Electric vehicle control method and electric vehicle control device

Abstract

An aspect of the present invention pertains to an electric vehicle control method for performing attitude control that controls the attitude in the front‐back direction, by adjusting driving force distribution between a front wheel and a rear wheel which are driving wheels. In this control method, the total driving force, which is a required driving force for an electric vehicle, is calculated on the basis of operation of an accelerator pedal. In addition, on the basis of this total driving force, estimated acceleration which is an estimated value for acceleration generated when an electric vehicle is driven with the total driving force, or estimated jerk which is an estimated value for jerk that is the time change rate of acceleration, is calculated. On the basis of the estimated acceleration or the estimated jerk, attitude control execution determination for switching the attitude control on or off is performed.

Description

Control method for electric vehicle and control device for electric vehicle

[0001] The present invention relates to a control method for an electric vehicle and a control device for an electric vehicle.

[0002] JP4876534B2 discloses a technique for reducing the pitch rate of an in-wheel motor vehicle, which does not fully utilize the effects of a suspension, when the vehicle passes over a bump or other obstacle in the road surface. More specifically, it discloses that different braking / driving forces are applied to the front and rear wheels, and when a pitch rate fluctuation is detected, different braking / driving forces are further applied to the left and right wheels at a predetermined cycle.

[0003] Conventionally, electric vehicles are known that perform attitude control by adjusting the distribution of driving force among a plurality of drive wheels. In such electric vehicles, attitude is generally controlled by adjusting the distribution of driving force through feedback control that feeds back detected values ​​such as pitch rate. That is, in conventional electric vehicles, attitude control works by following and, for example, canceling out attitude fluctuations that have already occurred, so that attitude control may not be able to be performed in time.

[0004] Furthermore, when the driving force distribution is adjustable, the driving force distribution is usually determined to optimize the electric fuel economy, taking into account driving stability. Therefore, when posture control based on driving force distribution is performed, the driving force distribution is shifted from the driving force distribution that optimizes the electric fuel economy to the driving force distribution that controls the attitude, which deteriorates the electric fuel economy. For this reason, it is desirable to keep the attitude control based on the adjustment of the driving force distribution off (non-executed state) as much as possible and to turn it on (executed state) only when necessary, even if it results in a trade-off with the electric fuel economy.

[0005] However, as described above, when attitude control is turned on / off as needed by adjusting the drive force distribution, if the start of attitude control is determined using detected values ​​such as the pitch rate, the start of attitude control is determined based on attitude fluctuations that have already occurred, resulting in a delay in the start decision itself. As a result, the delay in attitude control is exacerbated, and there are more cases where attitude control is not effectively completed in time. This is more noticeable when attitude control is performed using feedback control, but the same is true when attitude control is performed using feedforward control. In other words, when attitude control is turned on / off as needed by adjusting the drive force distribution, even if the attitude control is performed using feedforward control, if the start decision is made using detected values ​​such as the pitch rate, the delay in the start decision may result in attitude control not being effectively completed in time.

[0006] Therefore, an object of the present invention is to provide a control method for an electric vehicle and a control device for an electric vehicle that can reduce the delay in determining the start of attitude control and perform attitude control in a timely manner when attitude control through adjustment of driving force distribution is turned on / off as needed.

[0007] One aspect of the present invention is a control method for an electric vehicle that performs attitude control to control the vehicle's longitudinal attitude by adjusting the distribution of driving force between the front and rear drive wheels. In this control method, a total driving force, which is the driving force required for the electric vehicle, is calculated based on the operation of the accelerator pedal. Furthermore, an estimated acceleration, which is an estimate of the acceleration that occurs when the electric vehicle is driven by the total driving force, or an estimated jerk, which is an estimate of the jerk that is the time rate of change of acceleration, is calculated based on the total driving force. Then, an attitude control execution determination is made to switch the attitude control on or off based on the estimated acceleration or the estimated jerk.

[0008] FIG. 1 is an explanatory diagram showing a schematic configuration of an electric vehicle. FIG. 2 is an explanatory diagram showing a schematic structure of a chassis system. FIG. 3 is a graph showing a lower limit value of the pitch angle that an occupant of the electric vehicle can feel when accelerating. FIG. 4 is a block diagram showing the configuration of a controller 12 for attitude control. FIG. 5 is a block diagram showing the configuration of a attitude control execution determination unit. FIG. 6 is a flowchart showing the operation related to the on / off of attitude control in an electric vehicle. FIG. 7 is a time chart showing the transition of parameters when attitude control is switched from off to on. FIG. 8 is a block diagram showing the configuration of the attitude control execution determination unit in a second embodiment. FIG. 9 is a flowchart showing the operation related to the on / off of attitude control for an electric vehicle of the second embodiment. FIG. 10 is a block diagram showing a configuration for updating the coefficients of an approximation equation used to calculate running resistance. FIG. 11 is an explanatory diagram showing a specific mode of updating the coefficients of an approximation equation used to calculate running resistance. FIG. 12 is a flowchart showing the operation related to the update of the coefficients of an approximation equation used to calculate running resistance.

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0010] [First embodiment] <Configuration of electric vehicle> Fig. 1 is an explanatory diagram showing a schematic configuration of an electric vehicle 100. The electric vehicle 100 is, for example, an electric vehicle or a hybrid vehicle, and is a vehicle that can drive or brake one or more drive wheels using an electric motor. In particular, in this embodiment, the electric vehicle 100 is a so-called four-wheel drive (4WD) vehicle, and can control (adjust) the drive force generated in each of the multiple drive wheels. Specifically, as shown in Fig. 1, the electric vehicle 100 includes a front-wheel drive system 10, a rear-wheel drive system 11, and a controller 12.

[0011] The front-wheel drive system 10 is a system that controls front wheels 21, which are first drive wheels. The front-wheel drive system 10 includes a front inverter 22 and a front motor 23.

[0012] The front inverter 22 converts DC power output by a battery (not shown) into AC power and supplies it to the front motor 23, thereby driving the front motor 23. When the front motor 23 is rotated by the front wheels 21, the front inverter 22 converts regenerated AC power generated by the front motor 23 into DC power and inputs it to the battery, thereby charging the battery.

[0013] The front motor 23 is an electric motor that drives the front wheels 21. The front motor 23 is, for example, a three-phase AC synchronous motor. The torque generated by the front motor 23 is transmitted to the front wheels 21 via a front drive shaft 24, and a driving force (hereinafter referred to as a front wheel driving force F) is applied to the front wheels 21. Ｆ This generates a phenomenon called

[0014] The rear-wheel drive system 11 is a system that controls rear wheels 26, which are second drive wheels. The rear-wheel drive system 11 includes a rear inverter 27 and a rear motor 28.

[0015] The rear inverter 27 converts DC power output by the battery into AC power and supplies it to the rear motor 28, thereby driving the rear motor 28. Furthermore, when the rear motor 28 is rotated along with the rear wheels 26, the rear inverter 27 converts regenerated AC power generated by the rear motor 28 into DC power and inputs it to the battery, thereby charging the battery.

[0016] The rear motor 28 is an electric motor that drives the rear wheels 26. The rear motor 28 is configured, for example, by a three-phase AC synchronous motor similar to the front motor 23. The torque generated by the rear motor 28 is transmitted to the rear wheels 26 via a rear drive shaft 29, and a driving force (hereinafter referred to as rear wheel driving force F) is applied to the rear wheels 26. Ｒ This generates a phenomenon called

[0017] The controller 12 is configured with one or more computers that control the operation of the electric vehicle 100. The controller 12 is programmed to control the operation of the electric vehicle 100 at a predetermined control cycle. In this embodiment, the controller 12 is a control device for the electric vehicle 100 that performs attitude control to control the attitude in the longitudinal direction by adjusting the distribution of driving force between the front wheels 21 and the rear wheels 26, which are the driving wheels.

[0018] The controller 12 distributes the driving force (hereinafter referred to as total driving force TQ) requested by, for example, operating an accelerator pedal (not shown) to the front wheels 21 and rear wheels 26, which are the driving wheels. Then, the controller 12 calculates the front wheel driving force F according to the distribution. Ｆ and rear wheel driving force F Ｒ The front wheels 21 and the rear wheels 26 are driven by the front-wheel drive system 10 and the rear-wheel drive system 11, respectively, so that the following occurs: Furthermore, in this embodiment, the controller 12 is programmed to execute attitude control that controls the attitude of the electric vehicle 100 in the longitudinal direction by adjusting the distribution of driving force between the front wheels 21 and the rear wheels 26 as necessary.

[0019] When controlling the operation of the electric vehicle 100, the controller 12 can appropriately acquire various parameters that represent the operating state of the electric vehicle 100 by sensors (not shown) or by calculation. ＰＯ Therefore, the controller 12 detects the accelerator opening A ＰＯ The accelerator opening A can be obtained as needed. ＰＯ is a parameter that indicates the amount of accelerator pedal operation. The controller 12 also appropriately acquires the vehicle speed VSP of the electric vehicle 100 from a sensor (not shown) or by calculation.

[0020] Furthermore, in this embodiment, the electric vehicle 100 is configured to calculate the acceleration (hereinafter referred to as the actual acceleration G ａｃｔ Therefore, the controller 12 is provided with an acceleration sensor (not shown) that measures the actual acceleration G ａｃｔThe controller 12 can also acquire the actual acceleration G ａｃｔ By differentiating the above, the jerk (hereinafter referred to as the actual jerk J) which is the time rate of change of the acceleration occurring in the electric vehicle 100 is obtained. ａｃｔ (hereinafter referred to as "information") can be obtained as appropriate.

[0021] In addition, the controller 12 receives information from a car navigation system (not shown) about the current location of the electric vehicle 100 and the gradient of the road on which the electric vehicle 100 is traveling (hereinafter referred to as the road gradient φ ＬＳ The road surface gradient φ ＬＳ can be obtained by calculation based on the vehicle speed VSP and acceleration G of the electric vehicle 100, or changes therein. ＬＳ is obtained from the car navigation system.

[0022] <Principle of attitude control by driving force distribution> Fig. 2 is an explanatory diagram showing the schematic structure of the chassis system. As shown in Fig. 2, the front wheels 21 are connected to a vehicle shed 101, which is the part of the vehicle body where the passenger compartment and the like are formed, via a front suspension 31. Similarly, the rear wheels 26 are connected to the vehicle shed 101 via a rear suspension 32.

