JP4677953B2 - Traveling apparatus and control method thereof - Google Patents

Description

  The present invention is, for example, a traveling device suitable for use in a vehicle that travels under the control of two wheels that are independently driven and controlled so as to maintain front-rear stability between the two wheels. It relates to the control method. Specifically, the road surface frictional force is estimated, and the motor torque for driving the wheel is controlled according to the estimated value to prevent the wheel from idling.

  2. Description of the Related Art Conventionally, in a vehicle having a plurality of driving wheels, it is known to detect the idling of the wheels by determining the difference in speed between the wheels (see, for example, Patent Document 1). That is, in the invention disclosed in Patent Document 1, when the difference in speed between the wheels is significant, it is determined that one of the wheels is idling and control is performed.

  In some cases, the wheel idling is detected by determining the difference in wheel speed between the driving wheel and the non-driving wheel (see, for example, Patent Document 2). That is, in the invention disclosed in Patent Document 2, when the difference in wheel speed between the driving wheel and the non-driving wheel is significant, it is determined that the driving wheel is idling and control is performed.

However, all of the above-described techniques are controlled after the wheel idling occurs, and these techniques cannot prevent the idling from occurring.
Japanese Patent No. 3152785 JP 2005-51888 A

  For example, the present applicant has previously proposed a traveling device as described below (Japanese Patent Application No. 2005-117365) as a vehicle that travels on two wheels with a human being on board. First, an embodiment of the coaxial two-wheeled vehicle proposed by the present applicant will be described with reference to FIGS. 6A, 6B and 7.

  As shown in FIGS. 6A and 6B, the coaxial two-wheeled vehicle 10 previously proposed by the applicant of the present application has two wheels 11L and 11R provided in parallel, and these wheels 11L and 11R are respectively independent motors 12L and 11L. It is driven by 12R. Further, the drive of these motors 12L and 12R is controlled by the control device 13. An attitude sensor 14 is connected to the control device 13, and a drive torque (motor torque) value necessary for controlling the motors 12 </ b> L and 12 </ b> R is calculated according to a detection signal from the attitude sensor 14.

  On the other hand, in the vicinity of the wheels 11L and 11R, there are provided division tables 15L and 15R that show a specific example of a boarding unit on which the driver gets on. The divided tables 15L and 15R are held in a predetermined posture by a link mechanism (not shown). A handle lever 16 extending upward is provided between the divided tables 15L and 15R, and a battery 17 serving as a driving power source for the entire apparatus and a roll shaft angle detector (not shown) 21 are provided at the base thereof. Provided. Further, a grip portion 19 including a power switch 18 is provided on the handle lever 16.

  Then, as shown in FIG. 7, the driver 20 stands by putting his / her feet on the split tables 15 </ b> L and 15 </ b> R, grasps the grip 19 at the top of the handle lever 16, and rolls the power switch 18 and the handle lever 16. Manipulate the shaft angle. Further, the position of the center of gravity of the driver who has boarded the split tables 15L and 15R is detected by a built-in pressure sensor (not shown). The pressure detection signal and the table angle detection signal from the attitude sensor 14 are supplied to the control device 13 so that the traveling of the coaxial two-wheel vehicle 10 is controlled.

  Further, FIG. 8 schematically shows a specific configuration of a single-wheel model control device. In an actual two-wheeled vehicle, the table sensor is common. Further, the motor control connected to the wheels in the illustrated model is controlled by an independent control device for each wheel.

  In FIG. 8, pressure detection signals PS1, 2, 3, 4 from a pressure sensor (not shown) built in the table 15 and a table angle detection signal θ0 from a posture sensor 14 including a gyro sensor and an acceleration sensor. It is supplied to the attitude control unit 31 in the control device 13. These detection signals PS1 to PS4 and θ0 and an external table attitude command signal θREF [d (θREF) / dt] issued by a passenger or the like are calculated, and the calculated rotation command ωref is sent to the motor control unit 32. Supplied.

  Furthermore, the wheel 11 and the motor 12 are connected via a speed reducer 33. The motor 12 is provided with an encoder 34. Then, the rotor rotation angle position signal Θr from the encoder 34 is supplied to the motor control unit 32 in the control device 13. Thereby, the drive current to the motor 12 formed according to the rotation command ωref described above is feedback-controlled, and the driving of the wheels 11 is stabilized. In this way, the wheel 11 is driven stably, and the drive is controlled by the pressure detection signals PS1 to PS4 from a pressure sensor (not shown).

