Patent Grant
7222007
The present invention relates generally to a control apparatus for controlling a system of an automotive vehicle in response to sensed dynamic behavior, and more specifically, to a method and apparatus for controlling the system of the vehicle by determining attitude of the vehicle.
In recent years, many vehicle control systems have been developed to enhance vehicle stability and tracking performance in critical dynamic situations. Such vehicle control systems include yaw stability control systems, roll stability control systems, integrated vehicle dynamic control systems, etc. In these systems, knowledge of the vehicle roll and pitch attitude is very important. For example, in yaw stability control systems the effect of vehicle body roll and pitch, as well as the dynamically changing road super-elevations and road grades is significant because they directly influence both the vehicle lateral dynamics and lateral acceleration measurements. In roll stability control systems, roll angle is one of the most important variables used to construct feedback pressure command and combat the detected roll instability. Hence a successful vehicle dynamics control must involve an accurate determination of the vehicle roll and pitch attitude. However, these values are not directly measured on production vehicles and therefore must be estimated instead.
The vehicle state estimation algorithms implemented on a production vehicle for vehicle dynamic control purpose is normally based on dead reckoning sensors only, such as wheel/steering encoders and inertia sensors which are utilized to predict the vehicle's high frequency behavior. The vehicle state estimates can be obtained from a physical vehicle model, or via integration of the inertial sensor signals, or a combination of both. The estimation accuracy, however, can be very crude for a lot of maneuvers/road conditions, which in turn severally limits the control performance. One reason is that the vehicle model is only effective in the linear region. Another, perhaps more important, reason is that there is simply not enough inertia information. In order to accurately estimate vehicle states in all operating modes, a full six-degree-of-freedom inertial measurement unit (IMU) may be used. A typical IMU consists of three accelerometers and three gyroscopes mounted in a set of three orthogonal axes. The IMU measures the acceleration and the rotation rate of the vehicle in all three dimensions at a high sampling rate, typically at frequencies higher than 100 Hz. From this information, attitude and velocity of the vehicle can be derived via mathematical integration. Vehicle position and heading are generally not observable without external information.
Recent progress in the development of Micro-Electro Mechanical Systems (MEMS) has made it possible to put IMU on production vehicles because of their small size, low cost and ruggedness. The reduction in size and cost, especially cost, however, has also led to a drop in accuracy of the inertial unit as a whole. The predominant error sources in the inertial sensors, whether they are gyros or accelerometers, are bias, scale factors and random walk. These errors are added up via mathematical integration, and may lead to large drifts in the attitude and velocity estimates, unless external absolute sensors are used to constantly bound the errors.
In practice, all inertia sensing systems are aided in some way by low frequency external sensors, such as global positioning system (GPS), Doppler radar, star trackers to name a few. Due to the increasing popularity and decreasing cost of GPS, a lot of effort has been devoted to the development of GPS aided inertial systems for vehicle control purpose. While fairly good estimation accuracy can be attained in open sky environment using this approach, the performance deteriorates when the satellite signals bounce off of reflective surfaces such as tall buildings and other structures in the “urban canyon”. In the worst case, when fewer than three or four satellites can be “seen” (i.e., driving through a tunnel), the GPS provides no information to bound the errors associated with high frequency inertia sensors. Another disadvantage is that GPS device is not at all common and/or cost effective on current production vehicles.
Therefore, there is a significant need for a low-cost device that provides accurate and robust estimate of the vehicle global attitude.
It is the primary objective of the present invention to provide a methodology of estimating the vehicle global roll angle and pitch angle. The present invention determines an estimated vehicle attitude that can be used to initiate control commands for various subsystems including but not limited to powertrain controls, brake controls, steering controls, suspension controls and passive safety devices. Another application of vehicle global attitude is sensor fault detection.
One embodiment uses the following dead reckoning sensors: (i) a low-cost strapdown IMU sensor cluster, (ii) a steering wheel angle sensor, (iii) and wheel speed sensors. The proposed methodology utilizes the kinematic relationship among sensor signals, a bicycle model, and the nonholonomic constraints for a vehicle moving on a surface. The vehicle global attitude is obtained via a fusion of the data from all the sensors.