[0023] For example, the front wheel driving force F Ｆ , rear wheel driving force F Ｒ When the electric vehicle 100 accelerates due to the acceleration of the electric vehicle 100 caused by the acceleration of the electric vehicle 100, the load moves rearward (negative side in the X direction) of the electric vehicle 100. As a result, the center of gravity O Ｇ With pitch angle θ as the center Ｐ Therefore, when the electric vehicle 100 accelerates, the electric vehicle 100 assumes a posture in which the front portion, which is the portion on the positive side in the X direction, is lifted up (a so-called nose-up posture).

[0024] On the other hand, the front wheel driving force F Ｆ The torque of the front motor 23 (hereinafter referred to as "front torque") that generates acts on the vehicle shed 101 via the front suspension 31. Specifically, the front torque is generated by the rotation of the vehicle shed 101 relative to the virtual center of rotation O. Ｆ Around the pitch angle θＰ In other words, when the electric vehicle 100 accelerates, the front torque suppresses nose-up. Similarly, the rear wheel driving force F Ｒ The torque of the rear motor 28 (hereinafter referred to as rear torque) that generates the virtual rotation center O Ｒ Around the pitch angle θ Ｐ Therefore, when the electric vehicle 100 accelerates, the rear torque suppresses nose-up.

[0025] The magnitude of the front torque's effect of suppressing nose-up during acceleration depends on the anti-skid angle θ Ｆ Similarly, the magnitude of the rear torque's effect of suppressing nose-up during acceleration depends on the anti-skid angle θ Ｒ Therefore, if the drive force distribution to the front wheels 21 and the rear wheels 26 is adjusted so that the distribution to the drive wheels with relatively large anti-skid angles is increased, the effect of suppressing nose-up while maintaining the total drive force is increased. Therefore, in this embodiment, the controller 12 adjusts the drive force distribution to the front wheels 21 and the rear wheels 26 to adjust the longitudinal attitude of the electric vehicle 100 (i.e., the pitch angle θ Ｐ The system performs attitude control to control the speed of the vehicle (or its fluctuations).

[0026] In addition, the virtual rotation center O Ｆ is an instantaneous and virtual center of rotation that occurs in the vehicle body (particularly the vehicle shed 101) due to the transmission of front torque, and is determined in advance by the specific configuration of the front suspension 31, etc. Similarly, the virtual center of rotation O Ｒ is the instantaneous and virtual center of rotation of the vehicle body (particularly the vehicle shed 101) caused by the transmission of rear torque, and is determined in advance by the specific configuration of the rear suspension 32, etc. Ｆ is the distance between the center of rotation of the front wheel 21 and the virtual center of rotation O on the XZ plane. Ｆ The angle between the line connecting the two points and a line parallel to the road surface is the angle. Ｒis the distance between the rotation center of the rear wheel 26 and the virtual rotation center O on the XZ plane. Ｒ It is the angle between the line connecting the two and a line parallel to the road surface.

[0027] In this embodiment, as shown in FIG. 2, the anti-scatter angle θ of the rear suspension 32 Ｒ is the anti-skid angle θ of the front suspension 31 Ｆ Therefore, for example, when accelerating, the pitch angle θ Ｐ When suppressing or reducing the increase in the torque, the controller 12 relatively increases the drive force distribution to the rear wheels 26 .

[0028] Here, the relationship between the configuration of the chassis system and attitude control during acceleration has been described, but the controller 12 also performs attitude control during deceleration by adjusting the distribution of driving force between the front wheels 21 and the rear wheels 26. However, during deceleration, the electric vehicle 100 assumes an attitude in which the front portion sinks (a so-called nose dive attitude), which is the opposite of the above, so the controller 12 adjusts the distribution of driving force between the front wheels 21 and the rear wheels 26 accordingly. Furthermore, hereinafter, unless otherwise specified, the attitude of the electric vehicle 100 refers to the attitude in the fore-and-aft direction, i.e., the pitch angle θ Ｐ In other words, the attitude control by adjusting the driving force distribution is performed by adjusting the pitch angle θ Ｐ Control of pitch rate Δ Ｐ Control of pitch angle θ Ｐ and pitch rate Δ Ｐ The pitch rate Δ Ｐ is the pitch angle θ Ｐ is the time rate of change of

[0029] <Target Attitude> FIG. 3 shows the pitch angle θ that a passenger of the electric vehicle 100 can feel when accelerating. Ｐ 3 is a graph showing the lower limit value LL of the pitch angle θ that the occupant can feel during acceleration. Ｐ There is a lower limit LL for the pitch angle θ Ｐ When the pitch angle θ Ｐ It is difficult to detect the occurrence and fluctuations of

[0030] This lower limit LL is determined by the pitch rate Δ Ｐ Specifically, the pitch rate Δ Ｐ The larger the pitch angle θ Ｐ The lower limit value LL of the pitch rate Δ Ｐ is large, and the pitch angle θ Ｐ When the pitch angle θ Ｐ Even if the pitch angle θ Ｐ Conversely, the pitch rate Δ Ｐ is small, and the pitch angle θ Ｐ When the pitch angle θ Ｐ Even if the pitch angle θ Ｐ It is difficult to sense the fluctuations.

[0031] Pitch angle θ Ｐ The occurrence or fluctuation of the pitch angle θ does not usually cause any problems in terms of stable running (running stability) of the electric vehicle 100, but may deteriorate the riding comfort of the electric vehicle 100. For this reason, the controller 12 Ｐ Attitude control is performed by adjusting the distribution of driving force so that the value of the driving force is approximately equal to or less than the lower limit value LL.

[0032] However, in this embodiment, the pitch angle θ to be controlled by the attitude control Ｐ A predetermined upper limit value UL is set by experiment, simulation, or the like for the pitch rate Δ Ｐ It is a constant value that does not depend on

[0033] Then, the controller 12 calculates the pitch angle θ Ｐ is a value in the range of not less than the lower limit value LL and not more than the upper limit value UL (area E ２ａ When the pitch angle θ Ｐ is a value within the range below the lower limit LL (area E １ The posture of the electric vehicle 100 in the longitudinal direction is controlled so that the target posture of the electric vehicle 100 is the pitch angle θ ＰIn this embodiment, when performing attitude control, the controller 12 adjusts the pitch angle θ Ｐ is substantially zero or another predetermined angle (hereinafter referred to as the target pitch angle θ Ｐ ＊ The pitch angle θ is controlled to maintain the forward / backward attitude. Ｐ The predetermined angle may change depending on the specific running state of the electric vehicle 100.

[0034] On the other hand, pitch angle θ Ｐ is initially a value in the range below the lower limit value LL (area E １ When the pitch angle θ Ｐ is a value in the range exceeding the upper limit value UL (area E ２ｂ When the value of the driving force distribution is within the range of 0.5 to 1.0, the controller 12 does not execute attitude control by adjusting the driving force distribution.

[0035] That is, the controller 12 does not always execute attitude control, but performs an attitude control execution determination to switch attitude control on / off as needed. Specifically, when it is determined as a result of the attitude control execution determination that it is necessary to execute attitude control, the controller 12 turns on attitude control. As a result, attitude control is started, or the state in which attitude control is being executed is maintained. On the other hand, when it is determined as a result of the attitude control execution determination that it is not necessary to execute attitude control, the controller 12 turns off attitude control. As a result, attitude control is stopped, or the state in which attitude control is stopped is maintained.

[0036] More specifically, the controller 12 controls the pitch angle θ Ｐ is area E ２ａ When the pitch angle θ is within the range of θ 1 , it is determined that attitude control is necessary and attitude control is turned on. Ｐ is area E １ or area E ２ｂ When the value of the attitude control signal is reached, the controller 12 determines that it is not necessary to perform attitude control and turns off attitude control.

[0037] As shown in FIG. 3, the pitch angle θ Ｐhas a positive correlation with the acceleration G occurring in the electric vehicle 100, and the pitch rate Δ Ｐ has a positive correlation with the jerk J, which is the time rate of change of the acceleration G generated in the electric vehicle 100. Therefore, the controller 12 determines the pitch angle θ that the occupant can feel based on the acceleration G, the jerk J, or both. Ｐ Therefore, the controller 12 can determine whether to perform posture control based on the acceleration G, the jerk J, or both.

[0038] The controller 12 calculates the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ Therefore, the actual acceleration G ａｃｔ and / or actual jerk J ａｃｔ However, in this embodiment, the controller 12 can determine whether or not to perform posture control based on the actual acceleration G ａｃｔ and / or actual jerk J ａｃｔ Instead of the estimated acceleration G ｅｓｔ and / or estimated jerk J ｅｓｔ The attitude control execution decision is made based on the estimated acceleration G. This is to start attitude control without delay when it is necessary to perform it. ｅｓｔ is an estimated value of the acceleration G that will be generated in the electric vehicle 100. Similarly, the estimated jerk J ｅｓｔ is an estimated value of the jerk J that will be generated in the electric vehicle 100 from this.

[0039] The pitch angle θ that a passenger of the electric vehicle 100 may feel when accelerating or decelerating is Ｐ The lower limit of the road gradient φ ＬＳ Specifically, the road surface gradient φ ＬＳ When is large, the crew Ｐ Therefore, the controller 12 determines the estimated acceleration G ｅｓｔ and / or estimated jerk J ｅｓｔ In addition, the road gradient φ ＬＳ Specifically, the determination of whether to perform posture control can be made based on the road surface gradient φ ＬＳ is a predetermined threshold value TH ＬＳ(predetermined gradient), and the road surface on which the electric vehicle 100 is traveling is steeply inclined, and the occupant does not move at a pitch angle θ Ｐ When the change in the road surface gradient φ is difficult to perceive, the controller 12 determines that there is no need to perform the attitude control and turns off the attitude control. ＬＳ is the threshold TH ＬＳ is less than or equal to the estimated acceleration G ｅｓｔ and / or estimated jerk J ｅｓｔ When it is determined that attitude control is necessary based on the above, the controller 12 turns on attitude control.

[0040] <Configuration for Posture Control> Fig. 4 is a block diagram showing the configuration of the posture control controller 12. As shown in Fig. 4, the controller 12 includes a total driving force calculation unit 41, a basic distribution calculation unit 42, a posture control calculation unit 43, a driving force setting unit 44, a front motor control unit 45, and a rear motor control unit 46.

