CN107878462B - Speed prediction method and apparatus - Google Patents
Speed prediction method and apparatus Download PDFInfo
- Publication number
- CN107878462B CN107878462B CN201610875080.4A CN201610875080A CN107878462B CN 107878462 B CN107878462 B CN 107878462B CN 201610875080 A CN201610875080 A CN 201610875080A CN 107878462 B CN107878462 B CN 107878462B
- Authority
- CN
- China
- Prior art keywords
- wheel
- period
- wheel speed
- cycle
- error
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
- B60W40/105—Speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
- B60W2520/105—Longitudinal acceleration
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/28—Wheel speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2720/00—Output or target parameters relating to overall vehicle dynamics
- B60W2720/10—Longitudinal speed
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
This disclosure relates to a kind of speed prediction method and apparatus, this method comprises: obtaining wheel speed sample set and longitudinal acceleration of N number of wheel in current period of vehicle；Dynamic prediction model is established according to the wheel speed sample set and longitudinal acceleration of current period, and dynamic prediction model was modified according to the error correction parameter that a upper period obtains；State space vector is obtained according to preset state-space model and wheel speed sample set；According to the longitudinal acceleration of state space vector and current period, the speed prediction value of current period is obtained using revised dynamic prediction model；According to current period and the speed prediction value of M period acquisition before, the reference obtained with current period and M period take turns wheel speed measured value, obtain error sample set；The error correction parameter for being used for next period is obtained, according to error sample set with the calculating for next period.The disclosure can be improved the accuracy of speed prediction, reduce error.
Description
Technical Field
The disclosure relates to the technical field of measurement, in particular to a vehicle speed prediction method and device.
Background
Along with the increasing importance of the role played by the automobile in daily life, the improvement of the performance of the automobile in each aspect becomes a very much concerned problem, especially the prediction of the automobile speed under various scenes such as driving, braking, sideslip and the like, the predicted automobile speed is a very important state parameter for realizing the electronic control of the automobile, but the currently adopted automobile speed prediction method generally acquires the speeds of four wheels according to a wheel speed sensor, obtains the acceleration values of the four wheels and estimates the corresponding automobile speeds of the four wheels, then judges whether the automobile is a braking working condition or a driving working condition according to the wheel acceleration values, selects the maximum value or the minimum value in the corresponding automobile speeds of the four wheels as a predicted value according to different working conditions, does not fully utilize the information of the four wheels, and simultaneously, the phenomena of overlarge slip rates such as slip, idle running, sideslip and the like are frequently generated during the acceleration and the braking of the, it is easy to cause a large error in the calculation of the vehicle speed.
Disclosure of Invention
The invention aims to provide a vehicle speed prediction method and a vehicle speed prediction device, which are used for solving the problems of low universality and large error of the traditional speed measurement method.
In order to achieve the above object, according to a first aspect of an embodiment of the present disclosure, there is provided a vehicle speed prediction method including: acquiring a wheel speed sample set of N wheels of a vehicle in a current period and a longitudinal acceleration of the vehicle in the current period, wherein N is a positive integer;
establishing a dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period, and correcting the dynamic prediction model according to the error correction parameter obtained from the previous period and used for the current period;
acquiring a state space vector according to a preset state space model and the wheel speed sample set;
acquiring a predicted value of the vehicle speed in the current period by using a modified dynamic prediction model according to the state space vector and the longitudinal acceleration in the current period;
acquiring an error sample set according to the predicted vehicle speed value of the current period, the predicted vehicle speed values acquired in the previous M periods, the measured reference wheel speed value of the current period and the measured reference wheel speed values acquired in the M periods;
and acquiring error correction parameters for the next period according to the error sample set, and performing calculation of the next period by using the error correction parameters for the next period, wherein the period from the acquisition of the wheel speed sample set of the N wheels of the vehicle in the current period to the acquisition of the error correction parameters for the next period according to the error sample set is one period.
Optionally, the acquiring a set of wheel speed samples of N wheels of the vehicle in a current cycle and a longitudinal acceleration of the vehicle in the current cycle includes:
acquiring a plurality of wheel speed signals of N wheels of the vehicle, which are acquired within a preset time period;
filtering the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals;
selecting a wheel speed signal meeting a preset sample condition according to the filtered wheel speed signals to obtain a wheel speed sample set;
and acquiring a longitudinal acceleration measured value of the vehicle in the current period, which is acquired by an acceleration sensor of the vehicle, and filtering the longitudinal acceleration measured value of the current period by a preset filtering algorithm to obtain the longitudinal acceleration of the current period.
Optionally, the establishing a dynamic prediction model according to the wheel speed sample set of the current cycle and the longitudinal acceleration of the current cycle includes:
acquiring the deviation of the wheel speeds of the N wheels in the last period and the predicted vehicle speed value in the last period according to the wheel speed sample set in the last period and the predicted vehicle speed value in the last period;
respectively acquiring wheel acceleration of the N wheels according to the wheel speed sample set of the current period;
and establishing a dynamic prediction model according to the wheel speed sample set of the current period, the deviation, the wheel acceleration of the N wheels and the longitudinal acceleration of the current period.
Optionally, the obtaining an error correction parameter for a next cycle according to the error sample set includes:
establishing an error prediction model of the current period according to the error sample set;
and acquiring the error correction parameters for the next period by using the error prediction model.
Optionally, the obtaining an error sample set according to the predicted vehicle speed value of the current period and the predicted vehicle speed values obtained in M previous periods, and the reference measured vehicle speed value of the current period and the reference measured vehicle speed values obtained in M previous periods includes:
respectively obtaining errors of the predicted vehicle speed value and the measured reference wheel speed value of each period in the current period and the M periods to obtain a plurality of errors;
and selecting part or all of the errors as error samples to obtain the error sample set.
Optionally, the dynamic prediction model includes:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the right front wheel in the k-1 th period and the predicted value of the vehicle speed, and the deviation between the wheel speed of the right rear wheel in the k-1 th period and the predicted value of the vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel of the k-th cycle, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
Optionally, the state space vector is a vector using wheel speeds of the N wheels as elements, and the state space model includes:
wherein,respectively represents the wheel speed value, lambda, of any wheel of the vehicle in the k +1 th cycle and the k-th cycle1、λ2Represents a parameter value and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1.