In another embodiment, a system for controlling a safety system of an automotive vehicle includes a longitudinal acceleration sensor generating a longitudinal acceleration signal, a vehicle speed sensor generating a vehicle speed signal, a lateral acceleration sensor generating a lateral acceleration signal, a yaw rate sensor generating a yaw rate signal and a controller. The controller determines a reference pitch in response to the longitudinal acceleration signal and the vehicle speed signal and a reference roll angle in response to the yaw rate signal, the wheel speed signal and the lateral acceleration signal. The controller determines a roll stability index and a pitch stability index. The controller determines an adjusted pitch angle in response to the reference pitch angle and the pitch stability index and an adjusted roll angle in response to the reference roll angle and the roll stability index. The controller controls the safety system 44 in response to the adjusted roll angle and the adjusted pitch angle.
The present invention provides a technique of qualifying different sensor signals so that they can be fused to accurately estimate the vehicle attitude. A number of criteria are proposed for identifying cases that are not suitable for using one sensor signal but suitable for using others. As a result, the proposed sensing algorithm is robust to sensor bias and noise, vehicle maneuvers, vehicle parameter variation, road disturbances and the friction coefficient between the tires and the road.
The present invention allows the vehicle performance to be optimized for ride, safety and fuel economy by providing an accurate estimate of the vehicle attitude. Even in future vehicle models equipped with standard GPS devices, the proposed methodology is able to help achieve desired performance when sky-obstruction/GPS faults occur.
Other advantages and features of the present invention will become apparent when viewed in light of the detailed description of the preferred embodiment when taken in conjunction with the attached drawings and appended claims.
In the following figures the same reference numerals will be used to identify the same components.
The present invention may be used in conjunction with a rollover control system for a vehicle. However, the present invention may also be used with a deployment device such as airbag or roll bar. The present invention will be discussed below in terms of preferred embodiments relating to an automotive vehicle moving in a three-dimensional road terrain.
Referring to
As mentioned above, the system may also be used with active/semi-active suspension systems, anti-roll bar or other safety devices deployed or activated upon sensing predetermined dynamic conditions of the vehicle.
The sensing system 16 is coupled to a control system 18. The sensing system 16 may use a standard yaw stability control sensor set (including lateral accelerometer, yaw rate sensor, steering angle sensor and wheel speed sensor) together with a roll rate sensor and a longitudinal accelerometer. The various sensors will be further described below. The wheel speed sensors 20 are mounted at each corner of the vehicle, and the rest of the sensors of sensing system 16 are preferably mounted directly on the center of gravity of the vehicle body, along the directions x, y and z shown in
The angular rate sensors and the accelerometers are mounted on the vehicle car body along the body frame directions b1, b2 and b3; which are the x-y-z axes of the vehicle's sprung mass.
The longitudinal acceleration sensor is mounted on the car body located at the center of gravity, with its sensing direction along b1-axis, whose output is denoted as αx. The lateral acceleration sensor is mounted on the car body located at the center of gravity, with its sensing direction along b2-axis, whose output is denoted as αy.
The other frame used in the following discussion includes the road frame, as depicted in
In the following discussion, the Euler angles of the body frame b1b2b3 with respect to the road frame r1r2 r3 are denoted as θxbr, θybr and θzbr, which are also called the relative Euler angles.
The present invention estimates the relative Euler angles θxbr and θybr based on the available sensor signals and the signals calculated form the measured values.
Referring now to
In the preferred embodiment the sensors are located at the center of gravity of the vehicle. Those skilled in the art will recognize that the sensor may also be located off the center of gravity and translated equivalently thereto.
Lateral acceleration, roll orientation and speed may be obtained using a global positioning system (GPS). Based upon inputs from the sensors, controller 26 may control a safety device 44. Depending on the desired sensitivity of the system and various other factors, not all the sensors 28–38 may be used in a commercial embodiment. Safety device 44 is part of a vehicle subsystem control. Safety device 44 may control a passive safety device 46 such as an airbag or a steering actuator 48, a braking actuator 50 at one or more of the wheels 12a, 12b, 13a, 13b of the vehicle. Engine intervention 52 may act to reduce engine power to provide a safety function. Also, other vehicle components such as a suspension control 54 may be used to adjust the suspension to prevent rollover.
Roll rate sensor 34 and pitch rate sensor 37 may sense the roll condition of the vehicle based on sensing the height of one or more points on the vehicle relative to the road surface. Sensors that may be used to achieve this include a radar-based proximity sensor, a laser-based proximity sensor and a sonar-based proximity sensor.
