JP2007190941A - Driving force distribution device of four-wheel independent drive vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine the feedback control variable of driving force distribution of each wheel for making an error in relation to a target value of each of vehicle longitudinal force, vehicle lateral force and yaw moment small or zero. <P>SOLUTION: A controller 8 sets a feed forward component of the driving force distribution of each driving wheel 1 to 4, detects or estimates at least one or more of the vehicle longitudinal force, the vehicle lateral force and the yaw moment showing vehicle behaviors of a vehicle, and sets the target values for at least one vehicle behavior or more from the detected or estimated vehicle behaviors. As for each of the vehicle behaviors set with the target values, the error between the target value and the detected or estimated value is determined. From the determined error, each correction amount target value of the vehicle longitudinal force, the vehicle lateral force and the yaw moment is set, the feedback control variable of the driving force distribution of each of the driving wheels 1 to 4 for realizing the correction amount target values is determined, and the sum of the feed forward component of the driving force distribution and the feedback control variable of the driving force distribution is calculated as a target value of the driving force distribution of each of the driving wheels 1 to 4. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a driving force distribution control of a four-wheel independent drive vehicle that can independently control and drive left and right wheels.

In a four-wheel independent drive vehicle, if there is an error between the target yaw moment and the actual yaw moment determined according to the driver's operation due to disturbances such as crosswinds or changes in vehicle weight, this error should be reduced. Japanese Patent Application Laid-Open No. H10-228688 discloses a technique for providing a difference in driving force between left and right wheels.
Japanese Patent Laid-Open No. 5-221300

  In the above prior art, the error between the target value of the yaw moment and the actual value can be reduced. However, the lateral force and yaw moment of the vehicle may change due to a change in the tire lateral force of the wheel that generates the driving force difference, which may reduce steering stability.

An example of this will be described with reference to FIGS. In FIG. 1, the front wheel steering angle is turned to the left, while the left and right front wheel driving forces Fx ₁ and Fx ₂ (unit: N) and the driving force difference Fx ₂ −Fx ₁ are in a steady turning state to the left. The tire lateral forces Fy ₁ , Fy ₂ , Fy ₃ , and Fy ₄ (unit: N) of each wheel are generated leftward in the vehicle traveling direction.

In FIG. 1, there is no error between the vehicle longitudinal force Fx and the vehicle lateral force Fy with the target value, and there is an error ΔM between the target value M ^** (unit: N) of the yaw moment M applied to the vehicle center of gravity. = M ^** − M, in order to cancel this yaw moment error ΔM in the prior art, for example, driving forces Fb ₃ and Fb ₄ (unit: N) represented by the equation (1) as shown in FIG. ) On the left and right rear wheels. Lt is the tread length (unit: m) of the rear wheel.

However, since the tire lateral force decreases as the braking / driving force increases, the tire lateral forces Fy ₃ and Fy ₄ of the rear left and right wheels respectively decrease as shown in FIG. 2, but the tire lateral force of the front wheels does not change. Therefore, the vehicle lateral force is reduced. Further, in the case of FIG. 2, only the tire lateral force of the rear wheel is reduced, so that the yaw moment M applied to the vehicle center of gravity position also changes. A decrease in the lateral force of the vehicle and a change in the yaw moment M can cause a decrease in steering stability.

  The present invention has been made in view of such a problem, and feedback control of driving force distribution of each wheel that reduces or eliminates errors with respect to target values of vehicle longitudinal force, vehicle lateral force, and yaw moment. The purpose is to determine the quantity.

  The driving force distribution device according to the present invention sets a feedforward component of the driving force distribution of each wheel, and at least one of a vehicle longitudinal force, a vehicle lateral force, and a yaw moment representing a vehicle behavior of the vehicle. A target value is set for at least one vehicle behavior from the detected or estimated vehicle behavior. Then, an error between the target value and the detected or estimated value is obtained for the vehicle behavior for which the target value is set, and a correction amount target for the vehicle longitudinal force, the vehicle lateral force, and the yaw moment is obtained from the obtained error. A value is set, and a feedback control amount of the driving force distribution of each wheel that realizes the correction amount target value is obtained. Then, the sum of the feedforward component of the driving force distribution and the feedback control amount of the driving force distribution is calculated as a target value of the driving force distribution of each wheel, and the driving force of each wheel is independently calculated based on the target value of the driving force distribution. To control.

  According to the present invention, for example, even when there is an error from the target value only in the yaw moment as shown in FIG. 2, the yaw moment error is reduced or reduced without changing the vehicle longitudinal force and the vehicle lateral force. Therefore, it is possible to avoid a decrease in steering stability.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

  First, the theoretical background of the present invention and the effects of the present invention will be described. After that, an embodiment in which the present invention is applied to an electric vehicle will be described.

1. Theoretical Background of the Present Invention FIG. 3 shows the driving force, tire lateral force and steering angle of each wheel in a vehicle capable of independently driving the left front wheel 1, the right front wheel 2, the left rear wheel 3 and the right rear wheel 4, and the vehicle. It is a figure showing the yaw moment about the working front-back direction, a horizontal direction, and a gravity center.

_{δ 1, δ 2, δ 3} , δ 4 each wheel 1-4 respectively of the steering angle _{(unit: rad), Fx 1, Fx} 2, Fx 3, Fx 4 in the driving force of each wheel 1-4 (unit: N), Fy ₁ , Fy ₂ , Fy ₃ , Fy ₄ are tire lateral forces (unit: N) of the wheels 1 to 4. Further, Fx is a vehicle longitudinal component of the sum of tire forces (unit: N), Fy is a vehicle lateral component of the sum of tire forces (unit: N), and M is a vehicle center of gravity generated by the tire force of each wheel. This is the total yaw moment (unit: Nm). Lf is the distance from the vehicle center of gravity axis to the front wheel axle (unit: m), Lr is the distance from the vehicle center of gravity axis to the rear wheel axle (unit: m), and Lt is the tread length of the front and rear wheels (unit: m). It is. The wheel base length Ll (unit: m) is the sum of Lf and Lr.

The steering angles δ _i (i = 1 to 4) and M are positive when the vehicle is viewed from above, and δ _i is 0 when the rotational direction of each wheel coincides with the longitudinal direction of the vehicle. To do. Further, Fx _i is positive in the direction of accelerating the vehicle forward when δ _i is all 0, and the tire lateral force Fy ₁ (unit: N) is the direction of accelerating the vehicle in the left direction when δ _i is all 0. Positive.

Here, first, the vehicle longitudinal component Fx _i ′ and the vehicle lateral component Fy _i ′ of the resultant force (tire force) of the driving force and tire lateral force generated at each wheel will be considered. Then, as shown in FIG. 4, Fx _i ′ and Fy _i ′ when the steering angle of each wheel is turned by δ _i (i = 1 to 4) are expressed by the equations (2) and (3). However, Fx _i ′ is positive in the direction of accelerating the vehicle forward, and Fy _i ′ is positive in the direction of accelerating the vehicle in the left direction.

Therefore, when the tire lateral force variation amount when the driving force of each wheel is changed by DerutaFx _i and _{ΔFy i, Fx i ', Fy} i' variation DerutaFx of when the driving force of each wheel is changed by DerutaFx _{i i} ′ and ΔFy _i ′ are expressed by Expression (4) and Expression (5).

