Methods for estimating and adjusting the speed and acceleration of a vehicle

By defining the speed threshold and using adaptive filter technologies such as Kalman filters, the inaccuracy of vehicle speed and acceleration in the low-speed zone is solved, and the continuous state estimation from high speed to low speed is achieved, ensuring the control law Robustness and accuracy of acceleration values.

CN114026435BActive Publication Date: 2024-11-01AMPERE +1
View PDF 32 Cites 0 Cited by

Patent Information

Application Number
CN201980095723.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-04-23
Publication Date
2024-11-01
Estimated Expiration
2039-04-23

AI Technical Summary

Technical Problem

In the low speed range, sensors cannot accurately estimate the speed and acceleration of the vehicle, resulting in the instability of the control law in the low speed system, and the accelerometer value is inaccurate, due to factors such as factory installation location, road slope and body rolling, etc. Influence, resulting in discontinuity of velocity and acceleration information.

Method used

Define two speed thresholds SV1 and SV2, use adaptive filters (such as Kalman filters) to estimate the speed in the low speed zone, use wheel angular velocity sensors to measure the speed in the high speed zone, and ensure the speed in the middle zone by linear mixing method and the continuity of acceleration, avoid discontinuity.

Benefits of technology

Accurate estimate of vehicle speed and acceleration in the low speed zone is achieved, stability of control law is ensured, and discontinuity of speed and acceleration values ​​is avoided, and is suitable for vehicle state estimation in the range from high speed to low speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114026435B_ABST
    Figure CN114026435B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for estimating the speed of a motor vehicle, wherein: A. a first speed threshold value SV1 corresponding to a minimum speed value provided by a wheel angular velocity sensor of the vehicle is defined; B. a second speed threshold value SV2 greater than SV1 is defined; C. when the vehicle is traveling below SV1, a low speed value is estimated by using an estimation method of an adaptive filtering type; D. when the vehicle is traveling above SV2, a high speed value is measured by using the vehicle speed value provided by the wheel angular velocity sensor; E. in an intermediate area between SV1 and SV2, there is a mixture of high speed and low speed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of motor vehicles. More specifically, the present invention relates to a strategy for estimating speed and associated acceleration in a range from high speeds to very low speeds, eliminating the limitations of current sensors. Background Art

[0002] Knowing the exact velocity and the associated acceleration is very important in the context of control law development. For example, control laws used in ADAS (Advanced Driver Assistance Systems) systems and autonomous vehicles still require velocity and acceleration information.

[0003] In current vehicles, speed and acceleration can be accurately calculated above a certain threshold. If the actual speed is below this threshold, speed and acceleration information cannot be obtained. This speed range is usually called "low speed".

[0004] The main problem is that the speed cannot be estimated well below the speed threshold due to limitations of the sensors used.

[0005] Therefore, the control law used is not robust enough to control different low speed systems; such as for example:

[0006] ·Hand-held brake system (HFPB, which refers to non-manual parking brake, also known as automatic parking)

[0007] ACC (distance controller) system for "Stop & Start" situations

[0008] · Autonomous vehicles in traffic jams or TJP (Traffic Jam Pilot) systems.

[0009] The second problem is using the acceleration value from the accelerometer. This value is not very accurate (it may be offset) due to the following reasons:

[0010] Position of the accelerometer after factory installation

[0011] Some external quantities, such as the slope and camber of the road

[0012] Roll and pitch of the vehicle body.

[0013] Figure 1A diagram showing the problem encountered is shown. The current estimated vehicle speed is represented by curve 1, the accelerometer value is represented by curve 2, curve 3 represents the "peak" of the encoder wheel (i.e. the peak of the signal they send when the teeth pass, such wheels are also called "toothed wheels"), and between lines A and B is the low speed area where the vehicle is running below the 1 km / h threshold. The peak indicates whether the wheel has turned and gives a picture of the speed through its amplitude.

[0014] The speed is unknown in the area between lines A and B. For example, in the right part of the low speed zone, the wheel can be seen to be rotating (there is a peak of the encoder wheel), but the speed below the 1km / h threshold is not detected.