Optionally, the error prediction model includes:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,representing the estimated value of the error correction for the k-th cycle, Q representing the process noise covariance, zkRepresenting the actual value of the error in the k-th cycle, R representing the measurement noiseThe acoustic covariance is the sum of the acoustic covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkThe kalman gain in the k-th cycle is expressed, and I denotes an identity matrix.
Optionally, the filtering algorithm includes:
wherein,the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cycle, respectivelykIndicating the longitudinal acceleration value of the k-th cycle.
According to a second aspect of the embodiments of the present disclosure, there is provided a vehicle speed prediction device, the device including: the system comprises an acquisition module, a prediction model establishing module, a state space vector acquisition module, a vehicle speed prediction module, an error acquisition module and a correction parameter acquisition module;
the acquisition module is used for acquiring a wheel speed sample set of N wheels of a vehicle in a current period and a longitudinal acceleration of the vehicle in the current period, wherein N is a positive integer;
the prediction model establishing module is used for establishing a dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period, and correcting the dynamic prediction model according to the error correction parameter obtained in the previous period and used for the current period;
the state space vector acquisition module is used for acquiring a state space vector according to a preset state space model and the wheel speed sample set;
the vehicle speed prediction module is used for acquiring a vehicle speed prediction value of the current period by using a modified dynamic prediction model according to the state space vector and the longitudinal acceleration of the current period;
the error acquisition module is configured to acquire an error sample set according to the predicted vehicle speed value of the current period and the predicted vehicle speed values acquired in M previous periods, the measured reference wheel speed value of the current period and the measured reference wheel speed values acquired in M previous periods, where M is a non-negative integer;
and the correction parameter acquisition module is used for acquiring error correction parameters for the next period according to the error sample set and calculating the next period by using the error correction parameters for the next period, wherein the period from the acquisition of the wheel speed sample set of the N wheels of the vehicle in the current period to the acquisition of the error correction parameters for the next period according to the error sample set is one period.
Optionally, the obtaining module includes: the device comprises a signal acquisition module, a filtering module, a sample set acquisition module and an acceleration acquisition module;
the signal acquisition module is used for acquiring a plurality of acquired wheel speed signals of N wheels of the vehicle;
the filtering module is used for filtering the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals;
the sample set acquisition module is used for selecting a wheel speed signal meeting a preset sample condition according to the filtered wheel speed signals to obtain a wheel speed sample set;
the acceleration acquisition module is configured to acquire a longitudinal acceleration measured value of the vehicle in the current period, which is acquired by an acceleration sensor of the vehicle, and filter the longitudinal acceleration measured value of the current period through a preset filtering algorithm to obtain the longitudinal acceleration of the current period.
Optionally, the prediction model building module includes: the device comprises a deviation acquisition module, a wheel acceleration acquisition module and a modeling module;
the deviation obtaining module is used for obtaining the deviation between the wheel speed of the N wheels in the previous period and the predicted vehicle speed value of the previous period according to the wheel speed sample set of the previous period and the predicted vehicle speed value of the previous period;
the wheel acceleration acquisition module is used for respectively acquiring the wheel acceleration of the N wheels according to the wheel speed sample set of the current period;
the modeling module is used for establishing a dynamic prediction model according to the wheel speed sample set of the current period, the deviation, the wheel acceleration of the N wheels and the longitudinal acceleration of the current period.
Optionally, the correction parameter obtaining module includes:
the error model establishing module is used for establishing an error prediction model of the current period according to the error sample set;
and the parameter acquisition module is used for acquiring the error correction parameters for the next period by utilizing the error prediction model.
Optionally, the error obtaining module includes:
the error calculation module is used for respectively obtaining the errors of the vehicle speed predicted value and the reference wheel speed measured value in each period in the current period and the M periods to obtain a plurality of errors;
and the error sample screening module is used for selecting part or all of the errors as error samples to obtain the error sample set.
Optionally, the dynamic prediction model includes:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, and the k-1 th periodThe deviation between the right front wheel speed and the predicted vehicle speed, the deviation between the right rear wheel speed in the k-1 th cycle and the predicted vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel of the k-th cycle, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
Optionally, the state space vector is a vector using wheel speeds of the N wheels as elements, and the state space model includes:
wherein,respectively represents the wheel speed value, lambda, of any wheel of the vehicle in the k +1 th cycle and the k-th cycle1、λ2To representParameter value of and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1.
Optionally, the error prediction model includes:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,representing the estimated value of the error correction for the k-th cycle, Q representing the process noise covariance, zkRepresents the actual value of the error for the k-th cycle, R represents the measurement noise covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkThe kalman gain in the k-th cycle is expressed, and I denotes an identity matrix.
Optionally, the filtering algorithm includes:
wherein,the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cycle, respectivelykIndicating the longitudinal acceleration value of the k-th cycle.
Through the technical scheme, the wheel speeds of all wheels are integrated for modeling, the method can be suitable for various scenes in the running process of the vehicle, the problems of low universality and large errors which are easy to occur when the vehicle speed is calculated by using the wheel speed of a single wheel in the prior art can be solved, the universality can be improved, and meanwhile, the technical scheme provided by the disclosure can update the model for predicting the vehicle speed in real time, so that the interference caused by the wheel speed errors in the braking and sideslip processes can be effectively reduced, the accuracy of prediction is improved, and the errors are reduced.
Additional features and advantages of the disclosure are set forth in the detailed description which follows, and it is understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a method of predicting vehicle speed in accordance with an exemplary embodiment;
FIG. 2 is a flow chart illustrating another method of predicting vehicle speed in accordance with an exemplary embodiment;
FIG. 3 is a flow chart illustrating yet another method of vehicle speed prediction according to an exemplary embodiment;
FIG. 4 is a flow chart illustrating yet another method of vehicle speed prediction according to an exemplary embodiment;
FIG. 5 is a flow chart illustrating yet another method of vehicle speed prediction according to an exemplary embodiment;
FIG. 6 is a schematic diagram illustrating a vehicle electronics system configuration in accordance with an exemplary embodiment;
FIG. 7 is a block diagram illustrating a vehicle speed prediction device according to an exemplary embodiment;
FIG. 8 is a block diagram illustrating an acquisition module in accordance with an exemplary embodiment;
FIG. 9 is a block diagram illustrating a predictive model building module in accordance with an exemplary embodiment;
FIG. 10 is a block diagram illustrating a rework parameter acquisition module, according to an example embodiment;
FIG. 11 is a block diagram illustrating an error acquisition module in accordance with an exemplary embodiment.