Here, the relationship between the driving force and the tire lateral force is as shown in FIG. FIG. 5 is a graph showing the relationship between the driving force and the tire lateral force when there is no change in the wheel load and the road surface friction coefficient. The driving force is plotted on the horizontal axis and the tire lateral force is plotted on the vertical axis. Using the relationship of FIG. 5, the sensitivity of the tire lateral force with respect to the driving force change ΔFx _{i in} the current driving force FX _i and tire lateral force Fy _i of each wheel is set to k _i (i = 1 to 4). . That is, k _i is the value of equation (6) when ΔFx _i and ΔFy _i are very small as shown in FIG.

Then, if ΔFx _i and ΔFy _i are very small and the approximation of equation (6) is sufficiently established, ΔFy _i = k _i ΔFx _i , so that the driving force Fx _{i of} each wheel changes by a sufficiently small ΔFx _i. The change amounts ΔFx _i ′ and ΔFy _i ′ of Fx _i ′ and Fy _i ′ when expressed are expressed by equations (7) and (8).

  Here, in the state of FIG. 3, the vehicle longitudinal component Fx of the sum of tire forces, the vehicle lateral component Fy of the sum of tire forces, and the sum M of the yaw moment around the center of gravity of the vehicle generated by the tire force of each wheel. Can be represented by formula (9) to formula (11). However, M is positive in the counterclockwise direction when the vehicle is viewed from above as shown in FIG.

Therefore, when the braking / driving force of each wheel is changed by ΔFx _i , the changes ΔFx, ΔFy, ΔM of Fx, Fy, M are calculated using p _i and q _i in Equation (6) and Equation (7), respectively. It represents with Formula (12)-(14).

  The formulas (12) to (14) can be summarized by the formula (15).

When ΔFx ₁ is assumed to be known and ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ are solved, ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ are expressed by Expression (16).

However,

Therefore, from equation (16), when D ₁ ≠ 0, the driving force change amount ΔFx _{i for} each wheel that changes Fx, Fy, M by ΔFx, ΔFy, ΔM around the current operating point is χ Can be obtained as an arbitrary constant according to the equation (21).

Similarly, when equation (15) is solved assuming that any one of ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ is known, the current state is obtained when D ₂ ≠ 0, D ₃ ≠ 0, and D ₄ ≠ 0. Fx around the operating point, Fy, respectively M ΔFx, ΔFy, the formula for obtaining the driving force variation DerutaFx _i of each wheel to be only a change ΔM obtained. For example, when ΔFx ₄ is known and equation (15) is solved, equation (22) is obtained.

Accordingly, DerutaFx in formula (21) or equation (22), ΔFy, a ΔM Fx, Fy, the error between the target value of M, if the DerutaFx _i feedback control amount Fx _bi of the respective wheels, Fx, Fy, M A feedback control amount Fx _bi that corrects an error from the target value is obtained. This is the gist of the present invention.

  FIGS. 6 and 7 show an example of comparison with the prior art. In both the steady turning state of FIG. 1, a yaw moment that turns the vehicle body inside the turn due to a gust of wind (cross wind) is generated at time A. It is the simulation result performed supposing the case where it adds in steps from time to time B. FIG.

  FIG. 6 shows that the change in the yaw moment is suppressed only by the difference between the left and right driving forces of the rear wheels using the conventional technique, and FIG. 7 shows the vehicle behavior (Equation 21 or Equation 22) according to the present invention. This is a result when the feedback control amount is obtained so as to suppress changes in the vehicle longitudinal force, vehicle lateral force, and yaw moment). 6 and 7 show the time history response of the yaw rate, the lateral force, the driving force of each wheel, and the tire lateral force of each wheel in order from the top, and the unit of each axis in FIG. 6 and FIG. The range is the same.

  In the case of the prior art shown in FIG. 6, since the change in the tire lateral force when the left / right driving force difference is output at the rear wheel is not taken into account, not only the vehicle lateral force but also the yaw rate has changed greatly. . On the other hand, when the present invention of FIG. 7 is applied, it can be seen that the error is greatly reduced.

2. Effects of the Present Invention Based on the theoretical background and examples of the effects of the present invention, the effects of the present invention are summarized as follows.

In the first invention (the invention described in claim 1), first, the subject four-wheel independent drive vehicle is at least one vehicle selected from the vehicle longitudinal force Fx, the vehicle lateral force Fy, and the yaw moment M. Detect behavior. Then, a target value is set for the detected vehicle behavior, and Fx, Fy, and M correction amount target values dFx, dFy, and dM are set based on an error from the target value, and the dFx, dFy, and dM are corrected. The feedback control amount Fx _bi of each wheel to be obtained is obtained. As a result, if errors with respect to the target values of the vehicle behaviors Fx, Fy, and M are set in the correction amount target values dFx, dFy, and dM, the driving force distribution of each wheel that makes these vehicle behavior errors both small or zero. The feedback control amount can be obtained, and the effect that the steering stability can be improved can be expected.

  Further, in a vehicle in which all of the vehicle behaviors Fx, Fy, and M cannot be detected or estimated, for example, the detected or estimated vehicle behavior is asymptotically or coincident with the target value without changing the vehicle behavior that cannot be detected or estimated. The effect of being able to be expected. In order to obtain such an effect, for example, for a vehicle behavior that can be detected, an error from the target value may be set as the correction amount target value, and for a vehicle behavior that cannot be detected, 0 may be set as the correction amount target value.

According to the second invention (the invention described in claim 2), the feedback control amount Fx _bi (i = 1 to 4) and the feedforward component Fx _fi (i = 1 to 4) obtained once in the first invention. The correction amount target values dFx, dFy, dM are set once again from the error between the vehicle behavior realized by the sum Fx _fi + Fx _bi and the target vehicle behavior, and the process of compensating for the dFx, dFy, dM is further performed. . As a result, as shown in FIG. 7, it is possible to obtain the feedback control amount Fx _bi of the driving force distribution that further reduces the error from the target value of the vehicle behavior, and it is expected that the steering stability can be further improved. it can.

According to the third invention (the invention described in claim 3), in the first and second inventions, the sensitivity of the vehicle longitudinal force, the vehicle lateral force, and the yaw moment with respect to the driving force change is obtained, and based on this sensitivity. Thus, the feedback control amount Fx _bi for driving force distribution is obtained. As a result, the feedback control amount Fx _bi for driving force distribution can be obtained more accurately and simply around the current operating point, and improvement in steering stability and reduction in the calculation load of the control device can be expected.

According to the fourth invention (the invention described in claim 4), the feedback control amount Fx _bi of the driving force distribution is determined based on the sensitivity k _i of the tire lateral force to the driving force change of each wheel in the first to third. The configuration required. As a result, the feedback control amount Fx _bi for driving force distribution can be obtained more accurately and simply around the current operating point, and improvement in steering stability and reduction in the calculation load of the control device can be expected.