  FIG. 9 shows the mutual connection relationship of the systems. In FIG. 9, pressure detection signals PS <b> 1 to PS <b> 4 from the pressure sensors 41 to 44 and a roll shaft angle detection signal PM from the roll shaft angler (potentiometer) 21 of the handle lever 16 are supplied to the attitude sensor circuit 45. The attitude sensor circuit 45 has a gyro sensor 51 and an acceleration sensor 52 built therein. As a result, pressure detection signals PS1 to PS4, roll axis angle detection signal PM, and pitch angle ωp, yaw angle ωyaw, X, Y, and Z axis angle signals are extracted from the posture sensor circuit 45.

  These pressure detection signals PS1 to PS4, roll axis angle detection signal PM, and pitch angle ωp, yaw angle ωyaw, X, Y, and Z axis angle signals are supplied to the central controller 46 in the controller 13. Further, an operation signal from the power switch 18 is supplied to the central controller 46. As a result, the central controller 46 calculates the left and right wheel rotation commands ωref1 and ωref2 and supplies them to the motor control units 32L and 32R. Further, signals from the encoders 34L and 34R are supplied to the motor control units 32L and 32R, and the motors 12L and 12R are driven.

  Further, power from the battery 17 is supplied to the power supply circuit 47. For example, 24V motor power from the power supply circuit 47 is supplied to the motor control units 32L and 32R. For example, 5V control power is supplied to the attitude sensor circuit 45 and the central controller 46. The power supply circuit 47 is provided with a power switch 48 to control the supply of power to each unit. In this manner, the motors 12L and 12R are driven, and the wheels 11L and 11R are driven by these motors 12L and 12R, so that the coaxial two-wheel vehicle 10 travels.

  However, in such a coaxial two-wheeled vehicle, for example, when making a sudden turn, the vehicle is tilted as shown in FIG. Alternatively, when the occupant's center of gravity tilts on a road surface with a lateral slope, one wheel is brought into a non-contact state. Further, when the vehicle is lifted while a person is getting off, the vehicle is similarly ungrounded. In that case, in the prior art, the non-grounded tire rotates in synchronization with the grounded tire or rotates according to the magnitude of torque applied to the motor. For this reason, problems such as wasteful energy consumption occur in the non-grounded motor.

  Alternatively, when the road surface is wet with water or the like, the frictional force of the road surface is reduced and the tire is likely to slip. In this case, in the prior art, when the posture is inclined due to slip, the control is performed so that the motor generates more torque, so that the tire rotates at a high speed, and further, the slip state may be promoted. As a result, the posture becomes unstable, and there is a risk that good running cannot be performed. In addition, a large amount of energy is consumed thereby, resulting in a problem that the battery life is shortened. The present invention has been made in view of these points.

  By the way, the tire moment of inertia Is when the tire is driven is expressed, for example, as shown in FIG. Further, signals of respective parts of the control circuit relating to the tire inertia moment Is are shown in FIG. 11B. Furthermore, the control model of such a tire system can be expressed as shown in FIG. 12, for example. Such a control model is described, for example, in “SEV / Electric Vehicle Study Group, 38th Kanto Region Regular Study Group” “Composition and Prospect of Image Adhesion Control of Electric Vehicles” announced by Yoichi Hori of the University of Tokyo. (Http://mizugaki.iis.u-tokyo.ac.jp/staff/hori/paperPDF/SEV98.pdf).

In such a control model, the road surface frictional force Fdr is expressed by the following equation.
Fdr = μ × W [N] (Formula 1)
Tw = μ × r × W-Is * (dVb / dt) [Nm] (Formula 2)
Where μ: friction coefficient r: wheel effective radius [m]
W: Wheel load [kg] Is: Wheel inertia moment
Vb: Body speed [m / SEC] Tw: Drive torque [Nm]

  In this case, the slip is considered to occur when the tire driving torque Tw exceeds the road surface friction Fdr. Therefore, for example, if the road surface friction Fdr can be observed, the slip can be predicted and prevented by comparing with the tire torque Tw. Furthermore, since the tire can also predict the road surface and the non-grounded state based on the same principle, it is possible to suppress tire slipping. That is, it is considered that the road surface friction Fdr is observed by some means, and it can be determined that slip is generated when the tire driving torque Tw exceeds it, and control for suppressing idling of the tire can be performed.

