CN109118786B - Vehicle speed prediction method based on quantization adaptive Kalman filtering - Google Patents

Vehicle speed prediction method based on quantization adaptive Kalman filtering Download PDF

Info

Publication number
CN109118786B
CN109118786B CN201810947025.0A CN201810947025A CN109118786B CN 109118786 B CN109118786 B CN 109118786B CN 201810947025 A CN201810947025 A CN 201810947025A CN 109118786 B CN109118786 B CN 109118786B
Authority
CN
China
Prior art keywords
vehicle
time
follows
acceleration
roadside unit
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
Application number
CN201810947025.0A
Other languages
Chinese (zh)
Other versions
CN109118786A (en
Inventor
钱丽萍
冯安琪
黄玉蘋
冯旭
黄亮
吴远
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN201810947025.0A priority Critical patent/CN109118786B/en
Publication of CN109118786A publication Critical patent/CN109118786A/en
Application granted granted Critical
Publication of CN109118786B publication Critical patent/CN109118786B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

A vehicle speed prediction method based on quantization adaptive Kalman filtering comprises the following steps: firstly, in an intelligent network traffic system, automatically identifying running vehicles and acquiring related data through a DSRC technology, so as to realize information interaction between a vehicle-mounted system and a roadside unit; secondly, quantizing the azimuth angles of the roadside unit and the vehicle-mounted system by a quantization formula according to the collected related information, predicting the acceleration by an autoregressive moving average method, and finally predicting the speed by using adaptive Kalman filtering to reach a speed correction value; and finally, transmitting the processed information to other 3 roadside units through an optical cable so as to facilitate information interaction with the vehicle-mounted system next time. The invention provides a vehicle speed prediction method based on quantitative adaptive Kalman filtering in an intelligent internet traffic system.

