JP2010143379A - Vehicle attitude angle estimation device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To estimate an attitude angle with high precision even in the case of a road surface the vertical acceleration of which changes. <P>SOLUTION: An acceleration state quantity saturating characteristic correction means 30 corrects vertical acceleration state quantity obtained by using the estimation value of a roll angle and the estimation value of a pitch angle so that the vertical acceleration state quantity is provided with saturating characteristics. A state quantity deviation low pass filter value calculation means 26 calculates the deviation of longitudinal acceleration state quantity, the deviation of lateral acceleration state quantity, and the deviation of vertical acceleration state quantity to which low pass filter processing has been performed based on: the low pass filter processing values of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity, calculated by an attitude angle observer; and the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained by using a sensor signal and the estimation value of the longitudinal body speed and the estimation value of the lateral body speed. An attitude angle estimation means 28 estimates a roll angle and a pitch angle by feeding back the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a vehicle attitude angle estimation device and program, and in particular, a vehicle attitude for estimating a roll angle and a pitch angle, which are vehicle attitude angles with respect to a vertical axis as a vehicle physical quantity, by an observer using an estimated value of a pitch angular velocity. The present invention relates to an angle estimation device and a program.

  The relationship between the acceleration and angular velocity of the rigid body motion and the posture angle of the vehicle described by the Euler angle is known in Non-Patent Document 1, etc., and by performing an integral operation based on this relationship, the posture angle of the vehicle is determined. Can be derived. However, a simple integration operation accumulates sensor drift errors, and an accurate posture angle cannot be derived. In order to avoid such accumulation of errors in the integration calculation, it is necessary to configure an observer that corrects the output so that it substantially matches the output of other sensor signals.

Also, as a conventional technique for estimating the roll angle with respect to the vertical axis of the vehicle body, the roll angle is estimated from the relationship between the lateral component of the vehicle motion in the low-frequency region that is not affected by spin or the like and the effect of the vehicle attitude angle. A technique is known (Patent Document 1).
Introduction to aircraft mechanics (p12, Kato, Oya, Karasawa: University of Tokyo Press, 1982) JP 2008-94375 A

  However, in the technique described in Patent Document 1 described above, the posture angle observer uses a condition that “the vertical acceleration with a long time constant is equal to the gravitational acceleration”. If the vertical acceleration increases at the lowest point of the road gradient that changes from downhill to uphill, the condition is set assuming that the filter can remove short-term vertical acceleration changes due to The filter processing cannot sufficiently remove the influence due to the change in vertical acceleration, and thus there is a problem that the estimation accuracy is deteriorated.

  The present invention has been made to solve the above problems, and provides a vehicle attitude angle estimation device and program capable of accurately estimating an attitude angle even on a road surface in which vertical acceleration changes. For the purpose.

  In order to achieve the above object, a vehicle attitude angle estimation device according to a first aspect of the present invention includes a longitudinal speed estimation means for estimating a longitudinal vehicle body speed, which is a vehicle body speed in the vehicle longitudinal direction, based on the wheel speed of each wheel; The lateral velocity estimation means for estimating the lateral vehicle body speed, the pitch angular velocity estimation means for estimating the pitch angular velocity, and the detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of the vehicle motion Attitude angle estimation means for estimating a roll angle and a pitch angle, which are attitude angles with respect to a vertical axis of the vehicle body, based on a sensor signal corresponding to the vehicle body, an estimated value of the front and rear vehicle body speed, and an estimated value of the pitch angular velocity. A vehicle attitude angle estimation device, wherein the attitude angle estimation means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity obtained using the roll angle estimation value and the pitch angle estimation value, and Correction means for correcting at least one of the linear acceleration state quantities so as to have saturation characteristics, the longitudinal acceleration state quantity, the lateral acceleration state quantity, the vertical acceleration state quantity, the sensor signal, and the front and rear vehicle bodies Based on the estimated value of speed, the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained by using the estimated value of the lateral vehicle body speed, the deviation of the longitudinal acceleration state quantity, the lateral acceleration state, which has been subjected to low-pass filter processing A state quantity deviation calculating means for calculating a quantity deviation and a vertical acceleration state quantity deviation, and feeding back the calculated longitudinal acceleration state quantity deviation, lateral acceleration state quantity deviation, and vertical acceleration state quantity deviation. Thus, the roll angle and pitch angle are estimated.

  According to a second aspect of the present invention, there is provided a program for estimating a longitudinal vehicle body speed that is a vehicle body speed in the vehicle longitudinal direction based on a wheel speed of each wheel, and a lateral vehicle body speed that is a vehicle body speed in the vehicle lateral direction. Lateral velocity estimating means for estimating, pitch angular velocity estimating means for estimating pitch angular velocity, and sensor signals corresponding to detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of vehicle motion, and the longitudinal vehicle body velocity And a posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to a vertical axis of a vehicle body, based on the estimated value of the pitch angular velocity and the estimated value of the pitch angular velocity, The means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity, and a vertical acceleration state obtained using the roll angle estimation value and the pitch angle estimation value. Correction means for correcting at least one of the characteristics to have saturation characteristics, the longitudinal acceleration state quantity, the lateral acceleration state quantity, the vertical acceleration state quantity, the sensor signal, and an estimated value of the longitudinal vehicle body speed And the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained using the estimated value of the lateral vehicle body speed, the low-pass filter processing is performed, the longitudinal acceleration state quantity deviation, the lateral acceleration state quantity deviation, And a state quantity deviation calculating means for calculating a deviation of the vertical acceleration state quantity, and feeding back the calculated longitudinal acceleration state quantity deviation, lateral acceleration state quantity deviation, and vertical acceleration state quantity deviation, The roll angle and pitch angle are estimated.

  In the first invention and the second invention, at least one of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity obtained by using the roll angle estimation value and the pitch angle estimation value is obtained as a saturation characteristic. Correct so that Also, the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained by using the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity, the sensor signal, the estimated value of the longitudinal vehicle body speed, and the estimated value of the lateral vehicle body speed. Based on the above, the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to the low-pass filter processing are calculated. Then, the roll angle and the pitch angle are estimated by feeding back the calculated deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity.

  In the first invention and the second invention, at least one of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity obtained by using the roll angle estimation value and the pitch angle estimation value is obtained as a saturation characteristic. The deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity, which have been subjected to low-pass filter processing, are calculated and fed back, so that the vertical direction Even on a road surface where the acceleration changes, the posture angle can be accurately estimated.

  According to a third aspect of the present invention, there is provided a vehicle attitude angle estimation device, a longitudinal speed estimation means for estimating a longitudinal vehicle body speed, which is a vehicle body speed in the vehicle longitudinal direction, based on a wheel speed of each wheel, and a lateral vehicle body speed, in a lateral direction of the vehicle. Lateral velocity estimation means for estimating the vehicle body speed, pitch angular velocity estimation means for estimating the pitch angular velocity, and sensor signals corresponding to the detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of the vehicle motion, A posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to a vertical axis of the vehicle body, based on the estimated value of the front and rear vehicle body speeds and the estimated value of the pitch angular velocity; The posture angle estimating means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity, and a vertical acceleration state quantity obtained using the roll angle estimation value and the pitch angle estimation value, and Based on the sensor signal, the estimated value of the longitudinal vehicle body speed, and the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained using the estimated value of the lateral vehicle body speed, the deviation of the longitudinal acceleration state quantity, the lateral acceleration state quantity And a vertical acceleration state quantity deviation, and at least one of the longitudinal acceleration state quantity deviation, the lateral acceleration state quantity deviation, and the vertical acceleration state quantity deviation has saturation characteristics. And a low-pass filter processing unit that performs a low-pass filter process on the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity, Feedback of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to low-pass filter processing, Serial is obtained so as to estimate the roll angle and the pitch angle.