According to the fifth invention (the invention according to claim 5), in the third and fourth inventions, the feedback control amount for each wheel becomes smaller from the plurality of sets of feedback control amounts for driving force distribution. It was set as the structure which selects. The fact that a plurality of combinations of feedback control amounts for driving force distribution can be obtained is apparent from the fact that an arbitrary constant χ is included in the equations (21) and (22). Then, equation (21) or expression from the assumption that the change DerutaFx _i of longitudinal force Fx _i of each wheel in the derivation of (22) is sufficiently small, the driving force distribution more precisely by applying the fifth aspect of the present invention The feedback control amount can be obtained, and improvement in steering stability can be expected.

According to the sixth invention (the invention described in claim 6), the feedback control amount dFx with respect to the correction amounts ddFx, ddFy, ddM smaller than the correction amount target values dFx, dFy, dM in the third to fifth inventions. _It was set as the structure which _{calculates |} requires _bi repeatedly. As a result, the feedback control amount calculated at a time around the current operating point can be reduced, and therefore the change ΔFx _i of the braking / driving force Fx _i of each wheel is sufficiently small in the derivation of the equations (21) and (22). As is clear from the above assumption, the feedback control amount Fx _bi of the driving force distribution can be obtained more accurately, and it can be expected that the steering stability is improved and the calculation load of the control device is reduced.

According to the seventh invention (invention described in claim 7), the vehicle behavior Fx ^{# #} implemented by feedforward component of the driving force distribution _{Fx f1, Fx f2, Fx f3} , Fx f4, Fy ##, M # ^# Is estimated and obtained from the errors ΔFx, ΔFy, ΔM between the Fx ^## , Fy ^## , M ^## and the target vehicle behavior Fx ^** , Fy ^** , M ^** by the vehicle behavior error calculating means. The vehicle behavior correction amount target values dFx, dFy, and dM are set from the obtained error to obtain the feedback control amount Fx _bi . As a result, the feedback control amount Fx _bi that takes into account the error between the vehicle behavior realized by the feedforward component of the driving force distribution and the target vehicle behavior can be obtained, and the calculation load of the control device can be reduced.

According to the eighth invention (the invention described in claim 8), the driving that achieves the same vehicle behavior as the target values Fx ₁ ^** , Fx ₂ ^** , Fx ₃ ^** , and Fx ₄ ^{** of} the driving force distribution. A set of force distribution is obtained, and a driving force distribution is selected from the set. As a result, it is possible to select a driving force distribution that satisfies the range of the driving force that can be output from each wheel in the driving force distribution set that realizes the same vehicle behavior.

3. Embodiment of the Invention Next, a case where the present invention is applied to an electric vehicle will be described.

  FIG. 8 is a block diagram showing an example of a mechanical configuration of an electric vehicle to which the present invention is applied. The electric vehicle shown in FIG. 8 has a left front wheel 1 by a motor 11 driven by electric power supplied from a battery 9, a right front wheel 2 by a motor 12, a left rear wheel 3 by a motor 13, and a right rear wheel by a motor 14. 4 can be driven independently.

  The motors 11 to 14 are AC machines capable of powering operation and regenerative operation such as a three-phase synchronous motor and a three-phase induction motor, and the battery 9 is a nickel hydrogen battery or a lithium ion battery. The inverters 31 to 34 convert the alternating current generated by the motors 1 to 4 into a direct current and charge the battery 9, or convert the direct current discharged by the battery 9 into an alternating current and supply the alternating current to the motors 1 to 4. The speed of each wheel is detected by the wheel speed sensors 11 to 14, and the detected rotation speed of each wheel is transmitted to the controller 8.

  The rotation radii of the wheels 1 to 4 are all equal to R, and the reduction ratio is 1, that is, the motors and the wheels are directly connected. Furthermore, when the wheel load, the skid angle, and the road surface friction coefficient are the same for the four wheels, the relationship between the driving force and the tire lateral force is the same for the four wheels, that is, the four wheels have the same tire characteristics. .

  The longitudinal acceleration and lateral acceleration of the vehicle are detected by the acceleration sensor 100 attached to the center of gravity of the vehicle, the yaw rate of the vehicle is detected by the yaw rate sensor 101, and the detected lateral acceleration and yaw rate of the vehicle are transmitted to the controller 8. The

  The steering angles of the front wheels 1 and 2 are mechanically adjusted via the steering gear 15 by the steering of the steering 5 by the driver. The steering angle change amount of the front wheels 1 and 2 is set to be 1/16 of the steering angle change amount of the steering 5. The steering angles of the wheels 1 to 4 are detected by the steering angle sensors 41 to 44, and the detected steering angles of the wheels are transmitted to the controller 8.

  The rotation angle of the steering 5 by the driver is detected by the steering angle sensor 25, and the depression amounts of the accelerator pedal 6 and the brake pedal 7 are detected by the accelerator stroke sensor 26 and the brake stroke sensor 27, respectively, and transmitted to the controller 8.

  The controller 8 includes a CPU, a ROM, a RAM, an interface circuit, an inverter circuit, and the like. The wheel speed sensors 21 to 24, the steering angle sensor 25, the accelerator stroke sensor 26, the brake stroke sensor 27, the acceleration sensor 100, the yaw rate sensor 101, and the like. The detected signals are received, and control such as distributing torque to the motors 1 to 4 is performed based on these signals.

  Next, the control contents of the controller 8 will be described.

  The flowchart of FIG. 9 shows torque distribution control to the motors 1 to 4 executed by the controller 8, and corresponds to the first invention and the third to eighth inventions.

  According to this, first, in step S10, the rotational speeds ω1, ω2, ω3, and ω4 (unit: rad / s) of the wheels 1 to 4 are detected by the wheel speed sensors 21 to 24, respectively, and the radius R of each wheel is determined. The speeds V1, V2, V3, and V4 (unit: m / s) of each wheel are obtained by multiplication, and the vehicle speed V (unit: m / s) is obtained by equation (23).

Further, the longitudinal acceleration α _x (unit: m / s ² ) and the lateral acceleration α _y (unit: m / s ² ) of the vehicle are detected by the acceleration sensor 100, and the yaw rate γ (unit: rad / s) is detected. Detected by sensor 101.

Further, the accelerator stroke sensor 26 and the brake stroke sensor 27 detect the depression amounts AP (unit:%) and BP (unit:%) of the accelerator pedal 6 and the brake pedal 7 respectively, and the steering angle sensor 25 rotates the rotation angle of the steering wheel 5. θ (unit: rad) is detected, and the steering angles δ ₁ , δ ₂ , δ ₃ , and δ ₄ of the wheels 1 to 4 are detected by the steering angle sensors 41 to 44.

V and V1 to V4 are positive in the forward direction of the vehicle, the rotation angle θ of the steering 5 is positive in the counterclockwise direction, α _x is positive in the direction in which the vehicle accelerates forward, and α _y is the vehicle when the vehicle is turning left. The direction from the center of gravity to the turning center is positive, and γ is positive in the counterclockwise direction when the vehicle is viewed from vertically above.

In a vehicle that does not have a steering angle sensor, the steering angle of each wheel is obtained from the rotation angle θ of the steering 5. For example, in the electric vehicle having the configuration of the present embodiment, the steering angles δ ₁ and δ ₂ of the front wheels ₁ and ₂ are set to δ ₁ = δ ₂ = θ / 16, and the steering angles δ ₃ and δ ₄ of the rear wheels 3 and ₄ are set. Let δ ₃ = δ ₄ = 0. In such a case, it is preferable that the steering angle of each wheel can be corrected in consideration of the influence of the suspension such as compliance steer and roll steer.