Description

Vehicle speed prediction method based on quantization adaptive Kalman filtering
Technical Field
The invention belongs to the field of traffic, and particularly relates to a vehicle speed prediction method based on quantitative adaptive Kalman filtering in an intelligent networked traffic system.
Background
China is the country with the most population in the world, the reform is open, along with the rapid development of the economy of China, the living standard of people is increasingly improved, private cars begin to enter each family, and the traveling of the family is facilitated well. However, the popularization and popularization of vehicles also make the urban traffic environment worsen day by day, and traffic phenomena such as unbalanced traffic flow, traffic congestion, rear collision, side collision and the like occur. Along with the weak infrastructure and the traffic network congestion, the number of road traffic accidents is increasing, and the high incidence of the traffic accidents is sounding an alarm to the whole society, so that the road traffic safety is greatly concerned. In recent years, although the road infrastructure and the traffic network are greatly improved in China, the number of traffic accidents and casualties are reduced, but the total number of accidents and the incidence rate are still high.
Compared with the traditional road traffic system, the intelligent internet traffic system is more prone to a dynamic system in which information interaction is carried out by people, roads, vehicles, road traffic facilities and the like. According to a large amount of statistical research in various countries, it is found that driver error is a main factor causing traffic accidents. Therefore, under the condition that the current road infrastructure can not be improved, the work of acquiring the state information of the vehicles in other lanes of the road, processing and broadcasting the state information to the current vehicle is not slow, so that a driver can better take corresponding remedial measures, and traffic accidents caused by errors of the driver are reduced.
Disclosure of Invention
In order to overcome the defects of low safety and high traffic accident occurrence probability of the conventional road traffic system, the invention provides a vehicle speed prediction method based on quantitative adaptive Kalman filtering in an intelligent internet traffic system.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a vehicle speed prediction method based on quantization adaptive Kalman filtering comprises the following steps:
1) in an intelligent network traffic system, a vehicle is automatically identified through a DSRC technology, relevant data are obtained, and information interaction between a vehicle-mounted system and a roadside unit is realized, wherein the information interaction comprises the following steps:
step 1.1: when a running vehicle enters the range covered by the directional antenna, the vehicle-mounted system and the roadside unit realize two-way communication through the DSRC technology, so that the two sides can simultaneously transmit information on the storage units of the vehicle-mounted system, wherein the information transmitted by the vehicle-mounted system comprises the current speed, the current position and the time stamp of the vehicle, and the information transmitted by the roadside unit comprises the predicted position of the vehicle on other lanes, the direction in which the vehicle is located, several lanes and the acceleration;
step 1.2: the roadside unit sends the acquired vehicle information to an edge cloud server to perform a series of operation operations;
2) the edge cloud server calculates an azimuth angle and performs corresponding quantization processing according to the azimuth difference between the roadside unit and the vehicle-mounted system, and quantizes the vehicle driving direction according to the position information, wherein the quantization process comprises the following steps:
step 2.1: converting the position information into digital information which is present at a viewing angle, wherein the actual bearing angle of the vehicle
Figure BDA0001770498450000021
The definition with respect to a roadside unit is:
Figure BDA0001770498450000022
Figure BDA0001770498450000031
here, the parameters are defined as follows:
Figure BDA0001770498450000032
azimuth angle between roadside unit and vehicle system at time t;
θt: converting the vehicle position at the time t into an inverse trigonometric function of the azimuth angle;
σt: bearing error noise caused by signal reflection at time t;
(xt,yt): the current position of the vehicle at time t;
step 2.2: using the center of the crossroad as the origin of coordinates and the angle of opposite direction
Figure BDA0001770498450000033
And performing quantization processing to determine the direction of the vehicle, wherein the quantization formula is as follows:
Figure BDA0001770498450000034
here, the parameters are defined as follows:
bt: the quantitative direction of the vehicle at the crossroad at the moment t;
q (·): a quantization function;
i: direction identification of the crossroad;
step 2.3: to specify the direction information, the lane of the direction is quantized, and the actual bearing angle is set
Figure BDA0001770498450000035
Rename it to
Figure BDA0001770498450000036
It is recorded as:
Figure BDA0001770498450000037
Figure BDA0001770498450000038
here, the parameters are defined as follows:
Figure BDA0001770498450000041
azimuth angle between roadside unit and vehicle system at time t;
Figure BDA0001770498450000042
at the moment t, converting the relative position of the vehicle and the roadside unit into an inverse trigonometric function of the azimuth angle;