  According to a fourth aspect of the present invention, there is provided a program for estimating a longitudinal vehicle body speed that is a vehicle body speed in a vehicle longitudinal direction based on a wheel speed of each wheel, and a vehicle body speed that is a vehicle body speed in a vehicle lateral direction. Lateral velocity estimating means for estimating, pitch angular velocity estimating means for estimating pitch angular velocity, and sensor signals corresponding to detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of vehicle motion, and the longitudinal vehicle body velocity And a posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to a vertical axis of a vehicle body, based on the estimated value of the pitch angular velocity and the estimated value of the pitch angular velocity, The means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity, and a vertical acceleration state obtained using the roll angle estimation value and the pitch angle estimation value. And a longitudinal acceleration state quantity deviation and a lateral acceleration based on the sensor signal, the estimated value of the longitudinal vehicle body speed, and the longitudinal acceleration state quantity and the lateral acceleration state quantity obtained using the estimated value of the lateral vehicle body speed. The deviation of the state quantity and the deviation of the vertical acceleration state quantity are calculated, and at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity is calculated with a saturation characteristic. Correction means for correcting to have, and low-pass filter processing means for performing low-pass filter processing on the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity. The feedback of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to the low-pass filter processing is performed. Te, in which so as to estimate the roll angle and the pitch angle.

  In the third and fourth aspects of the invention, the longitudinal acceleration state quantity, the lateral acceleration state quantity, the vertical acceleration state quantity, the sensor signal, and the longitudinal vehicle body speed obtained using the estimated roll angle value and the estimated pitch angle value. Of the longitudinal acceleration state quantity, the lateral acceleration state quantity deviation, and the vertical acceleration state quantity of the vertical acceleration state quantity, Calculate the deviation. Then, at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity is corrected so as to have saturation characteristics, and the deviation of the longitudinal acceleration state quantity and the lateral acceleration state quantity are corrected. The low-pass filter process is performed on the deviation and the vertical acceleration state quantity deviation. Then, the roll angle and the pitch angle are estimated by feeding back the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to the low-pass filter processing.

  In the third and fourth inventions, at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity is corrected so as to have a saturation characteristic, and the low-pass Since the filter processing is performed and the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are calculated and fed back, even on the road surface where the vertical acceleration changes, The posture angle can be estimated with high accuracy.

  The vehicle posture angle estimation device may include a sensor that detects each of the longitudinal acceleration, lateral acceleration, vertical acceleration, yaw angular velocity, and roll angular velocity of the vehicle motion and outputs a sensor signal corresponding to the detected value. .

  Further, the pitch angular velocity includes an estimated value of the longitudinal vehicle body speed, an estimated value of the lateral vehicle body speed, a detected value of vertical acceleration, a detected value of the roll angular velocity, a previous estimated value of the roll angle estimated by the attitude angle estimating means, and an attitude angle. The pitch angle estimated by the estimating means can be estimated based on the previous estimated value. As a result, the pitch angular velocity is estimated from the balance of the vertical movement of the vehicle body, and the roll angle and the pitch angle, which are posture angles with respect to the vertical axis of the vehicle body, are estimated using the estimated value of the pitch angular velocity. The roll angle and pitch angle can be estimated with high accuracy even when the bank is running.

  Further, the program of each invention described above can be provided by being stored in a storage medium.

  As described above, according to the present invention, at least one of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity obtained by using the roll angle estimation value and the pitch angle estimation value is determined as a saturation characteristic. Or a low-pass filter by correcting at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity so as to have a saturation characteristic. Since the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are processed and fed back, the accuracy is obtained even on the road surface where the vertical acceleration changes. The effect that the posture angle can be estimated well is obtained.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, the present invention is applied to a vehicle attitude angle estimation device that estimates a pitch angle and a roll angle, which are attitude angles of a vehicle with respect to a vertical axis.

  As shown in FIG. 1, the vehicle attitude angle estimation device of the first embodiment includes a lateral speed estimation means 10 that estimates a lateral vehicle body speed V that is a vehicle body speed in the lateral direction of the vehicle, and a wheel speed of each wheel. A longitudinal speed estimation means 12 for estimating a longitudinal vehicle body speed U, which is a vehicle body speed in the longitudinal direction of the vehicle, is provided.

  The lateral vehicle body speed V can be estimated using a Kalman filter or the like, or can be estimated from the detection value of the lateral acceleration sensor.

  The wheel speed of each wheel can be detected by a wheel speed sensor provided corresponding to each wheel, and the front and rear vehicle body speed U is determined from the wheel speed of each wheel or the wheel speed of each wheel and the differential value of the wheel speed. Can be estimated. For example, the maximum value of the wheel speeds of four wheels can be output as the front and rear vehicle body speed U during braking, and the average value of the wheel speeds of the driven wheels can be output as the front and rear vehicle body speed U during driving.

  In addition, the vehicle attitude angle estimation device according to the present embodiment includes a vertical acceleration sensor 14, a lateral acceleration sensor 16, and a vertical acceleration sensor 16 that detect vertical acceleration Gz, lateral acceleration Gy, and longitudinal acceleration Gx, which are xyz 3-axis accelerations of vehicle motion. A longitudinal acceleration sensor 18 is also provided. As shown in FIG. 2, the x axis corresponds to the vehicle longitudinal direction, the y axis corresponds to the vehicle width direction (lateral direction), and the z axis corresponds to the vehicle vertical direction.

  Further, the vehicle attitude angle estimation device of the present embodiment is provided with a roll angular velocity sensor 20 that detects the roll angular velocity P and a yaw angular velocity sensor 22 that detects the yaw angular velocity R.

  The lateral speed estimation means 10, the longitudinal speed estimation means 12, the vertical acceleration sensor 14, and the roll angular speed sensor 20 are connected to a pitch angular speed estimation means 24 that estimates the pitch angular speed Q.

  The lateral velocity estimation means 10, the longitudinal velocity estimation means 12, the lateral acceleration sensor 16, the longitudinal acceleration sensor 18, and the yaw angular velocity sensor 22 are subjected to a low-pass filter process for passing only a motion to be noticed. It is connected to state quantity deviation low-pass filter value calculation means 26 for calculating the low-pass filter processing values of the deviation, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity.

  The state quantity deviation low-pass filter value calculation means 26 is connected to a posture angle estimation means 28 for estimating a roll angle φ and a pitch angle θ, which are posture angles with respect to the vertical axis of the vehicle body. The posture angle estimating means 28 is further connected with a vertical acceleration sensor 14, a lateral acceleration sensor 16, a longitudinal acceleration sensor 18, a roll angular velocity sensor 20, a yaw angular velocity sensor 22, and a pitch angular velocity estimating means 24.

  Further, the posture angle estimating means 28 further includes acceleration state quantity saturation characteristic correcting means 30 for calculating a vertical acceleration state quantity and correcting the vertical acceleration state quantity so as to have a saturation characteristic, and longitudinal, lateral and lateral directions. The feedback gain setting means 32 for setting the feedback gain of the low-pass filter processing value of the deviation of the vertical acceleration state quantity is connected.

