual rotation speed > target rotation speed" which is on the acceleration side is allowed, and the case where "actual rotation speed < target rotation speed" which is on the deceleration side is not allowed, and the opposite may be true when the vehicle is decelerating. A target rotation speed difference ΔN_tgt which corresponds to the tolerance limit of engagement shock may also be set depending on other driving situations and road conditions.

[0138] (f) In the eighth embodiment, if the prediction execution condition is not met, prediction of future engagement times based on the one previous engagement time may not be performed, but instead control may be switched to calculating an approximation line based on multiple previous engagement times using the conventional technology of Patent Document 1. Even if prediction accuracy cannot be ensured using only information on the one previous engagement time, using information on multiple previous engagement times may make it possible to engage the clutch earlier.

[0139] As described above, the present disclosure is not limited to the above-described embodiments, and can be embodied in various forms without departing from the spirit of the present disclosure.

[0140] The following disclosures regarding the dog clutch engagement control systems corresponding to the third to eighth embodiments among the above embodiments may be combined with each other.

[0141] "A dog clutch engagement control system in which, during a synchronization operation, a control device acquires an actual rotation speed of the input shaft at the time of one previous engagement, and calculates a corrected estimated rotation speed by multiplying the time axis of an estimated rotation speed map by a correction factor so as to bring the estimated rotation speed closer to the actual rotation speed."

[0142] "A dog clutch engagement control system in which a control device learns the relationship between a drive time and an actual rotation speed when increasing the rotation speed of an input shaft relative to a target rotation speed with the dog clutch in a released state, and updates an estimated rotation speed map."

[0143] "A dog clutch engagement control system in which, during execution of the synchronization operation, the control device increases the target rotation speed of the input shaft along a polygonal linear profile divided into a plurality of periods including a first period and a second period, and the second gradient in the second period is set to be smaller than the first gradient in the first period." corresponding to the fifth embodiment

[0144] "A dog clutch engagement control system that maintains the rotation speed of the input shaft at a predetermined idle rotation speed when the dog clutch is in a released state."

[0145] "A dog clutch engagement control system in which a control device estimates a delay time of the operation of the clutch actuator, and starts driving the clutch actuator earlier than the predicted future engagement time by an advance operation time set in accordance with the delay time."

[0146] Corresponding to the eighth embodiment, "a dog clutch engagement control system in which, when it is determined that the prediction execution condition is met according to the driving state of the vehicle, the control device predicts the future engagement time and drives the clutch actuator, and, when it is determined that the prediction execution condition is not met according to the driving state of the vehicle, the control device does not predict the future engagement time and drives the clutch actuator based on the difference in rotation speed between the input shaft and the output shaft."

[0147] The control device and methods described herein may be implemented by a special-purpose computer configured with a processor and memory programmed to perform one or more functions embodied in a computer program. Alternatively, the control device and methods described herein may be implemented by a special-purpose computer configured with a processor comprising one or more dedicated hardware logic circuits. Alternatively, the control device and methods described herein may be implemented by one or more special-purpose computers configured with a processor and memory programmed to perform one or more functions in combination with a processor configured with one or more hardware logic circuits. Furthermore, the computer program may be stored in a computer-readable non-transitory tangible storage medium as instructions executed by a computer.

[0148] The present disclosure has been described based on the embodiments. However, the present disclosure is not limited to the embodiments and structures. The present disclosure also encompasses various modifications and variations within the scope of equivalents. Furthermore, various combinations and forms, including only one element, more than one element, or less than one element, are also within the scope and spirit of the present disclosure.

Claims

1. A mesh clutch (10) having a first clutch member (11) connected to an input shaft (3) and having a plurality of first engagement teeth (13) arranged in the circumferential direction, and a second clutch member (12) connected to an output shaft (4) and having a plurality of second engagement teeth (14) arranged in the circumferential direction and capable of engaging with the first engagement teeth directly or via an intermediate member, the first clutch member and the second clutch member being switched between an engaged state and a released state; a clutch actuator (5) that moves the first clutch member and the second clutch member relatively in the axial direction, or, when the intermediate member is used, moves the intermediate member relatively in the axial direction with respect to the first clutch member and the second clutch member; a phase difference sensor (6) that detects a phase difference between the first clutch member and the second clutch member; and a control device (7) that outputs a drive command to the clutch actuator in accordance with an engagement time at which the mesh clutch is engageable, the control device comprising: a mesh clutch engagement control system which, when a predetermined prediction execution condition is currently satisfied while a synchronization operation is being performed in which the rotation speed of the input shaft is increased toward the rotation speed of the output shaft so as to synchronize the rotation of the input shaft and the output shaft in response to an engagement command to engage the mesh clutch in a released state, detects the engagement time based on an output of the phase difference sensor, predicts a future engagement time that is the future engagement time from the engagement time of a past instance and an estimated rotation speed of the input shaft, and drives the clutch actuator in accordance with the predicted future engagement time.