(xi,yi): the fixed position of the roadside unit in the direction i;
step 2.4: according to the quantization formula, performing second quantization on the direction of the vehicle to determine the lane of the vehicle;
Figure BDA0001770498450000043
here, the parameters are defined as follows:
qt: the quantized direction of the lane at time t;
ji: the jth lane in direction i;
n: total number of lanes;
3) assuming that only the latest p +1 vehicle speeds are used for acceleration estimation, the pth acceleration calculation method is:
Figure BDA0001770498450000044
here, the parameters are defined as follows:
Δ τ: a sampling time interval;
Δ v: the difference in velocity between the next time and the previous time;
vt-p: the speed of the trolley at the time t-p;
τt-p: a timestamp of the trolley at the time t-p;
at-p: the p-th acceleration value;
thereafter, vehicle acceleration prediction is performed by using an autoregressive moving average method according to the p acceleration values, wherein the prediction formula is as follows:
Figure BDA0001770498450000045
here, the parameters are defined as follows:
at: acceleration of the trolley at time t;
p: the autoregressive order, namely the total acceleration;
q: moving average order, i.e., total number of slips;
β, undetermined coefficient not equal to zero;
Figure BDA0001770498450000051
undetermined coefficients other than zero;
ξt: an error term independent at time t;
4) the method comprises the following steps that the edge cloud server predicts the speed of a running vehicle by using an adaptive Kalman filtering algorithm according to collected vehicle information and an acceleration predicted value, wherein the calculation formula of the speed is as follows:
vt+1＝vt+at×Δτ；
converting the above formula into a state equation and an observation equation by using a state space model, wherein the equations are as follows:
vt+1＝Atvt+Btut+ωt；
zt＝Ctvt+t；
here, the parameters are defined as follows:
vt: the speed of the trolley at time t;
At,Bt,Ct: a state transition matrix at time t;
ut: acceleration of the trolley at time t;
ωt: the systematic error at time t obeys the Gaussian distribution N (0, Q)t) Wherein Q istIs the process noise variance at time t;
t: measurement error at time t obeys Gaussian distribution N (0, R)t) Wherein R istIs the measured noise variance at time t;
zt: a state observation of the system at time t;
and then, predicting the vehicle speed by using an adaptive Kalman filtering algorithm according to the state space model, wherein the vehicle speed is updated by the following steps:
step 4.1: giving an initial value
Figure BDA0001770498450000061
And
Figure BDA0001770498450000062
wherein,
Figure BDA0001770498450000063
showing the predicted value of the horizontal position of the trolley at the moment t,
Figure BDA0001770498450000064
representing the error covariance of the trolley at the time t;
step 4.2: according to a given initial value
Figure BDA0001770498450000065
Calculating Kalman gain value K at t momenttWherein
Figure BDA0001770498450000066
step 4.3: predicted value according to time t
Figure BDA0001770498450000067
And the observed value ztThe corrected value of the current state can be obtained
Figure BDA0001770498450000068
Wherein, the formula is as follows:
zt＝Ctvt+Rt
Figure BDA0001770498450000069
step 4.4: updating error covariance
Figure BDA00017704984500000610
In preparation for predicting the error covariance at time t +1, wherein,
Figure BDA00017704984500000611
i is a unit array;
step 4.5: correction value according to time t
Figure BDA00017704984500000612
And acceleration utTo predict the speed of the car at the time t +1
Figure BDA00017704984500000613
Wherein,
Figure BDA00017704984500000614
step (ii) of4.6: prediction of error covariance from the error covariance of the previous time instant
Figure BDA00017704984500000615
To predict the error covariance of the next time instant
Figure BDA00017704984500000616
Wherein,
Figure BDA00017704984500000617
μt+1is an adaptive forgetting factor;
step 4.7: calculating the acceleration u at time t +1t+1Wherein the calculation method is the autoregressive moving average method in the 3) of claim 1;
step 4.8: updating the iteration number k to k +1, and returning to the step 4.2 to start a new round of calculation;
5) and finally, the edge cloud server transmits the processed information to the roadside unit through an optical cable so as to facilitate the information interaction with the vehicle-mounted system next time.
Further, in step 4.6, the calculation formula of the adaptive forgetting factor is as follows:
μt+1＝max{1,tr(Nt+1)/tr(Mt+1)}；
Figure BDA0001770498450000071
Figure BDA0001770498450000072
Figure BDA0001770498450000073
Figure BDA0001770498450000074
here, the parameters are defined as follows:
Nt+1: error at time t +1Variance, ensuring its value is symmetric and positive;
Lt+1: covariance matrix of innovation at time t + 1;
et+1: the difference between the measured value and the predicted value at the moment t +1 is the innovation;
Figure BDA0001770498450000075
transposing the state transition matrix C at time t + 1;
Mt+1: error variance at time t +1, ensuring error covariance
Figure BDA0001770498450000076
The values of (a) are symmetrical and positive; max {. cndot ]: and taking the maximum value after comparison.
Furthermore, in the step 1), in the intelligent internet traffic system, the roadside units are installed on traffic lights at the crossroad and are attached with the cloud server and the directional antenna, wherein the emission angle of the routing antenna is 60 degrees, so that the roadside units can better perform information interaction with a vehicle-mounted system in a vehicle.
