JP4835054B2 - Vehicle stabilization control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle stabilization control system capable of solving the problem in which a control instruction selected by a driving assistance system is different from an operation selected by a driver, and the driver feels discomfort due to mismatching with his/her feeling. <P>SOLUTION: This system comprises an environmental information providing means providing external environmental information of a vehicle; a current position calculation means calculating a current position of the vehicle as a coordinate value of geometric coordinate system; a vehicle speed detection means detecting a vehicle speed of the vehicle; a conversion means converting the environmental information to an image obtained by projecting the environmental information on the driver's retinae using the current position and the vehicle speed and outputting the image as converted environmental information; a sensory quantity calculation means calculating and outputting stimulation felt by the driver from the converted environmental information and a change quantity of the converted environmental information as a sensory quantity; and a control instruction setting means determining a control instruction based on the sensory quantity. <P>COPYRIGHT: (C)2007,JPO&amp;INPIT

Description

  The present invention relates to a vehicle stabilization control system based on environmental information processing based on a human visual stimulus model.

  Conventional driving support systems such as navigation cooperative shift control, ACC, and lane keep control disclosed in Japanese Patent Application Laid-Open No. 11-210870 are physically and geometrically environmental information (road shape, up to obstacles) by sensors. Distance) is measured, and vehicle control and driver driving operations are supported based on the measured values.

Further, as in navigation cooperative shift control disclosed in Japanese Patent Application Laid-Open No. 2003-194202, there is a system in which a control system functions discontinuously and outputs a control command to assist a driver.
Japanese Patent Laid-Open No. 11-210870 JP 2003-194202 A

  When humans visually process surrounding environmental information, they do not capture physical quantities such as distance and speed, but rather perform sensory evaluation using changes (size, direction) of the image projected on the retina. ing. In the following, the evaluation result obtained by humanly evaluating the surrounding environment is referred to as a stimulus. For example, when the area of the image projected on the human retina changes, the greater the area of this image, the greater the stimulation. When the stimulus is the vertical axis and the distance is the horizontal axis, the relationship between the stimulus and the distance is similar to the Weber-Fechner's law as shown in FIG. A human selects a driving operation based on a stimulus as shown in FIG.

  On the other hand, the conventional driving support system determines the control command by directly using the speed and distance of the vehicle measured by the sensor. That is, a human driver selects an operation based on an evaluation result evaluated sensuously, whereas a driving support system selects a control command based on speed and distance. Thus, in a certain environment, there is a problem that the control command selected by the system and the operation selected by the driver are different from each other, and the driver feels uncomfortable because they do not match their own senses.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a vehicle stabilization control system that matches human senses.

In order to solve the above-mentioned problem, the invention described in claim 1 is directed to an environment information providing means (16) for providing road shape information as environment information outside the vehicle, and a current position of the vehicle in a geometric coordinate system. Using the current position calculation means (31) for calculating the coordinate value, the vehicle speed detection means (4a, 4b) for detecting the vehicle speed of the vehicle, the road shape information, the current position, and the vehicle speed, A virtual object having a predetermined area is inside or on the side of the road and is scheduled to pass after a specified time at an angle perpendicular to the road extending direction and perpendicular to the road plane. Assuming that the area of the object is converted into the area of the object when the entire road including the object is projected onto the retina of the driver, the velocity vector of the object is also output. Conversion hand And (33), determine the variation of the conversion area from said conversion area outputted to the velocity vector from said converting means (33), senses the stimulus felt by the driver from the change in conversion area to the conversion area Sensory quantity calculation means (34) for calculating and outputting a quantity, and control command setting means (2h, 2i, 2j) for determining a control command based on the sensory quantity.

Generally, environmental information is obtained by storing road signs and the like in advance as template images and detecting road signs that match the template images by pattern matching or the like. On the other hand, by using road shape information as environmental information , it has a predetermined area in the road or on the side of the road and is scheduled to pass after the specified time, and in the direction of road extension. A virtual object that stands perpendicular to the road plane at an orthogonal angle can be set. When this virtual object is used , it is not necessary to store a template image or the like, but the sensory amount can be calculated. For this reason, it is possible to calculate the sensory amount with a simple configuration. Based on the virtual object, a sensory amount as a stimulus felt by the driver is calculated, and a control command that matches the driver's sense can be determined by selecting a control command based on the sensory amount.

The invention according to claim 2 is a basic required driving force calculation for calculating a physical quantity corresponding to the basic required driving force in order to generate the basic required driving force required by the driver for the driving wheels provided in the vehicle. Part (2a), front and rear wheel load calculation part (2g) for detecting the load applied to each of the front and rear wheels in the vehicle, and environmental information providing means for providing road shape information as environmental information outside the vehicle (16) , a current position calculating means (31) for calculating the current position of the vehicle as a coordinate value of a geometric coordinate system, a vehicle speed detecting means (4a, 4b) for detecting the vehicle speed of the vehicle, and the road shape Using the information, the current position, and the vehicle speed , a virtual object having a predetermined area is located on the road or near the road where the virtual object has a predetermined area. Total A converted area obtained by converting the area of the object into the area of the object when the entire road including the object is projected onto the driver's retina, assuming that the object is standing perpendicular to the road plane at an angle orthogonal to the direction and outputs a conversion means (33) also outputs a velocity vector in which the object has to obtain the variation of the conversion area from said conversion area output from said converting means (33) and the velocity vector, the Sensory amount calculation means (34) for calculating and outputting a sensory amount as a stimulus felt by the driver from a change amount of the conversion area with respect to the conversion area , and the vehicle so that the sensory amount is constant based on the sensory amount. Based on the calculation results of the virtual turning radius estimation unit (2h) for estimating the virtual turning radius of the vehicle, the front and rear wheel load calculation unit (2g), and the virtual turning radius estimation unit (2h) A target value calculation unit (2i) for calculating the target value of the stability factor, and the basic required driving force calculation unit (2a) so as to follow the target value calculated by the target value calculation unit (2i). And a vibration suppression correction control unit (2j) that corrects the physical quantity corresponding to the basic required driving force calculated by the above-described calculation, and the driving force corresponding to the corrected physical quantity corrected by the vibration suppression correction control unit (2j) It is characterized by being generated in the drive wheel.

  The above configuration calculates the quasi-static grounding load applied to each of the front and rear wheels, and calculates the stimulus that the driver is or will receive using external environmental information and the current position of the vehicle as sensory quantities. To do. Then, a target value of the stability factor is set based on the load and the sensory amount, and the physical quantity corresponding to the basic required driving force is corrected so as to follow the target value. For this reason, it is possible to stabilize the stability factor so that the turning radius becomes ideal that matches the driver's feeling corresponding to the surrounding environment that changes from moment to moment, and for example, pitching vibration energy is suppressed. For example, each state quantity inside the vehicle is stabilized, and the traveling state of the vehicle can be stabilized.

  Therefore, it is possible to suppress the fluctuation of the vehicle behavior due to the road surface and other external factors (road surface disturbance) from disturbing the vehicle body posture and the traveling track. In addition, the vehicle body posture disturbance and vibration caused by these factors are suppressed so as to coincide with the driver's senses, so the driver's steering operation to correct the vehicle body posture disturbance and vibration is no longer necessary. It is possible to prevent the disturbance component from being induced by the steering operation.

The invention according to claim 3 is characterized in that the basic required driving force calculating section (2a) calculates a basic required torque as a physical quantity corresponding to the basic required driving force.

According to a fourth aspect of the present invention, the front and rear wheel ground load calculating unit (2g) is configured to use the front and rear wheel static ground as the front and rear wheel ground load when the vehicle is in a steady running state. The load value is calculated, and the target value calculation unit (2i) calculates the target value of the stability factor based on the front and rear wheel static ground loads.

The invention according to claim 5 includes an estimated road gradient calculation unit (2d) that estimates a gradient of a road surface on which the vehicle is traveling, and the front and rear wheel ground load calculation unit (2g) is configured as the front and rear wheel ground load. Calculating the front and rear wheel static ground loads applied to the front wheels and the rear wheels when the vehicle is in a steady running state, and based on the estimated road gradient calculated by the estimated road gradient calculation unit (2d), The fluctuation amount of the front and rear wheel ground load when the vehicle travels on the estimated road gradient is calculated, and a value obtained by estimating the fluctuation due to the estimated road gradient from the front and rear wheel static ground load is obtained as the front and rear wheel ground load. The target value calculation unit (2i) determines the stability factor based on the front and rear wheel ground contact load that allows for the fluctuation of the front and rear wheel installation load due to the estimated road gradient. Characterized in that it adapted to calculate the value.