  The posture angle estimating means 28 feeds back the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to the low-pass filter using a feedback gain, and the longitudinal acceleration Gx of the vehicle motion. Roll angle, which is a posture angle with respect to the vertical axis of the vehicle body, based on sensor signals corresponding to the detected values of lateral acceleration Gy, vertical acceleration Gz, roll angular velocity P, and yaw angular velocity R, and an estimated value of pitch angular velocity Q Estimate φ and pitch angle θ. Further, the posture angle estimation unit 28 inputs the estimated values of the roll angle φ and the pitch angle θ to the pitch angular velocity estimation unit 24, the acceleration state quantity saturation characteristic correction unit 30, and the feedback gain setting unit 32 as previous values. The pitch angular velocity estimating means 24, the acceleration state quantity saturation characteristic correcting means 30, and the feedback gain setting means 32 are connected. Further, the posture angle estimation means 28 inputs the low-pass filter processed longitudinal acceleration state quantity, lateral acceleration state quantity, and vertical acceleration state quantity output from the observer to the state quantity deviation low-pass filter value calculation means 26. In addition, the state quantity deviation low-pass filter value calculation means 26 is connected.

  In the present embodiment, the state quantity deviation low-pass filter value calculation means 26, the attitude angle estimation means 28, the acceleration state quantity saturation characteristic correction means 30, and the feedback gain setting means 32 correspond to the attitude angle estimation means of the present invention. The state quantity deviation low-pass filter value calculation means 26 corresponds to the state quantity deviation calculation means of the present invention, and the acceleration state quantity saturation characteristic correction means 30 corresponds to the correction means of the present invention.

  The lateral speed estimating means 10, the longitudinal speed estimating means 12, the pitch angular speed estimating means 24, the state quantity deviation low pass filter value calculating means 26, the attitude angle estimating means 28, the acceleration state quantity saturation characteristic correcting means 30, and the feedback gain setting means 32 are: It can be composed of one or a plurality of computers or one or a plurality of electronic circuits for realizing the function of each means.

  Next, estimation of the pitch angular velocity will be described. The equation of motion of the rigid body that expresses the relationship between the sensor signal output from the three-axis sensor that detects the three-axis acceleration and the three-axis angular velocity fixed to the rigid body and the motion state quantity can be described as follows.

  However, Gx: longitudinal acceleration, Gy: lateral acceleration, Gz: vertical acceleration, P: roll angular velocity, Q: pitch angular velocity, R: yaw angular velocity, U: longitudinal velocity, V: lateral velocity, W: vertical velocity, φ: roll Angle, θ: Pitch angle, g: Gravitational acceleration.

  When the vehicle is a rigid body in the present embodiment, the longitudinal acceleration Gx, the lateral acceleration Gy, the vertical acceleration Gz, the roll angular velocity P, and the yaw angular velocity R are respectively the longitudinal acceleration sensor 18, the lateral acceleration sensor 16, and the vertical acceleration sensor 14. , And a roll angular velocity sensor 20 and a yaw angular velocity sensor 22. The lateral speed V can be estimated by the lateral speed estimating means 10 as described above, and the longitudinal speed U is estimated by the longitudinal speed estimating means 12 as described above based on the wheel speed of each wheel. be able to.

  In the present embodiment, as shown in FIG. 2, the coordinates are described in a right-hand system with the upward direction of the vehicle body as the positive direction of the z axis, and the angles are expressed by Euler angles.

While the bank is running, the vertical acceleration Gz increases due to centrifugal force. The increase in the vertical acceleration Gz corresponds to an increase in -QU on the left side in the above equation (3). By utilizing this property, the change in the vertical velocity W is ignored in the equation (3) (W = 0) while) to the previous value of a roll angle and a pitch angle, and the estimated value of the lateral speed, by the use of the estimated value V _s0 of the longitudinal vehicle body speed as longitudinal speed U, the following estimates of the pitch angular velocity It can be obtained from the formula. The pitch angular velocity estimating means 24 obtains an estimated value of the pitch angular velocity by executing the calculation of the following equation (6).

When the pitch angular velocity is estimated as described above, the pitch angle can be estimated simultaneously with the roll angle by the attitude angle observer. A description will be given of the constraint condition of the motion unique to the automobile when the attitude angle observer is used. When a posture angle observer is configured using the above equation of motion, feedback of a physical quantity that can be measured is required so that the state quantities of the speed and angle estimated by the integral calculation do not diverge. In the present embodiment, characteristics unique to automobile motion are used as physical quantities to be fed back as follows. In the following formulas, each of the previous estimate the roll angle and pitch angle, estimated value of the lateral vehicle body speed as the horizontal velocity is used respectively estimate V _s0 of the longitudinal vehicle body speed as longitudinal speed U.

The estimated value V _s0 of the longitudinal vehicle body is differentiated and substituted into the above equation (1), and the characteristic that “the wheel slip does not continue to increase for a long time” is used in the longitudinal direction of the vehicle body as a characteristic characteristic of the vehicle motion. Then, from the above equation (1), the condition shown in the following equation (7) used for feedback ignoring the change in the vertical speed is obtained.

The above equation (7) describes the relationship between the pitch angle θ and the difference between the acceleration estimated from the wheel speed (the differential value of the estimated value _Vs0 of the longitudinal vehicle body speed) and the longitudinal acceleration Gx.

  Further, regarding the lateral direction of the vehicle body, the roll angular velocity and the lateral acceleration can be ignored due to the property that “the slip angle does not continue to increase for a long time”. From the above equation (2), the following equation (8) used for feedback: The following conditions are obtained.

  Furthermore, in consideration of a general road gradient, it can be regarded that “the acceleration in the vertical direction substantially matches the acceleration of gravity”. This condition can be described by the algebraic equation shown in the following equation (9) using the longitudinal acceleration Gx, the lateral acceleration Gy, the vertical acceleration Gz, the pitch angle θ, and the roll angle φ.

  Next, the configuration of a basic nonlinear observer will be described. Here, the value u including the sensor signal output from the sensor and the estimated value of the pitch angular velocity is expressed as the following equation (10).

  Further, if the target vehicle motion is expressed as the following equation (11) and the physical quantity that can be measured to constitute the observer is expressed as the equation (12), the nonlinear observer is expressed by the following equation (13) and It can be described by the nonlinear equation of motion of equation (14).

  However, the tilde of x and the tilde of y represent the estimated values of x and y, respectively, and k (tilde of x, u) represents the observer gain to be designed.

  Since the above equations (7) to (9) are all satisfied when a certain amount of time is taken into consideration, both sides of the above equations (7) to (9) are used for feedback of the measured amount of the observer. A value obtained by low-pass filtering each of the above is used.

  By solving the differential equations shown in the following equations (15) to (17), the right side of the above equations (7) to (9), that is, before and after obtained using the estimated value of the roll angle and the estimated value of the pitch angle A value obtained by subjecting the acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity to low-pass filter processing is calculated.

However, τ _x , τ _y , and τ _g represent time constants of several seconds to several tens of seconds or more of the low-pass filter considered in the equations (7) to (9).

  Since the state equation regarding the angle in the above equations (4) and (5) does not include the speed state quantity, the roll angle φ and the pitch angle θ are estimated as angles separately from the speed estimation. be able to.

  For this reason, first, an observer for estimating the roll angle and the pitch angle is configured including the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity obtained by the low-pass filter. In the present embodiment, the state quantity of the observer is expressed by the following equation (18).

  Note that the tilde of each of the above variables means an estimated value.

  The state quantity as the observer output used for feedback is expressed by the following equation (19).

  Further, a state quantity y calculated from the sensor signal or the like and comprising the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity as the vehicle output is expressed by the following equation (20).