  In step S20, the vehicle longitudinal force Fx, vehicle lateral force Fy, and yaw moment M applied to the vehicle are estimated.

First, a vehicle longitudinal direction force Fx, which is the mass m (Unit: kg) of the vehicle longitudinal acceleration alpha _x determined in step S10 is multiplied by, determined as Fx = m.alpha _x. When obtaining Fx, it is more preferable to consider running resistance such as air resistance and rolling resistance according to the vehicle speed.

Then, the vehicle lateral force Fy is a lateral acceleration α of the vehicle in _{the y} mass m (Unit: kg) obtained in step S10 is multiplied by, determined as Fy = m.alpha _y.

Finally, the yaw moment M is obtained as M = Iγ ′ by multiplying the yaw angular acceleration γ ′ obtained by time differentiation of the yaw rate γ obtained in step S10 by the yaw inertia moment I (unit: kg · m ² ) of the vehicle.

Note that the symbols Fx, Fy, and M are the same as α _x , α _y , and γ, respectively.

In step S30, the sideslip angles β ₁ , β ₂ , β ₃ , β ₄ (unit: rad) of each wheel 1 to 4 are estimated. As the estimation method, for example, the method described in Japanese Patent Laid-Open No. 10-329689 is used, and the lateral acceleration α _y , yaw rate γ, vehicle speed V, each wheel steering angle δ _i detected and estimated in step S10 and the steering angle δ _i The vehicle body side slip angles β and β _i are estimated from the rotation angle θ. The sign of β _i is positive when the angle from the front-rear direction of the wheel to the direction of the wheel speed is counterclockwise when viewed from vertically above.

In step S30, the wheel loads W ₁ , W ₂ , W ₃ , W ₄ (unit: N) of each of the wheels 1 to 4 are obtained by equations (24) to (27).

However, Lf is the distance (unit: m) from the center of gravity of the vehicle to the front wheel axle, Lr is the distance (unit: m) from the center of gravity of the vehicle to the rear wheel axle in the yaw rotation direction, and Lt is the tread length of the front and rear wheels (unit: m). Unit: m), L1 is the wheelbase length (unit: m), Ll = (Lf + Lr), m is the mass of the vehicle (unit: kg), and g is the acceleration of gravity (unit: m / s ² ).

Furthermore, in step S30, the road surface friction coefficients μ ₁ , μ ₂ , μ ₃ , and μ ₄ (unit: none) of each wheel 1 to 4 are estimated. As the estimation method, for example, the method described in Japanese Patent Laid-Open No. 6-98418 is used to estimate the reaction force that each wheel receives from the road surface, and the road surface reaction force and the wheel load W _{i of} each wheel obtained in step S40. To estimate μ _i .

In step S40, the vehicle longitudinal force, the vehicle lateral force, and the yaw moment are calculated from the depression amounts AP and BP of the accelerator pedal 6 and the brake pedal 7, the rotation angle θ of the steering wheel 5, and the vehicle speed V (driving state). Dynamic target values Fx ^** , Fy ^** , and M ^** are set.

Further, in step S40, the dynamic target values Fx ^** , Fy ^** , and M ^** of the vehicle behavior are further allocated to the driving force distribution of each wheel that is realized when there is no disturbance such as a cross wind or a change in the vehicle weight. determination of the feedforward component _{Fx f1, Fx f2, Fx f3} , Fx f4.

Fx ^**, Fy ^**, Determination of M ^** and _{Fx f1, Fx f2, Fx f3} , Fx f4 will be described later with reference to a flowchart of FIG. When applying the seventh invention, in step S40, the basic value Fx _i ^## of the driving force distribution obtained in step S1070 in the flowchart of FIG. 10 is set as the feedforward component Fx _fi of the driving force distribution.

In step S50, the vehicle longitudinal force and vehicle lateral force are ^calculated from Fx, Fy, M estimated in step S20 and the dynamic target values Fx ^** , Fy ^** , M ^{** of} the vehicle behavior set in step S40. The yaw moment correction amount target values dFx, dFy, dM are set according to equations (28) to (30).

  When the seventh invention is applied, in step S50, the vehicle behavior correction amount target values dFx, dFy, dM are expressed by the following equation (31) using the errors ΔFx, ΔFy, ΔM obtained in step S1090 in the flowchart of FIG. -It sets with Formula (33).

In step S60, dFx determined in step S50, dFy, realizing dM, Fx _fi and step S20 that the feedback control amount _{Fx b1, Fx b2, Fx b3} , Fx b4 each wheel power distribution, determined in step S40 Is obtained by referring to the driving force correction amount map from β _i estimated in (1). This driving force correction amount map is set, for example, as shown in FIGS. 11a and 11b for each driving force and side slip angle that each wheel can take.

This driving force correction amount map extracts all combinations of the driving force and side slip angle of each wheel that can be taken by the vehicle, and the vehicle behavior correction amount target values dFx, dFy, dM and the distribution of each wheel driving force in each combination. The relationship with the feedback control amount Fx _bi is obtained in advance through experiments or simulations.

11a and 11b, the maps as shown in FIGS. 11a and 11b should be provided for all combinations of driving force and side slip angle that can be taken by the vehicle. Keep it. At this time, the wheel load W _i of each wheel, the rudder angle δ _i of each wheel, the road surface friction coefficient μ _i of each wheel, etc., for all possible combinations, dFx, dFy, dM and Fx _bi It is more preferable to obtain the relationship in advance through experiments or simulations and map it.

In step S60, the sixth invention is applied according to the flowchart shown in FIG. 12 or 13 when applying the third invention, and according to the flowchart shown in FIG. 13 when applying the fourth invention. In this case, Fx _bi is obtained according to the flowchart shown in FIG. A method of obtaining Fx _bi according to the flowcharts shown in FIGS. 12 to 14 will be described later.

In step S70, the sum of the feedforward component Fx _fi of the driving force distribution and the feedback control amount Fx _bi is set as a target value Fx _i ^** = Fx _fi + Fx _bi of the driving force distribution of each wheel.

When the eighth invention is applied, in Step S70, the target values Fx ₁ ^** , Fx ₂ ^** , Fx ₃ ^** of the driving force distribution for the left front wheel, the right front wheel, the left rear wheel, and the right rear wheel, respectively. , Find a set of four-wheel drive force distribution that realizes the same vehicle longitudinal force, vehicle lateral force, and yaw moment as Fx ₄ ^**, and from among the multiple drive force distributions in this drive force distribution set The driving force distribution satisfying the range of driving force that can be output for each wheel is selected, and the target values Fx ₁ ^** , Fx ₂ ^** , Fx ₃ ^** , Fx ₄ ^** of the driving force distribution of the four wheels are selected. Reset it.

In step S80, control is performed so that the motors 11 to 14 of each wheel output a value obtained by multiplying Fx _i ^** by the radius R of each wheel.

  Next, a case where torque distribution control is performed based on the second to eighth inventions will be described. This torque distribution control is performed instead of the torque distribution described above.