  In this way, the fluctuation amount of the front and rear wheel grounding load according to the road gradient is obtained, and the amount of change in the front and rear wheel grounding load due to the road gradient is subtracted from the static grounding load of each of the front and rear wheels, thereby obtaining the road gradient. It is possible to obtain the corresponding front and rear wheel ground loads.

The invention according to claim 6 is characterized in that the vibration damping correction control unit (2j) has a state equation indicating a state quantity in the vehicle based on a sprung vibration model in the vehicle, and the state equation. And the stability factor calculated from the output equation and the state quantity, and the target value calculation unit (2i) calculates the stability factor. The physical quantity corresponding to the basic required driving force is corrected based on the difference between the stability factor and the target value.

  Thus, based on the sprung vibration model in the vehicle, the state equation indicating the state quantity in the vehicle and the output equation expressing the stability factor in the state quantity based on the state equation are stored in the vibration suppression correction control unit in advance. In addition, the stability factor can be obtained from the output equation and the state quantity, and the physical quantity corresponding to the basic required driving force can be corrected based on the difference from the target value of the stability factor. Thus, the physical quantity corresponding to the basic required driving force is corrected so that the stability factor obtained from the output equation and the state quantity can follow the target value.

The invention according to claim 7 has a running resistance disturbance estimation unit (2f) for estimating a running resistance disturbance applied to the wheel in the vehicle, and the vibration suppression correction control unit (2j) estimates the running resistance disturbance. In view of the running resistance disturbance estimated by the part (2f), the state quantity in the state equation is obtained, the stability factor is obtained based on the obtained state quantity and the output equation, and the stability factor A difference from the target value is obtained.

In this manner, the travel disturbance resistance is estimated by the travel disturbance estimation unit, and the state quantity in the state equation can be obtained in view of the travel disturbance resistance. For example, as shown in claim 8 , the running disturbance resistance referred to here is obtained by multiplying, for example, these based on the differential value of the wheel speed of the front wheel and the weight of the vehicle.

According to an eighth aspect of the present invention, the travel resistance disturbance estimation unit (2f) is configured to detect the front wheel as the travel resistance disturbance based on a differential value of a wheel speed of a front wheel provided in the vehicle and a weight of the vehicle. It is characterized in that a running resistance disturbance is obtained.

In the invention according to claim 9, the conversion means assumes that the virtual object stands on both sides of the road at or near the point where the road is scheduled to pass after the specified time. The conversion area of virtual objects and the velocity vector of those objects are output .

In the invention according to claim 10, a plurality of the sensory quantities are calculated using a plurality of virtual objects assumed to be standing at points corresponding to a plurality of different specified times, and the control The determination of the command is based on a plurality of the sensory quantities.

Since the driver is driving in consideration of the future state of the vehicle, in this way, by estimating the stimulus received by the driver from a point where the driving is predicted in the future or an object in the vicinity thereof, the driver's Control commands that match the senses can be output.

The invention according to claim 11 is characterized in that the virtual object is newly installed at every predetermined control period .

[Embodiment 1]
A vehicle stabilization control system according to an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a schematic configuration of a vehicle stabilization control system in the present embodiment. In the present embodiment, the description will be made assuming that the driving mode of the vehicle is a rear wheel drive, but of course, the present invention can be applied to a front wheel drive vehicle or a four wheel drive vehicle.

  The vehicle stabilization control system of the present embodiment adjusts the driving torque generated by the engine 1 provided in the vehicle, and stabilizes the fluctuation of the stability factor based on the movement of the front and rear wheel loads caused by pitching vibration energy and the like. This stabilizes the body posture and vehicle characteristics. In this embodiment, it is assumed that the engine 1 is a gasoline engine that operates on gasoline.

  In the vehicle stabilization control system shown in FIG. 1, the engine 1 is controlled by an engine ECU 2. The engine ECU 2 receives detection signals from an accelerator stroke sensor 3, front wheel speed sensors 4a and 4b, a steering angle sensor 5, an intake air amount sensor 6, an engine rotation sensor 7, and a G / yaw rate / gyro sensor 8. At the same time, a reception signal from the GPS receiver 17 and road information from the road shape information providing means 16 are input.

  The accelerator stroke sensor 3 outputs a detection signal corresponding to the operation amount of the accelerator pedal 9. The engine ECU 2 obtains the accelerator operation amount based on the detection signal from the accelerator stroke sensor 3, and obtains the requested axle torque that becomes the required driving force according to the accelerator operation amount.

  The front wheel speed sensors 4a and 4b are respectively provided corresponding to the front wheels 10a and 10b serving as steering wheels, and are configured by a wheel speed sensor 4a for the right front wheel and a wheel speed sensor 4b for the left front wheel. ing. Each of these wheel speed sensors 4a is composed of a known sensor such as an electromagnetic pickup type that outputs a different detection signal according to the rotation of the teeth of a gear-type rotor provided on the axle, and corresponds to the rotation of each front wheel. Generate a detection signal.

  The rudder angle sensor 5 is a well-known sensor that outputs a detection signal corresponding to the steering angle of the vehicle front wheel, that is, the inclination angle of the vehicle front wheel with respect to the vehicle longitudinal axis. For example, when the steering shaft rotates in accordance with the steering operation, it is converted into steering of the vehicle front wheel via the steering mechanism. Therefore, the rudder angle sensor 5 detects the amount of steering rotation, thereby detecting the vehicle front wheel. A detection signal corresponding to the steering angle is output.

  The intake air amount sensor 6 and the engine rotation sensor 7 are both well-known ones installed in the engine 1, and the intake air amount sensor 6 outputs a detection signal corresponding to the intake air amount sucked into the engine. The engine rotation sensor 7 outputs a detection signal corresponding to the engine rotation speed.

  The road shape information providing means 16 and the G / yaw rate / gyro sensor 8 are provided in the main body 11 of the navigation system, for example. For example, the road shape information providing means 16 notifies the engine ECU 2 of road information stored in an information recording medium such as a hard disk or DVD provided in the main body 11. The G / yaw rate / gyro sensor 8 detects the acceleration and yaw angle of the vehicle and transmits them to the engine ECU 2.

  The GPS receiver 17 is connected to the main body 11 of the navigation system, calculates the current position of the vehicle in the geometric coordinate system from the received wave received from the GPS satellite, and transmits it to the engine ECU 2 as a received signal.

  The engine ECU 2 performs various calculations based on the detection signals from these sensors 3, 4 a, 4 b, 5 to 8, the received signal sent from the GPS receiver 17, and the road information of the road shape information providing means 16. And the intake air amount is adjusted based on the calculation result. As a result, the axle torque transmitted to the rear wheels 15a and 15b serving as drive wheels via the transmission 12, the final reduction gear 13 and the drive shaft 14 is adjusted.

  FIG. 2 schematically shows a block configuration of the engine ECU 2, and details of the engine ECU 2 will be described with reference to FIG.

  The engine ECU 2 is configured by a microcomputer that includes a CPU, a RAM, a ROM, an I / O, and the like. Then, the engine control program stored in the ROM is executed by the CPU and various calculations are performed to control the intake air amount to the engine 1.

  Specifically, as shown in FIG. 2, the engine ECU 2 includes a basic required torque calculation unit 2a, an estimated axle torque calculation unit 2b, a front wheel speed calculation unit 2c, an estimated road gradient calculation unit 2d, and a steering angle calculation unit 2e. A front wheel running resistance disturbance estimation unit 2f, front and rear wheel static contact load calculation unit 2g, virtual turning radius estimation calculation unit 2h, target value calculation unit 2i, vibration suppression correction control unit 2j, and visual space model processing unit 2k. It is configured.

  The basic required torque calculation unit 2a receives the detection signal output from the accelerator stroke sensor 3, obtains the accelerator operation amount as a physical quantity based on the detection signal, and obtains the basic request driving force corresponding to the operation amount. Find the torque. The basic required torque obtained here is the torque used for acceleration and deceleration of the vehicle, that is, the basically required axle torque. And the calculation result in this basic required torque calculation part 2a is output to the vibration suppression correction control part 2j.

  The estimated axle torque calculation unit 2b calculates an estimated axle torque, that is, an axle torque that will be generated when a detection result is obtained, based on detection signals from the intake air amount sensor 6 and the engine rotation sensor 7. To do. The calculation result in the estimated axle torque calculation unit 2b is output to the vibration suppression correction control unit 2j and the front and rear wheel static ground load calculation unit 2g.

  The front wheel speed calculation unit 2c calculates the wheel speeds of both front wheels serving as steering wheels based on detection signals from the both wheel speed sensors 4a and 4b. The front wheel speed calculation unit 2c is output to the front wheel running resistance disturbance estimation unit 2f and the virtual turning radius estimation calculation unit 2h.