Detailed Description
Reference will now be made in detail to the embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Before describing the vehicle speed prediction method and device provided by the present disclosure, an application scenario related to various embodiments of the present disclosure will be described first. The application scenario may be any vehicle, such as an automobile, the automobile is not limited to a conventional automobile, a pure electric automobile or a hybrid automobile, and may also be applicable to other types of automobiles or non-automobiles.
FIG. 1 is a flow chart illustrating a method of predicting vehicle speed, as shown in FIG. 1, according to an exemplary embodiment, including:
step 101, acquiring a wheel speed sample set of N wheels of a vehicle in a current period and a longitudinal acceleration of the vehicle in the current period, wherein N is a positive integer.
For example, N wheels of the vehicle may be all wheels of the vehicle, and in the case of an automobile, the automobile is generally provided with 4 wheels, so in the present embodiment, N may be 4.
And 102, establishing a dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period, and correcting the dynamic prediction model according to the error correction parameters for the current period obtained in the previous period.
The output of the dynamic prediction model is a predicted vehicle speed value of the vehicle, so the dynamic prediction model can be understood as a function model which takes the wheel speeds and longitudinal accelerations of the above-mentioned N wheels as variables and takes the predicted vehicle speed value as an output, and a coefficient of each wheel speed in the dynamic prediction model can be obtained according to a deviation of the wheel speed of each wheel in the previous period and the predicted vehicle speed value in the previous period.
The error correction parameter obtained in the previous cycle for the current cycle refers to the error correction parameter obtained in the previous execution of step 101-106.
Step 103, acquiring a state space vector according to a preset state space model and a wheel speed sample set.
The state space vector is a vector in which the wheel speeds of N wheels are elements, and the state space vector includes a left front wheel speed and a left rear wheel speed, taking 4 wheels of a general vehicle as an example. The right front wheel speed and the right rear wheel speed. Due to the complexity of the operating environment of the vehicle, the processing circuit for acquiring the wheel speed inevitably receives external strong electromagnetic interference, so that transient distortion of the wheel speed signal may be caused, and therefore the wheel speed sample set acquired in step 101 may be processed through a preset state space model, and the more accurate wheel speeds of the N wheels can be acquired under the conditions of distorted wheel speed and undistorted wheel speed respectively through the state space model, so as to obtain the state space vector.
Therefore, the state space vector is a vector having the wheel speeds of the N wheels as elements, and the wheel speed of each wheel can be calculated according to the following formula, so that the state space vector can be obtained, and the state space model can include:
wherein,respectively representing vehicles in the k +1 th cycle and the k < th > cycleOf any one wheel, λ1、λ2Represents a parameter value and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1. Wherein λ is1、λ2The specific value of (a) can be obtained from experimental calibration, e.g. can be determined empirically as1＝3/4，λ2＝1/4。
And 104, acquiring a predicted vehicle speed value of the current period by using the modified dynamic prediction model according to the state space vector and the longitudinal acceleration of the current period.
And 105, acquiring an error sample set according to the predicted vehicle speed value of the current period and the predicted vehicle speed values acquired in M previous periods, the measured wheel speed value of the reference wheel of the current period and the measured wheel speed values acquired in M previous periods, wherein M is a non-negative integer.
For example, the value of M may be set according to actual requirements, and the higher the value of M, the more error samples included in the error sample set are represented, which is more beneficial to improving the accuracy of the error correction parameter, but it is worth mentioning that too large data volume may cause too large storage space, occupy memory volume, and have no real-time property; if the error sample set included in the error sample set is too small, the error prediction value has a large deviation and the usability is reduced, so that the specific value of M can be determined according to the actual requirement.
In various embodiments of the present disclosure, the reference wheel-speed actual measurement value is used only for a correction use of the vehicle speed prediction. For example, the actual measured value of the wheel speed of the reference wheel may be a maximum value of the wheel speed of each wheel in the driving condition, a minimum value of the wheel speed of each wheel in the braking condition, an average value of the maximum wheel speeds of the left and right sides of the vehicle in the normal turning and the braking turning, and an average value of the minimum wheel speeds of the left and right sides of the vehicle in the driving turning.
And 106, acquiring error correction parameters for the next period according to the error sample set. And, the calculation of the next cycle can be performed using the error correction parameter for the next cycle.
Wherein, the period from step 101 to step 106 is one period.
FIG. 2 is a flow chart illustrating another vehicle speed prediction method according to an exemplary embodiment, wherein the step of obtaining a set of wheel speeds of N wheels of the vehicle in a current cycle and the longitudinal acceleration of the vehicle in the current cycle, as shown in FIG. 2, in step 101, may include the steps of:
at step 1011, a plurality of wheel speed signals of the N wheels of the vehicle are obtained.
Step 1012, filtering the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals.
And 1013, selecting wheel speed signals meeting preset sample conditions according to the filtered wheel speed signals to obtain a wheel speed sample set.
For example, the wheel speed of each wheel during the running of the vehicle can be monitored by a wheel sensor, and when the wheel rotates, a pulse signal with the same phase as the rotating body is generated and transmitted to an electronic control assembly of the vehicle as a wheel speed signal. According to the method, the electronic assembly can receive a plurality of wheel speed signals of all wheels, then the wheel speed signals are filtered, and wheel speed signals meeting sample conditions are selected from the wheel speed signals to obtain a wheel speed sample set, wherein the wheel speed sample set comprises wheel speeds of all wheels of the vehicle.