[0041] The total driving force calculation unit 41 calculates the total driving force TQ based on the operation of the accelerator pedal. The total driving force TQ is the driving force required for the electric vehicle 100. For example, the total driving force calculation unit 41 calculates the total driving force TQ based on the operation of the accelerator pedal A. ＰＯ and the total driving force TQ, and by referring to this map, the accelerator opening A ＰＯ The total driving force TQ corresponding to the above is calculated.

[0042] The total driving force calculation unit 41 calculates the accelerator opening A ＰＯ Instead of calculating the total driving force TQ based on the above, the total driving force TQ can be calculated based on commands from an ADAS (Advanced Drive Assistance System) or an AD (Autonomous Driving) system. These systems are systems that replace the driver's operation of the accelerator pedal, and therefore the calculation of the total driving force TQ that the total driving force calculation unit 41 performs based on commands from these systems is essentially a calculation based on the operation of the accelerator pedal.

[0043] The basic distribution calculation unit 42 distributes the total driving force TQ to the front wheels 21 and the rear wheels 26 in accordance with the basic distribution. The basic distribution is a driving force distribution determined so as to optimize the electricity consumption within a range that ensures driving stability, and is determined in advance through experiments, simulations, etc. For example, if the front motor 23 and the rear motor 28 are the same type and the electric vehicle 100 travels at a constant speed on a flat road, the basic distribution is front wheels:rear wheels = 50:50. The basic distribution may change depending on the specific driving state of the electric vehicle 100 (such as the steering state).

[0044] In this embodiment, the basic distribution calculation unit 42 calculates the first front torque target value T based on the basic distribution and the total driving force TQ. Ｆ１ ＊ , and the first rear torque target value T Ｒ１ ＊ The first front torque target value T Ｆ１ ＊ is the front wheel driving force F according to the basic distribution Ｆ represents the front motor torque generated on the front wheels 21. Ｒ１ ＊ is the rear wheel driving force F according to the basic distribution Ｒ represents the rear torque generated on the rear wheels 26.

[0045] The attitude control calculation unit 43 determines whether attitude control is necessary, and calculates a corrective drive force distribution that is a drive force distribution for attitude control that should be set when attitude control is necessary. The corrective drive force distribution is determined so as to cause the attitude of the electric vehicle 100 to approach or maintain a target attitude. Therefore, by distributing drive force to the front wheels 21 and the rear wheels 26 in accordance with the corrective drive force distribution, the attitude of the electric vehicle 100 is controlled to the target attitude.

[0046] More specifically, the attitude control calculation unit 43 of this embodiment includes an attitude control execution determination unit 47 and a correction unit 48 .

[0047] The posture control execution determination unit 47 performs a posture control execution determination to switch on / off posture control by adjusting the drive force distribution through feedforward control based on the total drive force TQ and the vehicle speed VSP. In this embodiment, the posture control execution determination unit 47 determines whether to execute posture control by adjusting the drive force distribution through feedforward control based on the total drive force TQ and the vehicle speed VSP. ＬＳ The attitude control execution determination is made based on the above. The result of the attitude control execution determination is represented by an attitude control flag FLG. The attitude control flag FLG is a flag that is set to "1" (on) when attitude control is required and set to "0" (off) when attitude control is not required, for example.

[0048] The correction unit 48 calculates the driving forces of the front wheels 21 and the rear wheels 26 (hereinafter referred to as "basic driving forces") for posture control by correcting the driving forces of the front wheels 21 and the rear wheels 26 distributed in accordance with the basic distribution (hereinafter referred to as "basic driving forces"). In this embodiment, the correction unit 48 calculates the first front torque target value T Ｆ１ ＊ and the first rear torque target value T Ｒ１ ＊ Based on this, the second front torque target value T corresponding to the corrected driving force is calculated. Ｆ２ ＊ and the second rear torque target value T Ｒ２ ＊ Correction unit 48 calculates the corrected driving force by feedforward control based on, for example, a predetermined vehicle model of electric vehicle 100. In this embodiment, the corrected distribution is determined as a result of calculating a specific corrected driving force.

[0049] The driving force setting unit 44 sets the driving force generated by the front wheels 21 and rear wheels 26 to either a basic driving force or a corrected driving force in accordance with the attitude control flag FLG.

[0050] Specifically, when the posture control execution determination determines that posture control is unnecessary and the posture control flag FLG is "0", the driving force setting unit 44 sets the driving force generated by the front wheels 21 and the rear wheels 26 to the basic driving force. In other words, when the posture control flag FLG is "0", the driving force setting unit 44 sets the first front torque target value T Ｆ１＊ is input to the front motor control unit 45, and the first rear torque target value T Ｒ１ ＊ is input to the rear motor control unit 46. This turns off the attitude control.

[0051] On the other hand, when the posture control execution determination determines that posture control is necessary and the posture control flag FLG is "1", the driving force setting unit 44 sets the driving force generated by the front wheels 21 and the rear wheels 26 to the corrected driving force. In other words, when the posture control flag FLG is "1", the driving force setting unit 44 sets the second front torque target value T Ｆ２ ＊ is input to the front motor control unit 45, and the second rear torque target value T Ｒ２ ＊ is input to the rear motor control unit 46. This turns on the attitude control.

[0052] The front motor control unit 45 controls the front motor 23 via the front inverter 22 so that the driving force set by the driving force setting unit 44 is generated at the front wheels 21. A first front torque target value T Ｆ１ ＊ When this is input, the front motor control unit 45 controls the front motor 23 to generate the first front torque target value T Ｆ１ ＊ On the other hand, a front torque corresponding to a second front torque target value T Ｆ２ ＊ When this is input, the front motor control unit 45 controls the front motor 23 to generate the second front torque target value T Ｆ２ ＊ This generates a front torque corresponding to the front wheel drive force F Ｆ is controlled to the basic driving force or the corrected driving force.

[0053] The rear motor control unit 46 controls the rear motor 28 via the rear inverter 27 so that the driving force set by the driving force setting unit 44 is generated at the rear wheels 26. A first rear torque target value T Ｒ１ ＊When this is input, the rear motor control unit 46 controls the rear motor 28 to generate the first rear torque target value T Ｒ１ ＊ On the other hand, a second rear torque target value T Ｒ２ ＊ When this is input, the rear motor control unit 46 controls the rear motor 28 to generate the second rear torque target value T Ｒ２ ＊ This generates a rear torque corresponding to the rear wheel driving force F Ｒ is controlled to the basic driving force or the corrected driving force.

[0054] The front motor control unit 45 and the rear motor control unit 46 constitute a driving force control unit that controls the driving forces of the front motor 23 and the rear motor 28 in accordance with the basic distribution or the corrected distribution.

[0055] Fig. 5 is a block diagram showing the configuration of the attitude control execution determination unit 47. As shown in Fig. 5, the attitude control execution determination unit 47 includes a running resistance calculation unit 51, an acceleration estimation unit 52, a jerk estimation unit 53, a gradient determination unit 54, a pitch state determination unit 55, and a flag setting unit 56.

[0056] The running resistance calculation unit 51 calculates the running resistance RL [N] of the electric vehicle 100 based on the vehicle speed VSP [m / s]. The running resistance RL is composed of air resistance, rolling resistance, acceleration resistance, etc., and can be approximated by a quadratic function of the vehicle speed VSP as shown in the following equation (1). The coefficient A of the quadratic term ０ , coefficient A of the first-order term １ , and a coefficient A representing a term (constant term) that does not depend on the vehicle speed VSP ２ can be determined in advance by, for example, an experiment or a simulation. ０ , A １ , A ２ The running resistance RL corresponding to the vehicle speed VSP is calculated using the above equation.

[0057]

[0058] The acceleration estimation unit 52 calculates an estimated acceleration G based on the total driving force TQ [Nm] and the running resistance RL. ｅｓｔ [m / sec ２ The acceleration estimation unit 52 calculates the estimated acceleration G ｅｓｔ That is, the acceleration estimation unit 52 calculates the estimated acceleration G ｅｓｔ is the total driving force TQ, the running resistance RL, the weight of the electric vehicle 100 (hereinafter referred to as vehicle weight M) [kg], and the gravitational acceleration g [m / sec ２ ], and road surface gradient φ ＬＳ [deg] to estimate the acceleration G ｅｓｔ The estimated acceleration G calculated by the acceleration estimation unit 52 is ｅｓｔ is an estimated value for the acceleration G that occurs when the electric vehicle 100 is driven by the total driving force TQ.

[0059]

[0060] The jerk estimation unit 53 estimates the acceleration G ｅｓｔ Based on this, the estimated jerk J ｅｓｔ [m / sec ３ Specifically, the jerk estimation unit 53 calculates the estimated acceleration G ｅｓｔ By differentiating with time, the estimated jerk J ｅｓｔ The estimated jerk J calculated by the jerk estimator 53 is ｅｓｔ is an estimated value for the time rate of change (jerk J) of acceleration G that occurs when electric vehicle 100 is driven by total driving force TQ.

[0061]

[0062] The gradient determination unit 54 determines the road surface gradient φ ＬＳ Specifically, the gradient determination unit 54 determines whether or not posture control is possible based on the road surface gradient φ ＬＳ and a predetermined threshold value TH ＬＳ Then, compare the road surface gradient φ ＬＳ is the threshold TH ＬＳ On the other hand, when the road surface gradient φ is equal to or less than φ, the gradient determination unit 54 determines that the posture control can be performed. ＬＳ is the threshold THＬＳ When the road surface gradient φ is greater than φ, the gradient determination unit 54 determines that it is not necessary to perform posture control. ＬＳ The threshold value TH ＬＳ is determined by fitting through experiments or simulations.

[0063] The pitch state determination unit 55 determines the pitch angle θ that occurs when the electric vehicle 100 is driven by the total driving force TQ. Ｐ The pitch state determination unit 55 determines whether or not attitude control is necessary based on the pitch angle θ Ｐ Estimated acceleration G, which is positively correlated with ｅｓｔ , pitch rate Δ Ｐ The estimated jerk J is positively correlated with ｅｓｔ , or both of them, the pitch state determination unit 55 can determine whether or not posture control is necessary. ｅｓｔ and estimated jerk J ｅｓｔ Based on this, it is determined whether attitude control is necessary.