  This application has been made in view of the above points, and the problem to be solved is that, in the conventional apparatus, for example, the vehicle is tilted by a centrifugal force during a sudden turn, and one wheel is not grounded. If a slip occurs when the road surface is wet with water or the like, the posture becomes unstable, and good running cannot be performed. In addition, there have been problems such as tires slipping a large amount of energy due to tire idling and shortening the battery life, but no means for solving such a situation has been proposed.

  Therefore, in the present invention, an arbitrary road surface friction force estimation value is applied to the mathematical model of the wheel drive system, and the rotation value of the arbitrary part of the wheel drive system of the actual vehicle and the simulated rotation of the corresponding part of the mathematical model Control the estimated frictional force value of the road so that it is equal to the value, and limit the motor torque command signal so that the motor torque of the wheel drive system does not exceed the estimated frictional force value applied to the mathematical model Therefore, according to this, it is possible to perform control for suppressing the idling of the tires satisfactorily.

According to the first aspect of the present invention, the rotation value of an arbitrary part of the mathematical model of the wheel drive system is applied to the mathematical model so that the simulated rotation value of the corresponding part of the wheel drive system of the actual vehicle is equal. Control the estimated value of the road surface friction force, limit the magnitude of the motor torque command signal so that the motor torque of the wheel drive system does not exceed the estimated value of the road surface friction force , and rotate any part of the wheel drive system When the calculated motor torque is greater than the estimated road friction force value during a period in which the difference between the value and the simulated rotation value of the corresponding part of the mathematical model is large, (estimated road friction force value / calculated A means for multiplying the motor torque command value by the motor torque command signal is provided to limit the magnitude of the motor torque command signal, and the rotation value of any part of the wheel drive system and the corresponding part of the mathematical model are simulated Difference from rotation value Small period, when the motor torque calculated is larger than the road friction force estimated value, by performing a limitation of the magnitude of the motor torque command signal is provided with means for multiplying a value of 0 to the motor torque command signal Thus, it is possible to perform control that suppresses tire slipping satisfactorily in a transient state and a steady state .

Further, according to the invention of claim 2 , the rotation value of an arbitrary part of the mathematical model of the wheel drive system is applied to the mathematical model so that the simulated rotation value of the corresponding part of the wheel drive system of the actual vehicle is equal. Control the estimated value of the road surface friction force, limit the magnitude of the motor torque command signal so that the motor torque of the wheel drive system does not exceed the estimated value of the road surface friction force , The limitation is that when the calculated motor torque is greater than the estimated value of road friction force during the period when the difference between the rotation value of any part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model is large In addition, the value of (estimated road surface friction force / calculated motor torque) is multiplied by the motor torque command signal, and the rotation value of an arbitrary part of the wheel drive system and the simulated rotation of the corresponding part of the mathematical model Small difference from the value There period, when the motor torque calculated is larger than the road friction force estimated value, by performing by multiplying the value of 0 to the motor torque command signal, the better idling of the tire in the transient state and steady state It is possible to realize a control method that suppresses.

  As a result, in conventional devices, for example, when the vehicle is tilted due to centrifugal force during a sudden turn and one wheel is not in contact with the ground, or when the road surface is wet with water, the posture is not correct. It becomes stable and can not run well. Further, according to the present invention, it is possible to provide a means for easily solving these problems, such as a problem that a large amount of energy is consumed due to tire idling and a battery life is shortened. .