  However,

The numerator of the second term on the right side of the equation (21) is the left side of the equation (7), that is, the yaw angular velocity value R detected by the three-axis sensor from the differential value of the estimated value _Vs0 of the front and rear vehicle body speed. This is a longitudinal acceleration state quantity obtained by subtracting the sum of the product of the estimated value of the lateral vehicle body speed and the longitudinal acceleration value Gx detected by the three-axis sensor. The numerator of the second term on the right side of the equation (22) is the product of the left side of the equation (7), that is, the yaw angular velocity value R detected by the three-axis sensor and the estimated value V _s0 of the front and rear vehicle speeds. This is a lateral acceleration state quantity obtained by subtracting the lateral acceleration value Gy detected by the three-axis sensor.

  The longitudinal acceleration state quantity and the lateral acceleration state quantity obtained using the sensor signal, the estimated value of the longitudinal vehicle body speed, and the estimated value of the lateral vehicle body speed by solving the differential equations shown in the above equations (21) and (22). Is subjected to low-pass filter processing with a low-pass filter. Further, g in the above equation (20) is a vertical acceleration state quantity.

  The low-pass filter processing values of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity calculated from the expressions (15) to (17), and the longitudinal acceleration state quantity represented by the expression (20), Based on the low-pass filter processing values of the lateral acceleration state quantity and the vertical acceleration state quantity, the low-pass filter processing values of the longitudinal acceleration state quantity deviation, the lateral acceleration state quantity deviation, and the vertical acceleration state quantity deviation, respectively. Is calculated.

  In the present embodiment, an appropriate observer gain k (x tilde, u) is set to constitute a posture angle observer that estimates a roll angle and a pitch angle.

  Next, a posture angle observer for estimating the roll angle φ and the pitch angle θ will be described.

  First, as an example of the observer gain, the observer gain expressed by the following equation (23) is set so that the diagonal component when linearized has a negative coefficient in order to ensure the stability of the observer. The case where it is set will be described.

However, a _{K φy, K φg, K θx} , K θg, K x, K y, K g is proper positive constant.

  In this case, the posture angle observer serving as a nonlinear observer for estimating the roll angle and the pitch angle is described by the equation of motion shown in the following equation (24).

  However, the tilde of x is represented by the following formula (25).

  Here, the result of estimating the attitude angle when traveling on a mountain circuit course using the nonlinear observer shown in the above equation (24) will be described with reference to FIGS. 3 (A) and 3 (B). As shown in FIG. 3A, in the vicinity of 12 to 15 s, a large roll angle that cannot be assumed from road surface conditions such as an actual bank angle is estimated.

  Such an abnormality in the estimated roll angle value occurs at the lowest point of the road surface gradient changing from the downhill to the uphill, and it can be assumed that it is related to the increase in the vertical acceleration. The posture angle observer here uses the condition that “the vertical acceleration subjected to filtering with a long time constant is equal to the gravitational acceleration”. This condition is set under the assumption that a short-term change in vertical acceleration due to a change in road surface gradient can be eliminated by a filter. It is expected that the influence of the change in acceleration could not be removed sufficiently.

  As shown in FIGS. 4A to 4C, the temporal change of the vertical acceleration, the lateral acceleration, and the roll angle estimated value after the low-pass filter processing is obtained. As shown in FIG. Since the vertical acceleration after filtering increases in the vicinity, it can be seen that the change in vertical acceleration due to a change in road surface gradient or the like could not be removed by the filter.

Further, as can be seen from the above equation (24), when the vertical acceleration state quantity after filtering calculated by the above equation (17) increases, the first line of the above equation (24) for calculating the roll angle estimated value is Increase / decrease (the increase / decrease depends on the sign of the lateral acceleration). For example, as shown in FIGS. 4A and 4B, when the vertical acceleration increases (when the tilde of the feedback amount g-g _f to the observer becomes negative), the negative lateral acceleration When this occurs, the first line of the above equation (24) is corrected in the increasing direction, and the roll angle is estimated and calculated as a large positive value as shown in FIG.

  As described above, it is presumed that the divergence phenomenon (increase in absolute value) of the roll angle estimation value occurring at the lowest point of the road surface gradient changing from the downhill to the uphill is caused by the feedback of the condition regarding the vertical acceleration.

  Therefore, the divergence phenomenon (increase in absolute value) of the roll angle estimation value is suppressed by performing a correction that relaxes the feedback of the condition related to vertical acceleration at the lowest point of the road gradient that changes from downhill to uphill. It is possible to plan.

  In the present embodiment, a saturation characteristic is added to the input signal to the low-pass filter in order to remove a short-term change in vertical acceleration due to a change in road surface gradient or the like. That is, instead of the above equation (17), the low-pass filter processing value of the vertical acceleration state quantity obtained by the following equation (26) is used for feedback.

Here, g _m is a constant for limiting the vertical acceleration state quantity g _v input to the low pass filter of the above equation (26).

  As shown in the above equation (27), by giving saturation characteristics to the vertical acceleration state quantity, if the vertical acceleration state quantity is greater than or equal to the upper limit value, the vertical acceleration state quantity is corrected to the upper limit value and the vertical acceleration state quantity is corrected. If the state quantity is less than or equal to the lower limit value, the vertical acceleration state quantity is corrected to the lower limit value.

  The acceleration state quantity saturation characteristic correcting means 30 calculates the vertical acceleration state quantity according to the above equation (28), and calculates so as to give the vertical acceleration state quantity a saturation characteristic according to the above expression (27). The vertical acceleration state quantity is corrected. Further, as will be described later, the posture angle observer of the posture angle estimating means 28 calculates a low-pass filter processing value of the corrected vertical acceleration state quantity according to the above equation (26).

  The state quantity deviation low-pass filter value calculation means 26 is based on the detected values of the lateral acceleration sensor 16, the longitudinal acceleration sensor 18, and the yaw angular velocity sensor 22, the estimated value of the lateral vehicle body speed, and the estimated value of the longitudinal vehicle body speed. The longitudinal acceleration state quantity and the lateral acceleration state quantity subjected to the low-pass filter processing expressed by the above equation (20) are calculated, and the calculated longitudinal acceleration state quantity and lateral acceleration state quantity and the attitude angle observer (15) The longitudinal acceleration state quantity and the lateral acceleration state quantity subjected to the low-pass filter processing calculated according to the formula (16) and the low-pass filter processing calculated according to the formula (26) in the attitude angle observer were performed. Based on the vertical acceleration state quantity, each of the longitudinal acceleration state quantity deviation, lateral acceleration state quantity deviation, and vertical acceleration state quantity deviation Calculating a pass filtered value.

  Next, the principle of setting the observer gain in this embodiment will be described.

  First, the nonlinear observer shown in the above equation (24) is linearized around the operating point, and the stability of the linearized system is analyzed. The A matrix when the above equation (24) is linearized (represented as tilde of dx / dt = Ax + Bu) is represented by the following equation (29).

  In the above equation (29), the subscript 0 of each variable represents a value at the operating point.

  In addition, regarding a roll angle, a pitch angle, a pitch angular velocity, a yaw angular velocity, a longitudinal acceleration, and a lateral acceleration, a region in which vehicle motion can be assumed is set, and the results of examining the stability at an arbitrary point in this region will be described. .

  FIG. 5 shows the analysis result of this stability. In FIG. 5, the pitch angle is displayed on the horizontal axis and the roll angle is displayed on the vertical axis, and all other parameters are searched within the assumed range. The symbol is displayed. Further, in FIG. 5, the value of the maximum real part of the unstable pole is indicated by contour lines. When the observer gain shown in the above equation (23) is used, it can be seen that an area in which the observer may become unstable is generated as the posture angle increases.