[0064] Specifically, the pitch state determination unit 55 calculates the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ The pitch angle θ corresponding to Ｐ and pitch rate Δ Ｐ (hereinafter referred to as pitch state) is something that the occupant can feel and is within a range that requires control. ２ａ (See FIG. 3) and determine whether the pitch state is within the range of region E. ２ａ When the pitch state is in the range of region E, the pitch state determination unit 55 determines that the posture control needs to be executed (ON). ２ａ If the pitch state is not within the range, the pitch state determination unit 55 determines that it is not necessary to perform attitude control.

[0065] The above determination by the pitch state determination unit 55 is made based on the pitch angle θ Ｐ is equal to or greater than the lower limit value LL and equal to or less than the upper limit value UL. ｅｓｔThe pitch angle θ corresponding to Ｐ , the estimated jerk J ｅｓｔ The pitch rate Δ Ｐ By comparing the calculated value with a lower limit value LL that is set in advance depending on the target position, it is possible to determine whether or not attitude control is required.

[0066] In this embodiment, the pitch state determination unit 55 determines whether or not posture control is required by a particularly simple or approximate method, as follows. ｅｓｔ A predetermined threshold value TH Ｇ (acceleration threshold) and estimate jerk J ｅｓｔ A predetermined threshold value TH Ｊ By setting the jerk threshold, the region where posture control is particularly necessary is determined. ｅｓｔ is the threshold TH Ｇ (and not more than the upper limit value UL), and the estimated jerk J ｅｓｔ is the threshold TH Ｊ When the estimated acceleration G ｅｓｔ is the threshold TH Ｇ or when the estimated jerk J ｅｓｔ is the threshold TH Ｊ When the difference is smaller than , the pitch state determination unit 55 determines that it is not necessary to perform posture control.

[0067] In addition, the estimated jerk J ｅｓｔ The threshold value TH Ｊ is the estimated acceleration G ｅｓｔ Similarly, the threshold value can be a variable threshold that changes depending on the estimated acceleration G ｅｓｔ The threshold value TH Ｇ is the estimated jerk J ｅｓｔ It can be a variable threshold that changes depending on

[0068] Furthermore, the pitch angle θ at which the pitch state determination unit 55 determines that it is necessary to perform (ON) attitude control is Ｐ The range of the estimated acceleration G ｅｓｔ is the threshold THＧ or more (and less than the upper limit value UL), or the estimated jerk J ｅｓｔ is the threshold TH Ｊ When any of the above conditions is satisfied, it may be determined that the posture control needs to be performed (ON). In this case, the estimated acceleration G ｅｓｔ is the threshold TH Ｇ and the estimated jerk J ｅｓｔ is the threshold TH Ｊ When the difference is smaller than , the pitch state determination unit 55 determines that it is not necessary to perform posture control.

[0069] The flag setting unit 56 sets the attitude control flag FLG based on the determination results of the gradient determination unit 54 and the pitch state determination unit 55. Specifically, when the gradient determination unit 54 determines that attitude control is not required, the flag setting unit 56 sets the attitude control flag FLG to "0" (off), regardless of the determination result of the pitch state determination unit 55. When the gradient determination unit 54 determines that attitude control is available, the flag setting unit 56 sets the attitude control flag FLG based on the determination result of the pitch state determination unit 55. That is, when the gradient determination unit 54 determines that attitude control is available and the pitch state determination unit 55 determines that attitude control needs to be performed (on), the flag setting unit 56 sets the attitude control flag FLG to "1" (on). On the other hand, when the gradient determination unit 54 determines that attitude control is available but the pitch state determination unit 55 determines that attitude control is not required, the flag setting unit 56 sets the attitude control flag FLG to "0" (off).

[0070] <Operation> Hereinafter, the operation of the attitude control of the electric vehicle 100 configured as described above, particularly the operation related to turning on / off the attitude control, will be described.

[0071] 6 is a flowchart showing the operation related to the on / off of the attitude control in the electric vehicle 100. As shown in FIG. 6, the total driving force calculation unit 41 calculates the accelerator opening A ＰＯ When the accelerator opening degree A is acquired, in step S11, ＰＯNext, in step S12, the running resistance calculation unit 51 calculates the running resistance RL of the electric vehicle 100 based on the total driving force TQ and the vehicle speed VSP. Then, in step S13, the acceleration estimation unit 52 calculates the estimated acceleration G ｅｓｔ In step S14, the jerk estimation unit 53 calculates the estimated acceleration G ｅｓｔ Using this, the estimated jerk J ｅｓｔ Calculate the following.

[0072] Then, in step S15, the gradient determination unit 54 determines the road surface gradient φ ＬＳ and obtain the road surface gradient φ ＬＳ Based on this, it is determined whether or not posture control is possible. ＬＳ is the threshold TH ＬＳ is larger than the road gradient φ ＬＳ Therefore, the crew sets the pitch angle θ Ｐ If the change is not felt, the process proceeds to step S19, where the attitude control is stopped or the stopped state of the attitude control is maintained. ＬＳ is the threshold TH ＬＳ or less, and the occupant is at a pitch angle θ Ｐ If the fluctuations can be felt and there is a possibility that the riding comfort of the electric vehicle 100 will be deteriorated, the process proceeds to step S16, and the generated pitch state (θ Ｐ , Δ Ｐ ) based on which it is determined whether attitude control is necessary.

[0073] In step S16, the pitch state determination unit 55 calculates the estimated acceleration G ｅｓｔ is the threshold value TH Ｇ Compare with the estimated acceleration G ｅｓｔ is the threshold TH Ｇ If so, the process proceeds to step S17, where the pitch state determination unit 55 further calculates the estimated jerk J ｅｓｔ is the threshold value TH Ｊ Then, the estimated jerk J ｅｓｔ is the threshold TH Ｊ or more, i.e., the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ are all thresholds TH Ｇ , T.H. ＪIf the above is true, it is determined that the execution (ON) of the attitude control is necessary, and the process proceeds to step S18, where the attitude control is executed or the execution of the attitude control is maintained.

[0074] On the other hand, in step S16, the estimated acceleration G ｅｓｔ is the threshold TH Ｇ or when the estimated jerk J ｅｓｔ is the threshold TH Ｊ If the difference is smaller than , it is determined that the posture control does not need to be performed, and therefore the process proceeds to step S19, where the posture control is stopped or maintained in a stopped state.

[0075] 7A and 7B are time charts showing the transition of parameters when the attitude control is switched from OFF to ON. FIG. 7A shows the transition of the vehicle speed VSP. FIG. 7B shows the transition of the accelerator opening A. ＰＯ 7(C) shows the transition of the total driving force TQ. FIG. 7(D) shows the transition of the acceleration G of the electric vehicle 100. In FIG. 7(D), the solid line indicates the estimated acceleration G. ｅｓｔ The dashed line indicates the actual acceleration G ａｃｔ 7(E) shows the transition of the jerk J of the electric vehicle 100. In FIG. 7(E), the solid line indicates the estimated jerk J. ｅｓｔ The dashed line indicates the actual jerk J ａｃｔ Shows.

[0076] 7(F) shows the transition of the attitude control flag FLG. In FIG. 7(F), the solid line indicates the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ 7(F) shows the attitude control flag FLG in this embodiment, in which the attitude control execution determination is performed by feedforward control based on the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ , or the actual pitch angle θ Ｐ and pitch rate Δ Ｐ 7(G) shows the attitude control flag FLG in a comparative example in which attitude control execution determination is performed by feedback control based on the pitch angle θＰ In FIG. 7G, the solid line indicates the transition of the pitch angle θ when the attitude control execution determination of this embodiment is performed. Ｐ In FIG. 7G, the dashed line indicates the pitch angle θ Ｐ The horizontal axis of each time chart in FIG. 7 represents time [sec].

[0077] As shown in Fig. 7A, the electric vehicle 100 is initially in a stopped state. Then, as shown in Fig. 7B, at time t １ The accelerator pedal is depressed from the accelerator opening A ＰＯ When the step increase occurs, the total driving force TQ is generated and increases accordingly, as shown in FIG. 7C. As a result, as shown in FIG. 7A, the control delay increases the time t １ Time t, which is delayed by a predetermined amount from ４ This increases the vehicle speed VSP of the electric vehicle 100. The control delay is a delay in control due to a communication delay, a torque response delay of the front motor 23 and the rear motor 28, and the like.

[0078] At this time, as shown in FIG. 7(D), the estimated acceleration G ｅｓｔ is substantially immediately followed by time t １ Therefore, as shown in FIG. 7(E), the estimated jerk J ｅｓｔ However, substantially without delay, at time t １ In addition, as shown in FIG. 7D, the estimated acceleration G ｅｓｔ is the time t １ and time t ４ Time t between ２ At the threshold TH Ｇ As shown in FIG. 7(E), the estimated jerk J ｅｓｔ is the time t ２ and time t ４ Time t between ３ In this case, the threshold value TH Ｊ Therefore, at time t ３ In this case, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ are all thresholds TH Ｇ , T.H. ＪTherefore, as shown in FIG. 7F, in this embodiment (solid line), the attitude control flag FLG is ３ The signal transitions from "0" (off) to "1" (on), and attitude control begins.

[0079] On the other hand, as shown in FIG. 7(D), the actual acceleration G ａｃｔ is controlled by a control delay at time t ４ Therefore, as shown in FIG. 7(E), the actual jerk J ａｃｔ However, due to control delay, ４ As shown in FIG. 7(D), the actual acceleration G ａｃｔ is the threshold TH Ｇ The time t ５ is the time t ４ Here, as shown in FIG. 7(E), the actual jerk J ａｃｔ is the threshold TH Ｊ As a result, the time t ６ In this case, the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ are all thresholds TH Ｇ , T.H. Ｊ Therefore, as shown in FIG. 7F, in the comparative example (dotted line), the attitude control flag FLG is set to 0 at least from time t ４ Time t later than ６ The signal transitions from "0" (off) to "1" (on), and attitude control begins.

[0080] Therefore, when comparing this embodiment with the comparative example, in this embodiment, the time t ３ When the posture control execution determination is completed and the acceleration G or the like actually occurs in the electric vehicle 100, the posture control is started. In contrast, in the comparative example, the posture control execution determination is completed at time t ４ For this reason, in the comparative example, it is unavoidable that a delay of at least the amount of control delay occurs after acceleration G or the like actually occurs in electric vehicle 100 until attitude control is started.