Still further, in step 1.2, the data in the server is cleared every other week in consideration of the limited storage capacity of the edge cloud server.
The technical conception of the invention is as follows: firstly, in an intelligent network traffic system, a vehicle is automatically identified through a DSRC technology, and relevant data are acquired, so that information interaction between a vehicle-mounted system and a roadside unit is realized. Secondly, quantizing the azimuth angles of the roadside unit and the vehicle-mounted system by using a quantization formula according to the collected related information; predicting the acceleration by using an autoregressive moving average method; and (4) carrying out speed prediction by using the adaptive Kalman filtering to reach a speed correction value. And finally, sending the processed information to other 3 roadside units through an optical cable so as to facilitate information interaction with the vehicle-mounted system next time.
The invention has the advantages that 1, the lanes in which the current vehicle is positioned can be clearly known through quantifying the azimuth difference between the roadside unit and the vehicle-mounted system. 2. The prediction of acceleration and speed is realized by combining an autoregressive moving average method and an adaptive Kalman filtering method, and the result is transmitted to a driver, so that the driver can make proper judgment and behavior according to the relevant information of the vehicle and own experience, and the occurrence rate of traffic accidents is effectively reduced.
Drawings
Fig. 1 is a schematic diagram of mobile internet traffic system information interaction.
Detailed Description
The present invention is described in further detail below with reference to the attached drawing figures.
Referring to fig. 1, a vehicle speed prediction method based on quantization adaptive kalman filtering, the present invention is based on an information interaction model (as shown in fig. 1) under DSRC technology communication. In an intelligent internet traffic system, firstly, an azimuth angle of a roadside unit and an azimuth angle of a vehicle-mounted system are quantized through a quantization formula, secondly, acceleration is predicted through an autoregressive moving average method, and finally, speed prediction is carried out through adaptive Kalman filtering to reach a speed correction value, wherein the prediction method comprises the following steps:
1) in an intelligent network traffic system, a vehicle is automatically identified through a DSRC technology, relevant data are obtained, and information interaction between a vehicle-mounted system and a roadside unit is realized, wherein the information interaction comprises the following steps:
step 1.1: when a running vehicle enters the range covered by the directional antenna, the vehicle-mounted system and the roadside unit realize two-way communication through the DSRC technology, so that the two sides can simultaneously transmit information on the storage units of the vehicle-mounted system, wherein the information transmitted by the vehicle-mounted system comprises the current speed, the current position and the time stamp of the vehicle, and the information transmitted by the roadside unit comprises the predicted position of the vehicle on other lanes, the direction in which the vehicle is located, several lanes and the acceleration;
step 1.2: the roadside unit sends the acquired vehicle information to an edge cloud server to perform a series of operation operations;
2) the edge cloud server calculates an azimuth angle and performs corresponding quantization processing according to the azimuth difference between the roadside unit and the vehicle-mounted system, and quantizes the vehicle driving direction according to the position information, wherein the quantization process comprises the following steps:
step 2.1: converting the position information into digital information which is present at a viewing angle, wherein the actual bearing angle of the vehicle
Figure BDA0001770498450000091
The definition with respect to a roadside unit is:
Figure BDA0001770498450000092
Figure BDA0001770498450000093
here, the parameters are defined as follows:
Figure BDA0001770498450000094
azimuth angle between roadside unit and vehicle system at time t;
θt: converting the vehicle position at the time t into an inverse trigonometric function of the azimuth angle;
σt: bearing error noise caused by signal reflection at time t;
(xt,yt): the current position of the vehicle at time t;
step 2.2: using the center of the crossroad as the origin of coordinates and the angle of opposite direction
Figure BDA0001770498450000101
And performing quantization processing to determine the direction of the vehicle, wherein the quantization formula is as follows:
Figure BDA0001770498450000102
here, the parameters are defined as follows:
bt: the quantitative direction of the vehicle at the crossroad at the moment t;
q (·): a quantization function;
i: direction identification of the crossroad;
step 2.3: to specify the direction information, the lane of the direction is quantized, and the actual bearing angle is set
Figure BDA0001770498450000103
Rename it to
Figure BDA0001770498450000104
It is recorded as:
Figure BDA0001770498450000105
Figure BDA0001770498450000106
here, the parameters are defined as follows:
Figure BDA0001770498450000107
azimuth angle between roadside unit and vehicle system at time t;
Figure BDA0001770498450000108
at the moment t, converting the relative position of the vehicle and the roadside unit into an inverse trigonometric function of the azimuth angle; (x)i,yi): the fixed position of the roadside unit in the direction i;