  The longitudinal acceleration and lateral acceleration in the first row, third column, and second row, third column in the above equation (23) corresponding to the feedback gain of the deviation of the vertical acceleration state quantity are intended to set the gain according to the sign of each sensor signal. On the other hand, in this embodiment, in order to stabilize the unstable region, the observer gain is set as shown in the following equation (30).

However, with _{regard to} K _φg (G _y ) and K _θg (G _x ) in the first row and the third column and the second row and the third column of the above equation (30) corresponding to the feedback gain of the deviation of the vertical acceleration state quantity, As shown in equations (32) and (32), saturation characteristics with respect to translational acceleration (longitudinal acceleration and lateral acceleration) are provided to prevent the observer gain from becoming excessive.

K _φg0 and K _θg0 are constant gains, G _ym and G _xm are upper limit values of longitudinal acceleration and lateral acceleration for suppressing increase in gain, and −G _ym and −G _xm are These are the lower limit value of the longitudinal acceleration and the lower limit value of the lateral acceleration for suppressing an increase in gain.

The feedback gain of the deviation of the vertical acceleration state quantity in the first row of the above equation (24) is K _φg G _y, which is proportional to the value of the lateral acceleration G _y , but in this embodiment, As shown in equation (30) above, the divergence phenomenon (increase in absolute value) of the roll angle estimation value is suppressed by giving saturation characteristics corresponding to the lateral acceleration to the feedback gain of the deviation of the vertical acceleration state quantity. can do. The above equations (31) and (32) are feedback gains for the vertical acceleration state quantity deviation, and by suppressing the gain increase, the increase in the posture angle error when the vertical acceleration state quantity deviation is increased is suppressed. can do.

Further, focusing on the fact that the observer stability in the region with a large posture angle can be secured by changing the feedback gain of the deviation of the longitudinal and lateral acceleration state quantities according to the posture angle, as shown in FIG. The larger the absolute value of the estimated value and the estimated value of the roll angle, the larger the absolute value of the estimated value of the roll angle. _.phi.x, it increases the K _{[theta] y,} the front and rear, to increase the absolute value of the feedback gain of the lateral acceleration state variable deviation signal. As a _result , the control gains of K _φx and K _θy can be set large in the unstable region of FIG. 5 to stabilize the observer, and the observer can be stabilized in a wide range as shown in FIG. is there.

The feedback gain setting means 32 sets K _φg (G _y ) and K _θg (G _x ) in the feedback gain of the deviation of the vertical acceleration state quantity so as to have a saturation characteristic for the longitudinal acceleration and a saturation characteristic for the lateral acceleration. At the same time, K _φx and K _θy are set so that the absolute value of the feedback gain of the deviation of the lateral acceleration state quantity increases as the absolute value of the estimated value of the pitch angle and the estimated value of the roll angle increases.

  Next, a nonlinear observer for estimating the roll angle φ and the pitch angle θ used in the present embodiment will be described. The nonlinear observer for estimating the roll angle and pitch angle is described by the equation of motion shown in the following equation (33).

  The posture angle estimation means 28 calculates the tilde of the roll angle differential amount dφ and the pitch angle differential amount dθ tilde, which is the posture angle with respect to the vehicle body vertical direction, by using the above equation (33), and calculates the calculated roll By integrating the tilde of the angular differential amount dφ and the tilde of the pitch angle differential amount dθ, the tilde of the roll angle φ and the tilde of the pitch angle θ can be calculated. Further, the posture angle estimation means 28 calculates the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity subjected to the low-pass filter process by using the above equation (33).

  The computer includes a lateral speed estimating means 10, a longitudinal speed estimating means 12, a pitch angular speed estimating means 24, a state quantity deviation low-pass filter value calculating means 26, an attitude angle estimating means 28, an acceleration state quantity saturation characteristic correcting means 30, and a feedback gain setting. Information processing by a program that functions as each means of the means 32 can be realized by a posture angle estimation processing routine shown in the flowchart of FIG. The computer includes a CPU, a ROM, and a RAM connected to each other by a bus, and an HDD connected as necessary. These programs are recorded on a recording medium such as a ROM or an HDD connected to the CPU of the computer. To be recorded.

  Explaining this attitude angle estimation processing routine, in step 100, the lateral vehicle body speed is estimated as described above, and in step 102, the front and rear vehicle body speeds are estimated based on the wheel speed of each wheel.

  In step 104, it is determined whether or not this is the first calculation. In the case of the first calculation, the previous estimated values of the roll angle and the pitch angle and the previous values of the front and rear, lateral and vertical acceleration state quantities do not exist. A predetermined initial value is set as the previous value of the previous estimated values of the roll angle and pitch angle and the low-pass filter processing values of the front, rear, side, and vertical acceleration state quantities at 106, as described above at step 110. Then, the estimated value of the pitch angular velocity Q is calculated using the above equation (6).

  On the other hand, in the second and subsequent calculations, the estimated values of the roll angle and pitch angle and the low-pass filter processing values of the front / rear, lateral and vertical acceleration state quantities are calculated in step 118 described later. As the previous estimated values of the roll angle and the pitch angle and the low-pass filter processing values of the longitudinal, lateral, and vertical acceleration state quantities, the previously calculated values are set, and in step 110, as described above, the above expression (6) Is used to calculate the estimated value of the pitch angular velocity Q.

In step 112, as described above, based on the sensor signal corresponding to the detected values of the longitudinal acceleration Gx, the lateral acceleration Gy, and the yaw angular velocity R of the vehicle motion, and the estimated value _Vs0 of the longitudinal vehicle body velocity U, 21), the low-pass filter processing values of the longitudinal acceleration state quantity and the lateral acceleration state quantity are calculated according to the equations (22) and (22), and the previous calculation is performed as the values of the longitudinal, lateral, and vertical acceleration state quantities subjected to the low-pass filter processing. Using the previous value, the deviation of the front, back, lateral and vertical acceleration state quantities subjected to the low-pass filter processing is calculated.

  In the next step 114, as explained above, saturation characteristics are given to the longitudinal acceleration Gx and the lateral acceleration Gy, and the feedback gain is set according to the absolute values of the previous estimated values of the roll angle and the pitch angle. To do.

  In step 116, based on the sensor signal corresponding to the detected values of the longitudinal acceleration Gx, the lateral acceleration Gy, and the vertical acceleration Gz of the vehicle motion and the previous estimated values of the roll angle and the pitch angle, the above equation (28) The vertical acceleration state quantity is calculated according to the above, and the calculated vertical acceleration state quantity is corrected so as to have a saturation characteristic according to the above equation (27).

  In the next step 118, the estimated value of the pitch angular velocity Q calculated in step 110, the low-pass filter processing value calculated in step 112, the feedback gain set in step 114, and the calculation in step 116 are performed. Using the vertical acceleration state quantity, the roll angle φ and the pitch angle θ, which are posture angles with respect to the vertical axis of the vehicle body, are estimated according to the above equation (33), and the estimated values of the roll angle φ and the pitch angle θ are stored in the computer memory. Store and return to step 100 above. In step 118, the low-pass filter processing values of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity are calculated according to (33) and stored in the memory of the computer.