  The estimated road gradient calculation unit 2d receives road information from the road shape information providing means 16, receives a reception signal from the GPS receiver 17, and extracts road gradient information included in the road information by extracting road information. Estimate the slope. And the calculation result in this estimated road gradient calculation part 2d is output to the front-and-rear wheel static ground load calculation part 2g. Here, the road gradient is estimated based on the road information stored in the road shape information providing means 16, but the vehicle longitudinal driving force acceleration component is removed from the detection signal from the acceleration sensor installed in the vehicle, and the gravity is The road gradient may be estimated by a known method such as calculating the road gradient based on the acceleration component.

  The steering angle calculation unit 2e calculates the steering angle based on the detection signal from the steering angle sensor 5, and outputs the calculation result to the virtual turning radius estimation calculation unit 2h and the target value calculation unit 2i.

  The visual space model processing unit 2k will be described below.

  FIG. 3 is a detailed block diagram of the visual space model processing unit 2k that performs environmental information processing based on the human visual stimulus model. The input to the current position calculation unit 31 is the sensor value of the G / yaw rate / gyro sensor 8 of the vehicle, the GPS reception signal, and the road information of the road shape information providing means 16. The current position calculation unit 31 calculates the positional relationship between the road and the vehicle, that is, the detailed position and travel angle of the current vehicle on the map, and outputs them to the virtual signboard generation unit 32. The virtual signboard generation unit 32 generates and outputs a virtual signboard based on a virtual signboard generation algorithm, which will be described later with reference to FIG. 4, from the map information, the current position calculation unit 31, and the outputs of the wheel speed sensors 4a and 4b. The geometric coordinate system / retinal coordinate system coordinate conversion unit 33 converts the virtual signboard generated by the virtual signboard generation unit 32 into an estimated image estimated to be projected on the retina of the driver and outputs the estimated image. The visual sensation amount calculation unit 34 uses the estimated image output from the geometric coordinate system / retinal coordinate system coordinate conversion unit 33 to calculate the sensation amount τ using Equation 58 described later.

  Hereinafter, the virtual signboard generation unit 32 of FIG. 3 will be described. The driver continuously captures the entire road as environmental information, and predicts the driving line that he wants to drive in the future as a continuous point based on the captured environmental information. In order to imitate an algorithm that continuously captures the entire road of the driver, the virtual signboard generation unit 32 installs virtual signboards (hereinafter referred to as virtual signboards) at regular time intervals on both sides of the road. It is assumed that the virtual signboard has a quadrangular shape having the same area A, and stands perpendicular to the road plane at an angle perpendicular to the road extending direction like a road sign.

  Hereinafter, based on FIG. 4, processing of the virtual signboard generation unit 32 of FIG. 3 and input and output to the virtual signboard generation unit 32 will be described. The calculation in the virtual signboard generation unit 32 when the vehicle position at the current calculation timing is the position P (0) and the vehicle speed at that time is the speed v will be described. Assuming that the vehicle is traveling on the center of the road, a position P (1) that is a geometric coordinate system coordinate value that is an arrival position after t1 seconds due to the speed v at the position P (0) is calculated. Next, the left virtual signboard S (1) L41 and the right virtual signboard S (1) R42 are installed at a point of a distance 1 in the vertical direction passing through the position P (1) with respect to the center line of the road. Similarly, the position P (2), which is the arrival position after t2 seconds by the speed v at the position P (0), is calculated, and the vertical distance l passing through the position P (2) with respect to the center line of the road Left virtual signboard S (2) L43 and right virtual signboard S (2) R44 are installed. Further, the position P (3), which is the arrival position after t3 seconds due to the speed v at the position P (0), is calculated, and the vertical distance l passing through the position P (3) with respect to the center line of the road is calculated. The left virtual signboard S (3) L45 and the right virtual signboard S (3) R46 are installed at the point. Next, using the vehicle speed output by the vehicle speed sensor, a speed vector that all the installed virtual signs have for the host vehicle is calculated. The virtual signboard generation unit 32 is executed every fixed control period t4 seconds, and all virtual signboards are newly installed every t4 seconds. The output from the virtual signboard generation unit 32 that has executed such processing is the coordinate value of each virtual signboard expressed in the geometric coordinate system and the velocity vector of each virtual signboard. The geometric coordinate system may be, for example, a world geodetic system or a coordinate system set in the map information.

  Hereinafter, the geometric coordinate system / retinal coordinate system coordinate conversion unit 33 of FIG. 3 will be described with reference to FIG. In the central vision visual model shown in FIG. 5, when the coordinate value of the center of the eyeball of the driver at the position P (0) is set as (x, y, z) = (0, 0, 0), it is projected on the retina. The elevation angle ψ and the azimuth angle φ with respect to the coordinate value (x, y, z) = (xs, ys, zs) of the virtual signboard can be expressed as Equation 56 and Equation 57. Hereinafter, the coordinate system represented by the elevation angle ψ and the azimuth angle φ is referred to as a retinal coordinate system.

（数５７）

幾何学座標系・網膜座標系座標変換部３３では、幾何学座標系から、ドライバの網膜に投射された際の座標系である網膜座標系に変換する。以下、詳細を説明する。まず、幾何学座標系・網膜座標系座標変換部３３では、仮想看板生成部３２から入力された各仮想看板の幾何学座標系座標値に対して、面積Ａを持たせるように面積を表す幾何学座標系座標値群とする。 (Formula 56)

(Equation 57)

The geometric coordinate system / retinal coordinate system coordinate conversion unit 33 converts the geometric coordinate system into a retinal coordinate system that is a coordinate system projected onto the driver's retina. Details will be described below. First, in the geometric coordinate system / retinal coordinate system coordinate conversion unit 33, a geometry that represents an area so as to have an area A with respect to the geometric coordinate system coordinate value of each virtual signboard input from the virtual signboard generation unit 32. A coordinate value group of academic coordinate systems.

  Then, the geometric coordinate system coordinate value group representing the area of each virtual signboard is converted into a coordinate value group in the retinal coordinate system using Expression 56 and Expression 57.

  Further, the velocity vectors of the left virtual signboard S (n) L and the right virtual signboard S (n) R are obtained from the velocity vectors of the left virtual signboard S (n) L and the right virtual signboard S (n) R in the retinal coordinate system. Convert to However, n = 1, 2, 3.

  Hereinafter, the visual sensation amount calculation unit 34 in FIG. 3 will be described. The driver generally does not use physical quantities such as distance and speed in the geometric coordinate system, but changes in the size and direction of the image projected on the retina (hereinafter referred to as the projected image), such as the area and area change rate. The surrounding environment is evaluated sensuously from changes in the direction of movement of the projected image. In this way, when the stimulus felt by the driver is quantified as the sensory amount τ, it can be expressed as Equation 58.

ドライバが投射像の面積変化率から周囲の環境を評価していることから、（数式５８）における分母である感覚（視覚）刺激を投射像の面積Ａ’とし、分子である感覚（視覚）刺激変化量を投射像の速度ベクトルと面積とを用いて表現される面積Ａ’の変化量ΔＡ’とする。 (Formula 58)

Since the driver evaluates the surrounding environment from the area change rate of the projected image, the sensory (visual) stimulus that is the denominator in (Formula 58) is defined as the projected image area A ′, and the sensory (visual) stimulus that is a numerator. Let the amount of change be the amount of change ΔA ′ of the area A ′ expressed using the velocity vector and area of the projected image.

  The visual sensation amount calculation unit 34 in FIG. 3 calculates the area A ′ of the projected image as in Expression 59 using the elevation angle ψ and the azimuth angle φ calculated in Expression 56 and Expression 57 described above.

(Equation 59)
A '= ψ × φ
Thereby, the area A′nL of the left virtual signboard S (n) L and the area A′nR of the right virtual signboard S (n) R that are estimated to be projected onto the retina of the driver can be calculated. Further, the visual sensation amount calculation unit 34 senses the stimulus given to the driver by the left virtual signboard S (n) L, and senses τ (n) L gives the stimulus given to the driver by the right virtual signboard S (n) R. n) Output as R.