Step 1014, acquiring a longitudinal acceleration measured value of the vehicle in the current period, acquired by an acceleration sensor of the vehicle, and filtering the longitudinal acceleration measured value in the current period by a preset filtering algorithm to obtain the longitudinal acceleration in the current period. Wherein steps 1011 to 1013, and step 1014 may be performed simultaneously.
As an example, the preset algorithm may be kalman (kalman) filtering, clipping filtering, etc., and in the case of clipping filtering, the filtering algorithm may include:
wherein,respectively represents the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cyclekRepresents the longitudinal acceleration value, delta, of the k-th cycle0A threshold value representing the difference between the longitudinal accelerations of two adjacent cycles. It can be understood that, due to the complexity of the operating environment of the vehicle, the processing circuit for acquiring the wheel speed inevitably receives strong electromagnetic interference from the outside, which may cause transient distortion of the wheel speed signal, and therefore, after obtaining the longitudinal acceleration measured value of the current period, it needs to determine whether to modify the longitudinal acceleration measured value of the current period, and the determination may be based on whether the difference between the longitudinal acceleration measured value of the current period and the longitudinal acceleration measured value of the previous period is smaller than the threshold δ0When the difference is smaller than the threshold value delta0When the measured value of the longitudinal acceleration in the current period is not distorted, the measured value can be directly used as the longitudinal acceleration of the vehicle in the current period, and when the difference value is larger than the threshold value delta0In the meantime, it can be considered that the measured longitudinal acceleration value of the current period is distorted, and the correction needs to be performed through the above filtering algorithm.
FIG. 3 is a flow chart illustrating yet another vehicle speed prediction method according to an exemplary embodiment, wherein the step 102 of building a dynamic prediction model based on the wheel speed sample set of the current cycle and the longitudinal acceleration of the current cycle, as shown in FIG. 3, may include the steps of:
and 1021, acquiring the deviation between the wheel speed of the N wheels in the previous period and the predicted vehicle speed value of the previous period according to the wheel speed sample set of the previous period and the predicted vehicle speed value of the previous period.
And step 1022, respectively acquiring wheel accelerations of the N wheels according to the wheel speed sample set of the current period.
And step 1023, establishing a dynamic prediction model according to the wheel speed sample set of the current period, the deviation between the wheel speed of the N wheels in the previous period and the predicted vehicle speed value of the previous period, the wheel acceleration of the N wheels and the longitudinal acceleration of the current period.
The dynamic prediction model may be understood as a function model that takes the wheel speeds and the longitudinal accelerations of the above-mentioned N wheels as variables and takes the predicted value of the vehicle speed as an output, and the coefficient of the wheel speed of each wheel in the dynamic prediction model may be obtained according to the deviation of the wheel speed of each wheel in the previous period from the predicted value of the vehicle speed in the previous period.
For example, taking a 4-wheel vehicle as an example, the dynamic prediction model can be expressed as:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the right front wheel in the k-1 th period and the predicted value of the vehicle speed, and the deviation between the wheel speed of the right rear wheel in the k-1 th period and the predicted value of the vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel of the k-th cycle, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
It is understood that, of the 5 formulas for vehicle speed prediction provided by the dynamic prediction model described above, which is specifically used for vehicle speed prediction may be determined based on the difference between the wheel acceleration of the 4 wheels and the longitudinal acceleration, respectively, and the determination conditions shown in the dynamic prediction model described above.
Optionally, fig. 4 is a flowchart illustrating another vehicle speed prediction method according to an exemplary embodiment, and as shown in fig. 4, the step 105 of obtaining an error sample set according to the predicted vehicle speed value of the current cycle and the predicted vehicle speed values obtained in M previous cycles, the reference measured vehicle speed value of the current cycle and the reference measured vehicle speed values obtained in M previous cycles may include the following steps:
and 1051, respectively obtaining errors of the predicted value of the vehicle speed and the measured value of the reference wheel speed in each period in the current period and the M periods to obtain a plurality of errors.
The method comprises the steps of obtaining an error between a predicted vehicle speed value of a current period and a measured wheel speed value of a reference wheel of the current period, obtaining a predicted vehicle speed value of a previous period and a measured wheel speed value of a reference wheel of the previous period, and so on until obtaining an error of an Mth period before the current period, thereby obtaining a plurality of errors.
Step 1052, selecting part or all of the plurality of errors as error samples to obtain an error sample set.
Optionally, fig. 5 is a flowchart illustrating a further vehicle speed prediction method according to an exemplary embodiment, and as shown in fig. 5, the step 106 of obtaining the error correction parameter for the next cycle according to the error sample set and calculating the next cycle by using the error correction parameter for the next cycle may include the following steps:
step 1061, establishing an error prediction model of the current period according to the error sample set.
Building the error prediction model for the current cycle from the set of error samples may be understood as fitting the error prediction model for the current cycle from a plurality of errors in the set of error samples. It should be noted that, since the error between the predicted vehicle speed value and the actual measured wheel speed value of the reference wheel in each period is obtained in each period and M previous periods, the elements in the error sample set obtained in each period are all updated (it can be understood that the error of the current period is added and the error of the earliest one of the M previous periods is removed), so that the error prediction model can be kept updated in real time.
Step 1062, obtaining the error correction parameters for the next cycle by using the error prediction model.
For example, the establishing of the error prediction model according to the error sample set may be performed by interpolation, fitting, kalman algorithm, a soberger observer, and the like, where the kalman algorithm is taken as an example, the error prediction model may include:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,representing the estimated value of the error correction for the k-th cycle, Q representing the process noise covariance, zkRepresents the actual value of the error for the k-th cycle, R represents the measurement noise covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkThe kalman gain in the k-th cycle is expressed, and I denotes an identity matrix.
The method is exemplified when the vehicle whose speed is to be predicted is a four-wheel vehicle, fig. 6 is a schematic view showing the structure of an electronic system of a vehicle according to an exemplary embodiment, as shown in fig. 6, the electronic control system of the vehicle includes: the system comprises an electronic control assembly and a hydraulic assembly, wherein the electronic control assembly comprises a calculating unit, a control unit, a monitoring unit and the like, a master cylinder of the hydraulic assembly is connected, a brake pedal is connected with the master cylinder, a solenoid valve in the hydraulic assembly is controlled by the control unit, and the system further comprises an acceleration sensor and wheel speed sensors which are respectively arranged for four wheels. From step 101 to step 106, there is a period T (i.e., the time required to complete step 101 to step 106).