[0081] As a result, as shown in FIG. 7G, in this embodiment (solid line), the pitch angle θ Ｐ The time t at which the change occurs ４ After that, the pitch rate Δ Ｐ (inclination) while suppressing the pitch angle θ Ｐ is the target pitch angle θ Ｐ ＊ That is, in this embodiment, the pitch angle θ Ｐ Attitude control can be activated from the very beginning when fluctuations occur.

[0082] On the other hand, in the comparative example (dashed line), at time t ４ After that, the attitude control starts at time t ６ During this time, a steep pitch rate Δ Ｐ and pitch angle θ Ｐ rises and the target pitch angle θ Ｐ ＊ Then, at time t ６ Attitude control is initiated at the pitch angle θ Ｐ is the target pitch angle θ Ｐ ＊ Therefore, in the comparative example, at time t ４ From time t ６ During the period when attitude control is not working, the pitch angle θ Ｐ That is, in the comparative example, the attitude control is not able to keep up.

[0083] Second Embodiment In the first embodiment, the posture control execution determination unit 47 determines the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ The necessity of attitude control is determined based on the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ However, the specific configuration of the posture control execution determination unit 47 for determining whether posture control is necessary is not limited to this. ｅｓｔ and / or estimated jerk J ｅｓｔ In addition to this, the actual acceleration G ａｃｔ , Real Jerk J ａｃｔ, or both of them, the necessity of posture control can be determined. In the second embodiment described below, as an example, the posture control execution determination unit 47 determines whether posture control is necessary or not based on the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ The actual acceleration G is used to determine whether or not posture control is necessary based on ａｃｔ and Real Jerk J ａｃｔ In this embodiment, the necessity of attitude control is determined comprehensively by combining the necessity of attitude control based on the above.

[0084] 8 is a block diagram showing the configuration of a posture control execution determination unit 47 in the second embodiment. As shown in FIG. 8, the posture control execution determination unit 47 in the second embodiment includes a running resistance calculation unit 51, an acceleration estimation unit 52, a jerk estimation unit 53, and a gradient determination unit 54, similar to those in the first embodiment. The posture control execution determination unit 47 in the second embodiment also includes a first pitch state determination unit 201 and a second pitch state determination unit 202, instead of the pitch state determination unit 55 in the first embodiment. The posture control execution determination unit 47 in the second embodiment also includes a flag setting unit 203, instead of the flag setting unit 56 in the first embodiment.

[0085] The first pitch state determination unit 201, like the pitch state determination unit 55 of the first embodiment, calculates the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ Based on this, the pitch state (θ Ｐ , Δ Ｐ ) and determines whether or not it is necessary to perform attitude control. Here, the first pitch state determination unit 201 determines whether or not attitude control is necessary using a simple or approximate method, similar to the first embodiment. Specifically, the first pitch state determination unit 201 determines whether or not attitude control is necessary based on the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ for each threshold TH Ｇ , T.H. Ｊ is set in advance. Then, the estimated acceleration G ｅｓｔ is the threshold TH Ｇ (and not more than the upper limit value UL), and the estimated jerk J ｅｓｔ is the threshold TH ＪWhen the estimated acceleration G is equal to or greater than the above, the first pitch state determination unit 201 determines that the posture control needs to be executed (ON). ｅｓｔ is the threshold TH Ｇ or when the estimated jerk J ｅｓｔ is the threshold TH Ｊ When the difference is smaller than , the first pitch state determination unit 201 determines that it is not necessary to perform posture control.

[0086] The second pitch state determination unit 202 determines the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ and determines whether or not it is necessary to perform attitude control based on these. That is, the second pitch state determination unit 202 determines whether or not it is necessary to perform attitude control based on the actually occurring pitch state (θ Ｐ , Δ Ｐ ) and determine whether or not attitude control is necessary. ｅｓｔ and estimated jerk J ｅｓｔ Instead of actual acceleration G ａｃｔ and Real Jerk J ａｃｔ Except for using the above, the second pitch state determination unit 202 can determine whether or not attitude control is required in the same manner as the pitch state determination unit 55 of the first embodiment or the first pitch state determination unit 201.

[0087] Here, the second pitch state determination unit 202 determines the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ That is, the second pitch state determination unit 202 determines whether or not posture control is necessary based on the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ for each threshold TH Ｇ , T.H. Ｊ Set the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ The threshold value TH Ｇ , T.H. Ｊ is the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ The threshold value TH Ｇ , T.H. ＪHowever, for simplicity, the same threshold value TH Ｇ , T.H. Ｊ is set. Then, the actual acceleration G ａｃｔ is the threshold TH Ｇ (and lower than the upper limit value UL), and the actual jerk J ａｃｔ is the threshold TH Ｊ When the actual acceleration G ａｃｔ is the threshold TH Ｇ When the actual jerk J is smaller than ａｃｔ is the threshold TH Ｊ When the difference is smaller than , the second pitch state determination unit 202 determines that it is not necessary to perform posture control.

[0088] The flag setting unit 203 sets the attitude control flag FLG based on the determination results of the gradient determination unit 54 , the first pitch state determination unit 201 , and the second pitch state determination unit 202 .

[0089] Specifically, when the gradient determination unit 54 determines that posture control does not need to be performed, the flag setting unit 203 sets the posture control flag FLG to "0" (off) regardless of the determination results of the first pitch state determination unit 201 and the second pitch state determination unit 202.

[0090] On the other hand, when the gradient determination unit 54 determines that attitude control is executable, the flag setting unit 203 sets the attitude control flag FLG based on the determination results of the first pitch state determination unit 201 and the second pitch state determination unit 202. Specifically, when the gradient determination unit 54 determines that attitude control is executable and the first pitch state determination unit 201 determines that execution (ON) of attitude control is necessary, the flag setting unit 203 sets the attitude control flag FLG to "1" (ON) regardless of the determination result of the second pitch state determination unit 202. Also, when the gradient determination unit 54 determines that attitude control is executable but the first pitch state determination unit 201 determines that execution of attitude control is not necessary, the flag setting unit 203 sets the attitude control flag FLG to "1" (ON) when the second pitch state determination unit 202 determines that execution (ON) of attitude control is necessary. Then, when the gradient determination unit 54 determines that attitude control can be performed, but both the first pitch state determination unit 201 and the second pitch state determination unit 202 determine that attitude control does not need to be performed, the flag setting unit 203 sets the attitude control flag FLG to "0" (off).

[0091] That is, when gradient determination unit 54 determines that attitude control is executable, and either first pitch state determination unit 201 or second pitch state determination unit 202 determines that execution (on) of attitude control is necessary, flag setting unit 203 sets attitude control flag FLG to "1" (on). When gradient determination unit 54 determines that attitude control is executable and both first pitch state determination unit 201 and second pitch state determination unit 202 determine that execution of attitude control is not necessary, flag setting unit 203 sets attitude control flag FLG to "0" (off).

[0092] Hereinafter, the operation of the electric vehicle 100 of the second embodiment configured as described above, relating to the on / off of the attitude control, will be described.

[0093] 9 is a flowchart showing the operation of turning on / off posture control for the electric vehicle 100 of the second embodiment. As shown in FIG. 9, in step S20, the total driving force calculation unit 41 calculates the accelerator opening A ＰＯ The second pitch state determination unit 202 obtains the actual acceleration G ａｃｔand Real Jerk J ａｃｔ Next, in step S21, the total driving force calculation unit 41 calculates the total driving force TQ based on the accelerator opening APO, and the running resistance calculation unit 51 calculates the running resistance RL of the electric vehicle 100 based on the total driving force TQ and the vehicle speed VSP. Also in step S21, the acceleration estimation unit 52 calculates the estimated acceleration G ｅｓｔ The jerk estimation unit 53 calculates the estimated acceleration G ｅｓｔ Estimate the jerk J using ｅｓｔ Calculate the following.

[0094] Then, in step S22, the gradient determination unit 54 determines the road surface gradient φ ＬＳ The possibility of posture control is judged based on the road surface gradient φ ＬＳ is the threshold TH ＬＳ is larger than the road gradient φ ＬＳ Therefore, the crew sets the pitch angle θ Ｐ If the change is not felt, the process proceeds to step S28, where the attitude control is stopped or the stopped state of the attitude control is maintained. ＬＳ is the threshold TH ＬＳ or less, and the occupant is at a pitch angle θ Ｐ If the fluctuations can be felt and there is a possibility that the riding comfort of the electric vehicle 100 will be deteriorated as a result, the process proceeds to step S23, and the pitch state (θ Ｐ , Δ Ｐ ), or the actual pitch state (θ Ｐ , Δ Ｐ ) based on which it is determined whether attitude control is necessary.

[0095] In step S23, the first pitch state determination unit 201 calculates the estimated acceleration G ｅｓｔ is the threshold value TH Ｇ Compare with the estimated acceleration G ｅｓｔ is the threshold TH Ｇ If so, the process proceeds to step S24, where the first pitch state determination unit 201 further calculates the estimated jerk J ｅｓｔ is the threshold value TH Ｊ Then, the estimated jerk J ｅｓｔ is the threshold TH Ｊ or more, i.e., the estimated acceleration Gｅｓｔ and estimated jerk J ｅｓｔ are all thresholds TH Ｇ , T.H. Ｊ If the above is true, it is determined that the execution (ON) of the attitude control is necessary, and the process proceeds to step S27, where the attitude control is executed or the execution of the attitude control is maintained.

[0096] On the other hand, in step S23, the estimated acceleration G ｅｓｔ is the threshold TH Ｇ or when the estimated jerk J is smaller than ｅｓｔ is the threshold TH Ｊ If the actual acceleration G is smaller than the actual acceleration G, the process proceeds to step S25. ａｃｔ is the threshold value TH Ｇ Then, compare with the actual acceleration G ａｃｔ is the threshold TH Ｇ If the actual jerk J is equal to or greater than the predetermined value, the process proceeds to step S26, where the second pitch state determination unit 202 further ａｃｔ is the threshold value TH Ｊ Compare with.

[0097] In step S25, the actual acceleration G ａｃｔ is the threshold TH Ｇ In step S26, the actual jerk J ａｃｔ is the threshold TH Ｊ If the estimated acceleration G is equal to or greater than the estimated acceleration G, it is determined that the posture control needs to be performed (ON). Therefore, the process proceeds to step S27, where the posture control is performed or the execution of the posture control is maintained. ｅｓｔ and estimated jerk J ｅｓｔ Even if it is determined that posture control is not required as a result of the determination based on the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ If it is determined that the attitude control needs to be performed (ON) as a result of the determination based on the above, the attitude control is turned ON.