  When torque distribution control is performed based on the second to eighth inventions, steps S50 to S70 in the flowchart of FIG. 9 may be replaced with the flowchart of FIG.

The flowchart of FIG. 15 will be described. First, in step S100, the vehicle longitudinal force eFx, the vehicle lateral force eFy, and the yaw moment eM realized by the feedforward component Fx _fi of the driving force distribution are expressed by the equations (34) to ( 36).

However,

Note that Fy _fi is a tire lateral force generated when Fx _fi is applied to each wheel in the current running state, and is based on the current side slip angle β _i , wheel load W _i , and road surface friction coefficient μ _i of each wheel. It is set from a tire characteristic map representing the relationship between the driving force and the tire lateral force used in step S700 of the flowchart of FIG.

In step S110, the vehicle longitudinal force is ^calculated from Fx, Fy, M estimated in step S20 of the flowchart of FIG. 9 and the dynamic target values Fx ^** , Fy ^** , M ^{** of} the vehicle behavior set in step S40. The vehicle lateral force and yaw moment correction amount target values dFx, dFy, and dM are set according to equations (38) to (40).

However, in the flowchart of FIG. 15, each of dFx, dFy, and dM is multiplied by correction amount convergence coefficients φ _x , φ _y , and φ _m that are greater than 0 and equal to or less than 1, and are set according to equations (41) to (43). May be.

In step S130, if the absolute values of dFx and dFy are both 10 [N] or less and the absolute value of dM is 10 [Nm] or less, the process proceeds to step S140, and the target value Fx _i ^** of the driving force distribution of each wheel. Is set to Fx _i ^** = Fx _fi , the flowchart of FIG. 15 is terminated, and the process proceeds to step S80 of FIG.

At step S150, the performs the same processing as step S60 in the flowchart of FIG. 9, determined dFx, dFy, the feedback control amount _{Fx b1, Fx b2, Fx b3} , Fx b4 each wheel power distribution to realize dM.

However, when Fx _bi is obtained according to the flowchart shown in FIG. 13 by applying the fourth invention in step S150, all of D ₁ , D ₂ , D ₃ , and D ₄ obtained in step S710 of the flowchart of FIG. If it becomes 0 and Fx _b1 = Fx _b2 = Fx _b3 = Fx _b4 = 0 in step S800, Fx _i ^** = Fx _fi is immediately passed through the flowchart of FIG. 15 and proceeds to step S80 of FIG.

In step S160, the feedforward component Fx _fi of the driving force distribution is updated by equation (44), and the process proceeds to step S100.

Next, the flowchart of FIG. 12 for obtaining the feedback control amount Fx _bi for each wheel driving force distribution based on the third invention in step S60 of the flowchart of FIG. 9 will be described below.

In step S500, the sensitivity K _ix (unit: none), K _iy (unit: none), K _iM (unit: rad · m) of the vehicle longitudinal force, the vehicle lateral force, and the yaw moment with respect to changes in the driving force of each wheel. )) (For example, the sensitivity of the vehicle front-rear direction force to the driving force change of the left front wheel 1 is K _1x and the sensitivity of the yaw moment to the driving force change of the right rear wheel 4 is K _4M ), and the vehicle behavior from Fx _fi and β _i Obtained by referring to the sensitivity map. For example, the vehicle behavior sensitivity map is set as shown in FIG. 16 (only the map of the left front wheel 1 is shown as an example in FIG. 16).

  This vehicle behavior sensitivity map extracts all combinations of driving force and side slip angle that can be taken by this vehicle, and before and after the vehicle when driving force of any one wheel is changed by 1 [N] in each combination. The amount of change in directional force, vehicle lateral force, and yaw moment is obtained and mapped.

In step S510, three wheels with [K _ix K _iy K _iM ] representing the sensitivity of each of the vehicle longitudinal force, the vehicle lateral force, and the yaw moment with respect to the driving force change of each wheel as vectors are linearly independent from each other. Choose a combination. How to choose is as follows.

First, the left front wheel 1 Consider the matrix K ₁ of formula (45) by arranging the vectors of the remaining three wheels to vertical than, the determinant det of the matrix K ₁ | K ₁ | if is not 0, the combination of the wheel The remaining three wheels other than the left front wheel 1 are set, and 1 is set in the flag flg.

If det | K ₁ | is 0, a matrix K _{2 in} which the vectors of the remaining 3 wheels other than the right front wheel 2 are vertically arranged is considered in the same manner as in equation (45), and the determinant det | K ₂ | If not, the combination of the wheels is the remaining three wheels other than the right front wheel 2, and 2 is set in the flag flg. If det | K ₂ | is also 0, this time, a matrix K _{3 in} which the vectors of the remaining three wheels other than the left rear wheel 3 are vertically arranged is considered in the same manner as in equation (45), and the determinant det | K ₃ | If it is not 0, the combination of the wheels is the remaining 3 wheels other than the left rear wheel 3, and 3 is set in the flag flg. If det | K ₃ | is also 0, a matrix K _{4 in} which the vectors of the remaining three wheels other than the right rear wheel 3 are vertically arranged is considered in the same way as the equation (45), and the determinant det | K ₄ | If it is not 0, the combination of the wheels is the remaining three wheels other than the left rear wheel 4, and 4 is set in the flag flg. If the determinant det | K ₄ | is also 0 and det | K _i | is all 0, the flag flg is set to 0 with no combination.

In step S520 the flag flg proceeds to 0, step S540, and _{Fx b1 = Fx b2 = Fx b3} = Fx b4 = 0. Otherwise, the process proceeds to step S530.

In step S530, Fx _bi is obtained by the calculation method according to the flag flg obtained in step S510, and the flowchart ends. For example, when the flag flg is 1, Fx _bi is obtained by Expression (46). That is, the driving force correction amounts of the three wheels whose [K _ix K _iy K _iM ] selected in step S510 are linearly independent from each other are obtained by multiplying dFx, dFy, dM by the inverse matrix of K _i , The driving force correction amount of the left front wheel 1 that has not been selected is set to zero. The same applies when the flag flg is 2-4.

Next, the flowchart of FIG. 13 for obtaining Fx _bi that realizes dFx, dFy, and dM based on the third or fourth invention in step S60 of the flowchart of FIG. 9 will be described below.

In step S700, the tire lateral force sensitivity k _i for the driving force change of each wheel is obtained. A method of obtaining k _i will be described by taking the case of the left front wheel 1 as an example.

The tire lateral force Fy _f1 + dFy ₁ corresponding to the driving force Fx _f1 + dFx ₁ is determined based on the current side slip angle β ₁ of each wheel, the wheel load W _{i of} each wheel, the road lateral friction coefficient μ _i , Is obtained by referring to a tire characteristic map representing the above relationship, and k ₁ is obtained by equation (47) as shown in FIG. This tire characteristic map is common to each wheel, and is set as shown in FIG.

Note that dFx ₁ (unit: N, dFx ₁ > 0) is a sufficiently small value compared to the wheel load that can be taken by the left front wheel of the vehicle, and is 10 [N] in this embodiment (other wheels). DFx _{i of the} same 10 “N”). That is, by obtaining the change amount dFy ₁ of Fy _f1 when Fx _f1 changes by a minute dFx ₁ , the sensitivity k ₁ of the tire lateral force with respect to the change in the drive force when the drive force is Fx _f1 is expressed by the equation (47). )

Similarly, k ₂ to k ₄ are obtained for the wheels 2 to 4.