The wheel speed signals of four wheels of the automobile can be collected through the wheel speed sensor, the wheel speed signals of the four wheels are subjected to filtering processing, and the filtering algorithm can be arithmetic average filtering, first-order lag filtering or amplitude limiting filtering, so that interference can be effectively removed. Setting a sample condition to screen the filtered wheel speed signals to obtain a wheel speed sample set { omega [ [ omega ] ]n}. The obtaining of the sample condition may be, for example, selecting a fixed threshold value or a movable value window according to a large amount of empirical data obtained during the daily driving of the vehicle. Meanwhile, the method for acquiring the longitudinal acceleration of the vehicle through the longitudinal acceleration of the vehicle acquired by the acceleration sensor may be as shown in step 1014 described above. The computing unit may then derive a set of wheel speed samples { ωnThe dynamic prediction model is built with the longitudinal acceleration as shown in step 1023, and the output of the dynamic prediction model is the current k-th cycle (denoted as T) for the vehiclek) Running speed Y ofkWhere k and n are positive integers, the initial value of k is 1, and i is less than n. And then may be based on the last cycle, i.e., the k-1 cycle (denoted as T)k-1) The obtained error correction parameters are used for correcting the dynamic prediction model, and the obtained state space vector and the longitudinal acceleration are used as the dynamic state after correctionInput of a prediction model, so that the current period T can be calculatedkPredicted value Y of vehicle speedk. When k is 1, namely the first period, since there is no previous period, the error correction parameter of the previous period is 0, and the corresponding dynamic prediction model is not updated. Then, the current period T can be obtained according to the obtained current period TkAnd errors between the vehicle speed predicted value output by the dynamic prediction model of each period in the previous k-1 periods and the reference wheel speed measured value are obtained, so that k errors are obtained, and part or all of the k errors are selected as error samples, so that an error sample set is obtained. So that the current period T can be established from the set of error sampleskAnd using the error prediction model to obtain the error prediction model for the next cycle, i.e., the (k + 1) th cycle (denoted as T)k+1) Error correction parameter of (2), which can be used for the next cycle Tk+1From step 101 to step 106, i.e. according to the cycle for the next period Tk+1The error correction parameter of (2) is corrected when step (101) to step (106) are performed next time, so that the next cycle T is obtained by using the corrected dynamic prediction modelk+1Predicted value Y of vehicle speedk+1. Therefore, the current vehicle speed predicted value of the vehicle can be obtained in real time by continuously and circularly executing the steps 101 to 106, and the vehicle speed predicted value can be more and more accurate by correcting the dynamic prediction model in each period.
The predicted value of the vehicle speed can be fed back to an electronic control module of the vehicle, and reliable reference information can be provided for driving, braking, turning and electronic stability control of the vehicle.
In conclusion, the present disclosure synthesizes wheel speeds of all wheels for modeling, may be applicable to various scenes in the vehicle driving process, and may avoid the problems of low generality and large error that may easily occur when calculating the vehicle speed by using the wheel speed of a single wheel in the prior art, so that the universality may be improved.
FIG. 7 is a block diagram illustrating a vehicle speed prediction device, as shown in FIG. 7, including: the system comprises an acquisition module 401, a prediction model establishing module 402, a state space vector acquisition module 403, a vehicle speed prediction module 404, an error acquisition module 405 and a correction parameter acquisition module 406;
the obtaining module 401 is configured to obtain a set of wheel speed samples of N wheels of a vehicle in a current cycle, and a longitudinal acceleration of the vehicle in the current cycle, where N is a positive integer.
And a prediction model establishing module 402, configured to establish a dynamic prediction model according to the wheel speed sample set in the current period and the longitudinal acceleration in the current period, and correct the dynamic prediction model according to the error correction parameter obtained in the previous period and used in the current period.
A state space vector obtaining module 403, configured to obtain a state space vector according to a preset state space model and a wheel speed sample set.
And a vehicle speed prediction module 404, configured to obtain a vehicle speed prediction value in the current period by using the modified dynamic prediction model according to the state space vector and the longitudinal acceleration in the current period.
An error obtaining module 405, configured to obtain an error sample set according to the predicted vehicle speed value in the current period and the predicted vehicle speed values obtained in M previous periods, and the wheel speed measured value of the reference wheel in the current period and the wheel speed measured value of the reference wheel obtained in M previous periods, where M is a non-negative integer.
And a correction parameter obtaining module 406, configured to obtain an error correction parameter for a next period according to the error sample set, and perform a calculation of the next period by using the error correction parameter for the next period, where a period from obtaining a wheel speed sample set of N wheels of the vehicle in a current period, and a longitudinal acceleration of the vehicle in the current period to obtaining the error correction parameter for the next period according to the error sample set is one period.
Optionally, fig. 8 is a block diagram illustrating an obtaining module according to an exemplary embodiment, and as shown in fig. 8, the obtaining module 401 may include:
the signal acquisition module 4011 is configured to obtain a plurality of wheel speed signals of N wheels of a vehicle.
The filtering module 4012 is configured to filter the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals.
The sample set obtaining module 4013 is configured to select a wheel speed signal that meets a preset sample condition according to the filtered wheel speed signals, so as to obtain a wheel speed sample set.
The acceleration obtaining module 4014 is configured to obtain a longitudinal acceleration measured value of the vehicle in the current period, which is acquired by an acceleration sensor of the vehicle, and filter the longitudinal acceleration measured value in the current period through a preset filtering algorithm to obtain a longitudinal acceleration in the current period.
Optionally, fig. 9 is a block diagram illustrating a prediction model building module according to an exemplary embodiment, and as shown in fig. 9, the prediction model building module 402 may include: a deviation acquisition module 4021, a wheel acceleration acquisition module 4022, and a modeling module 4023.