  Next, a description will be given of an estimation result obtained by the method for providing saturation characteristics according to this embodiment. In addition, in order to confirm the effect of the method of giving saturation characteristics, the posture angle when traveling on a mountain circuit was estimated. As explained in the present embodiment, the saturation characteristics are given to the vertical acceleration state quantity, and the saturation characteristics are given to the feedback gain of the deviation of the vertical acceleration state quantity. The posture angle was estimated for each of the cases where it was excluded.

  As shown in FIGS. 9A and 9B, in the conventional method excluding the saturation characteristic, the error of the estimated value of the roll angle in the vicinity of 12 to 15 seconds has increased. On the other hand, when the saturation specification is given, it can be seen that the error of the estimated value of the roll angle in the vicinity of 12 to 15 seconds is improved.

  As described above, according to the vehicle attitude angle estimation device according to the first embodiment, the vertical acceleration state quantity obtained using the estimated roll angle value and the estimated pitch angle value is given saturation characteristics. Since the deviation of the vertical acceleration state quantity subjected to the low-pass filter processing is calculated and fed back, the posture angle can be accurately estimated even on the road surface where the vertical acceleration changes. it can.

  Also, as the absolute value of the estimated value of the roll angle and the estimated value of the pitch angle are larger, the absolute value of the feedback gain of the deviation of the longitudinal acceleration state quantity and the deviation of the lateral acceleration state quantity is set larger. Even on a steep slope where the posture angle increases, the roll angle and pitch angle can be estimated with high accuracy.

  In addition, since the pitch angular velocity is estimated from the balance of the vertical movement of the vehicle body and the roll angle and pitch angle are estimated using the estimated value of the pitch angular velocity, the roll can be accurately rolled even during banking where the vertical acceleration increases. The angle and pitch angle can be estimated.

  Next, a second embodiment of the present invention will be described. In addition, since the structure of the vehicle attitude angle estimation apparatus which concerns on 2nd Embodiment is the structure similar to 1st Embodiment, it attaches | subjects the same code | symbol and abbreviate | omits description.

  In the second embodiment, not only the vertical acceleration state quantity but also the longitudinal acceleration state quantity and the lateral acceleration state quantity are subjected to low-pass filter processing with saturation characteristics. This is mainly different from the first embodiment.

  In the second embodiment, the acceleration state quantity saturation characteristic correcting means 30 calculates the vertical acceleration state quantity according to the above equation (28) and calculates the saturation characteristic according to the above expression (27). The vertical acceleration state quantity is corrected. The acceleration state quantity saturation characteristic correcting means 30 is represented by the longitudinal acceleration state quantity represented by the second term numerator on the right side of the equation (21) and the second term numerator on the right side of the equation (22). The calculated lateral acceleration state quantity is corrected, and the calculated longitudinal acceleration state quantity and lateral acceleration state quantity are corrected so as to have saturation characteristics as in the case of the vertical acceleration state quantity.

  Further, the posture angle observer of the posture angle estimating means 28 calculates the low-pass filter processing values of the front / rear, lateral and vertical acceleration state quantities using the corrected front / rear, lateral and vertical acceleration state quantities.

  The state quantity deviation low-pass filter value calculation means 26 calculates the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity subjected to the low-pass filter processing shown in the above equation (19), and the computed longitudinal acceleration state quantity. Based on the lateral acceleration state quantity, the vertical acceleration state quantity and the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity that have been subjected to the low-pass filter processing calculated as described above in the attitude angle observer. Thus, the low-pass filter processing values of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are calculated.

  In addition, about the other structure and effect | action of the vehicle attitude angle estimation apparatus which concern on 2nd Embodiment, since it is the same as that of 1st Embodiment, description is abbreviate | omitted.

  Thus, the front-rear, lateral, and vertical acceleration state quantities obtained using the estimated roll angle value and the estimated pitch angle value are corrected so as to have saturation characteristics, and low-pass filter processing has been performed. Since the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are calculated and fed back, the posture angle is accurately estimated even on a road surface where the vertical acceleration changes. be able to.

  In the above-described embodiment, the case where each of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity is corrected so as to have saturation characteristics has been described as an example, but the present invention is not limited thereto. Instead of this, at least one of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity may be corrected so as to have saturation characteristics.

  Next, a third embodiment of the present invention will be described. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  The third embodiment is mainly different from the first embodiment in that after calculating the deviation of the vertical acceleration state quantity, the correction is performed so as to have a saturation characteristic.

  As shown in FIG. 10, in the third embodiment, the lateral velocity estimating means 10, the longitudinal velocity estimating means 12, the lateral acceleration sensor 16, the longitudinal acceleration sensor 18, and the yaw angular velocity sensor 22 are different in the longitudinal acceleration state quantity. And a state quantity deviation calculating means 326 for calculating each of the deviation of the lateral acceleration state quantity and the deviation of the vertical acceleration state quantity.

  The state quantity deviation calculating means 326 is connected to state quantity deviation saturation characteristic correcting means 328 for correcting the deviation of the vertical acceleration state quantity so as to have a saturation characteristic.

  The state quantity deviation calculating means 326 and the state quantity deviation saturation characteristic correcting means 328 are subjected to low-pass filter processing for passing only the motion to be noticed, the longitudinal acceleration state quantity deviation, the lateral acceleration state quantity deviation, and the vertical acceleration. It is connected to low-pass filter value calculation means 330 for calculating each low-pass filter processing value of the state quantity deviation.

  The low pass filter value calculation means 330 is connected to the attitude angle estimation means 28. The posture angle estimating means 28 estimates the roll angle φ and the pitch angle θ, which are posture angles with respect to the vertical axis of the vehicle body, and is represented by the right side of the above equations (7) to (9) as state quantities used for feedback. The longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity are calculated.

  In addition, the posture angle estimation unit 28 inputs the estimated values of the roll angle φ and the pitch angle θ to the pitch angular velocity estimation unit 24 and the feedback gain setting unit 32 as previous values, and inputs the pitch angular velocity estimation unit 24 and the feedback. The gain setting means 32 is connected. In addition, the posture angle estimation unit 28 inputs the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity output from the observer to the state quantity deviation calculation unit 326 so as to input the state quantity deviation calculation unit 326. It is connected to the.

  In this embodiment, the state quantity deviation calculating means 326, the state quantity deviation saturation characteristic correcting means 328, the low-pass filter value calculating means 330, the attitude angle estimating means 28, and the feedback gain setting means 32 are the attitude angles of the present invention. Corresponding to the estimating means, the state quantity deviation calculating means 326 and the state quantity deviation saturation characteristic correcting means 328 correspond to the correcting means of the present invention, and the low-pass filter value calculating means 330 corresponds to the low-pass filter processing means of the present invention. ing.

  The state quantity deviation calculating means 326 is based on the detected values of the lateral acceleration sensor 16, the longitudinal acceleration sensor 18, and the yaw angular velocity sensor 22, the estimated value of the lateral vehicle body speed, and the estimated value of the longitudinal vehicle body speed (7). ) Formulas to (8), the longitudinal acceleration state quantity and the lateral acceleration state quantity are calculated, and the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state calculated by the attitude angle estimation means 28 are calculated. Using the amount, the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are calculated.

  The state quantity deviation saturation characteristic correcting unit 328 corrects the deviation of the vertical acceleration state quantity so as to have a saturation characteristic with respect to the deviation of the vertical acceleration state quantity in accordance with the following equation (34).

Here, u _gf is a deviation of the corrected vertical acceleration state quantity, and g _m is a constant for limiting the deviation g−g _v of the vertical acceleration state quantity.