A specific example of processing in the visual sensation amount calculation unit 34 will be described with reference to FIGS. 4 and 6. When the coordinate value and velocity vector of each virtual signboard expressed in the retinal coordinate system are input from the geometric coordinate system / retinal coordinate system coordinate conversion unit 33, the visual sense amount calculation unit 34 receives the position P of the geometric coordinate system. The area A ′ (1) L of the left virtual signboard S (1) L41 estimated to be projected onto the retina of the driver at (0), its area change amount ΔA ′ (1) L, and the right virtual signboard S (1 ) The area A ′ (1) R of R42 and its area change amount ΔA ′ (1) R, the area A ′ (2) L of the left virtual signboard S (2) L43 and its area change amount ΔA ′ (2) L, Area A ′ (2) R of the right virtual signboard S (2) R44 and its area change amount ΔA ′ (2) R, Area A ′ (3) L of the left virtual signboard S (3) L45 and its area change amount ΔA '(3) L, area A' (3) R of the right virtual signboard S (3) R46, and its area change amount ΔA '(3) R are calculated. Then, from the calculated area A ′ (1) L and the area change amount ΔA ′ (1) L, the sensory amount τ (1) L is changed to τ (1) L = ΔA ′ (1) L / A ′ (1). L is calculated, and the sensory quantity τ (1) R is calculated from the area A ′ (1) R and the area change amount ΔA ′ (1) R by τ (1) R = ΔA ′ (1) R / A ′ (1 ) R, and the sensory amount τ (2) L is calculated from the area A ′ (2) L and the area change amount ΔA ′ (2) L by τ (2) L = ΔA ′ (2) L / A ′ ( 2) Calculated as L, and the sensory amount τ (2) R is calculated from the area A ′ (2) R and the area change amount ΔA ′ (2) R to τ (2) R = ΔA ′ (2) R / A ′ (2) Calculated as R, and the sensory amount τ (3) L is calculated from the area A ′ (3) L and the area change amount ΔA ′ (3) L by τ (3) L = ΔA ′ (3) L / A '(3) L is calculated, and the sensory amount τ (3) R is calculated from the area A ′ (3) R and the area change amount ΔA ′ (3) R by τ ( ) R = ΔA '(3) R / A' (3) for calculating as R.

  The front wheel running resistance disturbance estimation unit 2f in FIG. 2 estimates a front wheel running resistance disturbance based on the calculated front wheel speed. A running resistance corresponding to the wheel speed is generated on the front wheels. For this reason, the running resistance disturbance is estimated from the wheel speed. For example, the force [N / m] in the translational direction is obtained by multiplying the vehicle weight by the differential value of the wheel speed, and further multiplied by the radius of the rolling wheel to thereby multiply the running resistance disturbance to the moment acting on the rolling wheel [ N].

  By determining the running resistance disturbance based on such a single differentiation of the wheel speed, it is possible to determine how much running resistance disturbance has entered as a result, regardless of what caused the running resistance disturbance. Become. That is, the running resistance disturbance is caused by, for example, cornering drag generated by steering by the driver or by road surface unevenness, but in any case, the wheel speed changes as a result. If the running resistance disturbance is calculated from the change (differential value) of the wheel speed, the running resistance disturbance received by the rolling wheels can be obtained regardless of the cause.

  As for the running resistance disturbance, the characteristics of the wheel speed and the running resistance disturbance are stored in advance in a memory in the engine ECU 2 and the running resistance disturbance corresponding to the wheel speed calculated based on the characteristics is selected. By doing so, it is also possible to estimate.

  The front and rear wheel static ground load calculation unit 2g in FIG. 2 will be described. The front and rear wheel static ground load calculating unit 2g calculates a static ground load on the front and rear wheels based on the calculated estimated axle torque and estimated road gradient.

  First, changes in the ground contact load of the front and rear wheels will be described. The variation in the ground contact load between the front and rear wheels is caused by, for example, pitching vibration. Pitching vibration is vibration that occurs around the left and right axis of the vehicle around the center of gravity of the vehicle, and the energy generated by this pitching vibration is called pitching vibration energy.

  Pitching vibration is generated by nose diving during driving (acceleration), braking (deceleration), and steering (turning). FIG. 7 shows these states.

  As shown in FIG. 7 (a), during driving (acceleration), the vehicle body side cannot follow the wheel rotation and is left behind, so the vehicle front side (nose) rises around the center of gravity of the vehicle, and a scouring occurs. To do. Further, as shown in FIG. 7B, when braking (decelerating), the vehicle body cannot follow the deceleration of the wheel due to inertia when braking force is generated on the wheel. A nose dive that sinks forward (nose) occurs. Then, as shown in FIG. 7C, cornering drag is generated during steering (turning), so that the wheel is decelerated based on the cornering drag and nose dive is generated as in braking (deceleration).

  Such rotational vibration generated around the center of gravity of the vehicle such as a squat and nose dive is pitching vibration, and the energy that generates it is pitching vibration energy. Such pitching vibration energy is always generated while the vehicle is running.

  Then, due to such pitching vibration or the like, the relationship between the ground load of the front and rear wheels and the force applied to the wheels fluctuates with respect to the steady running. That is, as shown in FIG. 7A, the front wheel ground load Wf is smaller and the rear wheel ground load Wr is larger at the time of squat than at the time of steady running, and the driving torque reaction force is increased. As shown in FIG. 7B, the front wheel ground load Wf is larger and the rear wheel ground load Wr is smaller during nose dive due to deceleration than during steady running. Therefore, the front wheel braking force is large and the rear wheel braking force is small. Further, as shown in FIG. 7C, the front wheel ground load Wf is larger and the rear wheel ground load Wr is smaller even during nose dive by turning than during steady running.

  Since the ground loads Wf and Wr are thus changed, the cornering power is changed, and thus the turning of the vehicle is not stable, and the driver is burdened with an operation burden such as a steering correction operation.

  When the relationship between the pitching vibration, the front and rear wheel ground loads, and the front and rear wheel cornering power is shown in a timing chart, the relationship is as shown in FIG. That is, when the pitching vibration as shown in FIG. 8 (a) occurs, as shown in FIG. 8 (b), the front wheel and rear wheel loads Wf, Wr are the front wheel and rear wheel loads Wfo, This is expressed by Formula 1 in which ΔWf and ΔWr, which are variations due to pitching vibration, are added to Wro.

(Equation 1)
Wf = Wfo + ΔWf, Wr = Wro + ΔWr
Therefore, the front wheel and rear wheel loads Wf and Wr have waveforms corresponding to the waveform of the pitching vibration. As shown in FIG. 8 (c), the front wheel and rear wheel cornering powers Kcf and Kcr are also the product of the front and rear wheel loads Wf and Wr and the coefficient Cw in the linear region of the tire characteristics. Thus, the waveforms of the cornering powers Kcfo and Kcro in the steady state are the same as the front wheel and rear wheel loads Wf and Wr.

  Therefore, if the axle torque generated by the engine is corrected so that the steering characteristics that govern the lateral movement of the vehicle are stabilized by using the movement of the front and rear wheel ground loads caused by pitching vibration, etc., the attitude of the vehicle body In addition, it is considered that disturbance of vehicle characteristics can be effectively suppressed.

  Next, the state quantity in the vehicle will be described with reference to the schematic diagram of the sprung vibration model shown in FIG.

  The sprung vibration model shown in FIG. 9 is assumed to receive a torque reaction force change ΔTr based on an arbitrary steady state and generate a vibration around the pitching center in the sprung portion. Here, assuming that the vehicle body is a flat plate made of an arbitrary reference plane parallel to the horizontal direction, the flat plate is provided with tires supported by a suspension, and the engine and transmission etc. are mounted on the vehicle body via a mount having a spring element. It is assumed that the sprung vibration is supported.

  In this sprung vibration model, each constant is set as follows. First, for each of the front and rear wheels provided on the reference plane B, the suspension spring constant is Kf, Kr, the suspension damping coefficient is Cf, Cr, the weight of the engine and transmission is m, and the spring constant of the engine mount is Ke and the attenuation coefficient are Ce.

  Further, assuming that the tire radius is r, the mass of the vehicle body [kg] on the spring is M, the mass [kg] of the engine and transmission (T / M) is m, the wheel base [m] is L, the center of gravity of the vehicle and the front axle. The distance [m] between the vehicle center of gravity and the rear wheel axle is Lr, the distance [m] between the vehicle center of gravity and the engine and T / M center of gravity is Le, and the vehicle body reference line (reference The distance [m] between the height of the plane and the height of the center of gravity of the vehicle is hc.

  The pitching inertia moment [kgm2] of the vehicle body is Ip, and the gravitational acceleration [m / s2] is g.

  On the other hand, regarding the independent variables, the vertical displacement [m] of the vehicle body on the spring is x, the vertical displacement of the engine and T / M is xe, and the pitch angle [rad] around the virtual pitching center is θp.

  First, since the virtual pitch angle with respect to the pitching center is expressed as θp, the amount of displacement around the pitch center at the front wheel axis that is Lf away from the pitching center is Lfθp, and the pitch at the rear wheel axis that is Lr away from the pitching center. The amount of displacement around the center is Lrθp. For this reason, the vertical displacement of the vehicle body is x + Lfθp on the front wheel side and x−Lfθp on the rear wheel side.

  Therefore, the equation of motion around the pitch center of the vehicle body is expressed as Equation 2.