In step S710, D ₁ , D ₂ , D ₃ , and D ₄ represented by the equations (48) to (51) are obtained using the k _i and the steering angle δ _i of each wheel.

  However,

In step S720, if D ₁ ≠ 0, the process proceeds to step S730, and the feedback control amount Fx _b1 of each wheel driving force distribution that realizes dFx, dFy, and dM using equation (53) corresponding to equation (21). , Fx _b2 , Fx _b3 , Fx _b4 are obtained. Otherwise, the process proceeds to step S740.

  Here, 0 is set as the arbitrary constant χ in the equation (53).

When applying the fifth invention in step S720, for example, the arbitrary constant χ of the equation (53) is changed to an equation (54) that minimizes the square sum of Fx _b1 , Fx _b2 , Fx _b3 , Fx _b4. ) To set.

Also, when applying the fifth aspect of the present invention at step S720, the minimum _{Fx b1, Fx b2, Fx b3} , the chi 4 sum of Fx _b4 is minimized or _{Fx b1, Fx b2,, Fx} b3, Fx b4 Χ that minimizes the difference between the value and the maximum value may be selected.

In step S740, if D ₄ ≠ 0, the process proceeds to step S750, and feedback control amount Fx _b1 of each wheel driving force distribution that realizes dFx, dFy, dM using equation (55) corresponding to equation (22), Fx _b2 , Fx _b3 , and Fx _b4 are obtained. Otherwise, the process proceeds to step S760.

Note that the arbitrary constant χ in the equation (55) is set to 0 as in step S730. When the fifth technique is applied, for example, as in step S730, χ that minimizes the sum of squares of Fx _b1 , Fx _b2 , Fx _b3 , and Fx _b4 is set by Expression (56).

In step S760, if D ₂ ≠ 0, the process proceeds to step S770, and Fx _bi that realizes dFx, dFy, and dM is obtained from an equation obtained by solving equation (15) with Fx _b2 known. Note that the value corresponding to the arbitrary constant χ in the equations (53) and (55) is set in the same manner as in steps S730 and S750. If D ₂ = 0, the process proceeds to step S780.

In step S780, the process proceeds to step S790 if D3 ≠ 0, from the equation obtained by solving equation (15) as known to Fx _b3, determined dFx, dFy, the Fx _bi to realize dM. Note that the value corresponding to the arbitrary constant χ in the equations (53) and (55) is set in the same manner as in steps S730 and S750. If D ₃ = 0, the process proceeds to step S800, and Fx _b1 = Fx _b2 = Fx _b3 = Fx _b4 = 0.

Next, the flowchart of FIG. 14 for obtaining Fx _bi that realizes dFx, dFy, and dM based on the sixth invention in step S60 of the flowchart of FIG. 9 will be described.

In step S805, the feed forward component Fx _fi for driving force distribution is set in the feed forward component recording variable FFx _fi for driving force distribution. Also, 0 is set to the calculation number counter Cn.

  In step S810, values obtained by dividing dFx, dFy, and dM by 10 are defined as ddFx, ddFy, and ddM.

Here, dFx, dFy, and dM are each divided by 10, but the number to be divided is not limited to this. Dividing by a number of 10 or more increases the calculation accuracy of Fx _bi , but the calculation time increases. Therefore, this division number is increased as long as the calculation performance of the controller 8 permits.

In step S820, the control amounts dFx _b1 , dFx _b2 , dFx _b3 , dFx _b4 of each wheel that realize ddFx, ddFy, ddM are applied according to the flowchart shown in FIG. 12 or FIG. 13 when the third invention is applied. When the fourth invention is applied, it is obtained according to the flowchart shown in FIG.

In step S830, a feedforward component Fx _fi driving force distribution, and updating the _{Fx fi ← Fx fi + dFx bi} , 1 is added to the number of operations counter Cn.

In step S840, if the calculation number counter Cn is 10 or more, the process proceeds to step S850, and the feedback control amount Fx _bi of the driving force distribution is set to Fx _bi = Fx _fi -FFx _fi .

In step S860, the feed forward component recording variable FFx _fi is reset to the feed forward component Fx _fi of the driving force distribution.

Next, in step S40 of the flowchart of FIG. 9, the dynamic target values Fx ^** , Fy ^** , M ^** for the vehicle longitudinal force, the vehicle lateral force, and the yaw moment, and the dynamic target of this vehicle behavior. value Fx ^**, Fy ^**, the M ^**, a feed-forward component of the driving force distribution for the wheels to be achieved if there is no change in the disturbance or vehicle weight etc. crosswinds such _{Fx f1, F xf2, F xf3} , F The flowchart of FIG. 10 for _{obtaining xf4} will be described.

First, in step S1030, the static target value Fx ^* of the vehicle longitudinal force is obtained by the equation (57) based on the depression amounts AP and BP of the accelerator pedal 6 and the brake pedal 7 and the vehicle speed V.

In Formula (57), Fa _x ^* is a value obtained by referring to the target driving force map based on the depression amount AP of the accelerator pedal 6 and the vehicle speed V, and Fb _x ^* is the depression amount BP of the brake pedal 7. Based on the target braking force map.

The target driving force map and the target braking force map are set as shown in FIGS. 18 and 19, for example. Further, Fx ^*, Fa _x ^*, both Fb _x ^* to the direction to accelerate the vehicle in front is positive.

In step S1040, the static target value Fy ^* of the vehicle lateral force is set with reference to the target vehicle lateral force map based on Fx ^* , the rotation angle θ of the steering wheel 5 and the vehicle speed V set in step S1030. . Further, the static target value γ ^* of the yaw rate is set with reference to the target yaw rate map based on Fx ^* , θ, and V.

  The target vehicle lateral force map and the target yaw rate map are maps set, for example, as shown in FIGS. 20 and 21, respectively. A method for setting these two maps will be described in step S1050 described later.

In step S1050, static target values Fx ₁ ^* , Fx ₂ ^* , Fx ₃ ^* , and Fx ₄ ^* for driving force distribution are set with reference to the static driving force distribution map based on θ, V, and Fx ^*. To do. This static driving force distribution map is set as shown in FIGS. 22a and 22b, for example.

  Here, how to obtain this static driving force distribution map and the target vehicle lateral force map and target yaw rate map used in step S1040 will be described below.

Four wheel driving force sum Fx _all (unit: N), left and right wheel driving force difference ΔFx _all (unit: N), front wheel driving force distribution η (unit: none), left and right wheel driving force difference front wheel distribution Δη (unit: N) None) is defined by the equations (58) to (61). In the present embodiment, η and Δη are always set to 0.6 (allocation to the front wheels is 60%).

Then, the following simulation or experiment is performed on all combinations of four parameters of Fx _all , ΔFx _all steering 5 rotation angle θ, and vehicle front / rear direction static target value Fx ^* that the vehicle can take, A static driving force distribution map, a target vehicle lateral force map, and a target yaw rate map are created.

First, the selected, Fx _all, determined by the driving force distribution Fx _i of each wheel formula (62) to (65) from DerutaFx _all, the steering angle of the front wheels 1 and 2 from the selected theta '[delta] ₁ = It is assumed that δ ₂ = θ ′ / 16 (the steering gear ratio is 1/16).

Next, the set driving force distribution Fx _i and the front wheel steering angle [delta] _1, [delta] ₂ (steering angle [delta] ₃ of the rear wheels 3, 4, [delta] ₄ is 0) to drive the vehicle in FIG. 8, and -Fx ^* Is added in the vehicle longitudinal direction at the vehicle center of gravity. Then, the vehicle lateral force Fy and the yaw rate γ when the vehicle speed V ′ becomes constant (steady state) after a sufficient time has elapsed are obtained. In this experiment or simulation, the running resistance elements such as air resistance and rolling resistance are excluded, and in the case of simulation, non-linearity such as driving force of each wheel and tire lateral force. This is done using a vehicle model that fully considers performance.

And finally, static driving force distribution map, the target vehicle sideways force map, the target yaw rate map of ^{V, θ, Fx *, Fy} *, γ *, the Fx _i ^*, V when each was now simulation ' ^{, θ, Fx *, Fy *} , γ *, and Fx _i ^*, static driving force distribution map, the target vehicle sideways force map, continue to set the target yaw rate map.

In step S1060, the dynamic target value Fx ^{** of} the vehicle longitudinal force, the dynamic target value Fy ^** of the vehicle lateral force, and the dynamic target value γ ^** of the yaw rate can be realized by distributing the driving force of each wheel. It is obtained by subjecting static target values Fx ^* , Fy ^* , γ ^* to a smoothing process (time delay element) so that the maneuverability of the vehicle is suitable within a certain range. Here, Fx ^** , Fy ^** , and γ ^** are obtained by using a second-order lag transfer function for Fx ^{* and} corresponding transfer functions for Fy ^* and γ ^* . In particular, a time delay element is added so that the responses of Fy ^** and γ ^** can be realized by the tire force of each wheel.

In step S1060, the obtained γ ^** is differentiated and multiplied by the yaw inertia moment I (unit: kg · m ² ) of the vehicle to obtain the dynamic target value M ^** of the yaw moment.

In step S1070, based on the static target value Fx _i ^* of the driving force distribution set in step S1050, the driving force distribution that substantially realizes the dynamic target values Fx ^** , Fy ^** , and γ ^** of the vehicle behavior. The basic values Fx ₁ ^## , Fx ₂ ^## , Fx ₃ ^## , and Fx ₄ ^{## of the above} are obtained by Expressions (66) to (69).

However, as shown in step S1050, η = Δη = 0.6, and Fx _all ^## is the same as Fx _i when Fx ^* is annealed to Fx ^{** in} step S1060. ^This is a value applied to the sum of Fx ₁ ^* + Fx ₂ ^* + Fx ₃ ^* + Fx ₄ ^* .

ΔFx _all ^## is a linear two-wheel model that approximates the vehicle linearly ("Motor Movement and Control", Chapter 3 Section 3.2.1, (Author) Masato Abe, (Publishing) Sankaido). Considering the case where the driving force difference ΔFx _all ^## is added, the model following control system (Chapter 3, Section 3.2 of “Vehicle Control”) is designed so that the yaw rate response of this linear two-wheel model is γ ^**. , Authors: Kimio Kanai, Tokumasa Ochi, Taketoshi Kawabuchi, publisher: Kashiwa Shoten), and corrected so as not to cause a deviation from the static target value Fx _i ^* of the driving force distribution in a steady state It is obtained from equation (70).

However,

However, in the equation (70), fr (s) is a transfer function of the smoothing process when the smoothing process is performed on γ ^* in step S1060 to obtain γ ^**, and Kf, Kr (unit: N) / Rad) is a cornering force per unit skid angle when the skid angle of the front and rear wheels is sufficiently small.

In step S1080, the driving force basic value Fx ₁ ^{# # _{distribution, Fx 2 ##, Fx 3 ##}} , realized by Fx ₄ ^{# #,} vehicle longitudinal direction force Fx ^{# #,} vehicle lateral force Fy ^{# #,} yaw The moment M ^## is obtained from the equations (72) to (74).

However,

Note that Fy ₁ ^## , Fy ₂ ^## , Fy ₃ ^## , Fy ₄ ^## are current vehicle states, and Fx ₁ ^## , Fx ₂ ^## , Fx ₃ ^## , Fx ₄ ^## are each The driving force and tire used in step S700 of the flowchart of FIG. 13 based on the current side slip angle β _i , wheel load W _i , and road surface friction coefficient μ _i of each wheel, which is the tire lateral force generated when applied to the wheel. Fy _i ^## is set from the tire characteristic map representing the relationship with the lateral force.

In step S1090, errors ΔFx, ΔFy, ΔM between Fx ^** , Fy ^** , M ^** and Fx ^## , Fy ^## , M ^## are obtained by Expressions (76) to (78).

In step S1100, ΔFx ₁ , ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ for correcting the errors ΔFx, ΔFy, and ΔM are obtained from ΔFx, ΔFy, ΔM, Fx _i ^##, and β _i with reference to the driving force correction amount map. . This driving force correction amount map is the same as the map of FIGS. 11a and 11b used in step S60 of the flowchart of FIG. 9, and dFx, dFy, dM are replaced with ΔFx, ΔFy, ΔM, Fx _b1 , Fx _b2 ,. Fx _b3 and Fx _b4 are used in place of ΔFx ₁ , ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ .

In step S1100, ΔFx ₁ , ΔFx ₂ , ΔFx ₃ , and ΔFx ₄ may be obtained according to the flowchart shown in FIG. In this case dFx even flowcharts, replacement dFy, DerutaFx the dM, DerutaFy, and _{ΔM, Fx b1, Fx b2,} Fx b3, ΔFx 1 to _{Fx b4, ΔFx 2, ΔFx 3} , used interchangeably with ΔFx _4.

In step S1110, the sum of ΔFx _i and Fx _i ^## is set as a feedforward component Fx _fi = Fx _i ^## + ΔFx _i of the driving force distribution of each wheel.

  Although the embodiment of the present invention has been described above, the present invention is different from the vehicle shown in FIG. 8 in that the vehicle behaviors Fx, Fy, and M need not be corrected or cannot be estimated. Can also be applied. For example, when the present invention is applied to a vehicle in which the vehicle lateral force Fy cannot be estimated, in step S50 in the flowchart of FIG. 9, the change in the vehicle lateral force Fy is estimated from the change in the yaw moment M, thereby correcting the vehicle lateral force. The quantity target value dFy may be set.

Further, the present invention is applied to a vehicle equipped with a steer-by-wire mechanism, such as a vehicle in which the rear wheels can be steered at an angle different from that of the front wheels, or a vehicle in which the steering angle of each wheel can be controlled independently of the rotation angle θ of the steering 5. Is also applicable. When applied to such a vehicle, for example, Fx _bi that realizes dFx, dFy, and dM, and a control amount dFx _bi that realizes ddFx, ddFy, and ddM are obtained by the flowchart of FIG. The steering angle control command value for each wheel may be used for the steering angle information used in the flowchart.