That is, the traveling device of the present invention is a traveling device that travels while controlling the driving of the wheels, and includes a control unit that calculates a motor torque in accordance with a supplied rotation command and generates a motor torque command signal; Detection means for detecting a rotation value of an arbitrary part of the wheel drive system driven by the generated motor torque command signal, and a mathematical model of the wheel drive system that is simulated by supplying the generated motor torque command signal In the mathematical model, an estimated value of road friction force is applied, and the road friction force applied so that the rotation value of any part of the wheel drive system is equal to the simulated rotation value of the corresponding part of the mathematical model. controls estimate, in the control unit, the motor torque calculated limits the magnitude of the motor torque command signal so as not to exceed the road surface friction force estimated value, and the rotation value of any portion of the wheel drive system When the calculated motor torque is larger than the estimated road friction force value during the period when the difference between the simulated rotation value of the corresponding part of the scientific model is large, (estimated road friction force value / calculated motor torque) ) Is multiplied by the motor torque command signal to limit the magnitude of the motor torque command signal, and the rotation value of any part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model When the calculated motor torque is greater than the estimated value of the road surface frictional force, a means for multiplying the motor torque command signal by 0 is provided to limit the magnitude of the motor torque command signal. Do it.

The traveling device control method of the present invention is a traveling device control method that travels while controlling the driving of wheels, and calculates motor torque in accordance with a supplied rotation command, and outputs a motor torque command signal. Rotation value of any part of the wheel drive system that is generated and driven by the generated motor torque command signal is detected, and a mathematical model of the wheel drive system that is simulated by supplying the generated motor torque command signal is provided Apply the estimated frictional force value to the mathematical model, and apply the estimated frictional force value so that the rotation value of any part of the wheel drive system is equal to the simulated rotation value of the corresponding part of the mathematical model. controls, motor torque calculated limits the magnitude of the motor torque command signal so as not to exceed the road surface friction force estimated value, the motor torque command signal magnitude limit, the wheel driving system When the calculated motor torque is greater than the estimated value of road friction force during the period when the difference between the rotation value of the desired part and the simulated rotation value of the corresponding part of the mathematical model is large, Value / calculated motor torque) is multiplied by the motor torque command signal, and the difference between the rotational value of an arbitrary part of the wheel drive system and the simulated rotational value of the corresponding part of the mathematical model is small Is performed by multiplying the motor torque command signal by a value of 0 when the calculated motor torque becomes larger than the estimated value of the road frictional force .

  The present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing configurations of an embodiment of a road surface reaction force estimator and an anti-slip adjuster to which a traveling device and a control method thereof according to the present invention are applied.

  In FIG. 1, the upper part of the figure is the same as the control model of the tire system shown in FIG. That is, the rotation command ωref is supplied to the input terminal 1 in FIG. 1, and this rotation command ωref is supplied to the speed controller 102 through the subtractor 101 of the motor control device 100. The speed controller 102 performs PI (Proportional / Integral) control and has a characteristic of [Kp + (Ki / S)] as shown in the figure.

  The output signal of the speed controller 102 is supplied to the slip adjuster 103. The slip adjuster 103 multiplies the supplied signal by a coefficient [Ksadj]. As a result, the motor torque command Tref [Nm] is extracted from the slip adjuster 103, and the motor torque command Tref [Nm] is converted into the motor current Im [A] by the amplifier 104 having the gain [Kamp].

  Further, the motor current Im [A] converted by the amplifier 104 is supplied to the motor of the actual model 200. This motor is represented by a motor constant [Km] 201, and a motor torque output Tm [Nm] is extracted from the motor constant 201. The motor torque output Tm [Nm] is supplied to the subtractor 202, and the external force torque Td [Nm] is subtracted to extract the rotor torque Tr [Nm].

  The rotor torque Tr [Nm] is supplied to the motor rotor 203 having the characteristic of [1 / (JmS + Dm)], and the rotor rotational angular velocity ωm [rad / esc] is taken out. Then, the rotor rotation angular velocity ωm [rad / esc] is supplied to the computing unit 204 having the characteristic of [1 / S], and the rotor rotation angle Θm [rad] is calculated. The calculated rotor rotation angle Θm [rad] is taken out to the output terminal 2 and fed back to the subtractor 101 through the arithmetic unit 105 having the characteristic [S].

  Further, the rotor rotation angle Θm [rad] from the computing unit 204 is supplied to a [1 / N] converter 205 constituting a reduction gear having a reduction ratio N: 1, and the reduced signal is transmitted through a subtractor 206 to a spring. The constant [Ks] 207 is supplied to be converted into the tire torque Tw [Nm]. The tire torque Tw [Nm] from the spring constant [Ks] 207 is supplied to the subtractor 202 as the external force torque Td [Nm] through the [N] converter 208 constituting the reduction gear.