(Equation 2)
Ipθp ″ = − Lf {Kf (x + Lfθp) + Cf (x ′ + Lfθp ′)}
−Le {Ke (x + Leθp−xe) + Ce (x ′ + Leθp′−xe ′)}
+ Lr {Kr (x−Lrθp) + Cr (x′−Lrθp ′)}
+ HcθpMg + (hcg−hc) ΔTr / r + ΔTr
Further, the equation of the vertical motion of the vehicle body and the equation of the vertical motion of the engine 1 and T / M are expressed as Equation 3 and Equation 4, respectively.

(Equation 3)
Mx ″ = − Kf (x + Lfθp) −Cf (x ′ + Lfθp ′)
−Ke (x + Leθp−xe) −Ce (x ′ + Leθp′−xe ′)
−Kr (x−Lrθp) −Cr (x′−Lrθp ′)
(Equation 4)
mxe ″ = − Ke (xe−x−Leθp) −Ce (xe′−x′−Leθp ′)
Then, when xe ″, x ″, and θp ″ are obtained from Equations 2 to 4, Equations 5 to 7 are obtained, respectively.

(Equation 5)
xe ″ = − Ke / m · xe−Ce / m · xe ′ + Ke / m · x + Ce / m · x ′
+ KeLe / m · θp + CeLe / m · θp '
(Equation 6)
x ″ = Ke / M · xe + Ce / M · xe ′ − (Ke + Kf + Kr) / M · x
− (Ce + Cf + Cr) / M · x ′ − (KfLf + KeLe−KrLr) / M · θp
− (CfLf + CeLe−CrLr) / M · θp ′
(Equation 7)
θp ″ = KeLe / Ip · xe + CeLe / Ip · xe ′
− (KfLf + KeLe−KrLr) / Ip · x
− (CfLf + CeLe−CrLr) / Ip · x ′
− (KfLf2 + KeLe2 + KrLr2-hcMg) / Ip · θp
− (CfLf2 + CeLe2 + CrLr2) / Ip · θp ′
+ {1+ (hcg−hc) / r} / Ip · ΔTr
Accordingly, the state quantities are xe = x1, xe ′ = x2, x = x3, x ′ = x4, θp = x5, θp ′ = x5, ΔTr = u, and the coefficients of the variables in the above equations are a1 to a6. , B1 to b6, c1 to c6, and P (1), the above equations are converted as follows.

(Equation 8)
xe ″ = a1xe + a2xe ′ + a3x + a4x ′ + a5θp + a6θp ′
= A1x1 + a2x2 + a3x3 + a4x4 + a5x5 + a6x6
(Equation 9)
x ″ = b1xe + b2xe ′ + b3x + b4x ′ + b5θp + b6θp ′
= B1x1 + b2x2 + b3x3 + b4x4 + b5x5 + b6x6
(Equation 10)
θp ″ = c1xe + c2xe ′ + c3x + c4x ′ + c5θp + c6θp ′ + P (1) u
= C1x1 + c2x2 + c3x3 + c4x4 + c5x5 + c6x6 + P (1) u
However, in the above formulas 8 to 10, a1 = −Ke / m, a2 = −Ce / m, a3 = Ke / m, a4 = Ce / m, a5 = KeLe / m, a6 = CeLe / m, b1 = Ke / M, b2 = Ce / M, b3 = − (Ke + Kf + Kr) / M, b4 = − (Ce + Cf + Cr) / M, b5 = − (KfLf + KeLe−KrLr) / M, b6 = − (CfLf + CeLe−CrLr) / M, c1 = KeLe / Ip, c2 = CeLe / Ip, c3 =-(KfLf
+ KeLe-KrLr) / Ip, c4 =-(CfLf + CeLe-CrLr) / Ip, c5 =-(KfLf2 + KeLe2 + KrLr2-hcMg) / Ip, c6 =-(CfLf2 + CeLe2 + CrLr) / c (h) + c (1) / R} / Ip.

  In addition, since x1 to x6 are defined as described above, the following relationship also holds.

(Equation 11)
x′1 = xe ′ = x2
(Equation 12)
x′2 = xe ″ = a1x1 + a2x2 + a3x3 + a4x4 + a5x5 + a6x6
(Equation 13)
x′3 = x ′ = x4
(Equation 14)
x'4 = x '= b1x1 + b2x2 + b3x3 + b4x4 + b5x5 + b6x6
(Equation 15)
x′5 = θp ′ = x6
(Equation 16)
x′6 = θp ″ = c1x1 + c2x2 + c3x3 + c4x4
+ C5x5 + c6x6 + P (1) u
Therefore, when Expressions 11 to 16 are expressed in the state space, the state equation is expressed by a determinant of 6 rows and 6 columns like Expression 17, and when Expression 17 is simplified, it is expressed as Expression 18.

（数１８）
ｘ’＝Ａｘ＋Ｂｕ
このようにして、バネ上振動モデルの状態方程式が導出される。したがって、この状態方程式に基づいて、エンジンが発生させる車軸トルク（駆動力に相当する物理量）を補正すれば、車両の横運動を支配するステアリング特性（スタビリティファクタ）を能動的に制御することが可能となる。 (Equation 17)

(Equation 18)
x ′ = Ax + Bu
In this way, the state equation of the sprung vibration model is derived. Therefore, if the axle torque (physical quantity corresponding to the driving force) generated by the engine is corrected based on this state equation, the steering characteristic (stability factor) that governs the lateral movement of the vehicle can be actively controlled. It becomes possible.

  The front and rear wheel static ground load calculation unit 2g in FIG. 2 sets the state quantities xe ′, x ′, θp ′... Each state quantity in the steady state is calculated. Each state quantity at this time is expressed as Equations 19-21.

(Equation 19)
0 = a1xe + a3x + a5θp
(Equation 20)
0 = b1xe + b3x + b5θp
(Equation 21)
0 = c1xe + c3x + c5θp + P (1) ΔTr
Therefore, the determinant of 3 rows and 3 columns in the state quantities xe, x, and θp can be rewritten as Equation 22. However, xe_s, x_s, and θp_s are steady solutions of the state quantities xe, x, and θp. When A is a coefficient represented by a determinant of 3 rows and 3 columns such as Equation 23, the steady solutions xe_s, x_s, and θp_s are represented as Equation 24.

（数２３）

（数２４）

また、定常状態での前輪および後輪それぞれの静荷重Ｗｆ＿ｓ、Ｗｒ＿ｓは、図１０に示す振動モデルに基づいて、以下のように求められる。 (Equation 22)

(Equation 23)

(Equation 24)

Further, the static loads Wf_s and Wr_s of the front wheels and the rear wheels in the steady state are obtained as follows based on the vibration model shown in FIG.

  The vibration model in FIG. 10 shows physical quantities in consideration of the virtual center-of-gravity point movement amount ΔL with respect to the vibration model shown in FIG.

  The static loads Wf_s and Wr_s of the front wheels and the rear wheels in the steady state are expressed by Expressions 25 and 26, respectively. However, in these equations, Wfo and Wro indicate the static loads of the front wheels and the rear wheels, and ΔWf_s and ΔWr_s are total road surface reaction forces including the influence of all external factors such as braking force and running resistance. The static load change amount of the front wheel and the rear wheel when the driving force change amount ΔF acts is shown.

(Equation 25)
Wf_s = Wfo + ΔWf_s
(Equation 26)
Wr_s = Wro + ΔWr_s
When the distance between the center of gravity of the vehicle and the front wheel shaft at rest is Lfo, and the distance between the center of gravity of the vehicle and the rear wheel shaft is Lro, the stationary loads Wfo and Wro of the front and rear wheels are expressed by Equations 27 and 28. Indicated. Note that W corresponds to the gravity Mg applied to the vehicle weight M, that is, the total load applied to the front and rear wheels.

(Equation 27)
Wfo = (Lro / L) W
(Equation 28)
Wro = (Lfo / L) W
Further, the static load change amounts ΔWf_s and ΔWr_s of the front wheels and the rear wheels are obtained by multiplying the spring constants kf and kr of the front wheels or the rear wheels by the vertical displacement amounts x_s + Lfoθp_s and x_s−Lfoθp_s of the vehicle body on the front wheels or the rear wheels. Become. Since the steady solutions x_s and θp_s are expressed as in Expression 24, the static load change amounts ΔWf_s and ΔWr_s of the front wheels and the rear wheels are expressed as Expressions 29 and 30, respectively.

（数３０）

このため、前後輪静的接地荷重演算部２ｇにて、推定車軸トルク算出部２ｂにて演算された推定車軸トルクからその変化量ΔＴｒを求め、それを数式２９、３０に代入すれば、前輪および後輪の静的荷重変化量ΔＷｆ＿ｓ、ΔＷｒ＿ｓが算出される。そして、数式２５、２６におけるＷｆｏ、Ｗｒｏ、ΔＷｆ＿ｓ、ΔＷｒ＿ｓがそれぞれ数式２７〜３０で表されることから、それらを数式２９、３０に代入すれば、前輪および後輪の静止荷重Ｗｆｏ、Ｗｒｏを求めることができる。このようにして、前後輪における静的な接地荷重Ｗｆ＿ｓ、Ｗｒ＿ｓが求められる。 (Equation 29)

(Equation 30)

For this reason, if the change amount ΔTr is obtained from the estimated axle torque calculated by the estimated axle torque calculation unit 2b in the front and rear wheel static ground load calculation unit 2g and is substituted into the formulas 29 and 30, the front wheels and The rear wheel static load changes ΔWf_s and ΔWr_s are calculated. Since Wfo, Wro, ΔWf_s, and ΔWr_s in Formulas 25 and 26 are expressed by Formulas 27 to 30, respectively, by substituting them into Formulas 29 and 30, static loads Wfo and Wro of the front wheels and the rear wheels are obtained. be able to. In this way, the static ground loads Wf_s and Wr_s on the front and rear wheels are obtained.

  On the other hand, the amount of change in the front and rear wheel contact load due to the road gradient can be obtained based on the model shown in FIG.

  FIG. 11 shows the relationship between physical quantities when the vehicle is located at a predetermined road gradient θ. As shown in this figure, assuming that the height of the center of gravity of the vehicle from the road surface is the center of gravity height hcg and the vehicle body plane on which the center of gravity is located is Bp, the position where the front and rear wheel loads are applied (front and rear wheel ground contact position) ) From the intersection of the line extending in the vertical direction and the vehicle body plane Bp to the center of gravity is Lf + hcgtanθ and Lf−hcgtanθ.

  Further, assuming that the intersections when straight lines are drawn in the vertical direction from the center of gravity of the vehicle and the front and rear wheel ground contact positions with respect to an arbitrary horizontal line H are S1, S2, and S3, the distances La = Lcos θ, S1− between S1 and S3. The distance between S2 is Lf ′ = (Lf + hcgtanθ) cosθ, and the distance between S2 and S3 is Lr ′ = (Lr−hcgtanθ) cosθ.

  In this case, assuming that the ground contact loads in the vertical direction applied to the front and rear wheels are Wf ′ and Wr ′, these are expressed in the same manner as in Expressions 27 and 28.

(Equation 31)
Wf ′ = (Lr ′ / L ′) W
= {(Lr-hcgtanθ) cosθ / (Lcosθ)} W
= {(Lr / L- (hcg / L) tan θ) W
= Wfo-W (hcg / L) tanθ
(Expression 32)
Wr ′ = (Lf ′ / L ′) W
= {(Lf + hcgtanθ) cosθ / (Lcosθ)} W
= {(Lf / L + (hcg / L) tan θ) W
= Wro + W (hcg / L) tanθ
The ground load applied to the front and rear wheels in the vertical direction of the road surface corresponds to a vertical component of the road surface with respect to the vertical ground loads Wf ′ and Wr ′, and becomes Wf ′ cos θ and Wr ′ cos θ, respectively. , Wr ′ is expressed by Equations 31 and 32, the following is obtained.

(Expression 33)
Wf ′ cos θ = {Wfo−W (hcg / L) tan θ} cos θ
= Wfocosθ-W (hcg) sinθ
(Equation 34)
Wr ′ cos θ = {Wro + W (hcg / L) tan θ} cos θ
= Wrocos θ + W (hcg) sin θ
In this manner, the amount of change in the front and rear wheel ground load due to the road gradient is obtained. The change amount of the front and rear wheel installation load is a load amount that constantly changes during traveling on the road gradient. Accordingly, when the road gradient is obtained by the road gradient calculation unit 2d, the amount of change in the front and rear wheel grounding load at the time of the road gradient is obtained, and the front and rear wheel grounding due to the road gradient is determined from the static grounding load of each of the front and rear wheels. By subtracting the amount of change in the load, it is possible to obtain the front and rear wheel ground loads according to the road gradient.

  For this reason, LfKcf−LrKcr obtained by a target value calculation unit 2i described later is obtained as an ideal turning radius stability factor considering the road gradient. As described above, the front and rear wheel ground load is obtained by the front and rear wheel static ground load calculation unit 2g.

  Further, the virtual turning radius estimation unit 2h in FIG. 2 includes information on the front wheel speed calculated by the front wheel speed calculation unit 2c and the yaw angle (sensor value) sent from the G / yaw rate / gyro sensor 8 and visual information. Based on the sense amount τ sent from the space model processing unit 2k, a virtual turning radius suitable for the vehicle to travel is estimated. Specifically, first, a yaw angle correction value that matches the driver's sense, that is, keeps the visual stimulus change constant is calculated from the sense amount τ. Next, assuming that the sum of the yaw angle correction value and the yaw angle (sensor value) is a value of γsensor [rad / s], the turning radius ρ is expressed by the following equation from the yaw angle γsensor and the vehicle speed V: Shown in

(Equation 35)
ρ = V / γsensor
Therefore, the virtual turning radius estimation unit 2h obtains the vehicle speed V by, for example, obtaining the average value of both front wheel speeds from the front wheel speed, and substitutes the vehicle speed V and the yaw angle γsensor into the formula 35 to obtain the virtual speed. The turning radius is calculated.

  The target value calculation unit 2i determines a target value of LfKcf−LrKcr that determines the stability factor. Hereinafter, the target value of LfKcf−LrKcr will be described.

  At the time of steady circle turning, the slip angle β and the yaw rate γ are defined as follows. These will be described with reference to FIG.

  FIG. 12 schematically shows the size of each part of the vehicle. As shown in this figure, the vehicle body mass [kg] on the spring is M, the wheel base [m] is L, the distance [m] between the vehicle center of gravity and the front wheel axle is Lf, and the vehicle center of gravity and the rear wheel axle is The distance [m] between them is Lr. The vehicle speed is V [m / sec], the steering angle is δ [rad], and the tire cornering powers of the front and rear wheels are Kcf and Kcr [N / rad], respectively. Based on these, it is known that the turning radius ρ is expressed by the following equation.

なお、数式３６に示されるＡが上述したスタビリティファクタ（Ｓｔａｂｉｌｉｔｙ Ｆａｃｔｏｒ）であり、次式で示され、車両ごとに決まる値である。 (Equation 36)

In addition, A shown in Formula 36 is the above-described stability factor, which is expressed by the following formula and is determined for each vehicle.

この数式３７におけるスタビリティファクタ、具体的にはスタビリティファクタにおけるＬｆＫｃｆ−ＬｒＫｃｒの項の正負によってステアリング特性が確定する。このステアリング特性は、以下のように示される。 (Equation 37)

The steering characteristic is determined by the stability factor in Formula 37, specifically, the sign of LfKcf-LrKcr in the stability factor. This steering characteristic is shown as follows.

すなわち、ＬｆＫｃｆ−ＬｒＫｃｒの項をΔとした場合、図１３（ａ）の車両旋回状態を示す図に示されるように、円旋回が定常旋回（理想旋回）であると考えると、Δが負であればアンダーステア、０であれば定常旋回、Δが正であればオーバステアというステアリング特性となる。換言すれば、図１３（ｂ）に示す車速と旋回半径との関係図に示されるように、Δが負であれば車速に対して旋回半径が大きくなる状態（アンダーステア）、０であれば定常旋回、Δが正であれば車速に対して旋回半径が小さくなる状態（オーバステア）となる。 (Equation 38)

That is, assuming that the term of LfKcf−LrKcr is Δ, Δ is negative when the circular turn is considered to be a steady turn (ideal turn) as shown in the vehicle turning state of FIG. If it is, the steering characteristic is understeer, if it is 0, it is a steady turn, and if Δ is positive, it is a steering characteristic. In other words, as shown in the relationship diagram between the vehicle speed and the turning radius shown in FIG. 13B, when Δ is negative, the turning radius is larger than the vehicle speed (understeer), and when it is 0, it is steady. If the turn and Δ are positive, the turning radius becomes smaller with respect to the vehicle speed (oversteer).

  Therefore, depending on the setting of the term LfKcf−LrKcr in the stability factor, it becomes possible to allow the vehicle to make an ideal turn and to improve the steering characteristics.

  Therefore, the target value calculation unit 2i calculates the target value of LfKcf−LrKcr by performing a reverse calculation from Expression 36. Specifically, LfKcf−LrKcr is obtained from the following equation.

(Equation 39)
LfKcf−LrKcr = Kcf_sKcr_s (2L2 / MV2) · (1−ρδ / L)
Here, Kcf_s and Kcr_s are static cornering powers of the front and rear wheels considering only the road gradient and load movement due to acceleration / deceleration, δ is a steering angle, and ρ is a virtual turning radius.