[0098] On the other hand, in step S25, the actual acceleration G ａｃｔ is the threshold TH Ｇ or when the actual jerk J ａｃｔ is the threshold THＪ If the estimated acceleration G is smaller than the estimated acceleration G, it is determined that the posture control does not need to be performed. Therefore, the process proceeds to step S28, where the posture control is stopped or the stopped state of the posture control is maintained. ｅｓｔ and estimated jerk J ｅｓｔ As a result of the determination based on the above, it is determined that the execution of attitude control is not necessary, and the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ If it is determined that the attitude control does not need to be performed as a result of the determination based on the above, the attitude control is turned off.

[0099] In this way, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ and the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ When the necessity of attitude control is determined by combining the judgment based on Ｐ For example, if there is a change in the vehicle weight M or a change in air resistance due to the installation of aftermarket parts, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ There is a non-negligible calculation error in the running resistance RL required for the calculation of the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ Before it is determined that attitude control needs to be performed (ON) based on the pitch angle θ Ｐ In this case, as described above, the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ If the determination of whether attitude control is necessary based on the above is used in combination, the pitch angle θ Ｐ And when the fluctuation actually occurs, attitude control can be made to function.

[0100] [Third Embodiment] In the first and second embodiments, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ is calculated based on the running resistance RL, which is approximated by a quadratic function of the vehicle speed VSP.０ , A １ , A ２ is determined in advance through experiments, simulations, etc. However, the running resistance RL can change depending on the specific usage state of the electric vehicle 100. For example, when the vehicle weight M changes or when the air resistance changes due to the installation of aftermarket parts, the running resistance RL of the electric vehicle 100 changes. In this way, when there is a significant change in the running resistance RL, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ In order to maintain the accuracy of the attitude control execution judgment using ０ , A １ , A ２ It is desirable to update the coefficient A used to calculate the running resistance RL according to the specific usage state of the electric vehicle 100. In the third embodiment described below, ０ , A １ , A ２ An example of updating the parameter value in accordance with the specific usage state of the electric vehicle 100 will be described.

[0101] FIG. 10 shows the coefficient A of the approximation formula used to calculate the running resistance RL. ０ , A １ , A ２ 10 is a block diagram showing a configuration for updating coefficients 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 129, 130, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146,

[0102] The coefficient update unit 301 includes a flat road determination unit 311 , a steady running determination unit 312 , a first error calculation unit 313 , a running resistance storage unit 314 , a second error calculation unit 315 , and a coefficient update unit 316 .

[0103] The flat road determination unit 311 determines the road surface gradient φ ＬＳ In other words, the flat road determination unit 311 determines whether the electric vehicle 100 is traveling on a flat road in a state where the gradient resistance is not included in the running resistance RL. Specifically, the flat road determination unit 311 determines whether the electric vehicle 100 is traveling on a flat road based on the road surface gradient φ ＬＳ is a predetermined error ±φ ０ When it is close to zero below (-φ ０≦φ ＬＳ ≦+φ ０ ), it is determined that the electric vehicle 100 is traveling on a flat road.

[0104] The steady running determination unit 312 determines whether the electric vehicle 100 is running steadily at a constant speed without accelerating or decelerating (so-called road-load running). In other words, the steady running determination unit 312 determines whether the electric vehicle 100 is running in a state where the resistance due to acceleration or deceleration is not included in the running resistance RL. In this embodiment, the steady running determination unit 312 determines whether the electric vehicle 100 is running in a state where the resistance due to acceleration or deceleration is not included in the running resistance RL. ａｃｔ and obtain the actual acceleration G ａｃｔ Specifically, it is determined whether the steady running is continuing based on the actual acceleration G ａｃｔ is a predetermined error ±G ０ Below (-G ０ ≦G ａｃｔ ≦+G ０ ) the electric vehicle 100 is determined to be in steady running.

[0105] The first error calculation unit 313 calculates the estimated acceleration G ｅｓｔ and actual acceleration G ａｃｔ The first error G ｅｒｒ The first error calculation unit 313 calculates the estimated acceleration G ｅｓｔ However, the first error calculation unit 313 obtains the estimated acceleration G ｅｓｔ In this embodiment, the first error calculation unit 313 may calculate the estimated acceleration G ｅｓｔ From the actual acceleration G ａｔ By subtracting ｅｒｒ Calculate the following.

[0106] The first error calculation unit 313 calculates the first error G ｅｒｒ Based on this, the estimated acceleration G ｅｓｔ The period during which significant errors continue to occur (hereafter, the duration τ ｅｒｒ Specifically, the first error calculation unit 313 counts the calculated first error G ｅｒｒ A predetermined threshold value TH ｅｒｒ１ Compared with the first error G ｅｒｒ is the threshold TH ｅｒｒ１The time when the time is equal to or greater than the duration τ ｅｒｒ Let's say.

[0107] The running resistance storage unit 314 at least temporarily stores the running resistance RL when the electric vehicle 100 is running steadily on a flat road in association with the vehicle speed VSP at that time. Specifically, the running resistance storage unit 314 stores the first error G ｅｒｒ is the threshold TH ｅｒｒ１ The duration τ ｅｒｒ is a predetermined threshold τ ０ If the total driving force TQ is equal to or greater than the above, the running resistance storage unit 314 acquires the total driving force TQ at that time. Then, the running resistance storage unit 314 stores the acquired total driving force TQ as the actual running resistance RL (hereinafter referred to as the actual running resistance RL) during steady running on a flat road. ａｃｔ ) and stored in association with the vehicle speed VSP.

[0108] The second error calculation unit 315 calculates the actual running resistance RL stored in the running resistance storage unit 314. ａｃｔ and the running resistance RL calculated by the running resistance calculation unit 51, ｅｒｒ (Running resistance error) is calculated. ｅｒｒ The calculation of the second error RL is performed for each vehicle speed VSP. ｅｒｒ By calculating the variance and other statistical values ​​of the second error RL ｅｒｒ Evaluate the variation.

[0109] The coefficient update unit 316 updates the actual running resistance RL ａｃｔ Based on this, the running resistance calculation unit 51 calculates the coefficient A ０ , A １ , A ２ The coefficient update unit 316 updates (calculates) all or part of the coefficient A ０ , A １ , A ２ or by correcting the new coefficient A ０ , A １ , A ２ is calculated to obtain the pre-stored coefficient A ０ , A １ , A ２By replacing ０ , A １ , A ２ Update.

[0110] In this embodiment, the coefficient update unit 316 stores at least three actual running resistances RL ａｃｔ The second error RL ｅｒｒ The variation is a predetermined threshold TH ｅｒｒ２ When the coefficient A of the constant term is equal to or less than the above, the coefficient updating unit 316 ２ Specifically, the coefficient update unit 316 updates the second error RL ｅｒｒ Based on this, the coefficient A of the constant term ２ By correcting the coefficient A of the constant term ２ For example, the coefficient update unit 316 updates the coefficient A of the previous constant term. ２ , the second error RL ｅｒｒ By adding the average value of ２ Correct the following.

[0111] In this embodiment, the running resistance memory unit 314 stores at least three actual running resistances RL ａｃｔ Although the second error RL ｅｒｒ The variation is the threshold TH ｅｒｒ２ When the coefficient A of the quadratic and linear terms is larger than ０ , A １ Specifically, the coefficient update unit 316 updates the actual running resistance RL ａｃｔ Based on this, the coefficients A of the quadratic and linear terms ０ , A １ For example, the coefficient update unit 316 updates the coefficient A of the constant term by newly calculating ２ Using the previous value as the actual running resistance RL ａｃｔ By finding a quadratic regression curve based on ０ , A １ Calculate the following.

[0112] FIG. 11 shows the coefficient A of the approximation formula used to calculate the running resistance RL. ０ , A １ , A ２11 is an explanatory diagram showing a specific manner in which the coefficient A is updated. In FIG. 11, the data set indicated by the circle (○) legend is a data set obtained by updating the coefficient A stored in advance in the initial state of the electric vehicle 100. ０ , A １ , A ２ Furthermore, each data set indicated by a triangle (△) and a square (□) legend indicates the running resistance RL stored in the running resistance storage unit 314 in an independent running scene in which the specific usage situation of the electric vehicle 100 is different.

[0113] As shown in FIG. 11, when the running resistance RL indicated by the triangle (△) legend is obtained by steady running on a flat road, the second error RL ｅｒｒ Is E １Ａ , E １Ｂ , and E １Ｃ Then, these second errors RL ｅｒｒ (=E １Ａ , E １Ｂ , E １Ｃ ) are all of the same magnitude, and the variation is within the threshold value TH ｅｒｒ２ Therefore, the coefficient update unit 316 updates these second errors RL ｅｒｒ The average value of E １Ａ , E １Ｂ , E １Ｃ The average value of the previous coefficient A ２ By adding to the coefficient A of the constant term ２ This translates the quadratic curve that approximates the running resistance RL from the solid line to the dashed line. Therefore, the coefficient A of the constant term ２ This update is suitable when the running resistance components (rolling resistance, etc.) other than air resistance change due to a change in the vehicle weight M, for example.

[0114] When the running resistance RL indicated by the square (□) legend is obtained by steady running on a flat road, the second error RL at each data point is ｅｒｒ Is E ２Ａ , E ２Ｂ , and E ２Ｃ Then, these second errors RL ｅｒｒ (=E ２Ａ , E ２Ｂ , E２Ｃ ) are different in magnitude, and the variation is determined by the threshold value TH ｅｒｒ２ Therefore, the coefficient update unit 316 updates the coefficient A of the constant term ２ By finding a quadratic regression curve for the data points indicated by the square (□) legend while maintaining the coefficients A of the quadratic and linear terms, ０ , A １ This changes the shape of the quadratic curve that approximates the running resistance RL from a solid line to a dashed line. Therefore, the coefficient A of the quadratic and linear terms ０ , A １ This update is suitable when the air resistance of the electric vehicle 100 changes due to the installation or replacement of aftermarket parts.

[0115] FIG. 12 shows the coefficient A of the approximation formula used to calculate the running resistance RL. ０ , A １ , A ２ As shown in Fig. 12, in step S30, the flat road determination unit 311 determines whether the road surface gradient φ ＬＳ Whether or not the electric vehicle 100 is traveling on a flat road is determined based on the above. If it is determined that the electric vehicle 100 is traveling on a flat road, the process proceeds to step S31, where the steady-state traveling determination unit 312 determines whether or not the electric vehicle 100 is traveling at a steady state without accelerating or decelerating. If it is determined that the electric vehicle 100 is traveling at a steady state, the process proceeds to step S32.