  Further, the tire torque Tw [Nm] from the spring constant [Ks] 207 is added to the road surface friction torque Fdr × r [Nm] by the adder 209 and supplied to the tire 210 having the characteristic of [1 / (JtS + Dt)]. Is done. The tire 210 is supplied to the calculator 211 having a rotational angular velocity of [1 / S], and the tire rotational angle θt [rad] is taken out and supplied to the subtractor 206.

  Thereby, in the actual model 200, the motor constant 201 is driven according to the motor current Im [A], and the external torque Td [Nm] is subtracted from the motor torque output Tm [Nm], and the subtracted rotor torque Tr is subtracted. Based on [Nm], the rotor rotation angle Θm [rad] is extracted. In this case, the external force torque Td [Nm] subtracted from the motor torque output Tm [Nm] is affected by the road surface friction torque Fdr × r [Nm], and the road surface friction torque Fdr is applied to the output terminal 2. The rotor rotation angle Θm [rad] with respect to the motor current Im [A] under the influence of xr [Nm] is extracted.

  For such an actual model 200, a mathematical model 300 for simulating the above-described actual model 200 is provided in the motor control apparatus 100. The mathematical model 300 is a program that analyzes the function of the real model 200 and realizes the function by computer simulation. Therefore, although an embodiment of such a program takes the form of a program list or the like, the following description will be made using functional blocks shown in FIG. 1 for the purpose of facilitating understanding.

  In this mathematical model 300, the motor current Im [A] converted by the amplifier 104 is supplied to the motor model constant [Kmm] 301 of the mathematical model 300, and the motor torque output Tmm [Nm] is obtained from the motor model constant 301. It is taken out. The motor torque output Tmm [Nm] is supplied to the subtracter 302, and the external model estimated torque Tmd [Nm] is subtracted to extract the rotor model torque Tmr [Nm].

  The rotor model torque Tmr [Nm] is supplied to the motor rotor model 303 having the characteristic of [1 / (JmmS + Dmm)], and the rotor model rotational angular velocity ωmm [rad / esc] is extracted. The rotor rotation model angular velocity ωmm [rad / esc] is supplied to the calculator 304 having the characteristic [1 / S], and the rotor model rotation angle Θmm [rad] is calculated. The calculated rotor model rotation angle Θmm [rad] is taken out to the output terminal 3.

  Further, the rotor model rotation angle Θmm [rad] from the calculator 304 is supplied to the converter 305 of [1 / Nm] (where Nm = N) constituting a reduction gear model having a reduction ratio N: 1, and the speed reduction The obtained signal is supplied to the spring model constant [Kms] 307 through the subtractor 306 and converted into the tire model torque Tmw [Nm]. The tire model torque Tmw [Nm] from the spring model constant [Kms] 307 is supplied to the subtractor 302 as the external force estimated torque Tmd [Nm] through the [N] converter 308 constituting the reduction gear model.

  Further, the tire model torque Tmw [Nm] from the spring model constant 307 is added to the road surface frictional force estimated value Tidr [Nm] by the adder 309 and supplied to the tire model 310 having the characteristic of [1 / (JmtS + Dmt)]. The The tire model 310 is supplied to the calculator 311 having a rotational angular velocity of [1 / S], and the tire model rotational angle θmt [rad] is extracted and supplied to the subtractor 306.

  Thereby, in the mathematical model 300, the motor model constant 301 is driven according to the motor current Im [A], and the external force estimated torque Tmd [Nm] is subtracted from the motor model torque output Tmm [Nm]. Based on the rotor model torque Tmr [Nm], the rotor model rotation angle Θmm [rad] is extracted.

  In this case, the external force estimated torque Tmd [Nm] subtracted from the motor model torque output Tmm [Nm] is affected by the road surface frictional force estimated value Tidr [Nm]. The rotor model rotation angle Θmm [rad] with respect to the motor current Im [A] under the influence of the force estimation value Tidr [Nm] is extracted.

  That is, the rotor model rotation angle Θmm [rad] with respect to the motor current Im [A] under the influence of the road surface frictional force estimated value Tidr [Nm] is extracted from the output terminal 3 of the mathematical model 300. . On the other hand, the rotor rotation angle Θm [rad] with respect to the motor current Im [A] under the influence of the road surface friction torque Fdr × r [Nm] is extracted from the output terminal 2 of the actual model 200.