The deviation obtaining module 4021 is configured to obtain a deviation between a wheel speed of the N wheels in a previous period and a predicted vehicle speed value of the previous period according to the wheel speed sample set of the previous period and the predicted vehicle speed value of the previous period;
the wheel acceleration acquisition module 4022 is configured to acquire wheel accelerations of the N wheels according to the wheel speed sample set of the current period;
the modeling module 4023 is configured to establish a dynamic prediction model according to the wheel speed sample set of the current cycle, the deviation, the wheel accelerations of the N wheels, and the longitudinal acceleration of the current cycle.
Optionally, fig. 10 is a block diagram illustrating a modified parameter obtaining module according to an exemplary embodiment, and as shown in fig. 10, the modified parameter obtaining module 406 may include:
and an error model establishing module 4061, configured to establish an error prediction model of the current period according to the error sample set.
A parameter obtaining module 4062, configured to obtain an error correction parameter for the next cycle by using the error prediction model.
Optionally, fig. 11 is a block diagram illustrating an error obtaining module according to an exemplary embodiment, and as shown in fig. 11, the error obtaining module 405 may include:
the error calculating module 4051 is configured to obtain errors between the predicted vehicle speed value and the measured reference wheel speed value in each of the current cycle and M cycles, respectively, to obtain a plurality of errors.
The error sample screening module 4052 is configured to select a part or all of the multiple errors as an error sample, and obtain an error sample set.
Optionally, the dynamic prediction model may include:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the right front wheel in the k-1 th period and the predicted value of the vehicle speed, and the deviation between the wheel speed of the right rear wheel in the k-1 th period and the predicted value of the vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel of the k-th cycle, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
Optionally, the state space vector is a vector with the wheel speeds of N wheels as elements, and the state space model may include:
wherein,respectively represents the wheel speed value, lambda, of any wheel of the vehicle in the k +1 th cycle and the k-th cycle1、λ2Represents a parameter value and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1.
Optionally, the error prediction model may include:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,indicating the estimated value of the error correction value for the k-th cycle, Q indicating the processNoise covariance, zkRepresents the actual value of the error for the k-th cycle, R represents the measurement noise covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkThe kalman gain in the k-th cycle is expressed, and I denotes an identity matrix.
Optionally, the filtering algorithm for obtaining the vehicle acceleration may include:
wherein,respectively represents the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cyclekIndicating the longitudinal acceleration value of the k-th cycle.
The specific description of the functions implemented by the modules has been described in detail in the above method embodiments, and is not repeated here.
To sum up, this disclosure can avoid not having the information of four fast of make full use of among the prior art, the commonality is low, the great problem of error, consequently can improve the commonality, simultaneously, the technical scheme that this disclosure provided can update the model of prediction speed in real time to can effectively reduce the braking, sideslip in-process wheel speed error interference of bringing, improve the degree of accuracy of prediction, reduce the error.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made therein without departing from the scope thereof, and that any combination between the various embodiments of the present disclosure may also be considered as disclosed herein, unless it departs from the spirit of the present disclosure. The scope of the present disclosure is limited only by the appended claims.
Claims (16)
1. A vehicle speed prediction method, characterized by comprising:
acquiring a wheel speed sample set of N wheels of a vehicle in a current period and a longitudinal acceleration of the vehicle in the current period, wherein N is a positive integer;
establishing a dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period, and correcting the dynamic prediction model according to the error correction parameter obtained from the previous period and used for the current period; the establishing of the dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period comprises: acquiring the deviation of the wheel speeds of the N wheels in the last period and the predicted vehicle speed value in the last period according to the wheel speed sample set in the last period and the predicted vehicle speed value in the last period; respectively acquiring wheel acceleration of the N wheels according to the wheel speed sample set of the current period; establishing a dynamic prediction model according to the wheel speed sample set of the current period, the deviation, the wheel acceleration of the N wheels and the longitudinal acceleration of the current period;
acquiring a state space vector according to a preset state space model and the wheel speed sample set;
acquiring a predicted value of the vehicle speed in the current period by using a modified dynamic prediction model according to the state space vector and the longitudinal acceleration in the current period;
acquiring an error sample set according to the predicted vehicle speed value of the current period, the predicted vehicle speed values acquired in M periods before, the measured reference wheel speed values of the current period and the measured reference wheel speed values acquired in the M periods, wherein M is a non-negative integer;
and acquiring error correction parameters for the next period according to the error sample set, and performing calculation of the next period by using the error correction parameters for the next period, wherein the period from the acquisition of the wheel speed sample set of the N wheels of the vehicle in the current period to the acquisition of the error correction parameters for the next period according to the error sample set is one period.
2. The method of claim 1, wherein the obtaining a set of wheel speed samples of N wheels of a vehicle over a current cycle and a longitudinal acceleration of the vehicle over the current cycle comprises:
acquiring a plurality of wheel speed signals of N wheels of the vehicle, which are acquired within a preset time period;
filtering the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals;
selecting a wheel speed signal meeting a preset sample condition according to the filtered wheel speed signals to obtain a wheel speed sample set;
and acquiring a longitudinal acceleration measured value of the vehicle in the current period, which is acquired by an acceleration sensor of the vehicle, and filtering the longitudinal acceleration measured value of the current period by a preset filtering algorithm to obtain the longitudinal acceleration of the current period.
3. The method of claim 1, wherein obtaining error correction parameters for a next cycle from the set of error samples comprises:
establishing an error prediction model of the current period according to the error sample set;
and acquiring the error correction parameters for the next period by using the error prediction model.
4. The method according to claim 1, wherein the obtaining an error sample set according to the predicted vehicle speed value of the current cycle and the predicted vehicle speed values obtained in M previous cycles, and the reference measured vehicle speed value of the current cycle and the reference measured vehicle speed values obtained in M previous cycles comprises:
respectively obtaining errors of the predicted vehicle speed value and the measured reference wheel speed value of each period in the current period and the M periods to obtain a plurality of errors;
and selecting part or all of the errors as error samples to obtain the error sample set.