[0015] Finally, the graph of the accelerometer (curve 2) shows the offset in the low velocity region (the region where there is no peak and the accelerometer is always at a non-zero value).

[0016] Therefore, it is necessary to develop a strategy to estimate the speed and acceleration in the low-speed zone (between A and B) to supplement the speed values ​​already present on the car.

[0017] An example of such a strategy is known from the document "Improving the Response of a Wheel Speed ​​Sensor by Using a RLS Lattice Algorithm", W. Hernandez, Sensors, June 2006, pp. 64-79. This document discloses more specifically the use of an adaptive filter to solve the inaccuracy problem at low speeds, in particular the use of a Kalman filter.

[0018] The main advantage of this software solution based on adaptive filtering is its low cost.

[0019] However, there is a bigger problem beyond speed estimation, namely, the estimated speed and acceleration values ​​are discontinuous when transitioning from a high speed range above a threshold to a low speed range below a threshold.

[0020] The object of the present invention is mainly to solve this technical problem by proposing a method that enables estimation of the speed and / or acceleration of a vehicle at low speeds, while being suitable for accurately measuring the speed of the vehicle at medium and high speeds without any discontinuity in these values. Summary of the invention

[0021] To this end, the subject of the invention is a method for estimating the speed of a motor vehicle, wherein:

[0022] - defining a first speed threshold SV1, which corresponds to the minimum speed value provided by the vehicle wheel angular velocity sensor;

[0023] - defining a second speed threshold SV2 greater than SV1;

[0024] - when the vehicle is traveling below SV1, estimating a low speed value by using an estimation method of an adaptive filter type;

[0025] - when the vehicle is traveling above SV2, measuring a high speed value by using the vehicle speed values ​​provided by the wheel angular velocity sensors;

[0026] - In the intermediate region between SV1 and SV2, there is a mixture of high and low speeds.

[0027] According to the invention, three speed ranges are used: low speed, high speed and an intermediate mixed region. The use of a mixed range makes it possible to avoid discontinuities in speed and acceleration (essential for ensuring the stability of the control law).

[0028] Advantageously, the adaptive filter is a Kalman filter.

[0029] Advantageously, in the intermediate region between SV1 and SV2, the mixing is performed periodically at successive moments using a linear mixing method according to the following formula:

[0030]

[0031] This linear blending enables the use of velocity values ​​in the Kalman method and vehicle speed To calculate the mixing speed (velocity).

[0032] Vehicle speed is the speed measured using the wheel angular velocity.

[0033] Speed ​​is the mixed speed at the current time t, speed t-1 is the mixing speed at the previous mixing time t-1, is the velocity value calculated by the Kalman method at the current time t, and is the velocity value measured by the angle sensor at the current time t.

[0034] According to a feature of the present invention, the first threshold SV1 may be 1 km / h.

[0035] According to another feature of the invention, the second speed threshold SV2 may be 1.5 km / h.

[0036] Advantageously, in step C), acceleration values ​​are also estimated using a Kalman filter and in step E), there is also a blending of acceleration values ​​between SV1 and SV2.

[0037] An advantage of the present invention is that the estimated velocity has no discontinuities and also the associated acceleration values ​​can be taken into account. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The invention will be better understood on reading the following description of exemplary embodiments given as illustrative examples, with reference to the accompanying drawings, in which:

[0039] - Figure 1 A diagram showing the problems encountered at low speeds;

[0040] - Figure 2 The principle of the present invention is schematically shown;

[0041] - Figure 3 and Figure 4 An example showing the results of using the method of the present invention during vehicle starting and stopping phases. DETAILED DESCRIPTION

[0042] Figure 2 The general principle of the invention is schematically represented, wherein three speed zones are defined:

[0043] - A low speed zone below a first threshold value SV1, below which no speed values ​​and acceleration values ​​can be obtained.

[0044] Usually 1km / h.

[0045] In this region, the speed is measured according to the Kalman method. This method itself is known to the person skilled in the art, but is reviewed below in order to explain the invention more clearly.

[0046] - a high speed zone above a second threshold SV2, which is greater than the first threshold SV1 (e.g. 1.5 km / h). In this zone, speed values ​​and acceleration values ​​are provided by vehicle sensors; and

[0047] - A mixed region lying between these two thresholds SV1 and SV2.