  Therefore, the road surface frictional force estimated value Tidr [Nm] is controlled so that these rotor model rotation angle Θmm [rad] and rotor rotation angle Θm [rad] are equal. Thereby, the controlled road surface friction force estimated value Tidr [Nm] can be made equal to the road surface friction torque Fdr × r [Nm].

  This is because, in the above-described actual model 200 and mathematical model 300, the requirements other than the road surface friction torque Fdr × r [Nm] and the road surface friction force estimated value Tidr [Nm] are considered to be exactly the same, and the rotor rotation angle Θm This is because the difference between [rad] and the rotor model rotation angle Θmm [rad] is considered to be caused only by the difference between the road surface friction torque Fdr × r [Nm] and the road surface friction force estimated value Tidr [Nm].

However, in the relationship between the actual model 200 and the mathematical model 300 described above, the values of each element are Kmm = Km [Nm / A], Jmm = Jm [kg / sec ² ], Dmm = Dm [Nm / (rad / sec). )], Nm = N, Kms = Ks [Nm / rad], Jmt = Jt [kg / sec ² ], and Dmt = Dt [Nm / (rad / sec)]. Such control is referred to as model reference adaptive control.

  In order to realize the above-described control, in the configuration of FIG. 1, the rotor model rotation angle Θmm [rad] and the rotor rotation angle Θm [rad] are supplied to the subtractor 106 to extract the difference model error Em. The model error Em is supplied to the estimator 400 for gain [Kadp + (Kadi / S)], and the value of the road frictional force estimated value Tidr [Nm] formed by the estimator 400 is controlled.

  As a result, a road surface frictional force estimated value Tidr [Nm] equal to the road surface friction torque Fdr × r [Nm] is extracted. Further, in the configuration of FIG. 1, the road surface frictional force estimated value Tidr [Nm] taken out is supplied to a setting unit 107 for a coefficient [Ksadj] supplied to the slip adjuster 103 to set the coefficient [Ksadj]. .

  FIG. 2 is a flowchart showing an embodiment of processing executed by the motor control device 100 up to the setting of the coefficient [Ksadj].

  In FIG. 2, when the slip prevention control is started, the control parameters are set in the first step S1, and the control gains Kp and Ki of the speed controller 102 proportional to the load weight W [kg] are set. Next, in step S <b> 2, a motor torque command Tref is calculated by the speed controller 102 and the slip prevention adjuster 103 of the motor control device 100.

  In step S3, the actual motor is driven by the motor torque command Tref. In step S4, the model error Em is calculated by comparing the encoder signal (rotor rotation angle Θm [rad]) of the actual motor with the mathematical model signal (rotor model rotation angle Θmm [rad]). In step S5, a road surface reaction force (friction force) estimated value Tidr [Nm] is calculated by processing in the estimator 400.

In step S6, the coefficient [Ksadj] of the slip prevention adjuster 103 is varied. The coefficient [Ksadj] here is set by the setting device 107.
If | Tidr | <Tw0 then Ksadj = | Tidr / Tw0 |
Else Ksadj = 1
Ask for. However, the default value is a decimal value of Ksadj = 1, 0 ≦ Ksadj ≦ 1.

  In step S7, it is determined whether or not the model error | Em | is smaller than an arbitrary reference value. If not (NO), the process returns to step S1. Therefore, in this case, control is performed with the coefficient [Ksadj] determined in step S6 described above, and the actual motor is driven with a torque equal to or less than the road surface frictional force estimated value Tidr [Nm].

  If the model error | Em | is small in step S7 (YES), the motor torque and the road surface reaction force are further compared in step S8. If | Tw | <| Tidr | Returned to Therefore, also in this case, control is performed by the coefficient [Ksadj] determined in step S6.

  On the other hand, when | Tw | <| Tidr | is not satisfied in step S8 (NO), the slip prediction condition is set in step S9, Ksadj = 0 is set, and the process returns to step S1. Therefore, in this case, the motor torque command Tref [Nm] taken out from the slip prevention adjuster 103 is 0, and the actual motor is not driven.