  Among these, the independent variables are the vehicle speed V, the steering angle δ, and the virtual turning radius ρ. The vehicle speed V is obtained by the above-described method from the front wheel speed calculated by the front wheel speed calculator 2c, and the calculation result of the steering angle calculator 2e is used for the steering angle. As the virtual turning radius, the one obtained from Expression 35 is used. In this manner, the target value of LfKcf−LrKcr is obtained by the target value calculation unit 2i.

  Then, if the target value of LfKcf-LrKcr is calculated based on the target value of LfKcf-LrKcr obtained by the target value calculator 2i, the vibration suppression correction control unit 2j is the actual LfKcf− The amount of change from the target value of the stability factor is obtained from the difference from LrKcr, and the basic required torque calculated by the basic required torque calculation unit 2a is corrected so as to eliminate the change.

  Specifically, as shown in FIG. 2, the difference between the target value of the stability factor and the current actual stability factor is obtained by subtracting the output y obtained from the pitching vibration model from the target value. The value is multiplied by 1 / s and Ki. This is because the feedback of the servo system can be performed by setting the target value as an integral system, and even if the target value is not 0, the target value can be followed. Further, a running resistance disturbance is added to this, and each state quantity x (x, x ′, xe, xe ′, θp, θp ′) calculated using the pitching vibration model is obtained by a control system design method. The correction value of the drive torque can be obtained by subtracting the value multiplied by the predetermined state feedback gain Ks (K1 to K6).

  Here, Δ (= LfKcf−LrKcr) is as follows when the tire cornering powers Kcf and Kcr of the front and rear wheels depend on the front wheel and rear wheel ground loads Wf and Wr. Can be expressed as

(Equation 40)
Δ = LfKcf−LrKcr
= Cw (LfWf-LrWr)
As a result, the variation in Δ (= LfKcf−LrKcr) accompanying the variation in the cornering powers Wf and Wr at the front wheels and the rear wheels, that is, the variation in the steering characteristics is expressed by the following equation from Equations 25, 26, and 40. .

(Equation 41)
LfKcf-LrKcr = Cw (LfWf-FrWr)
= CwLf (Wfo + ΔWf) −CwLr (Wro + ΔWr)
= -Cw (KfLf-KrLr) x-Cw (CfLf-CrLr) x '
−Cw (KfLf2 + KrLr2) θp−Cw (CfLf2−CrLr2) θp ′
Therefore, Δ (= LfKcf−LrKcr), which is the governing parameter of the steering characteristic, is expressed by a linear combination of state quantities as in Equation 42 based on Equations 11 to 16, and Equation 41 is expressed in a six-dimensional manner. When expressed as a determinant, Equation 43 is obtained.

ただし、ｑ１〜ｑ４は係数であり、それぞれ、ｑ１＝−Ｃｗ（ＫｆＬｆ−ＫｒＬｒ）、ｑ２＝−Ｃｗ（ＣｆＬｆ−ＣｒＬｒ）、ｑ３＝−Ｃｗ（ＫｆＬｆ２＋ＫｒＬｒ２）、ｑ４＝−Ｃｗ（ＣｆＬｆ２−ＣｒＬｒ２）である。 (Equation 42)
LfKcf-LrKcr = q1x3 + q2x4 + q3x5 + q4x6
(Equation 43)

However, q1 to q4 are coefficients, and q1 = −Cw (KfLf−KrLr), q2 = −Cw (CfLf−CrLr), q3 = −Cw (KfLf2 + KrLr2), q4 = −Cw (CfLf2−CrLr2), respectively. is there.

  This is the output equation shown in FIG. 2. When this equation 42 is simplified, the sprung y = Cx + Du (however, since D = 0, it is substantially y = Cx). The output equation of the vibration model is derived.

  Therefore, by substituting each state quantity x, x ′, xe, xe ′, θp, θp ′ for this output equation, the difference between the target value and the actual stability factor is calculated. It is possible to perform target tracking control based on feedback of the servo system state in which feedback control is performed based on the feedback control.

  The axle torque thus determined is corrected in consideration of the estimated running resistance disturbance and the state quantity x so as to follow the target value of LfKcf−LrKcr. And since the correction value of this axle torque is an absolute value, the relative value of axle torque is calculated | required by subtracting it from the calculation result in the estimated axle torque calculation part 2b. That is, when the correction value of the axle torque is obtained using the pitching vibration model, the corrected axle torque calculated from the target value may not be accurate depending on the accuracy of the pitching vibration model. For this reason, in order to set the steady-state deviation to 0, the correction value is relativized by taking the difference between the estimated axle torque currently assumed to be actually generated and the correction value of the axle torque.

  Then, after the relative correction value is multiplied by the reduction ratio (differential ratio: 1 / Rd) in the final reduction gear, the value is subtracted from the basic required torque calculated by the basic required torque calculation unit 2a. . Thereby, the correction value of the axle torque is obtained as an absolute value with respect to the basic required torque, and this value is set as the corrected required torque.

  In this way, the axle torque to be generated in the engine 1 is determined, and the intake air amount and fuel injection amount of the engine 1 are adjusted so that the corrected required torque can be obtained, and the energy corresponding thereto is output. . This energy is transmitted as rotational energy to the drive wheels 15a and 15b via the transmission (T / M) 12 and the final reduction gear 13, and the drive wheels 15a and 15b generate the axle torque according to the request.

  As described above, according to the vehicle stabilization control system shown in the present embodiment, in view of the estimated running resistance disturbance and the state quantity x so as to follow the target value of LfKcf-LrKcr that is a stability factor. Thus, the axle torque is corrected. For this reason, it is possible to stabilize the stability factor so that it becomes an ideal turning radius from time to time in response to various surrounding environments that change from moment to moment, for example, pitching vibration energy is suppressed, etc. Each state quantity inside the vehicle is stabilized, and the traveling state of the vehicle can be stabilized.

  With the above configuration, when the control according to the present embodiment is performed, the amplitude of the state quantity is small as compared with the case where the control is not performed, and stabilization is possible at an early stage.

  Therefore, it is possible to suppress the fluctuation of the vehicle behavior due to the road surface and other external factors (road surface disturbance) from disturbing the vehicle body posture and the traveling track. In addition, the vehicle body posture disturbance and vibration caused by these factors are suppressed, so the driver's steering operation to correct the vehicle body posture disturbance and vibration becomes unnecessary, and the driver's steering operation induces disturbance components. Can be prevented.

  Therefore, it is possible to suppress the influence of driver operation disturbance and road surface disturbance, and it is possible to stabilize the vehicle body posture and vehicle characteristics.

  Furthermore, this vehicle stabilization control system uses an image projected on the retina of a driver obtained by converting environmental information expressed in a geometric coordinate system based on a human visual stimulus model. Using this image, the driver's sensation amount τ is obtained, and this sensation amount τ is used to calculate a virtual turning radius for calculating a stability factor for suppressing the influence of driver operation disturbance and road surface disturbance. Therefore, it is possible to achieve stabilization of vehicle characteristics that more closely match the driving feeling of the driver, and to reduce the driving burden on the driver.

  Conventionally, in Japanese Patent Application Laid-Open No. 5-26067, a drive wheel control device that obtains a yaw rate in a vehicle, detects that the vehicle is in an oversteer state or an understeer state from the yaw rate, and controls engine output to prevent it. Has been proposed. Also in the apparatus of this publication, it is possible to suppress the vehicle from being oversteered or understeered by adjusting the engine output. However, the yaw rate detects a rotational component about the vehicle vertical axis and does not detect a pitching vibration component that is a rotation around the center of gravity of the vehicle about the vehicle left-right axis. Even if the engine output is corrected, correction that suppresses the pitching vibration component cannot be performed. Therefore, by suppressing the pitching vibration component as shown in the present embodiment, it is not possible to stabilize the fluctuations in the front and rear wheel loads and thereby to stabilize the cornering power.

[Other Embodiments]
In the above-described embodiment, the vehicle position is calculated using the outputs of the G / yaw rate sensor / gyro sensor 8, the GPS receiver 17, and the road shape information providing means 16, but the vehicle position calculation method has these configurations. It is not limited. For example, it is possible to estimate the behavior using the map information of the road shape information providing means 16 and the GPS reception signal without using the internal sensors such as the G / yaw rate sensor / gyro sensor 8. Further, it is possible to estimate the behavior using the map information and the internal sensor without using the GPS receiver 17. Further, road-to-vehicle communication or camera images may be used.