[0048] I. Estimating Velocity Using the Kalman Method

[0049] I.1 Conventional Kalman Filter

[0050] The Kalman filter considers three state variables [x]:

[0051] x(1) is the distance traveled since the first moment t;

[0052] x(2) speed information;

[0053] x(3) Final acceleration.

[0054] The two sensor measurements [z] used to estimate the state variables are:

[0055] z(1) Average value of encoder wheel peak values ​​(WT). The peak signals of these 4 wheels are already present in the vehicle messaging system (CAN). This information enables the displacement of each wheel to be known by counting the number of teeth (typically 48 teeth) passed by the encoder at each sampling interval.

[0056] z(2) Angular velocity of the wheels (WS). The angular velocity signals of the four wheels are already available in the vehicle message system (CAN). The average value of the rear wheels (the axle of the non-driven wheels, i.e. the one with less slip during the start phase) will be used in the Kalman calculation.

[0057] The Kalman filter equations are:

[0058] 1) Prediction

[0059]

[0060]

[0061] 2) Calibration

[0062]

[0063]

[0064]

[0065]

[0066] P k|k =(IK k H k ) k|k-1

[0067] Used symbol as follows:

[0068] x: state of the system (vector)

[0069] z: sensor measurement value (vector)

[0070] P: estimated covariance matrix

[0071] ·F k : State transition matrix

[0072] ·u k : Command input

[0073] ·B k : Command transition matrix

[0074] H: measurement value transition matrix

[0075] Q: Model noise covariance matrix (precision)

[0076] R: measurement noise covariance matrix (precision)

[0077] I: Identity matrix

[0078] · The estimated value of the variable x

[0079] · The measured value of variable x

[0080] Notice : In the fitted Kalman filter, the vector u is zero, which simplifies the first equation.

[0081] I.2 Speed ​​Estimation

[0082] At the input of the system, there are two sensor data, corresponding to the wheel speed (WS) and the encoder wheel peak value (WT). These data are processed (DP: "Data Processing") and then enter the Kalman filter ("Estimation" block), from which the velocity and acceleration are generated.

[0083] Step 1 - "Data Processing":

[0084] Wheel Pulse: The encoder sends the position of the last tooth seen. We will use the increment of the number of teeth [WT] during the sampling interval [Te] of the system (the interval must be at the same rate as the sensor is recording). The average between these four wheels will then be used as the measurement of [WT]. The value equivalent to the linear velocity and using the wheel peak value is

[0085]

[0086] Here, it is assumed that the wheel radius [R] is constant and known (set parameter), and [nb_pic] is the number of teeth of the encoder.

[0087] WS: The average value of the angular velocity of the rear wheels (the axle of the non-driven wheels, i.e., the one with less slip during the start phase) will be used in the Kalman filter. This angular velocity [WS] will be converted to a linear velocity based on the following formula:

[0088]

[0089] Step 2: “Estimate”:

[0090] The Kalman model used is as follows:

[0091] Equation of state:

[0092]

[0093] d k 、v k and a k are the travel distance, speed and acceleration at the kth iteration of the filter, respectively.

[0094] Input data vector

[0095]

[0096] nb_pic=96, encoder increment.

[0097] Te=0.01s, sampling period.

[0098] This state equation represents the first line of the prediction step shown above. Here it is assumed that the acceleration changes constantly.

[0099] The input vector (z) corresponds to the sensor data inserted into the Kalman filter. The data [WT] corresponds to the sum of the peak values ​​of the encoder wheels divided by four (number of wheels). The variable [WS] itself is equal to the sum of the vehicle's rear wheel speeds divided by two.

[0100] Since the last data is not always available (falls to 0 below SV1), the matrix H is adapted in the Kalman filter (see Kalman Equations, Correction Phase).

[0101] II. Speed ​​Mixing

[0102] In the speed range between the first threshold value SV1 and the second threshold value SV2 (in Figure 2 In the example shown, it is between 1 km / h and 1.5 km / h), and there is a mixture of high speeds and low speeds.