It is a figure showing the tire force added to the vehicle which turns by the right-and-left driving force difference. It is a figure showing the correction method of the yaw moment by a prior art. It is a figure showing the driving force, tire lateral force, steering angle, etc. of each wheel in a four-wheel independent drive vehicle. It is a figure showing the tire force which is the driving force in one certain wheel, tire lateral force, and the resultant force. It is a figure showing the relationship between braking / driving force and tire lateral force. It is a figure showing an example of the simulation result of the driving force control by a prior art example. It is a figure showing an example of the simulation result of the driving force control by this invention. It is a figure which shows the structure of the electric vehicle provided with the motor which drives four wheels independently. It is a figure which shows the flowchart of torque distribution control in one Embodiment. It is a figure which shows the flowchart which calculates | requires the feedforward component of driving force distribution. It is the map which recorded the driving force variation | change_quantity of each wheel which changes a vehicle behavior. It is the map which recorded the driving force change amount of each wheel which similarly changes a vehicle behavior. It is a figure which shows the flowchart which calculates | requires the feedback control amount of driving force distribution. It is a figure which shows the flowchart which calculates | requires the feedback control amount of driving force distribution. It is a figure which shows the flowchart which calculates | requires the feedback control amount of driving force distribution. It is a figure which shows the flowchart of torque distribution control in one Embodiment. It is a map of the sensitivity of the vehicle behavior with respect to the driving force change of each wheel. It is a figure showing the relationship between the driving force and tire lateral force which change according to a side slip angle, a wheel load, and a road surface friction coefficient. It is a map showing the static target value of the vehicle longitudinal force according to the depression amount of the accelerator pedal and the vehicle speed. It is a map showing the static target value of the vehicle longitudinal force according to the depression amount of a brake pedal. It is a map showing the static target value of the vehicle lateral force according to the steering rotation angle, the vehicle speed, and the vehicle longitudinal force. It is a map showing the static target value of the yaw rate according to the steering rotation angle, the vehicle speed, and the vehicle longitudinal force. It is a map showing the static target value of the driving force distribution of each wheel. It is a map showing the static target value of the driving force distribution of each wheel.

Explanation of symbols

1-4: Wheel 5: Steering 6: Accelerator pedal 7: Brake pedal 8: Controller 9: Battery 11-14: Motor 15: Steering gear 21-24: Wheel speed sensor 25: Steering angle sensor 26: Accelerator stroke sensor 27: Brake stroke sensors 31 to 34: inverters 41 to 44: rudder angle sensor 100: acceleration sensor 101: yaw rate sensor

Claims (8)

In the vehicle driving force distribution device that drives the four wheels independently,
Feedforward driving force distribution setting means for setting a feedforward component of the driving force distribution of each wheel;
Vehicle behavior determination means for detecting or estimating at least one of vehicle longitudinal force, vehicle lateral force, and yaw moment representing the vehicle behavior of the vehicle;
Vehicle behavior target value setting means for setting a target value based on the driving state of the vehicle for at least one of the vehicle behaviors detected or estimated by the vehicle behavior determination means;
Vehicle behavior error calculating means for obtaining an error between the target value and the value detected or estimated by the vehicle behavior determining means for the vehicle behavior set by the vehicle behavior target value setting means;
Vehicle behavior correction amount target value setting means for setting a correction amount target value for vehicle longitudinal force, vehicle lateral force, and yaw moment from the error obtained by the vehicle behavior error calculation means;
A feedback control amount calculation means for obtaining a feedback control amount of the driving force distribution of each wheel that realizes a correction amount target value of the vehicle longitudinal force, vehicle lateral force, and yaw moment;
Target driving force distribution calculating means for calculating the sum of the feedforward component of the driving force distribution and the feedback control amount of the driving force distribution as a target value of the driving force distribution of each wheel;
Driving force control means for independently controlling the driving force of each wheel according to the target value of the driving force distribution;
A driving force distribution device comprising: Vehicle behavior estimation means for estimating a vehicle longitudinal force, vehicle lateral force, and yaw moment realized by the target value of the driving force distribution;
For the vehicle behavior for which the target value has been set by the vehicle behavior target value setting means, a value obtained by adding the error obtained by the vehicle behavior error calculation means to the target value of the vehicle behavior and the vehicle behavior estimation means are estimated. Vehicle behavior error recalculation means for obtaining an error from the measured value,
Means for setting a correction amount target value of the vehicle longitudinal force, vehicle lateral force, and yaw moment from the error obtained by the vehicle behavior error recalculation means by the vehicle behavior correction amount target value setting means;
Driving force distribution correction amount calculating means for obtaining a correction amount of driving force distribution of each wheel that realizes a correction amount target value of the vehicle longitudinal force, vehicle lateral force, and yaw moment;
The driving force distribution according to claim 1, further comprising: setting a sum of the target value of the driving force distribution and a correction amount of the driving force distribution as a target value of the driving force distribution. apparatus.   The feedback control amount calculation means obtains a feedback control amount for driving force distribution based on the sensitivity of the vehicle longitudinal force, vehicle lateral force, and yaw moment with respect to the driving force change of each wheel. The driving force distribution device according to 1 or 2.   The feedback control amount calculation means includes tire lateral force sensitivity detection means for obtaining the sensitivity of the tire lateral force with respect to a change in driving force in each wheel, and obtains a feedback control amount of the driving force distribution based on the sensitivity of the tire lateral force. The driving force distribution device according to any one of claims 1 to 3.   The feedback control amount calculation means selects a feedback control amount for driving force distribution that makes the feedback control amount for each wheel smaller from a plurality of sets of feedback control amounts for the driving force distribution. Item 5. The driving force distribution device according to Item 3 or 4.   The feedback control amount calculation means performs a process of calculating a minute control amount of the driving force distribution of each wheel that realizes a minute control amount smaller than the correction amount target value. 6. The method according to claim 3, wherein the control is repeatedly performed asymptotically or coincident with a correction amount target value, and the sum of the minute control amounts of the driving force distribution obtained repeatedly is used as the feedback control amount of the driving force distribution. The driving force distribution device according to claim 1. The vehicle behavior target value setting means sets a target value for each of the vehicle longitudinal force, vehicle lateral force, and yaw moment,
The feedforward driving force distribution setting means estimates vehicle longitudinal force, vehicle lateral force, and yaw moment realized by the feedforward component, and vehicle longitudinal force realized by the target value and the feedforward component, vehicle Calculate the error from the lateral force and yaw moment,
The vehicle behavior correction amount target value setting means is configured to calculate the vehicle longitudinal force, the vehicle lateral force, the yaw from the error obtained by the vehicle behavior error calculation means and the error obtained by the feedforward driving force distribution setting means. The driving force distribution device according to any one of claims 1 to 6, wherein a moment correction amount target value is set. A driving force distribution set calculating means for obtaining a set of driving force distributions of the four wheels that realizes the same vehicle longitudinal force, vehicle lateral force, and yaw moment as the target value of the driving force distribution of each wheel;
The driving force according to any one of claims 1 to 7, wherein a target value of the driving force distribution of each wheel is reset from a plurality of driving force distributions in the set of driving force distributions. Distribution device.