2. A mesh clutch (10) having a first clutch member (11) connected to an input shaft (3) and having a plurality of first engagement teeth (13) arranged in the circumferential direction, and a second clutch member (12) connected to an output shaft (4) and having a plurality of second engagement teeth (14) arranged in the circumferential direction and capable of engaging with the first engagement teeth directly or via an intermediate member, the first clutch member and the second clutch member being switched between an engaged state and a released state; a clutch actuator (5) for moving the first clutch member and the second clutch member relative to each other in the axial direction, or for moving the intermediate member relative to the first clutch member and the second clutch member in the axial direction when the intermediate member is used; a phase difference sensor (6) for detecting a phase difference between the first clutch member and the second clutch member; and a control device (7) for outputting a drive command to the clutch actuator in accordance with an engagement time when the mesh clutch is engageable, the control device comprising: a predetermined prediction execution condition is satisfied currently during execution of a synchronization operation in which the rotation speed of the input shaft is increased toward the rotation speed of the output shaft so as to synchronize the rotation of the input shaft and the output shaft in response to an engagement command to engage the dog clutch in a released state; an estimated phase difference between the input shaft and the output shaft is calculated based on an initial phase difference between the input shaft and the output shaft at the start of the synchronization operation and a cumulative phase difference converted from a time integral value of the rotation speed difference between the input shaft and the output shaft during execution of the synchronization operation; depending on whether the phase difference sensor can detect a phase difference, switching between detecting the engagement timing based on the output of the phase difference sensor or estimating the engagement timing from the estimated phase difference; predicting a future engagement timing, which is the engagement timing in the future, from the engagement timing of a past engagement and the estimated rotation speed of the input shaft, and driving the clutch actuator in accordance with the predicted future engagement timing.

3. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device stores an estimated rotation speed map that specifies the change in estimated rotation speed versus the drive time of the input shaft during the synchronization operation, and during the synchronization operation, the actual rotation speed of the input shaft at the past engagement time is obtained, and a correction factor is multiplied on the time axis of the estimated rotation speed map to calculate the corrected estimated rotation speed so as to bring the estimated rotation speed closer to the actual rotation speed.

4. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device stores an estimated rotation speed map that specifies the change in estimated rotation speed versus the drive time of the input shaft during the synchronization operation, and learns the relationship between the drive time and actual rotation speed when increasing the rotation speed of the input shaft toward a target rotation speed with the dog clutch in a released state, and updates the estimated rotation speed map.

5. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device, during execution of the synchronization operation, increases a target rotation speed, which is a target value of the rotation speed of the input shaft, along a broken line profile divided into a plurality of periods including a first period (Tup1) in which the rotation speed of the input shaft increases to the minimum rotation speed at which the phase difference can be detected by the phase difference sensor, and a second period (Tup2) following the first period, in which the rotation speed of the input shaft further increases from the minimum rotation speed at which the phase difference can be detected by the phase difference sensor, and a second gradient (a2) which is the time rate of change of the target rotation speed in the second period is set to be smaller than a first gradient (a1) which is the time rate of change of the target rotation speed in the first period.

6. The dog clutch engagement control system according to claim 5, wherein the dog clutch engagement control system is mounted on a vehicle (90) having an output shaft connected to a drive wheel (92), and the broken-line profile of the target rotational speed is switched depending on the driving state of the vehicle.

7. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device maintains the rotational speed of the input shaft at a predetermined idle speed greater than 0 and less than or equal to the rotational speed of the output shaft when the dog clutch is in a released state prior to the start of the synchronization operation.

8. The dog clutch engagement control system according to claim 7, wherein the dog clutch engagement control system is mounted on a vehicle (90) having an output shaft connected to a drive wheel (92), and the control device switches the idle speed in response to vehicle speed.

9. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device estimates a delay time in the operation of the clutch actuator from information on delay factors related to the drive of the clutch actuator, and starts driving the clutch actuator earlier than the predicted future engagement time by an advance operation time adjusted in accordance with the delay time.

10. A dog clutch engagement control system as described in claim 9, wherein the delay factor is a time constant of the filtering process in a filter section (71) that filters the output signal of the phase difference sensor, and the larger the time constant of the filtering process, the longer the advance operation time is set.

11. A dog clutch engagement control system as described in claim 1 or 2, wherein the control device is mounted on a vehicle (90) having an output shaft connected to a drive wheel (92), and when it is determined that the prediction execution condition is satisfied according to the driving state of the vehicle, the control device executes the prediction of the future engagement time and drives the clutch actuator, and when it is determined that the prediction execution condition is not satisfied according to the driving state of the vehicle, the control device does not execute the prediction of the future engagement time, and drives the clutch actuator based on the difference in rotational speed between the input shaft and the output shaft.

12. The dog clutch engagement control system as set forth in claim 11, wherein it is determined that the predicted execution condition is not satisfied when the vehicle speed is lower than a predetermined threshold.

Images