step 2.4: according to the quantization formula, performing second quantization on the direction of the vehicle to determine the lane of the vehicle;
Figure BDA0001770498450000109
here, the parameters are defined as follows:
qt: the quantized direction of the lane at time t;
ji: the jth lane in direction i;
n: total number of lanes;
3) assuming that only the latest p +1 vehicle speeds are used for acceleration estimation, the pth acceleration calculation method is:
Figure BDA0001770498450000111
here, the parameters are defined as follows:
Δ τ: a sampling time interval;
Δ v: the difference in velocity between the next time and the previous time;
vt-p: the speed of the trolley at the time t-p;
τt-p: a timestamp of the trolley at the time t-p;
at-p: the p-th acceleration value;
thereafter, vehicle acceleration prediction is performed by using an autoregressive moving average method according to the p acceleration values, wherein the prediction formula is as follows:
Figure BDA0001770498450000112
here, the parameters are defined as follows:
at: acceleration of the trolley at time t;
p: the autoregressive order, namely the total acceleration;
q: moving average order, i.e., total number of slips;
β, undetermined coefficient not equal to zero;
Figure BDA0001770498450000113
undetermined coefficients other than zero;
ξt: an error term independent at time t;
4) the method comprises the following steps that the edge cloud server predicts the speed of a running vehicle by using an adaptive Kalman filtering algorithm according to collected vehicle information and an acceleration predicted value, wherein the calculation formula of the speed is as follows:
vt+1＝vt+at×Δτ；
converting the above formula into a state equation and an observation equation by using a state space model, wherein the equations are as follows:
vt+1＝Atvt+Btut+ωt；
zt＝Ctvt+t；
here, the parameters are defined as follows:
vt: the speed of the trolley at time t;
At,Bt,Ct: a state transition matrix at time t;
ut: acceleration of the trolley at time t;
ωt: the systematic error at time t obeys the Gaussian distribution N (0, Q)t) Wherein Q istIs the process noise variance at time t;
t: measurement error at time t obeys Gaussian distribution N (0, R)t) Wherein R istIs the measured noise variance at time t;
zt: a state observation of the system at time t;
and then, predicting the vehicle speed by using an adaptive Kalman filtering algorithm according to the state space model, wherein the vehicle speed is updated by the following steps:
step 4.1: giving an initial value
Figure BDA0001770498450000121
And
Figure BDA0001770498450000122
wherein,
Figure BDA0001770498450000123
showing the predicted value of the horizontal position of the trolley at the moment t,
Figure BDA0001770498450000124
indicating the vehicle at time tAn error covariance;
step 4.2: according to a given initial value
Figure BDA0001770498450000125
Calculating Kalman gain value K at t momenttWherein
Figure BDA0001770498450000126
step 4.3: predicted value according to time t
Figure BDA0001770498450000127
And the observed value ztThe corrected value of the current state can be obtained
Figure BDA0001770498450000128
Wherein, the formula is as follows:
zt＝Ctvt+Rt
Figure BDA0001770498450000131
step 4.4: updating error covariance
Figure BDA0001770498450000132
In preparation for predicting the error covariance at time t +1, wherein,
Figure BDA0001770498450000133
i is a unit array;
step 4.5: correction value according to time t
Figure BDA0001770498450000134
And acceleration utTo predict the speed of the car at the time t +1
Figure BDA0001770498450000135
Wherein,
Figure BDA0001770498450000136
step 4.6: prediction of error covariance from the error covariance of the previous time instant
Figure BDA0001770498450000137
To predict the error covariance of the next time instant
Figure BDA0001770498450000138
Wherein,
Figure BDA0001770498450000139
μt+1is an adaptive forgetting factor;
step 4.7: calculating the acceleration u at time t +1t+1Wherein the calculation method is the autoregressive moving average method in the 3) of claim 1;
step 4.8: updating the iteration number k to k +1, and returning to the step 4.2 to start a new round of calculation;
5) and finally, the edge cloud server transmits the processed information to the roadside unit through an optical cable so as to facilitate the information interaction with the vehicle-mounted system next time.
Further, in step 4.6, the calculation formula of the adaptive forgetting factor is as follows:
μt+1＝max{1,tr(Nt+1)/tr(Mt+1)}；
Figure BDA00017704984500001310
Figure BDA00017704984500001311
Figure BDA00017704984500001312
Figure BDA00017704984500001313
here, the parameters are defined as follows:
Nt+1: the error variance at the moment of t +1 ensures that the values are symmetrical and positive;
Lt+1: covariance matrix of innovation at time t + 1;
et+1: the difference between the measured value and the predicted value at the moment t +1 is the innovation;
Figure BDA0001770498450000141
transposing the state transition matrix C at time t + 1;
Mt+1: error variance at time t +1, ensuring error covariance
Figure BDA0001770498450000142
The values of (a) are symmetrical and positive; max {. cndot ]: and taking the maximum value after comparison.
Furthermore, in the step 1), in the intelligent internet traffic system, the roadside units are installed on traffic lights at the crossroad and are attached with the cloud server and the directional antenna, wherein the emission angle of the routing antenna is 60 degrees, so that the roadside units can better perform information interaction with a vehicle-mounted system in a vehicle.
Still further, in step 1.2, the data in the server is cleared every other week in consideration of the limited storage capacity of the edge cloud server.