  Therefore, in the above-described embodiment, the rotation value of an arbitrary part of the mathematical model of the wheel drive system is applied to the mathematical model so that the simulated rotation value of the corresponding part of the wheel drive system of the actual vehicle is equal. By controlling the estimated value of the road surface frictional force and limiting the magnitude of the motor torque command signal so that the motor torque of the wheel drive system does not exceed the estimated value of the road surface frictional force, it is possible to suppress tire idling well. Control can be performed.

  As a result, in conventional devices, for example, when the vehicle is tilted due to centrifugal force during a sudden turn and one wheel is not in contact with the ground, or when the road surface is wet with water, the posture is not correct. It becomes stable and can not run well. Further, according to the present invention, it is possible to provide a means for easily solving these problems, such as a problem that a large amount of energy is consumed due to tire idling and a battery life is shortened. .

  Further, FIG. 3 shows a signal state of the apparatus of the present invention by simulation. FIG. 3 shows a change in which the road surface reaction force is estimated by the estimator when the road surface reaction force is changed in the simulation, and a state in which the road surface reaction force is estimated by the apparatus of the present invention.

  That is, in FIG. 3, the motor control device 100 includes a subtractor 151, amplifiers 152 and 153, an integrator 154, an adder 155, and a differentiator 156. The real model 200 includes subtracters 251 and 252, amplifiers 253, 254 and 255, calculators 256 and 257, integrators 258 and 259, and an adder 260.

  Further, the mathematical model 300 includes subtracters 351 and 352, amplifiers 353, 354, and 355, calculators 356 and 357, integrators 358 and 359, and an adder 360. The estimator 400 includes a subtracter 451, amplifiers 452 and 453, an integrator 454, and an adder 455.

  In such an apparatus, for example, a sine wave signal as shown in the waveform diagram A in the figure is supplied to the subtractor 151, and a road surface friction torque Fdr having a step waveform as shown on the upper side of the waveform diagram B in the figure. When xr [Nm] is supplied to the adder 260, a motor torque command Tref [Nm] as shown in the waveform diagram C in the figure is formed, and the rotor rotation angle Θm [rad] as shown in the waveform diagram D in the figure is formed. ] Is taken out. As a result, the road surface frictional force estimated value Tidr [Nm] is changed as shown on the upper side of the waveform diagram B in the figure.

  Therefore, when such control is performed, for example, when the normal road surface state is changed to the low road surface friction state in FIG. 4, the road surface friction force Fdr [Nm] is conventionally shown in FIG. FIG. 4B shows the case where the tire torque [Nm] is maintained as it is and the motor rotational speed [rad / esc] is increased even if the tire torque is lower than the tire torque [Nm]. Thus, the tire torque [Nm] is reduced, and the motor rotational speed [rad / esc] is controlled to decrease.

  For example, in FIG. 5, when the road surface contact state changes to the road surface non-contact state, conventionally, as shown in FIG. 5A, even if the road surface frictional force Fdr [Nm] is less than the tire torque [Nm], According to the present invention, the tire torque [Nm] is reduced to 0 as shown in FIG. 5B, where the tire torque [Nm] is maintained and the motor rotational speed [rad / esc] is increased. The control is performed so that the motor rotation speed [rad / esc] is zero.

  Thus, according to the traveling device of the present invention, the traveling device performs traveling while controlling the driving of the wheel, and includes control means for calculating the motor torque according to the supplied rotation command and generating the motor torque command signal. Detecting means for detecting a rotation value of an arbitrary part of the wheel drive system driven by the generated motor torque command signal, and a mathematical model of the wheel drive system simulated by supplying the generated motor torque command signal The mathematical model has a road surface friction force estimation value applied so that the rotational value of any part of the wheel drive system is equal to the simulated rotational value of the corresponding part of the mathematical model. By controlling the estimated frictional force value, the control means restricts the magnitude of the motor torque command signal so that the calculated motor torque does not exceed the estimated road frictional force value. It is possible to perform control to suppress rolling.