Roll rate sensor 34 and pitch rate sensor 37 may also sense the roll condition based on sensing the linear or rotational relative displacement or displacement velocity of one or more of the suspension chassis components which may include a linear height or travel sensor, a rotary height or travel sensor, a wheel speed sensor used to look for a change in velocity, a steering wheel position sensor, a steering wheel velocity sensor and a driver heading command input from an electronic component that may include steer by wire using a hand wheel or joy stick.
The roll condition may also be sensed by sensing the force or torque associated with the loading condition of one or more suspension or chassis components including a pressure transducer in an act of air suspension, a shock absorber sensor such as a load cell, a strain gauge, the steering system absolute or relative motor load, the steering system pressure of the hydraulic lines, a tire laterally force sensor or sensors, a longitudinal tire force sensor, a vertical tire force sensor or a tire sidewall torsion sensor.
The roll condition of the vehicle may also be established by one or more of the following translational or rotational positions, velocities or accelerations of the vehicle including a roll gyro, the roll rate sensor 34, the yaw rate sensor 28, the lateral acceleration sensor 32, a vertical acceleration sensor, a vehicle longitudinal acceleration sensor, lateral or vertical speed sensor including a wheel-based speed sensor, a radar-based speed sensor, a sonar-based speed sensor, a laser-based speed sensor or an optical-based speed sensor.
Steering control 48 may control the position of the front right wheel actuator, the front left wheel actuator, the rear left wheel actuator, and the right rear wheel actuator. Although as described above, two or more of the actuators may be simultaneously controlled. For example, in a rack-and-pinion system, the two wheels coupled thereto are simultaneously controlled. Based on the inputs from sensors 28 through 38, controller 26 determines a roll condition and controls the steering position of the wheels.
Speed sensor 20 may be one of a variety of speed sensors known to those skilled in the art. For example, a suitable speed sensor may include a sensor at every wheel that is averaged by controller 26. Preferably, the controller translates the wheel speeds into the speed of the vehicle. Yaw rate, steering angle, wheel speed and possibly a slip angle estimate at each wheel may be translated back to the speed of the vehicle at the center of gravity. Various other algorithms are known to those skilled in the art. Speed may also be obtained from a transmission sensor. For example, if speed is determined while speeding up or braking around a corner, the lowest or highest wheel speed may not be used because of its error. Also, a transmission sensor may be used to determine vehicle speed.
Controller 26 may include a reference signal generator 58 and a global attitude observer 60. While these functions are provided by controller 26, several controllers may be used to provide the same functions. The controller 26 may be programmed to provide both of the functions among other functions. A global roll and global pitch angle are provided by the global attitude observer 60 which is then provided to device 44. One, several, or all of the safety devices in the vehicle may use global pitch and global roll angles determined by the global attitude observer 60.
Referring now to
A reference roll calculation block 64 determines a reference roll angle in response to a yaw rate signal from a yaw rate sensor 28 and a lateral acceleration signal from a lateral acceleration sensor 32. A derivation of these signals will be further described below.
Referring now to
An IMU kinematic equation generator 70 is coupled to the roll rate sensor 34, the pitch rate sensor 37, and the yaw rate sensor 38. The IMU kinematic equation generator generates a plurality of kinematic signals that are integrated to determine the Euler rates of change of the pitch and roll of the vehicle. The output of the IMU kinematic equation generator 70 and the output 68C of tuning block 68 are provided to a summing block 72. The adjustments referred to below as Δθy, Δθx are summed together and provided to integration block 78. The integration block 78 determines the pitch and roll angles of the vehicle as will be further described below.
Referring now to
Using the kinematic relationship between the sensors (IMU output) and the rates of changes of the Euler angles, and assuming that the rate of rotation of the earth is negligible, the state equations in step 82 for vehicle motion can be written as
{dot over (θ)}x=ωx+(ωy·sin θx+ωz·cos θx)·tan θy, (1)
{dot over (θ)}y=ωy·cos θx−ωz·sin θx, (2)
{dot over (θ)}z=(ωy·sin θx+ωz·cos θx)·sec θy, (3)
{dot over (ν)}x=αx+ωz·νy−ωy·νz+g·sin θy, (4)
{dot over (ν)}y=αy−ωz·νx+ωx·νz−g·sin θx·cos θy, (5)
{dot over (ν)}z=αz+ωy·νx−ωx·νy−g·cos θy·cos θy, (6)
in which ν=[νx,νy,νz]T represent velocities, ω=[ωx,ωy,ωz]T represent angular velocities, α=[αx,αy,αz]T represent accelerations, all in body frame; θ=[θx,θy,θz]T represent the three Euler angles, roll, pitch and yaw, respectively; g is the gravitational constant which is assumed to be known. Equations (1)–(6) are the fundamental equations that govern the 3-D motion of the vehicle.