Claims (4)

1. A vehicle speed prediction method based on quantization adaptive Kalman filtering is characterized by comprising the following steps:
1) in an intelligent network traffic system, a vehicle is automatically identified through a DSRC technology, relevant data are obtained, and information interaction between a vehicle-mounted system and a roadside unit is realized, wherein the information interaction comprises the following steps:
step 1.1: when a running vehicle enters the range covered by the directional antenna, the vehicle-mounted system and the roadside unit realize two-way communication through the DSRC technology, so that the two sides can simultaneously transmit information on the storage units of the vehicle-mounted system, wherein the information transmitted by the vehicle-mounted system comprises the current speed, the current position and the time stamp of the vehicle, and the information transmitted by the roadside unit comprises the predicted position of the vehicle on other lanes, the direction in which the vehicle is located, several lanes and the acceleration;
step 1.2: the roadside unit sends the acquired vehicle information to an edge cloud server to perform a series of operation operations;
2) the edge cloud server calculates an azimuth angle and performs corresponding quantization processing according to the azimuth difference between the roadside unit and the vehicle-mounted system, and quantizes the vehicle driving direction according to the position information, wherein the quantization process comprises the following steps:
step 2.1: converting the position information into digital information existing at a viewing angle, wherein the azimuth angle between the roadside unit and the on-board system of the vehicle at the time t
Figure FDA0002484485270000011
The definition with respect to a roadside unit is:
Figure 1
Figure FDA0002484485270000021
here, the parameters are defined as follows:
Figure FDA0002484485270000022
azimuth angle between roadside unit and vehicle system at time t;
θt: converting the vehicle position at the time t into an inverse trigonometric function of the azimuth angle;
σt: bearing error noise caused by signal reflection at time t;
(xt,yt): the current position of the vehicle at time t;
step 2.2: using the center of the crossroad as the origin of coordinates, aiming at the azimuth angle between the roadside unit and the vehicle-mounted system at the moment t
Figure FDA0002484485270000023
And performing quantization processing to determine the direction of the vehicle, wherein the quantization formula is as follows:
Figure FDA0002484485270000024
here, the parameters are defined as follows:
bt: the quantitative direction of the vehicle at the crossroad at the moment t;
q (·): a quantization function;
i: direction identification of the crossroad;
step 2.3: in order to realize the concretization of the direction information, the lane of the direction is quantified, and the azimuth angle between the roadside unit and the vehicle-mounted system at the time t
Figure FDA0002484485270000025
Rename it to
Figure FDA0002484485270000026
It is recorded as:
Figure FDA0002484485270000031
Figure FDA0002484485270000032
here, the parameters are defined as follows:
θt': at the moment t, converting the relative position of the vehicle and the roadside unit into an inverse trigonometric function of the azimuth angle;
(xi,yi): the fixed position of the roadside unit in the direction i;
step 2.4: according to the quantization formula, performing second quantization on the direction of the vehicle to determine the lane of the vehicle;
Figure FDA0002484485270000033
here, the parameters are defined as follows:
qt: the quantized direction of the lane at time t;
ji: the jth lane in direction i;
n: total number of lanes;
3) assuming that only the latest p +1 vehicle speeds are used for acceleration estimation, the pth acceleration calculation method is:
Figure FDA0002484485270000034
here, the parameters are defined as follows:
Δ τ: a sampling time interval;
Δ v: the difference in velocity between the next time and the previous time;
vt-p: the speed of the trolley at the time t-p;
τt-p: a timestamp of the trolley at the time t-p;
at-p: the p-th acceleration value;
thereafter, vehicle acceleration prediction is performed by using an autoregressive moving average method according to the p acceleration values, wherein the prediction formula is as follows:
Figure FDA0002484485270000041
here, the parameters are defined as follows:
at: acceleration of the trolley at time t;
p: the autoregressive order, namely the total acceleration;
q: moving average order, i.e., total number of slips;
β, undetermined coefficient not equal to zero;
Figure FDA0002484485270000042
undetermined coefficients other than zero;
ξt: an error term independent at time t;