  As shown in the above equation (34), by giving saturation characteristics to the deviation of the vertical acceleration state quantity, if the deviation of the vertical acceleration state quantity is not less than the upper limit value, the deviation of the vertical acceleration state quantity is set to the upper limit value. If the deviation of the vertical acceleration state quantity is not more than the lower limit value, the deviation of the vertical acceleration state quantity is corrected to the lower limit value.

The low-pass filter value calculation means 330 performs low-pass filter processing on the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity according to the following formulas (36) to (38). was carried out, tilde low-pass filtered value _g xdf _{-g xdf} deviation of the longitudinal acceleration state quantity, lateral acceleration state amount of the deviation of the low-pass filtered value _g YDF _{-g YDF} tilde, and vertical acceleration state quantity of deviation lowpass The tilde of the filter processing value g _zdf -g _zdf is calculated.

  In addition, since the other structure of the vehicle attitude angle estimation apparatus which concerns on 3rd Embodiment is the same as that of 1st Embodiment, description is abbreviate | omitted.

  The computer includes a lateral speed estimating means 10, a longitudinal speed estimating means 12, a pitch angular speed estimating means 24, a state quantity deviation calculating means 326, a state quantity deviation saturation characteristic correcting means 328, a low-pass filter value calculating means 330, an attitude angle estimating means 28, Information processing by a program that functions as each means of the feedback gain setting means 32 can be realized by a posture angle estimation processing routine shown in the flowchart of FIG. The computer includes a CPU, a ROM, and a RAM connected to each other by a bus, and an HDD connected as necessary. These programs are recorded on a recording medium such as a ROM or an HDD connected to the CPU of the computer. To be recorded.

  This posture angle estimation processing routine will be described below. Note that the same processes as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted.

  First, in step 100, the lateral vehicle body speed is estimated as described above, and in step 102, the longitudinal vehicle body speed is estimated based on the wheel speed of each wheel.

  In step 104, it is determined whether or not this is the first calculation. In the case of the first calculation, in step 346, the previous estimated values of the roll angle and the pitch angle and the previous values of the front, rear, horizontal and vertical acceleration state quantities are determined in advance. In step 110, an estimated value of the pitch angular velocity Q is calculated.

  On the other hand, in the second and subsequent calculations, the estimated values of the roll angle and pitch angle and the front / rear, lateral, and vertical acceleration state quantities are calculated in step 356, which will be described later. In step 348, the roll angle and pitch are calculated. The previous calculated value of the angle and the previous calculated value of the front, rear, side, and vertical acceleration state quantities are set, and in step 110, the estimated value of the pitch angular velocity Q is calculated.

In step 350, as described above, based on the sensor signal corresponding to the detected values of the longitudinal acceleration Gx, the lateral acceleration Gy, and the yaw angular velocity R of the vehicle motion, and the estimated value _Vs0 of the longitudinal vehicle body speed U, According to the left side of equation (7) and the left side of equation (8) above, the longitudinal acceleration state quantity and the lateral acceleration state quantity are calculated, and the previous values of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity are used. The deviation of the longitudinal acceleration state quantity, the lateral acceleration state quantity deviation, and the vertical acceleration state quantity deviation are calculated.

  In step 352, the deviation of the vertical acceleration state quantity calculated in step 350 is corrected so as to have a saturation characteristic according to the above equation (34). In the next step 354, the low-pass filter processing values of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are calculated according to the above equations (36) to (38).

  In step 114, as described above, saturation characteristics are given to the longitudinal acceleration Gx and the lateral acceleration Gy, and a feedback gain is set according to the absolute values of the previous estimated values of the roll angle and the pitch angle. To do.

  In the next step 356, the vertical axis of the vehicle body is estimated using the estimated value of the pitch angular velocity Q calculated in step 110, the low-pass filter processing value calculated in step 354, and the feedback gain set in step 114. The roll angle φ and the pitch angle θ, which are the posture angles with respect to, are estimated, the estimated values of the roll angle φ and the pitch angle θ are stored in the memory of the computer, and the process returns to step 100. In step 356, each of the longitudinal acceleration state quantity, the lateral acceleration state quantity, and the vertical acceleration state quantity is calculated and stored in the memory of the computer.

  As described above, according to the vehicle attitude angle estimation device according to the third embodiment, the vertical acceleration state variable is corrected by correcting the deviation of the vertical acceleration state quantity so as to have the saturation characteristic, and is subjected to the low-pass filter processing. Since the deviation of the acceleration state quantity is calculated and fed back, the posture angle can be accurately estimated even on a road surface where the vertical acceleration changes.

  Next, a fourth embodiment of the present invention will be described. In addition, since the structure of the vehicle attitude angle estimation apparatus which concerns on 4th Embodiment is a structure similar to 3rd Embodiment, it attaches | subjects the same code | symbol and abbreviate | omits description.

  In the fourth embodiment, not only the deviation of the vertical acceleration state quantity but also the deviation of the longitudinal acceleration state quantity and the deviation of the lateral acceleration state quantity are corrected so as to have saturation characteristics, The main difference from the first embodiment is that low-pass filter processing is performed.

  In the fourth embodiment, the state quantity deviation saturation characteristic correction unit 328 corrects the deviation of the vertical acceleration state quantity so as to have a saturation characteristic with respect to the deviation of the vertical acceleration state quantity according to the above equation (34). To do. In addition, the state quantity deviation saturation characteristic correcting means 328, like the vertical acceleration state quantity deviation, has a longitudinal acceleration state quantity so as to have saturation characteristics with respect to the longitudinal acceleration state quantity deviation and the lateral acceleration state quantity deviation. And deviation of lateral acceleration state quantity are corrected.

  Further, the low-pass filter value calculation means 330 calculates the corrected low-pass filter processing value of the deviation of the front / rear, lateral and vertical acceleration state quantities.

  In addition, since it is the same as that of 3rd Embodiment about the other structure and effect | action of the vehicle attitude angle estimation apparatus which concerns on 4th Embodiment, description is abbreviate | omitted.

  In this way, the deviation of the longitudinal, lateral, and vertical acceleration state quantities is corrected so as to have saturation characteristics, and the deviation of the longitudinal, lateral, and vertical acceleration state quantities that have undergone the low-pass filter processing is calculated and fed back. Therefore, the posture angle can be accurately estimated even on a road surface in which the vertical acceleration changes.

  In the above embodiment, the case where the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity are corrected so as to have saturation characteristics has been described as an example. However, the present invention is not limited to this, and at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity may be corrected so as to have saturation characteristics. Good.

  In the above embodiment, the case where the pitch angular velocity, the roll angle, and the pitch angle are estimated using the previous value of each estimated value has been described as an example. The pitch angular velocity, the roll angle, and the pitch angle may be estimated by using the previous value.