[0116] In step S32, the first error calculation unit 313 calculates the estimated acceleration G ｅｓｔ and actual acceleration G ａｃｔ Based on this, the first error G ｅｒｒ and calculate the first error G ｅｒｒ The duration τ for which there is a significant error ｅｒｒ Then, in step S33, the duration τ ｅｒｒ is the threshold τ ０ If the total driving force TQ is equal to or greater than the actual running resistance RL, the process proceeds to step S34. ａｃｔ and stores it in association with the vehicle speed VSP.

[0117] Then, in step S35, the actual running resistance RL ａｃｔ If the number of errors is three or more, in step S36, the second error calculation unit 315 calculates the second error RL ｅｒｒ is calculated and its variation is evaluated.

[0118] Then, in step S37, the coefficient update unit 316 updates the second error RL ｅｒｒ The variation is the threshold TH ｅｒｒ２ Then, it is determined whether the second error RL ｅｒｒ The variation is the threshold TH ｅｒｒ２ If it is equal to or less than this, the process proceeds to step S38, and the coefficient updating unit 316 updates the coefficient A of the constant term in the approximation formula used to calculate the running resistance RL. ２ On the other hand, the second error RL ｅｒｒ The variation is the threshold TH ｅｒｒ２ If the coefficient A is greater than 1, the process proceeds to step S39, and the coefficient update unit 316 updates the coefficient A of the second-order and first-order terms in the approximation formula used to calculate the running resistance RL. ０ , A １ Update.

[0119] As described above, the coefficient A in the approximation formula used to calculate the running resistance RL is ０ , A １ , A ２ By updating the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ is accurately calculated. As a result, the estimated acceleration G ｅｓｔ and / or estimated jerk J ｅｓｔ The attitude control execution decision using the actual pitch angle θ Ｐ Therefore, the attitude control is performed accurately before the actual pitch angle θ Ｐ and will be initiated in a timely manner without delay in response to the occurrence of such fluctuations.

[0120] The third embodiment can be implemented in combination with either the first or second embodiment. In the third embodiment, the first error calculation unit 313 calculates the first error G ｅｒｒHowever, the first error calculation unit 313 is not limited to this, and may calculate a first error (J ｅｒｒ In this case, the subsequent processing is performed by calculating the first error (J ｅｒｒ ) in the same manner as in the third embodiment. The first error calculation unit 313 calculates an error (G ｅｒｒ ) and the error (J ｅｒｒ ) can be calculated as the first error. In this case, the subsequent processing is performed using the error (G ｅｒｒ ) and the error (J ｅｒｒ For example, the error (G ｅｒｒ ) is the threshold τ ０ and / or the error (J ｅｒｒ ) is the threshold τ ０ If this continues, the running resistance storage unit 314 stores the actual running resistance RL ａｃｔ The configuration can be such that the following is stored.

[0121] In the first, second, and third embodiments, the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ The posture control execution decision is made using both the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ Alternatively, only one of the actual acceleration G ａｃｔ and Real Jerk J ａｃｔ The attitude control execution decision is made using both the actual acceleration G ａｃｔ Or Real Jerk J ａｃｔ Only one of the above may be used for determining whether or not to execute attitude control. This also applies when the third embodiment is combined with the second embodiment.

[0122] In the first, second, and third embodiments, the corrector 48 performs the attitude control by feedforward control, but the present invention is not limited to this. Ｐ or pitch rate Δ Ｐ Even when the posture control is performed by feedback control that feeds back the above, it is preferable to determine whether to perform posture control by feedforward control, as in the first, second, and third embodiments.

[0123] As described above, the control methods for electric vehicles according to the first, second, and third embodiments are methods for controlling the electric vehicle 100 to perform attitude control, which controls the longitudinal attitude by adjusting the distribution of driving force between the front wheels 21 and the rear wheels 26, which are the driving wheels. In this control method for the electric vehicle 100, the total driving force TQ, which is the driving force required for the electric vehicle 100, is calculated based on the operation of the accelerator pedal, and the estimated acceleration G, which is an estimated value of the acceleration G that occurs when the electric vehicle 100 is driven with the total driving force TQ, is calculated based on the total driving force TQ. ｅｓｔ , or the estimated jerk J, which is an estimate of the jerk J, which is the time rate of change of the acceleration G. ｅｓｔ , is calculated. Then, the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ Based on this, a posture control execution determination is made to switch the posture control on / off.

[0124] In this way, the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ If the attitude control is turned on or off by the feedforward control based on Ｐ In other words, when attitude control by adjusting the drive force distribution is turned on / off as needed, the delay in determining the start of attitude control can be reduced, and attitude control can be performed in a timely manner. In particular, when attitude control is performed by feedforward control, the pitch angle θ ＰThe attitude control is activated before the fluctuation actually occurs. Furthermore, even when the attitude control is performed by feedback control, the control delay caused by the attitude control execution determination is reduced. Therefore, the delay in the start of the attitude control can be minimized. Therefore, as described above, the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ If the on / off of posture control is determined by feedforward control based on the above, posture control can be performed by adjusting the drive force distribution while suppressing deterioration in electricity consumption. In other words, deterioration in electricity consumption due to posture control is minimized, and the ride comfort of electric vehicle 100 is improved.

[0125] In the control methods for electric vehicles according to the first, second, and third embodiments, the estimated acceleration G is calculated based on the total driving force TQ. ｅｓｔ and estimated jerk J ｅｓｔ is calculated, and the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ In this way, the posture control is switched on / off based on the estimated acceleration G ｅｓｔ and estimated jerk J ｅｓｔ By using both of these, the accuracy of the attitude control execution decision is improved. Ｐ , Δ Ｐ ) and can turn on attitude control at the appropriate timing depending on the

[0126] In the control methods for electric vehicles according to the first, second, and third embodiments, in particular, the estimated acceleration G ｅｓｔ is a predetermined threshold value for the acceleration G. Ｇ ) or more, and the estimated jerk J ｅｓｔ is a predetermined threshold value for the jerk J, Ｊ ) or more, attitude control is turned on. Even with this simple method, it is possible to control the pitch state (θ Ｐ , Δ Ｐ ) and can turn on attitude control at the appropriate timing depending on the

[0127] In the control methods for electric vehicles according to the first, second, and third embodiments, the attitude control is performed by feedforward control, which corrects the driving force distribution based on a vehicle model. ｅｓｔ or estimated jerk J ｅｓｔ When the attitude control is controlled on / off by the attitude control execution judgment of the feedforward control based on the pitch angle θ Ｐ The posture control then functions before the fluctuation actually occurs. In other words, it is particularly easy to achieve both suppression of deterioration in power consumption and improvement of ride comfort through posture control.

[0128] In the control methods for electric vehicles according to the first, second, and third embodiments, the gradient (φ ＬＳ ) is obtained, and the gradient (φ ＬＳ ) and estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ In this way, the posture control is switched on / off based on the road surface gradient φ ＬＳ According to the attitude control execution decision that takes into account the pitch angle θ Ｐ Furthermore, when driving on a slope where the vehicle is less likely to feel the change in its condition, the attitude control can be kept off, thereby preventing the deterioration of fuel economy due to unnecessary attitude control.

[0129] In the control methods for electric vehicles according to the first, second, and third embodiments, the gradient (φ ＬＳ ) is a predetermined gradient (TH ＬＳ ) or less, the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ Based on this, it is determined whether attitude control is necessary, and the gradient (φ ＬＳ ) is the predetermined gradient (TH ＬＳ ), attitude control is turned off. Ｐ Furthermore, when traveling on a steeply inclined road surface where the fluctuations are difficult to feel, the attitude control can be kept off with particular reliability. Therefore, it is particularly easy to prevent deterioration in power consumption due to unnecessary execution of attitude control.

[0130] In the control method for an electric vehicle according to the second embodiment, the actual acceleration G occurring in the electric vehicle 100 is ａｃｔ , or the actual jerk J, which is the time rate of change of the actual acceleration G occurring in the electric vehicle 100 ａｃｔ is obtained. Then, the estimated acceleration G ｅｓｔ Or estimated jerk J ｅｓｔ Judgment based on, or actual acceleration G ａｃｔ Or Real Jerk J ａｃｔ When it is determined that attitude control is necessary by either the determination based on the actual acceleration G ａｃｔ Or Real Jerk J ａｃｔ When the determination by the above is combined, it is found that the running resistance RL includes a significant error depending on the specific usage situation of the electric vehicle 100, and the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ Even if the judgment by the pitch angle θ becomes inaccurate, the actual pitch angle θ Ｐ When the pitch angle θ Ｐ And if that fluctuation actually occurs, attitude control will function reliably.

[0131] In the control methods for electric vehicles according to the first, second, and third embodiments, the running resistance RL of the electric vehicle 100 is calculated based on the vehicle speed VSP, and the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ In this way, the estimated acceleration G is calculated based on the running resistance RL. ｅｓｔ or estimated jerk J ｅｓｔ By calculating the acceleration G, it is possible to accurately estimate the acceleration G before the actual acceleration G occurs. ｅｓｔ or estimated jerk J ｅｓｔ As a result, the accuracy of the attitude control execution determination is improved.

[0132] In particular, in the control method for an electric vehicle according to the third embodiment, it is determined whether the road surface on which the electric vehicle 100 is traveling is flat, and it is determined whether the electric vehicle 100 is traveling at a steady state without accelerating or decelerating. Furthermore, it is determined whether the total driving force TQ when the electric vehicle 100 is traveling at a steady state on a flat road is equal to or greater than the actual running resistance RL. ａｃｔ The actual running resistance RL is stored as ａｃｔ Based on this, the coefficient A used to calculate the running resistance RL ０ , A １ , A ２ In this way, the coefficient A used in the calculation of the running resistance RL is updated. ０ , A １ , A ２ By updating the running resistance RL, it is possible to accurately calculate the running resistance RL even if the running resistance RL changes depending on the specific usage state of the electric vehicle 100. As a result, the accuracy of the attitude control execution determination is particularly improved.