5. The method according to any of claims 1-4, wherein the dynamic predictive model comprises:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the right front wheel in the k-1 th period and the predicted value of the vehicle speed, and the deviation between the wheel speed of the right rear wheel in the k-1 th period and the predicted value of the vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,a left front wheel speed, a right front wheel speed, a left rear wheel speed, a right rear wheel speed, a left rear wheel speed,Speed of right rear wheel, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
6. The method of any of claims 1-4, wherein the state space vector is a vector having wheel speeds of the N wheels as elements, and wherein the state space model comprises:
wherein,respectively represents the wheel speed value, lambda, of any wheel of the vehicle in the k +1 th cycle and the k-th cycle1、λ2Represents a parameter value and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1.
7. The method of claim 3, wherein the error prediction model comprises:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,representing the estimated value of the error correction for the k-th cycle, Q representing the process noise covariance, zkRepresents the actual value of the error for the k-th cycle, R represents the measurement noise covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkKalman gain, I denotes the identity matrix, v, for the k-th cyclekRepresenting noise.
8. The method of claim 2, wherein the filtering algorithm comprises:
wherein,the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cycle, respectivelykRepresents the longitudinal acceleration value, delta, of the k-th cycle0A threshold value representing the difference between the longitudinal accelerations of two adjacent cycles.
9. A vehicle speed prediction apparatus, characterized by comprising: the system comprises an acquisition module, a prediction model establishing module, a state space vector acquisition module, a vehicle speed prediction module, an error acquisition module and a correction parameter acquisition module;
the acquisition module is used for acquiring a wheel speed sample set of N wheels of a vehicle in a current period and a longitudinal acceleration of the vehicle in the current period, wherein N is a positive integer;
the prediction model establishing module is used for establishing a dynamic prediction model according to the wheel speed sample set of the current period and the longitudinal acceleration of the current period, and correcting the dynamic prediction model according to the error correction parameter obtained in the previous period and used for the current period; the prediction model building module comprises: the device comprises a deviation acquisition module, a wheel acceleration acquisition module and a modeling module; the deviation obtaining module is used for obtaining the deviation between the wheel speed of the N wheels in the previous period and the predicted vehicle speed value of the previous period according to the wheel speed sample set of the previous period and the predicted vehicle speed value of the previous period; the wheel acceleration acquisition module is used for respectively acquiring the wheel acceleration of the N wheels according to the wheel speed sample set of the current period; the modeling module is used for establishing a dynamic prediction model according to the wheel speed sample set of the current period, the deviation, the wheel acceleration of the N wheels and the longitudinal acceleration of the current period;
the state space vector acquisition module is used for acquiring a state space vector according to a preset state space model and the wheel speed sample set;
the vehicle speed prediction module is used for acquiring a vehicle speed prediction value of the current period by using a modified dynamic prediction model according to the state space vector and the longitudinal acceleration of the current period;
the error acquisition module is configured to acquire an error sample set according to the predicted vehicle speed value of the current period and the predicted vehicle speed values acquired in M previous periods, the measured reference wheel speed value of the current period and the measured reference wheel speed values acquired in M previous periods, where M is a non-negative integer;
and the correction parameter acquisition module is used for acquiring error correction parameters for the next period according to the error sample set and calculating the next period by using the error correction parameters for the next period, wherein the period from the acquisition of the wheel speed sample set of the N wheels of the vehicle in the current period to the acquisition of the error correction parameters for the next period according to the error sample set is one period.
10. The apparatus of claim 9, wherein the obtaining module comprises: the device comprises a signal acquisition module, a filtering module, a sample set acquisition module and an acceleration acquisition module;
the signal acquisition module is used for acquiring a plurality of acquired wheel speed signals of N wheels of the vehicle;
the filtering module is used for filtering the plurality of wheel speed signals to obtain a plurality of filtered wheel speed signals;
the sample set acquisition module is used for selecting a wheel speed signal meeting a preset sample condition according to the filtered wheel speed signals to obtain a wheel speed sample set;
the acceleration acquisition module is configured to acquire a longitudinal acceleration measured value of the vehicle in the current period, which is acquired by an acceleration sensor of the vehicle, and filter the longitudinal acceleration measured value of the current period through a preset filtering algorithm to obtain the longitudinal acceleration of the current period.
11. The apparatus of claim 9, wherein the correction parameter obtaining module comprises:
the error model establishing module is used for establishing an error prediction model of the current period according to the error sample set;
and the parameter acquisition module is used for acquiring the error correction parameters for the next period by utilizing the error prediction model.
12. The apparatus of claim 9, wherein the error acquisition module comprises:
the error calculation module is used for respectively obtaining the errors of the vehicle speed predicted value and the reference wheel speed measured value in each period in the current period and the M periods to obtain a plurality of errors;
and the error sample screening module is used for selecting part or all of the errors as error samples to obtain the error sample set.
13. The apparatus according to any of claims 9-12, wherein the dynamic predictive model comprises:
wherein,respectively representing the deviation between the wheel speed of the left front wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the left rear wheel in the k-1 th period and the predicted value of the vehicle speed, the deviation between the wheel speed of the right front wheel in the k-1 th period and the predicted value of the vehicle speed, and the deviation between the wheel speed of the right rear wheel in the k-1 th period and the predicted value of the vehicle speed,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel in the k-1 th period,respectively showing the wheel speed of the left front wheel, the wheel speed of the right front wheel, the wheel speed of the left rear wheel and the wheel speed of the right rear wheel of the k-th cycle, VkIs a predicted value of the vehicle speed for the k-th cycle,indicating the left front wheel acceleration for the k-th cycle,indicating the right front wheel acceleration for the k-th cycle,representing the left rear wheel acceleration for the k-th cycle,representing the right rear wheel acceleration, ax, of the k-th cyclekRepresents the longitudinal acceleration of the vehicle for the k-th cycle, and Δ represents a threshold value for the difference between the wheel acceleration and the longitudinal acceleration.
14. The apparatus of any of claims 9-12, wherein the state space vector is a vector having wheel speeds of the N wheels as elements, and wherein the state space model comprises:
wherein,respectively represents the wheel speed value, lambda, of any wheel of the vehicle in the k +1 th cycle and the k-th cycle1、λ2Represents a parameter value and satisfies lambda1+λ2Δ denotes a deviation threshold of the state space vectors of two adjacent cycles, 1.