[0103] More specifically, the blending is performed using a linear blending method according to the following formula:

[0104]

[0105] This linear blending enables the use of the velocity value of the Kalman method and vehicle speed To calculate the value of the mixing speed (speed).

[0106] Vehicle speed is the speed calculated by using the wheel angular velocity. At low speeds, the high speed value of the vehicle cannot be obtained.

[0107] To ensure correct blending, the reference speed value used is the last speed value. This value is used to define the weight of each speed (the weight defined by the relative distance relative to the threshold). For example, the weight of the speed of the Kalman method is defined as

[0108]

[0109] The reference speed is chosen so that continuity during mixing is ensured.

[0110] The velocity estimated by the Kalman method cannot be used because the initial value may be greater than SV2 (due to the delay of the filter). The vehicle speed cannot be used either because it shows discontinuities at low speeds where the angular velocity is no longer available.

[0111] III. Examples of results obtained

[0112] III.1-Start-up phase

[0113] Figure 3 The green curve corresponds to the current system speed, curve 3 shows the WT, and curves 4 and 5 show the speed and acceleration calculated according to the method of the present invention, respectively.

[0114] Regarding speed, it can be seen that the Kalman filter gives an increasing speed 4, which corresponds to the currently used vehicle speed 1. The speed 4 calculated by the method of the invention starts to rise at the first detected wheel peak, ie the first peak of curve 3.

[0115] The same observations can be made regarding acceleration 5. The new estimate starts from the first peak detected and converges fairly well towards a value corresponding to the expected value for velocity 4.

[0116] The grey area corresponds to the transition zone between low and high speeds. It can be seen that there is no discontinuity and the estimated speed values ​​show a coherent transition with the vehicle's true speed dynamics.

[0117] III.2 Braking phase to stop

[0118] Figure 4 Represents the results obtained during the stopping phase.

[0119] Observing the speed, it can be seen that the speed curve followed by curve 4 is more consistent with the encoder wheel peak value than curve 1. Using the method of the present invention, the stop of the vehicle can also be detected more neatly.

[0120] Acceleration 5 seems to correspond to the given velocity 4 and stops at the same time as velocity 4.

[0121] The grey area corresponds to the transition area between high and low speeds. It can be seen that there is no discontinuity and the speed value 4 estimated by the hybrid shows a coherent transition with the Kalman speed dynamics.

Claims

1. A method for estimating the speed of a motor vehicle, wherein: A. defining a first speed threshold SV1, which corresponds to the minimum speed value provided by the vehicle wheel angular velocity sensor; B. defining a second speed threshold SV2 greater than SV1; C. when the vehicle is traveling below SV1, a low speed value is estimated by using an estimation method of an adaptive filter type; D. when the vehicle is traveling above SV2, measuring a high speed value by using the vehicle speed value provided by the wheel angular velocity sensor; E. In the intermediate region between SV1 and SV2, there is a mixture of high and low speeds.

2. The method according to claim 1, wherein: The adaptive filter is a Kalman filter.

3. The method according to claim 2, wherein: In the intermediate region between SV1 and SV2, the mixing is performed periodically at successive moments using a linear mixing method according to the following formula: Among them, speed is the mixed speed at the current time t, speed t-1 is the mixing speed at the previous mixing time t-1, is the velocity value calculated by the Kalman method at the current time t, and is the speed value measured by the wheel angular velocity sensor at the current time t.

4. The method according to any one of claims 1 to 3, wherein: The first speed threshold SV1 is 1 km / h.

5. The method according to any one of claims 1 to 3, wherein: The second speed threshold SV2 is 1.5 km / h.

6. The method according to claim 2 or 3, wherein: In step C), the Kalman filter is also used to estimate acceleration values, and in step E), there is also a mixing of acceleration values ​​between SV1 and SV2.

Citation Information

Patent Citations

  • Vehicle speed estimation method of motor vehicle self-adaption cruise system

    CN101655504A

  • Velocity calculation device,velocity calculation method, and navigation device

    CN101661048A

  • GPS and MEMS hybrid location-detection architecture

    CN103931248A

  • Whole-speed range control method of position sensor of elevator door motor

    CN104601072A

  • Estimation method and device for longitudinal vehicle speed of vehicle

    CN106394561A