[0133] In the control method for an electric vehicle according to the third embodiment, when the electric vehicle 100 is traveling steadily on a flat road, the estimated acceleration G ｅｓｔ and the actual acceleration G that is the actual acceleration G that occurs in the electric vehicle 100. ａｃｔ and the acceleration estimation error (G ｅｒｒ ) is calculated. ０ ) or more acceleration estimation error (G ｅｒｒ ) is a predetermined time (τ ０ ) or more continued running resistance RL ａｃｔ Based on this, the coefficient A used to calculate the running resistance RL ０ , A １ , A ２ is updated. In this way, the actual acceleration G ａｃｔ Estimated acceleration G ｅｓｔ When it is certain that there is a significant error, the coefficient A used to calculate the running resistance RL ０ , A １ , A ２ By updating the coefficient A ０ , A １ , A ２ As a result, the coefficient A ０ , A１ , A ２ Since the accuracy of the above is improved, the accuracy of the attitude control execution determination is particularly likely to be maintained.

[0134] In the control method for an electric vehicle according to the third embodiment, the calculated running resistance RL and the actual running resistance RL ａｃｔ and the running resistance error (RL ｅｒｒ ) is calculated, and this running resistance error (RL ｅｒｒ ) is a predetermined threshold value TH ｅｒｒ２ When the coefficient A is less than the above, the coefficient A is used to calculate the running resistance RL. ０ , A １ , A ２ Among them, the coefficient A that constitutes the constant term ２ is updated. In this way, the running resistance error (second error RL ｅｒｒ ) is the threshold value TH ｅｒｒ２ If the coefficient A is ２ By updating the running resistance RL, when factors of running resistance other than air resistance, such as rolling resistance, change, the running resistance RL can be accurately calculated according to the change. As a result, when the rolling resistance or the like changes, the accuracy of the attitude control execution determination is particularly likely to be maintained.

[0135] In the control method for an electric vehicle according to the third embodiment, the calculated running resistance RL and the actual running resistance RL ａｃｔ and the running resistance error (RL ｅｒｒ ) is calculated, and the running resistance error (RL ｅｒｒ ) is a predetermined threshold value TH ｅｒｒ２ When the coefficient A is larger than ０ , A １ , A ２ Among these, the coefficient A constituting the term including the vehicle speed VSP ０ , A １ is updated. In this way, the running resistance error (second error RL ｅｒｒ ) is the threshold value TH ｅｒｒ２ If the coefficient A is greater than ０ , A １By updating, when the air resistance of the electric vehicle 100 changes, it is possible to accurately calculate the running resistance RL according to the change. As a result, when the air resistance changes, the accuracy of the attitude control execution determination is particularly likely to be maintained.

[0136] The control device for an electric vehicle according to the first, second, and third embodiments is a control device (controller 12) for an electric vehicle 100 that performs attitude control to control the longitudinal attitude by adjusting the distribution of driving force between the front wheels 21 and the rear wheels 26, which are the driving wheels. This control device (controller 12) includes a total driving force calculation unit 41 that calculates a total driving force TQ, which is the driving force required for the electric vehicle 100, based on the operation of the accelerator pedal, and an estimated acceleration G , which is an estimated value of the acceleration G that occurs when the electric vehicle 100 is driven with the total driving force TQ, based on the total driving force TQ. ｅｓｔ , or the estimated jerk J, which is an estimate of the jerk J, which is the time rate of change of the acceleration G. ｅｓｔ , and an estimation unit (52, 53) that calculates the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ and a posture control execution determination unit 47 that determines whether to perform posture control based on the attitude control result. With this configuration, even if the attitude control is turned off due to power consumption or the like, the pitch angle θ Ｐ The attitude control can be started without delay before the fluctuation actually occurs. Therefore, the attitude control by adjusting the drive force distribution can be executed while suppressing the deterioration of the electric fuel consumption. In other words, the deterioration of the electric fuel consumption due to the attitude control is suppressed to a minimum, and the riding comfort of the electric vehicle 100 is improved.

[0137] The control program for the electric vehicle 100 according to the first, second, and third embodiments controls the control device (controller 12) of the electric vehicle 100 to include a total driving force calculation unit 41 that calculates a total driving force TQ, which is a driving force required for the electric vehicle 100, based on the operation of the accelerator pedal; an estimated acceleration G, which is an estimated value of the acceleration G that occurs when the electric vehicle 100 is driven with the total driving force TQ, based on the total driving force TQ; ｅｓｔ, or the estimated jerk J, which is an estimate of the jerk J, which is the time rate of change of the acceleration G. ｅｓｔ , an estimation unit (52, 53) that calculates the estimated acceleration G ｅｓｔ or estimated jerk J ｅｓｔ and a posture control execution determination unit 47 that performs a posture control execution determination to switch posture control on / off based on the posture control execution determination unit 47.

[0138] The above describes embodiments of the present invention, but the configurations described in the above embodiments and each modified example only show some of the application examples of the present invention and are not intended to limit the technical scope of the present invention.

Claims

1. A control method for an electric vehicle that performs attitude control to control the attitude in the longitudinal direction by adjusting the drive force distribution between the front and rear wheels, which are driving wheels, comprising: calculating a total drive force, which is a required drive force for the electric vehicle, based on an operation of an accelerator pedal; calculating an estimated acceleration, which is an estimated value of an acceleration that occurs when the electric vehicle is driven with the total drive force, and an estimated jerk, which is an estimated value of a jerk that is a time change rate of the acceleration, based on the total drive force; acquiring a gradient of a road surface on which the electric vehicle travels; when the gradient is equal to or less than a predetermined gradient, turning on the attitude control when the estimated acceleration is equal to or greater than an acceleration threshold, which is a predetermined threshold for the acceleration, and the estimated jerk is equal to or greater than a jerk threshold, which is a predetermined threshold for the jerk; when the gradient is greater than the predetermined gradient, turning off the attitude control; A control method for an electric vehicle.

2. A control method for an electric vehicle that performs attitude control to control the attitude in the longitudinal direction by adjusting the drive force distribution between the front and rear wheels, which are driving wheels, comprising: calculating a total drive force, which is a required drive force for the electric vehicle, based on an operation of an accelerator pedal; calculating a running resistance of the electric vehicle based on a vehicle speed; calculating an estimated acceleration, which is an estimated value of an acceleration that occurs when the electric vehicle is driven with the total drive force, and an estimated jerk, which is an estimated value of a jerk that is a time change rate of the acceleration, based on the total drive force and the running resistance; turning on the attitude control when the estimated acceleration is equal to or greater than an acceleration threshold, which is a predetermined threshold for the acceleration, and the estimated jerk is equal to or greater than a jerk threshold, which is a predetermined threshold for the jerk; determining whether a road surface on which the electric vehicle travels is a flat road; determining whether the electric vehicle is performing a steady running without accelerating or decelerating; storing the total drive force when the electric vehicle is steadily running on the flat road as an actual running resistance, which is the actual running resistance; updating a coefficient used for calculating the running resistance based on the actual running resistance; A control method for an electric vehicle.

3. The control method for an electric vehicle according to claim 1 or 2, comprising: performing the attitude control by feedforward control that corrects the drive force distribution based on a vehicle model. Control method for an electric vehicle.

4. The control method for an electric vehicle according to claim 2, wherein when the electric vehicle is traveling steadily on the flat road, an acceleration estimation error, which is an error between the estimated acceleration and the actual acceleration that occurs in the electric vehicle, is calculated, when the acceleration estimation error equal to or greater than a predetermined error continues for a predetermined time or longer, a coefficient used in the calculation of the running resistance is updated based on the actual running resistance, Control method for an electric vehicle.

5. The control method for an electric vehicle according to claim 4, wherein a running resistance error, which is an error between the calculated running resistance and the actual running resistance, is calculated, when the variation of the running resistance error is equal to or less than a predetermined threshold, a coefficient constituting a constant term among the coefficients used in the calculation of the running resistance is updated, Control method for an electric vehicle.

6. The control method for an electric vehicle according to claim 5, wherein a running resistance error, which is an error between the calculated running resistance and the actual running resistance, is calculated, when the variation of the running resistance error is greater than a predetermined threshold, a coefficient constituting a term including the vehicle speed among the coefficients used in the calculation of the running resistance is updated, Control method for an electric vehicle.

7. A control device for an electric vehicle that executes attitude control for controlling the attitude in the front-rear direction by adjusting the drive force distribution between the front and rear wheels, which are drive wheels, wherein a total drive force calculation unit that calculates a total drive force, which is a required drive force for the electric vehicle, based on an operation of an accelerator pedal, an estimation unit that calculates an estimated acceleration, which is an estimated value of an acceleration that occurs when the electric vehicle is driven by the total drive force, and an estimated jerk, which is an estimated value of a jerk that is a time change rate of the acceleration, based on the total drive force, when the gradient of the road surface on which the electric vehicle travels is equal to or less than a predetermined gradient, the attitude control is turned on when the estimated acceleration is equal to or greater than an acceleration threshold, which is a predetermined threshold for the acceleration, and the estimated jerk is equal to or greater than a jerk threshold, which is a predetermined threshold for the jerk, and when the gradient is greater than the predetermined gradient, an attitude control execution determination unit that turns off the attitude control, A control device for an electric vehicle, comprising the above.

8. A control device for an electric vehicle that executes attitude control for controlling the attitude in the front-rear direction by adjusting the drive force distribution between the front and rear wheels, which are drive wheels, wherein A total driving force calculation unit that calculates a total driving force, which is a required driving force for the electric vehicle, based on an operation of an accelerator pedal; A running resistance calculation unit that calculates a running resistance of the electric vehicle based on a vehicle speed; An estimation unit that calculates an estimated acceleration, which is an estimated value of an acceleration generated when the electric vehicle is driven by the total driving force, and an estimated jerk, which is an estimated value of a time change rate of the acceleration, based on the total driving force and the running resistance; An attitude control execution determination unit that turns on the attitude control when the estimated acceleration is equal to or greater than an acceleration threshold, which is a predetermined threshold for the acceleration, and the estimated jerk is equal to or greater than a jerk threshold, which is a predetermined threshold for the jerk; Comprising; The attitude control execution determination unit; Determines whether the road surface on which the electric vehicle travels is a flat road; Determines whether the electric vehicle is performing a steady running without accelerating or decelerating; Stores the total driving force when the electric vehicle is steadily running on the flat road as an actual running resistance, which is the actual running resistance; Updates a coefficient used in the calculation of the running resistance based on the actual running resistance; A control device for an electric vehicle.