It is a block diagram which shows the 1st Embodiment of this invention. It is explanatory drawing which shows the coordinate system of this Embodiment. It is a graph which shows the result of having estimated the posture angle at the time of mountain circuit course driving. (A) The graph which shows the time change of the vertical acceleration after a low-pass filter process, (B) The graph which shows the time change of the lateral acceleration after a low-pass filter process, and (C) The graph which shows the time change of a roll angle estimated value. . It is a figure which shows the analysis result of stability of a posture angle observer. (A) A graph showing the relationship between the pitch angle and roll angle and K _θx in the feedback gain of the deviation of the longitudinal acceleration state quantity, and (B) in the feedback gain of the deviation of the pitch angle and roll angle and the lateral acceleration state quantity. It is a graph which shows the relationship with _Kθx . It is a figure which shows the analysis result of stability of the attitude | position angle observer used in the 1st Embodiment of this invention. It is a flowchart which shows the process of the 1st Embodiment of this invention. It is a diagram which shows the estimation result of the roll angle and pitch angle in 1st Embodiment. It is a block diagram which shows the 3rd Embodiment of this invention. It is a flowchart which shows the process of the 3rd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Lateral velocity estimation means 12 Longitudinal speed estimation means 14 Vertical acceleration sensor 16 Lateral acceleration sensor 18 Longitudinal acceleration sensor 20 Roll angular velocity sensor 22 Yaw angular velocity sensor 24 Pitch angular velocity estimation means 26 State quantity deviation low-pass filter value calculation means 28 Attitude angle estimation means 30 Acceleration state quantity saturation characteristic correction means 32 Feedback gain setting means 326 State quantity deviation calculation means 328 State quantity deviation saturation characteristic correction means 330 Low-pass filter value calculation means

Claims (6)

Front-rear speed estimation means for estimating front-rear vehicle body speed, which is a vehicle body speed in the vehicle front-rear direction, based on the wheel speed of each wheel;
Lateral speed estimation means for estimating a lateral vehicle body speed which is a vehicle body speed in the lateral direction of the vehicle;
Pitch angular velocity estimation means for estimating the pitch angular velocity;
Based on the sensor signals corresponding to the detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of the vehicle motion, the estimated value of the longitudinal vehicle body velocity, and the estimated value of the pitch angular velocity, Posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to an axis;
A vehicle attitude angle estimation device including:
The posture angle estimation means has at least one of a longitudinal acceleration state quantity, a lateral acceleration state quantity, and a vertical acceleration state quantity obtained by using the roll angle estimation value and the pitch angle estimation value, and has saturation characteristics. Correction means for correcting the longitudinal acceleration state quantity, the lateral acceleration state quantity, the vertical acceleration state quantity, the vertical acceleration state quantity, the sensor signal, the estimated value of the longitudinal vehicle body speed, and the estimated value of the lateral vehicle body speed. Based on the longitudinal acceleration state quantity and lateral acceleration state quantity obtained by using the low-pass filter processing, the longitudinal acceleration state quantity deviation, lateral acceleration state quantity deviation, and vertical acceleration state quantity deviation are calculated. A state quantity deviation calculating means, which feeds back the calculated deviation in the longitudinal acceleration state quantity, deviation in the lateral acceleration state quantity, and deviation in the vertical acceleration state quantity, Vehicle attitude angle estimation apparatus and estimates the pitch angle. Front-rear speed estimation means for estimating front-rear vehicle body speed, which is a vehicle body speed in the vehicle front-rear direction, based on the wheel speed of each wheel;
Lateral speed estimation means for estimating a lateral vehicle body speed which is a vehicle body speed in the lateral direction of the vehicle;
Pitch angular velocity estimation means for estimating the pitch angular velocity;
Based on the sensor signals corresponding to the detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of the vehicle motion, the estimated value of the longitudinal vehicle body velocity, and the estimated value of the pitch angular velocity, Posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to an axis;
A vehicle attitude angle estimation device including:
The posture angle estimation means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity, a vertical acceleration state quantity obtained using the roll angle estimation value and the pitch angle estimation value, the sensor signal, and the front and rear vehicle body speeds. And the longitudinal acceleration state quantity, the lateral acceleration state quantity deviation, the vertical acceleration state quantity, and the vertical acceleration state quantity, Correcting means for correcting at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity so as to have a saturation characteristic; Low-pass filter processing means for performing low-pass filter processing on the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity, The roll angle and the pitch angle are estimated by feeding back the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity subjected to low-pass filter processing. A vehicle attitude angle estimation device.   The pitch angular velocity estimation means includes an estimated value of the longitudinal vehicle body speed, an estimated value of the lateral vehicle body speed, a detected value of the vertical acceleration, a detected value of the roll angular speed, and the roll angle estimated by the posture angle estimating means. The vehicle attitude angle estimation device according to claim 1 or 2, wherein the pitch angular velocity is estimated based on a previous estimated value and a previous estimated value of the pitch angle estimated by the attitude angle estimating means.     The sensor according to any one of claims 1 to 3, further comprising a sensor that detects each of longitudinal acceleration, lateral acceleration, vertical acceleration, yaw angular velocity, and roll angular velocity of the vehicle motion and outputs a sensor signal corresponding to the detected value. The vehicle attitude angle estimation device described. Computer
Front-rear speed estimating means for estimating front-rear vehicle body speed, which is a vehicle body speed in the vehicle front-rear direction, based on the wheel speed of each wheel;
Lateral speed estimation means for estimating a lateral vehicle body speed which is a vehicle body speed in the lateral direction of the vehicle,
Pitch angular velocity estimating means for estimating the pitch angular velocity, sensor signals corresponding to detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of vehicle motion, estimated values of the front and rear vehicle body speeds, and the pitch A program for functioning as a posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to a vertical axis of a vehicle body, based on an estimated value of angular velocity,
The posture angle estimation means has at least one of a longitudinal acceleration state quantity, a lateral acceleration state quantity, and a vertical acceleration state quantity obtained by using the roll angle estimation value and the pitch angle estimation value, and has saturation characteristics. Correction means for correcting the longitudinal acceleration state quantity, the lateral acceleration state quantity, the vertical acceleration state quantity, the vertical acceleration state quantity, the sensor signal, the estimated value of the longitudinal vehicle body speed, and the estimated value of the lateral vehicle body speed. Based on the longitudinal acceleration state quantity and lateral acceleration state quantity obtained by using the low-pass filter processing, the longitudinal acceleration state quantity deviation, lateral acceleration state quantity deviation, and vertical acceleration state quantity deviation are calculated. A state quantity deviation calculating means, which feeds back the calculated deviation in the longitudinal acceleration state quantity, deviation in the lateral acceleration state quantity, and deviation in the vertical acceleration state quantity, Program to estimate the pitch angle. Computer
Front-rear speed estimating means for estimating front-rear vehicle body speed, which is a vehicle body speed in the vehicle front-rear direction, based on the wheel speed of each wheel;
Lateral speed estimation means for estimating a lateral vehicle body speed which is a vehicle body speed in the lateral direction of the vehicle,
Pitch angular velocity estimating means for estimating the pitch angular velocity, sensor signals corresponding to detected values of longitudinal acceleration, lateral acceleration, vertical acceleration, roll angular velocity, and yaw angular velocity of vehicle motion, estimated values of the front and rear vehicle body speeds, and the pitch A program for functioning as a posture angle estimating means for estimating a roll angle and a pitch angle, which are posture angles with respect to a vertical axis of a vehicle body, based on an estimated value of angular velocity,
The posture angle estimation means includes a longitudinal acceleration state quantity, a lateral acceleration state quantity, a vertical acceleration state quantity obtained using the roll angle estimation value and the pitch angle estimation value, the sensor signal, and the front and rear vehicle body speeds. And the longitudinal acceleration state quantity, the lateral acceleration state quantity deviation, the vertical acceleration state quantity, and the vertical acceleration state quantity, Correcting means for correcting at least one of the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity so as to have a saturation characteristic; Low-pass filter processing means for performing low-pass filter processing on the deviation of the longitudinal acceleration state quantity, the deviation of the lateral acceleration state quantity, and the deviation of the vertical acceleration state quantity, Low-pass filtering is performed, deviation of the longitudinal acceleration state quantity, the lateral acceleration state amount of the deviation, and by feeding back the deviation of the vertical acceleration state quantity, a program for estimating the roll angle and the pitch angle.