For vehicle dynamic control purpose, the Euler yaw angle θz (or the heading) is not required. As can be seen, the yaw angle θz does not find its way into above equations except equation (3). Furthermore, since the vehicle is constrained to move on a surface, the vertical velocity νz is normally very small and can be neglected. Thus the estimation determination is based on the following reduced kinematic equations:
{dot over (θ)}x=ωx+(ωy·sin θx+ωz·cos θx)·tan θy, (7)
{dot over (θ)}y=ωy·cos θx−ωz·sin θx, (8)
{dot over (ν)}x=αx+ωz·νy+g·sin θy, (9)
{dot over (ν)}y=αy−ωz·νx−g·sin θx·cos θy. (10)
Theoretically, the vehicle attitude can be computed via mathematical integration of equations (7) and (8), supposing the initial condition is known and ωs are measured by gyro sensors. However in practice, direct integration intends to drift due to sensor bias and inevitable numerical errors. Absolute sensors such as GPS are always needed to constantly eliminate errors due to gyro integration. It is known to those skilled in the art that Kalman filters provide an optimal way to fuse IMU signals and absolute sensor signals. However, probabilistic information regarding the measurement and process noises is normally required.
As will be seen in this embodiment, a method that utilizes the measured accelerations, yaw rate, wheel speed and steering wheel angle to correct gyro integration is described. In other words, the dead reckoning sensors are used to provide information which is normally provided by absolute sensors such as GPS. As seen from equations (9) and (10), vehicle pitch and roll angle can be calculated if νx,{dot over (ν)}x,νy and {dot over (ν)}y were available:
Although it is possible to obtain fairly accurate νx and thus {dot over (ν)}x from wheel speed sensors when wheel slip is small, νy and {dot over (ν)}y are generally not available on current production vehicles. Thus, equations (11) and (12) cannot be implemented. Fortunately, for a lot of maneuvers, νy or {dot over (ν)}y may be small and thus can be neglected. In such cases, the so-called reference pitch and roll angles, {circumflex over (θ)}yref and {circumflex over (θ)}xref, respectively, are determined in step 84 and are set forth in the following equations:
where {circumflex over (ν)}x represents the wheel speed based longitudinal velocity calculation, and {circumflex over ({dot over (ν)}x its derivative; αxs and αys represent longitudinal and lateral accelerometers measurements. One embodiment of refining the vehicle reference roll angle of equation (13) using steer angle, yaw rate, lateral acceleration, and longitudinal velocity can be found in U.S. Pat. No. 6,073,065, which is incorporated by reference herein.
Note that both {circumflex over (θ)}yref and {circumflex over (θ)}xref are independent of gyro measurement. Therefore they can be thought as “pseudo measurements” of the vehicle attitude, and will be used to eliminate errors due to gyro integration. Unlike the GPS measurements, these two reference signals are maneuver dependent with their noise co-variances unknown, which makes it difficult to apply a traditional Kalman filter. The following discrete-time nonlinear observer in step 86 is used to solve this technical difficulty:
{circumflex over ({dot over (θ)}y(k)=ωys(k)·cos {circumflex over (θ)}x(k)−ωzs(k)·sin {circumflex over (θ)}x(k)+Δθy, (15)
{circumflex over ({dot over (θ)}x(k)=ωxs(k)+[ωys(k)·sin {circumflex over (θ)}x(k)+ωzs(k)·cos {circumflex over (θ)}x(k)]·tan {circumflex over (θ)}y(k)+Δθx, (16)
where k represents the sampling instance, ô represents computed quantities, o*s represents measured quantities, and the adjustment Δθy and Δθx are defined in step 88 as:
Δθy=Kθy(t)·({circumflex over (θ)}yref−{circumflex over (θ)}y), (17)
Δθx=Kθx(t)·({circumflex over (θ)}xref−{circumflex over (θ)}x), (18)
in which Kθy and Kθx are non-negative tunable observer gains and {circumflex over (θ)}y and {circumflex over (θ)}x are calculated from equations (11) and (12). The observer gains must then be determined in step 90. The observer gains correspond to the stability of the vehicle. In step 92 the roll and pitch attitudes are determined by integration, such as the trapezoidal or other appropriate integration method:
where Ts is the sampling period. It can be seen that when Kθx=Kθy=0, the above scheme is equivalent to gyro integration. When Kθx>0 and Kθy>0, the estimates {circumflex over (θ)}y and {circumflex over (θ)}x exponentially converge to their references {circumflex over (θ)}yref and {circumflex over (θ)}xref, respectively. The convergence rate and final accuracy can be adjusted by the observer gains.