4) the method comprises the following steps that the edge cloud server predicts the speed of a running vehicle by using an adaptive Kalman filtering algorithm according to collected vehicle information and an acceleration predicted value, wherein the calculation formula of the speed is as follows:
vt+1＝vt+at×Δτ；
converting the above formula into a state equation and an observation equation by using a state space model, wherein the equations are as follows:
vt+1＝Atvt+Btut+ωt；
zt＝Ctvt+t；
here, the parameters are defined as follows:
vt: the speed of the trolley at time t;
At,Bt,Ct: a state transition matrix at time t;
ut: acceleration of the trolley at time t;
ωt: the systematic error at time t obeys the Gaussian distribution N (0, Q)t) Wherein Q istIs the process noise variance at time t;
t: measurement error at time t obeys Gaussian distribution N (0, R)t) Wherein R istIs the measured noise variance at time t;
zt: a state observation of the system at time t;
and then, predicting the vehicle speed by using an adaptive Kalman filtering algorithm according to the state space model, wherein the vehicle speed is updated by the following steps:
step 4.1: giving an initial value
Figure FDA0002484485270000051
And
Figure FDA0002484485270000052
wherein,
Figure FDA0002484485270000053
showing the predicted value of the horizontal position of the trolley at the moment t,
Figure FDA0002484485270000054
representing the error covariance of the trolley at the time t;
step 4.2: according to a given initial value
Figure FDA0002484485270000055
Calculating Kalman gain value K at t momenttWherein
Figure FDA0002484485270000056
step 4.3: predicted value according to time t
Figure FDA0002484485270000057
And the observed value ztThe corrected value of the current state can be obtained
Figure FDA0002484485270000058
Wherein, the formula is as follows:
zt＝Ctvt+Rt
Figure FDA0002484485270000059
step 4.4: updating error covariance
Figure FDA00024844852700000510
In preparation for predicting the error covariance at time t +1, wherein,
Figure FDA00024844852700000511
i is a unit array;
step 4.5: correction value according to time t
Figure FDA00024844852700000512
And acceleration utTo predict the speed of the car at the time t +1
Figure FDA00024844852700000513
Wherein,
Figure FDA00024844852700000514
step 4.6: prediction of error covariance from the error covariance of the previous time instant
Figure FDA00024844852700000515
To predict the error covariance of the next time instant
Figure FDA00024844852700000516
Wherein,
Figure FDA00024844852700000517
μt+1is an adaptive forgetting factor;
step 4.7: calculating the acceleration u at time t +1t+1Wherein, the calculation method is the autoregressive moving average method in the 3);
step 4.8: updating the iteration number k to k +1, and returning to the step 4.2 to start a new round of calculation;
5) and finally, the edge cloud server transmits the processed information to the roadside unit through an optical cable so as to facilitate the information interaction with the vehicle-mounted system next time.
2. The vehicle speed prediction method based on the quantized adaptive kalman filter according to claim 1, wherein: in the step 1), in the intelligent network traffic system, the roadside units are installed on traffic lights of the crossroad and are attached with the cloud server and the directional antenna, wherein the emission angle of the routing antenna is 60 degrees, so that the roadside units can better perform information interaction with a vehicle-mounted system in a vehicle.
3. A vehicle speed prediction method based on quantization adaptive kalman filtering according to claim 1 or 2, characterized in that: in step 4.6, the calculation formula of the adaptive forgetting factor is as follows:
μt+1＝max{1,tr(Nt+1)/tr(Mt+1)}；
Figure FDA0002484485270000061
Figure FDA0002484485270000062
Figure FDA0002484485270000063
Figure FDA0002484485270000064
here, the parameters are defined as follows:
Nt+1: the error variance at the moment of t +1 ensures that the values are symmetrical and positive;
Lt+1: covariance matrix of innovation at time t + 1;
et+1: the difference between the measured value and the predicted value at the moment t +1 is the innovation;
Figure FDA0002484485270000065
transposing the state transition matrix C at time t + 1;
Mt+1: error variance at time t +1, ensuring error covariance
Figure FDA0002484485270000066
The values of (a) are symmetrical and positive;
max {. cndot ]: and taking the maximum value after comparison.
4. A vehicle speed prediction method based on quantization adaptive kalman filtering according to claim 1 or 2, characterized in that: in step 1.2, the data in the server is cleared every other week in consideration of the limited storage capacity of the edge cloud server.
CN201810947025.0A 2018-08-20 2018-08-20 Vehicle speed prediction method based on quantization adaptive Kalman filtering Active CN109118786B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810947025.0A CN109118786B (en) 2018-08-20 2018-08-20 Vehicle speed prediction method based on quantization adaptive Kalman filtering