It is a figure which shows schematic structure of the vehicle stabilization control system used in Example 1. FIG. 1 is a block diagram showing an outline of an engine ECU 2 used in Embodiment 1. FIG. 3 is a block diagram of a visual space model processing unit used in Example 1. FIG. It is a figure showing the installation place of the virtual signboard used in Example 1. FIG. It is a figure showing the visual model of the human central vision used in Example 1. FIG. It is a figure showing the virtual signboard projected on the retina of the driver used in Example 1. FIG. It is the figure which showed the nose dive at the time of the driving | running (acceleration) by the pitching vibration used in Example 1, the time of braking (deceleration), and the steering (turning). 3 is a timing chart showing the relationship between pitching vibration, front and rear wheel ground loads, and front and rear wheel cornering power used in Example 1; 3 is a schematic diagram of a sprung vibration model used in Example 1. FIG. It is a figure of the vibration model which showed the static load of each front wheel and rear wheel in the steady state used in Example 1. It is the figure which showed the variation | change_quantity of the front-and-rear wheel grounding load by the road gradient used in Example 1. FIG. FIG. 3 is a schematic diagram showing the size of each part of the vehicle used in the first embodiment. FIG. 13A is a diagram used in the first embodiment, FIG. 13A is a diagram illustrating a route when the vehicle is turning, and FIG. 13B is a diagram illustrating the vehicle speed when the vehicle is turning. It is the correlation figure which showed the relationship with a turning radius. It is a figure showing the relationship when a vertical axis | shaft is a human stimulus and a horizontal axis is distance.

Explanation of symbols

1 Engine 2 Engine ECU
2a Basic required torque calculation unit 2b Estimated axle torque calculation unit 2c Front wheel speed calculation unit 2d Estimated road gradient calculation unit 2e Steering angle calculation unit 2f Front wheel running resistance disturbance estimation unit 2g Front and rear wheel static contact load calculation unit 2h Virtual turning radius estimation Calculation unit 2i Target value calculation unit 2j Vibration suppression correction control unit 2k Visual space model processing unit 3 Acceleration stroke sensor 4a, 4b Wheel speed sensor for front wheels 5 Steering angle sensor 6 Intake air amount sensor 7 Engine rotation sensor 8 G / yaw rate / gyro Sensor 10a, 10b Front wheel 11 Main body 12 Transmission 13 Final reduction device 14 Drive shaft 15a, 15b Rear wheel 16 Road shape information providing means 17 GPS receiver 31 Current position calculation unit 32 Virtual signboard generation unit 33 Geometric coordinate system / retinal coordinate system Coordinate converter 34 Visual sense amount calculator 41 Left virtual signboard S ( ) L
42 Right virtual signboard S (1) R
43 Left virtual signboard S (2) L
44 Right virtual signboard S (2) R
45 Left virtual signboard S (3) L
46 Right virtual signboard S (3) R

Claims (11)

Environmental information providing means (16) for providing road shape information as environmental information outside the vehicle;
Current position calculation means (31) for calculating the current position of the vehicle as a coordinate value of a geometric coordinate system;
Vehicle speed detection means (4a, 4b) for detecting the vehicle speed of the vehicle;
Using the road shape information, the current position, and the vehicle speed , a virtual object having a predetermined area in a road or near a road that is scheduled to pass after a predetermined time, or on the roadside. Is perpendicular to the road plane and is perpendicular to the road plane, and the area of the object is the area of the object when the entire road including the object is projected onto the driver's retina. A conversion means (33) for outputting the converted area converted into, and also for outputting the velocity vector of the object;
Wherein said conversion area outputted from the conversion means (33) and determine the variation of the conversion area from said velocity vector, and calculates the sense of a stimulation felt by a driver from the change in conversion area to the conversion area Sensory amount calculation means (34) for outputting;
A vehicle stabilization control system comprising control command setting means (2h, 2i, 2j) for determining a control command based on the sense amount. A basic required driving force calculation unit (2a) for calculating a physical quantity corresponding to the basic required driving force in order to generate a basic required driving force required by the driver for the driving wheels provided in the vehicle;
Front and rear wheel load calculator (2g) for detecting a load applied to each of the front and rear wheels in the vehicle;
Environmental information providing means (16) for providing road shape information as environmental information outside the vehicle;
Current position calculation means (31) for calculating the current position of the vehicle as a coordinate value of a geometric coordinate system;
Vehicle speed detection means (4a, 4b) for detecting the vehicle speed of the vehicle;
Using the road shape information, the current position, and the vehicle speed , a virtual object having a predetermined area in a road or near a road that is scheduled to pass after a predetermined time, or on the roadside. Is perpendicular to the road extension direction and is perpendicular to the road plane, the area of the object is the object when the entire road including the object is projected onto the driver's retina. A conversion means (33) for outputting the converted area converted into the area of the object and outputting the velocity vector of the object;
Wherein said conversion area outputted from the conversion means (33) and determine the variation of the conversion area from said velocity vector, and calculates the sense of a stimulation felt by a driver from the change in conversion area to the conversion area Sensory amount calculation means (34) for outputting;
A virtual turning radius estimation unit (2h) that estimates a virtual turning radius in the vehicle based on the sense amount, so that the sense amount is constant;
A target value calculation unit (2i) for calculating a target value of the stability factor based on the calculation results in the front and rear wheel load calculation unit (2g) and the virtual turning radius estimation unit (2h);
Vibration suppression correction control for correcting a physical quantity corresponding to the basic required driving force calculated by the basic required driving force calculating unit (2a) so as to follow the target value calculated by the target value calculating unit (2i). Vehicle stabilization control, characterized in that a driving force corresponding to the corrected physical quantity corrected by the vibration suppression correction control unit (2j) is generated in the driving wheel. system. The vehicle stabilization control system according to claim 2 , wherein the basic required driving force calculation unit (2a) calculates a basic required torque as a physical quantity corresponding to the basic required driving force. The front and rear wheel ground load calculation unit (2g) calculates, as the front and rear wheel ground load, front and rear wheel static ground loads applied to the front and rear wheels when the vehicle is in a steady running state,
The vehicle according to claim 2 or 3 , wherein the target value calculation unit (2i) calculates the target value of the stability factor based on the front and rear wheel static ground loads. Stabilization control system. An estimated road gradient calculation unit (2d) for estimating a gradient of a road surface on which the vehicle is traveling;
The front / rear wheel ground load calculating unit (2g) calculates, as the front / rear wheel ground load, front / rear wheel static ground loads applied to the front wheels and the rear wheels when the vehicle is in a steady running state, and the estimated road Based on the estimated road gradient calculated by the gradient calculation unit (2d), the fluctuation amount of the front and rear wheel ground load when the vehicle travels on the estimated road gradient is calculated, and the front and rear wheel static ground load is calculated. The value considering the fluctuation due to the estimated road gradient is obtained as the front and rear wheel ground load,
The target value calculation unit (2i) calculates the target value of the stability factor based on the front and rear wheel ground contact load that allows for the fluctuation of the front and rear wheel installation load due to the estimated road gradient. The vehicle stabilization control system according to claim 2 or 3 , wherein The vibration suppression correction control unit (2j) has a state equation indicating a state quantity in the vehicle based on a sprung vibration model in the vehicle, and determines the stability factor based on the state equation. An output equation expressed by a state quantity, and a stability factor obtained from the output equation and the state quantity, and the target value of the stability factor calculated by the target value calculation unit (2i). The vehicle stabilization control system according to any one of claims 2 to 5 , wherein the physical quantity corresponding to the basic required driving force is corrected based on the difference. A running resistance disturbance estimation unit (2f) for estimating running resistance disturbance applied to the wheels in the vehicle;
The vibration suppression correction control unit (2j) obtains the state quantity in the state equation in view of the running resistance disturbance estimated by the running resistance disturbance estimation unit (2f), and obtains the obtained state quantity and the output equation. 7. The vehicle stabilization control system according to claim 6 , wherein the stability factor is obtained based on the difference and the difference between the stability factor and the target value is obtained. The running resistance disturbance estimation unit (2f) obtains the running resistance disturbance of the front wheels as the running resistance disturbance based on the differential value of the wheel speed of the front wheel provided in the vehicle and the weight of the vehicle. The vehicle stabilization control system according to claim 7 , wherein The conversion means assumes that the virtual objects are standing on both sides of the road at or near the point that is scheduled to pass after a specified time, and the conversion area of each virtual object, The vehicle stabilization control system according to claim 1 or 2, wherein a velocity vector of the objects is output . A plurality of the sensory quantities are calculated using a plurality of virtual objects assumed to be standing at points corresponding to a plurality of different specified times , and the determination of the control command is a plurality of the sensory senses. The vehicle stabilization control system according to any one of claims 1, 2, and 9, wherein the vehicle stabilization control system is based on a quantity. The vehicle stabilization control system according to any one of claims 1, 2, 9, and 10, wherein the virtual object is newly installed every fixed control period .

Images