The above scheme uses the observer equations (15)–(16) to blend the IMU gyro signals with the reference signals. Before integration, observer gains Kθy and Kθx are determined based on vehicle stability status. When the vehicle is stable, the reference signals are normally very accurate and the observer gains should be increased as a rule. In such cases, the reference signals are trusted more and the gyro integrations are trusted less. On the other hand, as the vehicle becomes unstable, the reference signals normally are not very reliable. The tunable observer gains should be reduced so that the estimates rely more on the gyro integrations. The variation in the amount of gain will vary by vehicle.
The determination of the vehicle stability status is needed in the proposed scheme. One embodiment of judging the lateral stability of a vehicle using steer angle, yaw rate, lateral acceleration, and longitudinal velocity is described in U.S. Pat. No. 6,073,065. A dynamic factor DNCF was proposed as an indicator of the magnitude of {dot over (ν)}y, the change of vehicle lateral velocity, and further as an indicator of the lateral stability of the vehicle. In the present invention, the DNCF calculation is further simplified, and a formula that reduces real-time computational load is given by:
where L is wheelbase, δ is front wheel steering angle and ku is understeer coefficient. It is well know to those skilled in the art that the expression in the bracket is zero when the vehicle is undergoing a stable corning if the bicycle model and its nominal understeer coefficient is accurate.
There are many other variables that can be used to determine the vehicle stability, i.e., linear side slip angle of each axle (βlin) steering wheel rate ({dot over (δ)}H), desired yaw rate (ωzd), measured yaw rate (ωzs) desired lateral acceleration (αyd), measured lateral acceleration (αys) wheel slip (λ), driver brake request, ABS-in-cycle flag, TCS-in-cycle flag, etc. The observer gains or pitch and roll stability indexes can be scheduled by certain fuzzy logic, or in general, can be any appropriate functions of these variables, i.e.,
Kθy(t)=f1(DNCF,{dot over (δ)}H,λ, . . . ), (22)
Kθx(t)=f2(DNCF,{dot over (δ)}H,λ, . . . ). (23)
It is seen that the proposed scheme effectively utilizes the available information of a vehicle: the kinematic relationship among sensor signals, a bicycle model, and the nonholonomic constraints for the vehicle moving on a surface. The bicycle model basically provides information about quality of the reference signals and gyro integrations which are both independent of vehicle model. Thus the estimation accuracy is not directly affected by model uncertainties. In step 94 one, some, or all of the vehicle safety systems are controlled by the controller.
While the invention has been described in connection with one or more embodiments, it should be understood that the invention is not limited to those embodiments. On the contrary, the invention is intended to cover all alternatives, modifications, and equivalents, as may be included within the spirit and scope of the appended claims.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6002974 | Schiffmann | Dec 1999 | A |
| 6002975 | Schiffmann et al. | Dec 1999 | A |
| 6073065 | Brown et al. | Jun 2000 | A |
| 6263261 | Brown et al. | Jul 2001 | B1 |
| 6654671 | Schubert | Nov 2003 | B2 |
| 6834218 | Meyers et al. | Dec 2004 | B2 |
| 6904350 | Lu et al. | Jun 2005 | B2 |
| 7079928 | Lu et al. | Jul 2006 | B2 |
| 7085639 | Lu et al. | Aug 2006 | B2 |
| 7109856 | Lu et al. | Sep 2006 | B2 |
| 7130735 | Brown et al. | Oct 2006 | B2 |
| 7132837 | Tao | Nov 2006 | B1 |
| 7136731 | Lu et al. | Nov 2006 | B2 |
| 20030158633 | Schubert | Aug 2003 | A1 |
| 20050149240 | Tseng et al. | Jul 2005 | A1 |
| Number | Date | Country |
|---|---|---|
| 60023640 | Dec 2005 | DE |
| 69930594 | May 2006 | DE |
| 983919 | Mar 2000 | EP |
| 1110835 | Jun 2001 | EP |
| 1110835 | Nov 2005 | EP |
| 983919 | Mar 2006 | EP |
| 2001219840 | Aug 2001 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20050149240 A1 | Jul 2005 | US |