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810947025.0A CN109118786B (en) 2018-08-20 2018-08-20 Vehicle speed prediction method based on quantization adaptive Kalman filtering

Publications (2)

Publication Number Publication Date
CN109118786A CN109118786A (en) 2019-01-01
CN109118786B true CN109118786B (en) 2020-08-04

Family

ID=64853523

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810947025.0A Active CN109118786B (en) 2018-08-20 2018-08-20 Vehicle speed prediction method based on quantization adaptive Kalman filtering

Country Status (1)

Country Link
CN (1) CN109118786B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110031646B (en) * 2019-05-10 2021-04-30 北京工业大学 Method for correcting driving speed by using driving direction vibration acceleration
CN110310516A (en) * 2019-06-17 2019-10-08 北京航空航天大学 A kind of highway merging area traffic conflict method of discrimination based on vehicle driving trace prediction
CN112256006B (en) * 2019-07-02 2023-04-28 中国移动通信集团贵州有限公司 Data processing method and device and electronic equipment
CN112798811B (en) * 2020-12-30 2023-07-28 杭州海康威视数字技术股份有限公司 Speed measurement method, device and equipment

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101604448A (en) * 2009-03-16 2009-12-16 北京中星微电子有限公司 A kind of speed-measuring method of moving target and system
CN107909815A (en) * 2017-12-07 2018-04-13 浙江工业大学 A kind of car speed Forecasting Methodology based on adaptive Kalman filter

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105405289B (en) * 2011-12-28 2017-11-07 三菱电机株式会社 Center side system and vehicle side system
CN103310637B (en) * 2013-05-16 2015-06-24 重庆大学 Vehicle data-based bus speed acquisition device and method
CN105374215B (en) * 2015-11-29 2017-06-13 吉林大学 A kind of highway trackside prior-warning device
CN105844904B (en) * 2016-04-22 2019-09-27 北京航空航天大学 A kind of detection of abnormal vehicle behavior and method for tracing based on DSRC
KR101847712B1 (en) * 2016-06-08 2018-04-11 울산대학교 산학협력단 Method and apparatus for allocating resource in vehicle communication and mobile communication network environment
CN107302576B (en) * 2017-06-19 2018-11-13 长安大学 Adaptive service data distribution method based on opportunistic communication under car networking environment

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101604448A (en) * 2009-03-16 2009-12-16 北京中星微电子有限公司 A kind of speed-measuring method of moving target and system
CN107909815A (en) * 2017-12-07 2018-04-13 浙江工业大学 A kind of car speed Forecasting Methodology based on adaptive Kalman filter

Also Published As

Publication number Publication date
CN109118786A (en) 2019-01-01

Similar Documents

Publication Publication Date Title
CN109275121B (en) Vehicle trajectory tracking method based on adaptive extended Kalman filtering
CN109147390B (en) Vehicle trajectory tracking method based on quantization adaptive Kalman filtering
CN109272745B (en) Vehicle track prediction method based on deep neural network
CN109190811B (en) Vehicle speed tracking method based on adaptive extended Kalman filtering
CN109118787B (en) Vehicle speed prediction method based on deep neural network
CN109118786B (en) Vehicle speed prediction method based on quantization adaptive Kalman filtering
CN109285376B (en) Bus passenger flow statistical analysis system based on deep learning
US11520331B2 (en) Methods and apparatus to update autonomous vehicle perspectives
CN108417087B (en) Vehicle safe passing system and method
CN108550262B (en) Urban traffic sensing system based on millimeter wave radar
CN112700470B (en) Target detection and track extraction method based on traffic video stream
CN103745603B (en) A kind of right turn lane vehicle road coordination signal control method and system
CN109345853A (en) A kind of unmanned vehicle safe driving optimization method based on GIS
CN104408915B (en) A kind of method of estimation of traffic state data and system
CN113112805B (en) Intersection monitoring and early warning method based on base station communication and intersection camera positioning
CN116206440A (en) Intelligent high-speed traffic flow acquisition, prediction, control and informatization pushing system and method
US10930145B2 (en) Traffic system for predicting and providing traffic signal switching timing
CN111402586A (en) Road meteorological environment forecasting and early warning control system and method based on Internet of vehicles
WO2021138316A1 (en) Generation of training data for verbal harassment detection
CN114283606A (en) Method, device, equipment and system for vehicle navigation and cloud control platform
CN108665084B (en) Method and system for predicting driving risk
CN109308806A (en) A kind of the traveling detection method and server of the vehicles
CN114782921A (en) Pedestrian and vehicle collision early warning system and method in internet connection environment based on pedestrian intention identification
CN117612127B (en) Scene generation method and device, storage medium and electronic equipment
CN109584549A (en) A kind of road traffic operation index detection method based on large scale floating vehicle data

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