  According to the traveling device control method of the present invention, the traveling device control method performs traveling while controlling the driving of the wheels, and calculates the motor torque in accordance with the supplied rotation command, and the motor torque command. Mathematical model of a wheel drive system that generates a signal, detects a rotation value of an arbitrary part of the wheel drive system driven by the generated motor torque command signal, and is simulated by supplying the generated motor torque command signal The road surface friction force applied to the mathematical model is applied so that the rotation value of any part of the wheel drive system is equal to the simulated rotation value of the corresponding part of the mathematical model. Control that suppresses tire slipping satisfactorily by controlling the estimated value and limiting the magnitude of the motor torque command signal so that the calculated motor torque does not exceed the estimated road frictional force value The law can be realized.

  The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

It is a block diagram which shows the structure of one Embodiment of the road surface reaction force estimator and the slip prevention adjuster to which the traveling apparatus by this invention and its control method are applied. It is a flowchart figure for description of the operation. It is a figure for the description. It is a figure for the description. It is a figure for the description. It is a block diagram which shows one Embodiment of the traveling apparatus with which this invention is applied. It is a figure for the description. It is a figure for the description. It is a figure for the description. It is a figure for the description. It is a figure for the description. It is a figure for the description.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Input terminal, 2, 3 ... Output terminal, 100 ... Motor controller, 101, 106 ... Subtractor, 102 ... Speed controller, 103 ... Slip adjuster, 104 ... Amplifier, 105 ... Calculator, 107 ... Coefficient Setting unit, 200 ... real model, 201 ... motor constant, 202, 206 ... subtractor, 203 ... motor rotor, 204, 211 ... calculator, 205, 208 ... converter, 207 ... spring constant, 209 ... adder, 210 ... Tire, 300 ... mathematical model, 301 ... motor model constant, 302, 306 ... subtractor, 203 ... motor model rotor, 204, 311 ... arithmetic unit, 205, 208 ... converter, 307 ... spring model constant [Ks], 309 ... Adder, 310 ... tire model, 400 ... estimator

Claims (2)

A traveling device that travels while controlling the driving of wheels,
Control means for calculating a motor torque in accordance with a supplied rotation command and generating a motor torque command signal;
Detecting means for detecting a rotation value of an arbitrary part of a wheel drive system driven by the generated motor torque command signal;
A mathematical model of the wheel drive system that is simulated by the supply of the generated motor torque command signal,
The mathematical model is applied with an estimated value of road frictional force so that the rotational value of an arbitrary part of the wheel drive system is equal to the simulated rotational value of the corresponding part of the mathematical model. Control the estimated road friction force,
In the control means,
Limiting the magnitude of the motor torque command signal so that the calculated motor torque does not exceed the road frictional force estimated value ;
During a period in which the difference between the rotation value of an arbitrary part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model is large, the calculated motor torque is larger than the estimated road friction force value. In some cases, a means for multiplying the motor torque command signal by a value of (the road frictional force estimated value / the calculated motor torque) is provided to limit the magnitude of the motor torque command signal,
During a period in which the difference between the rotation value of an arbitrary part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model is small, the calculated motor torque is greater than the estimated road friction force value. In some cases, the traveling device is characterized in that means for multiplying the motor torque command signal by a value of 0 is provided to limit the magnitude of the motor torque command signal . A method of controlling a traveling device that travels while controlling the driving of wheels,
Calculate motor torque according to the supplied rotation command and generate a motor torque command signal,
While detecting the rotation value of any part of the wheel drive system driven by the generated motor torque command signal,
Providing a mathematical model of the wheel drive system that is simulated by the supply of the generated motor torque command signal;
Applying an estimated road friction force to the mathematical model, and applying the applied road friction so that the rotation value of any part of the wheel drive system is equal to the simulated rotation value of the corresponding part of the mathematical model Control force estimates,
Limiting the magnitude of the motor torque command signal so that the calculated motor torque does not exceed the road frictional force estimated value ;
The limit of the magnitude of the motor torque command signal is
During a period in which the difference between the rotation value of an arbitrary part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model is large, the calculated motor torque is larger than the estimated road friction force value. When (the road frictional force estimated value / the calculated motor torque) is multiplied by the motor torque command signal,
During a period in which the difference between the rotation value of an arbitrary part of the wheel drive system and the simulated rotation value of the corresponding part of the mathematical model is small, the calculated motor torque is greater than the estimated road friction force value. In some cases, the control method of the traveling device is performed by multiplying the motor torque command signal by a value of 0 .

Images