15. The apparatus of claim 11, wherein the error prediction model comprises:
xk＝xk-1+vk
wherein x iskAn error correction value indicating the k-th cycle,a predicted value indicating an error correction value for the k-th cycle,representing the estimated value of the error correction for the k-th cycle, Q representing the process noise covariance, zkRepresents the actual value of the error for the k-th cycle, R represents the measurement noise covariance,represents the estimation error covariance for the k-th cycle,represents the prediction error covariance of the K-th cycle, KkKalman gain, I denotes the identity matrix, v, for the k-th cyclekRepresenting noise.
16. The apparatus of claim 10, wherein the filtering algorithm comprises:
wherein,the longitudinal acceleration measured values, ax, of the vehicle in the k-th cycle and the k-1 th cycle, respectivelykIndicating the longitudinal acceleration value of the k-th cycle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610875080.4A CN107878462B (en) | 2016-09-30 | 2016-09-30 | Speed prediction method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610875080.4A CN107878462B (en) | 2016-09-30 | 2016-09-30 | Speed prediction method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107878462A CN107878462A (en) | 2018-04-06 |
| CN107878462B true CN107878462B (en) | 2019-11-22 |
Family
ID=61769523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610875080.4A Active CN107878462B (en) | 2016-09-30 | 2016-09-30 | Speed prediction method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107878462B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110231497A (en) * | 2019-06-19 | 2019-09-13 | 浙江吉利控股集团有限公司 | A kind of speed calibrating installation to be tested the speed based on laser radar and method |
| CN112550300B (en) * | 2019-09-25 | 2022-05-13 | 比亚迪股份有限公司 | Vehicle speed detection method and device, storage medium, electronic equipment and vehicle |
| CN110920629B (en) * | 2019-12-13 | 2021-08-27 | 华人运通(江苏)技术有限公司 | Vehicle speed calculation method and device and computer readable storage medium |
| CN111071259B (en) * | 2019-12-27 | 2020-09-29 | 清华大学 | Vehicle speed prediction method, vehicle speed prediction device, vehicle control device, and storage medium |
| CN112660143B (en) * | 2020-12-25 | 2023-01-24 | 浙江吉利控股集团有限公司 | Longitudinal vehicle speed estimation method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102009002360A1 (en) * | 2009-04-14 | 2010-10-21 | Robert Bosch Gmbh | Method for determining reference speed of all-wheel-driven vehicle, involves comparing model-based model reference speed with correction speed, and setting reference speed to smaller value of model reference speed and correction speed |
| CN104742888A (en) * | 2015-02-06 | 2015-07-01 | 中国第一汽车股份有限公司 | Full-driven vehicle reference vehicle speed real-time detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2923596B1 (en) * | 2007-11-09 | 2010-01-15 | Michelin Soc Tech | SYSTEM FOR MONITORING THE BEHAVIOR OF THE GROUND OF A VEHICLE COMPRISING A DETERMINATION OF THE SLOPE OF THE BEARING FLOOR 1 |
-
2016
- 2016-09-30 CN CN201610875080.4A patent/CN107878462B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102009002360A1 (en) * | 2009-04-14 | 2010-10-21 | Robert Bosch Gmbh | Method for determining reference speed of all-wheel-driven vehicle, involves comparing model-based model reference speed with correction speed, and setting reference speed to smaller value of model reference speed and correction speed |
| CN104742888A (en) * | 2015-02-06 | 2015-07-01 | 中国第一汽车股份有限公司 | Full-driven vehicle reference vehicle speed real-time detection method |
Non-Patent Citations (1)
| Title |
|---|
| 基于模糊扩展卡尔曼滤波的轮毂电机驱动车辆纵向速度估计算法;王志福等;《西南交通大学学报》;20151202;第50卷(第6期);第1095页右栏第2段至第1097页右栏第4段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107878462A (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107878462B (en) | Speed prediction method and apparatus | |
| EP1641639B1 (en) | Tire pressure estimation | |
| US10597039B2 (en) | Method for estimating variables affecting the vehicle dynamics and corresponding virtual sensor | |
| EP2145779B1 (en) | Apparatus and method for detecting decrease in tire air pressure and program for detecting decrease in tire air pressure | |
| CN113002549A (en) | Vehicle state estimation method, device, equipment and storage medium | |
| CN109684677A (en) | A kind of gradient evaluation method based on Kalman filtering algorithm | |
| CN113434967B (en) | Method and device for estimating wheel speed of vehicle, control method and equipment and storage medium | |
| JP2009274639A (en) | Tire pressure drop detecting device, method, and tire pressure drop detecting program | |
| CN108931233B (en) | Road side slope value detection method and device | |
| CN111323848A (en) | Method for correcting vehicle rainfall sensor and storage medium | |
| CN110400331B (en) | Depth map processing method and device | |
| CN112417750B (en) | Virtual sensor for estimating on-line unmeasurable variables via continuous time derivative | |
| JP2002160512A (en) | Tire pressure estimation device | |
| CN109910527B (en) | Method and device for determining tire pressure of automobile tires | |
| CN111942399A (en) | Vehicle speed estimation method and system based on unscented Kalman filtering | |
| CN108287934A (en) | A Robust Estimation Method of Vehicle Center of Mass Side Slip Angle Based on Longitudinal Force Observer | |
| CN106004881A (en) | Road adhesion coefficient estimation method based on frequency domain fusion | |
| CN110843550B (en) | Method for eliminating low-speed jitter of electric automobile | |
| US20040153216A1 (en) | Method for estimating a vehicle's velocity | |
| CN115284809B (en) | Intelligent internet fleet active suspension control method and system and computer equipment | |
| CN117962909A (en) | Whole vehicle quality estimation method, system, vehicle and equipment | |
| JP2009119958A (en) | Vehicle state estimation unit | |
| JP2016040721A (en) | Vehicle data recording device | |
| CN116729412A (en) | Method, device, equipment and medium for adapting kinematic model in vehicle positioning | |
| CN113358371B (en) | Sprung acceleration estimation method, storage medium, and electronic apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |