CN115688682A - Vehicle track data compression method and device based on fuzzy prediction - Google Patents
Vehicle track data compression method and device based on fuzzy prediction Download PDFInfo
- Publication number
- CN115688682A CN115688682A CN202211706702.2A CN202211706702A CN115688682A CN 115688682 A CN115688682 A CN 115688682A CN 202211706702 A CN202211706702 A CN 202211706702A CN 115688682 A CN115688682 A CN 115688682A
- Authority
- CN
- China
- Prior art keywords
- fuzzy
- track
- prediction
- vehicle
- track data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013144 data compression Methods 0.000 title claims abstract description 28
- 238000007906 compression Methods 0.000 claims abstract description 11
- 230000006835 compression Effects 0.000 claims abstract description 11
- 238000006073 displacement reaction Methods 0.000 claims description 43
- 230000006870 function Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 7
- 238000012217 deletion Methods 0.000 claims description 6
- 230000037430 deletion Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005242 forging Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
Abstract
The invention discloses a vehicle track data compression method and device based on fuzzy prediction, which comprises the following steps: (1) And acquiring track data in the edge Internet of vehicles environment, determining a membership function, and fuzzifying the track data. (2) And (3) track fuzzy prediction, namely predicting the driving intention through original track data, and then performing fuzzy prediction according to the fuzzy data to obtain fuzzy track data. (3) And (4) calculating errors and reserving track points, comparing the original fuzzy data with the predicted fuzzy data, and if the errors are larger than a set threshold value, reserving the fuzzy track points. The method is based on the fuzzification expression of the track data, and has a light-weight privacy protection function. And track data expression based on fuzzy characters is adopted, so that the communication traffic of edge Internet of vehicles data is remarkably reduced, and the bandwidth consumption is reduced. The method further compresses the track data based on fuzzy prediction of the track, and greatly improves the compression effect.
Description
Technical Field
The invention relates to the technical field of track compression, fuzzy theory and track prediction, in particular to a vehicle track data compression method and device based on fuzzy prediction.
Background
The mining and analysis of the trajectory data of the vehicles is helpful for users or city planners to make better decisions, and has wide application scenarios. Such as optimization and design of traffic routes, identification of dangerous driving behavior, traffic prediction in cities, etc. Through on-vehicle GPS equipment, can gather the orbit data of vehicle in real time, transmit and save to high in the clouds data center. However, a huge amount of data is generated along with the accumulation of time and space, and bandwidth resources and space storage resources are consumed seriously. Therefore, the track data needs to be compressed and then transmitted to the cloud. Under the environment of the Internet of vehicles, the computing power of widely distributed edge equipment can be fully utilized, and the track data can be quickly compressed, so that the communication bandwidth consumption in the track data transmission is reduced.
On the other hand, with the wide application of the internet of vehicles, the security and privacy problem of vehicle information is more and more emphasized by people. If the data are directly transmitted in the vehicle networking environment, an attacker can easily acquire the information of the user by forging a base station, masquerading a legal terminal and other attack means, thereby revealing the privacy of the vehicle user and even threatening the personal safety of the vehicle user. One technique currently employed to protect user's security and privacy is a security authentication technique, which implements identity authentication through digital signature and encryption. However, this approach not only increases the communication burden of the vehicle networking environment, which is inherently resource-limited, but also exposes the privacy of the vehicle user when decrypted by an attacker.
In addition, in most of the existing prediction-based track data compression methods, some prediction models are simple in calculation, but the prediction error is large, for example, the prediction models based on the linear track. Some predictive models predict with higher accuracy but require more computing power and more computing time, such as trajectory predictive models based on deep learning. Therefore, the above models have difficulty in achieving a balance between the accuracy of prediction and the prediction computation power.
In order to solve the above problems, in a marginal internet of vehicles environment, how to design a reasonable trajectory data structure to represent vehicle trajectory data, and create a trajectory prediction model on the basis of the reasonable trajectory data structure to achieve balance between prediction accuracy and prediction calculation power is a problem to be solved urgently.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a vehicle track data compression method and device based on fuzzy prediction.
The purpose of the invention is realized by the following technical scheme: in a first aspect, the present invention provides a vehicle trajectory data compression method based on fuzzy prediction, which includes the following steps:
(1) Track data fuzzification: acquiring vehicle track data, and determining parameters of a membership function according to statistical data of a vehicle displacement increment sequence so as to define the membership function; dividing a plurality of domains according to the membership function, expressing the domains by using fuzzy characters, and converting the track data into the fuzzy characters;
(2) Fuzzy prediction of the trajectory: predicting the driving intention of the target vehicle through a neural network, and selecting a corresponding multi-order fusion Markov model according to the driving intention to perform fuzzy prediction of a future fuzzy displacement increment sequence;
(3) Residual calculation and track point deletion: and according to the comparison between the fuzzy character of fuzzy prediction and the original fuzzy character, deleting the track points of which the residual errors are smaller than the set threshold value, and reserving the track points of which the residual errors are larger than the set threshold value.
Further, in step (1), the detailed steps of the fuzzification of the trajectory data are as follows:
(1-1) track data coordinate transformation: the vehicle sends original track data to a nearby edge gateway (RSU) through an Internet of vehicles network, and the longitude and latitude coordinate sequence of the vehicle is converted into a two-dimensional coordinate sequence taking the edge gateway as an origin by taking the position of the gateway as a coordinate origin;
(1-2) fuzzification of trajectory data based on fuzzy characters: respectively calculating displacement increment sequences in the longitude x direction and the latitude y direction; the method specifically comprises the following steps: calculating displacement sequences in the x direction and the y direction, and extracting symbols to obtain a displacement increment sequence and a displacement symbol sequence in the x direction and a displacement increment sequence and a displacement symbol sequence in the y direction; and for the displacement increment sequence in the x direction, determining fuzzy granularity according to the accuracy of the road type information or the track data, further creating r fuzzy sets, converting the value of the displacement increment sequence into corresponding fuzzy character representation, and converting the displacement increment sequence into a fuzzy character string.
Further, mapping the original track sequence to a two-dimensional coordinate with the edge gateway as an origin; the vehicle acquires its own position in real time by the vehicle-mounted GPS device, and obtains raw trajectory data expressed in longitude and latitude at each time.
Further, the fuzzy granularity is determined according to the accuracy of the road type information or the track data, and then r fuzzy sets are created.
Further, in step (2), the track fuzzy prediction comprises the following detailed steps:
(2-1) prediction of driving intention: inputting the information of the target vehicle and the information of surrounding vehicles into a network overlapped by a GRU unit, and outputting driving intentions of three target vehicles, namely left lane changing, straight lane changing and right lane changing;
(2-2) track position prediction: and selecting a corresponding multi-order fusion Markov model according to the driving intention of the target vehicle, and predicting a future fuzzy displacement increment sequence according to the historical fuzzy displacement increment sequence.
Further, the information of the target vehicle comprises historical track position, speed and the number of the lane where the target vehicle is located; the information of the surrounding vehicles includes the time of collision of the target vehicle with the surrounding vehicles and the number of lanes in which the surrounding vehicles are located.
Further, in the step (3), the detailed steps of residual calculation and track point deletion are as follows:
(3-1) residual calculation based on fuzzy characters: the model predicted fuzzy character needs to be compared with the real fuzzy character, and the residual error of the fuzzy character is calculated as follows:
wherein i is the serial number of the fuzzy character set to which the predicted fuzzy character belongs, j is the serial number of the fuzzy character set to which the original fuzzy character belongs, and r is the number of the fuzzy characters;
(3-2) removing track points according to the set threshold: setting a threshold value
，
If the prediction result does not deviate from the real track position, namely the residual error is smaller than the threshold value, the track position is removed, and if the prediction result deviates from the real track position, namely the residual error is larger than the threshold value, the track position needs to be reserved, so that the purpose of compression is achieved.
Further, when the threshold is set to 0, it indicates that prediction is not allowed to have an error, and as long as there is an error between the predicted value and the original value, the original value is added to the compression trajectory; when the threshold is set to 1, it means that any prediction is within the error tolerance, and at this time, no original value is added to the compression trajectory; thus, the threshold value
The larger the prediction is, the larger the allowable error range is.
In a second aspect, the present invention provides a vehicle trajectory data compression device based on fuzzy prediction, which includes a memory and one or more processors, where the memory stores executable codes, and the processors execute the executable codes to implement the steps of the vehicle trajectory data compression method based on fuzzy prediction.
In a third aspect, the present invention provides a computer-readable storage medium, on which a program is stored, which, when executed by a processor, performs the steps of the method for compressing vehicle trajectory data based on fuzzy prediction.
The invention has the following beneficial effects:
(1) The track data is subjected to fuzzy coding, and the light-weight privacy protection function is achieved.
(2) And the track data expression based on fuzzy characters obviously reduces the communication traffic of edge Internet of vehicles data and reduces the bandwidth consumption.
(3) Track data is further compressed based on fuzzy prediction of the track, and the compression effect is greatly improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a track data compression method based on fuzzy prediction according to the present invention.
Fig. 2 is a flow chart of track fuzzy prediction.
Fig. 3 is a structural diagram of a track data compression device based on fuzzy prediction according to the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
Referring to fig. 1, the invention provides a track data compression method based on fuzzy prediction, comprising the following steps:
(1) Track data fuzzification: collecting original track data of the vehicle, and determining parameters of the membership function according to statistical data of the track data of the vehicle, thereby defining the membership function. And dividing a domain of discourse according to the membership function, expressing the domain of discourse by using fuzzy characters, and converting the track data into the fuzzy characters. The detailed steps of the track data fuzzification of the vehicle are as follows:
(1-1) coordinate transformation of track data: and mapping the original track sequence to a two-dimensional coordinate with the edge gateway as an origin. The vehicle acquires the position of the vehicle in real time through the vehicle-mounted GPS equipment and establishes the slave
Is timed to
Time of day, raw track data
Indicating that the vehicle is
At the moment of time to
The position of the moment of time is,
raw track data of time of day
Indicating that the vehicle is
Longitude (Longitude) of the time of day vehicle is
Latitude (Latitude) is
. Sending it to a nearby edge gateway (RSU) via an in-vehicle network, the edge gatewaykIn the position of
Then, the position of the gateway is taken as the origin of coordinates, and the longitude and latitude coordinate sequence of the vehicle is converted into a two-dimensional coordinate sequence with the edge gateway as the origin
Wherein
To represent
Two-dimensional coordinates of time of day.
(1-2) fuzzification of trajectory data based on fuzzy characters: the displacement increment sequences in the x and y directions are calculated respectively. The method specifically comprises the following steps: is calculated atxAndythe displacement sequence of the direction is extracted to obtain displacement symbolsxIncremental sequence of directional displacements
And shift the symbol sequence
Sequence of incremental displacements in y-direction
And a sequence of shifted symbols
。
Representt 1 Relative to time of dayt 0 At the moment of timexThe incremental displacement in the direction(s) is,
to representt n Relative to time of day locust n-1 At a moment in timexThe incremental displacement in the direction is increased by an amount,
to represent
The sign of the displacement of (a);
to representt 1 Relative to time of dayt 0 At a moment in timeyThe incremental displacement in the direction is increased by an amount,
to representt n Relative to time of day locust n-1 At a moment in timeyThe incremental displacement in the direction is increased by an amount,
to represent
The sign of the displacement of (a). For thexDirectional displacement increment sequence, and r fuzzy sets are created by formulas (1) to (3)
Wherein the parameters of all blur sets are determined by equation (4). Fuzzy character set
，
In which the particle size is blurred
The more the number of fuzzy sets and fuzzy characters, the finer the granularity of the fuzzification, determined by the accuracy of the road type information or trajectory data.yIncremental blurring of directional displacementxAnd (4) direction. Converting the value of the displacement increment sequence into corresponding fuzzy character representation, and then displacingThe sequence of increments translates into a fuzzy string.
Wherein, the first and the second end of the pipe are connected with each other,
is the membership function of 1 st to 5 th fuzzy sets,
the fuzzy set membership function parameters are determined according to historical data distribution.
，
And
is divided intoxAndythe minimum value, the maximum value and the average value of the displacement increment of two adjacent positions are oriented.
(2) Fuzzy prediction of the trajectory: the driving intention of the target vehicle is predicted through a Gate Recovery Unit (GRU) network, and then a corresponding multi-level fusion Markov model is selected according to the driving intention to perform fuzzy prediction. Referring to fig. 2, the track fuzzy prediction comprises the following detailed steps:
(2-1) prediction of driving intention: the information of the target vehicle comprises historical track position, speed and the number of the lane where the target vehicle is located, the information of the surrounding vehicles comprises the collision time of the target vehicle and the surrounding vehicles and the number of the lane where the target vehicle and the surrounding vehicles are located, the information is input into a network overlapped by a GRU unit, and three driving intentions, namely left lane changing, straight driving and right lane changing, are output.
The input data is the position and speed of the target vehicle
Acceleration of
Lane marking
(ii) a Time to collision TTC with surrounding vehicles, lane markings
Representing the TTC and the lane number of the j vehicle at time i with the target vehicle. The total input data of the driving intention prediction module is
，
And T is the total sampling time. The network structure adopts multilayer GRU superposition, a Droupout layer is added to prevent overfitting, a softmax layer is connected, and the probabilities of left lane changing, straight traveling and right lane changing of each driving intention are output. The TTC (Time to Collision) is a safety index for measuring lane change, and the calculation formula of the TTC between the target vehicle and the surrounding vehicles is shown in formula (5).
Wherein
To represent
The time of collision TTC with the surrounding vehicle j,
the lane number representing the target vehicle is the same as the lane number of vehicle j,
is the relative distance between the target vehicle and the vehicle j in the x directionAfter the separation, the water is separated from the water,
the relative speed of the target vehicle and the vehicle j in the x direction,
the relative distance in the y-direction between the target vehicle and vehicle j,
the relative speed of the target vehicle and vehicle j in the y direction,
and
respectively, the traveling direction of the target vehicle and the vehicle j is the angle with the x direction. The TTC comprises the distance relation and the speed relation between the target vehicle and other surrounding vehicles, and the position and the speed of the surrounding vehicles are converted into the time TTC of collision between the surrounding vehicles and the target vehicle, so that the input dimension is effectively reduced, the parameters of a network are reduced, and the calculation amount is reduced.
The GRU unit contains two gates, a reset gate and an update gate. Can be calculated by the equations (6) to (9).
Wherein the content of the first and second substances,r t indicating the number of layers of the reset gate,
a sigmoid activation function is represented,h t-1 is the output at time t-1,
an input representing the time of the t-instant,
a weight matrix representing the reset gate layer. In the use of the reset gate, the new memory will store history related information using the reset gate, as shown in equations (7) and (8).
Wherein
The update of the door level is indicated,
is the new output at time t. Finally, the output of the GRU is updated as shown in equation (9).
Wherein the content of the first and second substances,
representing element by element multiplication (elementary product).
(2-2) track position prediction: and selecting a corresponding multi-step fusion Markov Model (Adaboost-Markov Model) according to the driving intention of the target vehicle, predicting a future fuzzy displacement increment sequence by the historical fuzzy displacement increment sequence, wherein the calculation formulas of the multi-step fusion Markov Model are shown in (10) to (13).
In the formulas (10) to (13),
is thatmAnd the weight of the training sample corresponding to the order Markov model.
Is thatmOrder Markov model corresponding training samplei The prediction result determiner of (1).
Is thatmWeight coefficients of the order Markov model.
Is composed ofmThe order of the normalization factor is such that,
is composed ofm-Probability distribution of 1 st order Markov model.
The normalized m-order Markov model weight coefficient is obtained, k represents the order,
an m-order Markov model.
And (5) obtaining a final Markov model.
(3) Residual calculation and track point deletion: and according to the comparison between the fuzzy character of fuzzy prediction and the original fuzzy character, deleting the track points of which the residual errors are smaller than the set threshold value, and reserving the track points of which the residual errors are larger than the set threshold value. The detailed steps of given fuzzy character strings to be predicted, residual calculation and track point deletion are as follows:
(3-1) residual calculation based on fuzzy characters: the model predicted fuzzy character needs to be compared with the real fuzzy character, and the residual error of the fuzzy character is calculated as shown in formula (14). Wherein i is the serial number in the fuzzy character set to which the predicted fuzzy character Pred belongs, j is the serial number in the fuzzy character set to which the original fuzzy character belongs, and r is the number of the fuzzy characters.
(3-2) removing track points according to the set threshold: setting a threshold value
If the deviation between the prediction result and the real track position is smaller than the threshold, the track position is removed, and if the deviation is larger than the threshold, the track position needs to be reserved, so that the purpose of compression is achieved. When the threshold value is set to 0, it means that prediction is not allowed to have an error, and the original value is added to the compression trace as long as there is an error between the predicted value and the original value. When the threshold is set to 1, it indicates that any prediction is within the error tolerance, and no original value is added to the compressed track. Thus, the threshold value
The larger the prediction error range, the better the compression effect (compression ratio).
Corresponding to the embodiment of the track data compression method based on fuzzy prediction, the invention also provides an embodiment of a track data compression device based on fuzzy prediction.
Referring to fig. 3, an apparatus for compressing trajectory data based on fuzzy prediction according to an embodiment of the present invention includes a memory and one or more processors, where the memory stores executable codes, and the processors execute the executable codes to implement a method for compressing trajectory data based on fuzzy prediction in the foregoing embodiment.
The track data compression device based on fuzzy prediction of the present invention can be applied to any device with data processing capability, such as a computer or other devices or apparatuses. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a device in a logical sense, a processor of any device with data processing capability reads corresponding computer program instructions in the nonvolatile memory into the memory for operation. From a hardware aspect, as shown in fig. 3, the present invention is a hardware structure diagram of any device with data processing capability in which a track data compression apparatus based on fuzzy prediction is located, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 3, in an embodiment, any device with data processing capability in which the apparatus is located may also include other hardware according to an actual function of the any device with data processing capability, which is not described again.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the invention. One of ordinary skill in the art can understand and implement it without inventive effort.
Embodiments of the present invention further provide a computer-readable storage medium, on which a program is stored, where the program, when executed by a processor, implements the trajectory data compression method based on fuzzy prediction in the foregoing embodiments.
The computer readable storage medium may be an internal storage unit, such as a hard disk or a memory, of any data processing capability device described in any of the foregoing embodiments. The computer readable storage medium may also be any external storage device of a device with data processing capabilities, such as a plug-in hard disk, a Smart Media Card (SMC), an SD Card, a Flash memory Card (Flash Card), etc. provided on the device. Further, the computer readable storage medium may include both internal storage units and external storage devices of any data processing capable device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the arbitrary data processing-capable device, and may also be used for temporarily storing data that has been output or is to be output.
The above-described embodiments are intended to illustrate rather than limit the invention, and any modifications and variations of the present invention are within the spirit and scope of the appended claims.
Claims (10)
1. A vehicle track data compression method based on fuzzy prediction is characterized by comprising the following steps:
(1) Track data fuzzification: acquiring vehicle track data, and determining parameters of a membership function according to statistical data of a vehicle displacement increment sequence so as to define the membership function; dividing a plurality of domains according to the membership function, expressing the domains by using fuzzy characters, and converting the track data into the fuzzy characters;
(2) Fuzzy prediction of the trajectory: predicting the driving intention of a target vehicle through a neural network, and selecting a corresponding multi-order fusion Markov model according to the driving intention to perform fuzzy prediction on a future fuzzy displacement increment sequence;
(3) Residual calculation and track point deletion: and according to the comparison between the fuzzy character of fuzzy prediction and the original fuzzy character, deleting the track points of which the residual errors are smaller than the set threshold value, and reserving the track points of which the residual errors are larger than the set threshold value.
2. The vehicle track data compression method based on fuzzy prediction as claimed in claim 1, characterized in that in step (1), the detailed step of the fuzzification of the track data is as follows:
(1-1) coordinate transformation of track data: the vehicle sends the track data to a nearby edge gateway RSU through a vehicle-to-vehicle network, and the longitude and latitude coordinate sequence of the vehicle is converted into a two-dimensional coordinate sequence taking the edge gateway as an origin by taking the position of the gateway as a coordinate origin;
(1-2) fuzzification of trajectory data based on fuzzy characters: respectively calculating displacement increment sequences in the directions of longitude x and latitude y; the method specifically comprises the following steps: calculating displacement sequences in the x direction and the y direction, and extracting symbols to obtain a displacement increment sequence and a displacement symbol sequence in the x direction and a displacement increment sequence and a displacement symbol sequence in the y direction; and for the displacement increment sequences in the x and y directions, r fuzzy sets are created, the values of the displacement increment sequences are converted into corresponding fuzzy character representations through the fuzzy sets and fuzzy division, and then the displacement increment sequences are converted into fuzzy character strings.
3. The vehicle trajectory data compression method based on fuzzy prediction as claimed in claim 2, wherein the original trajectory sequence is mapped to two-dimensional coordinates with an edge gateway as an origin; the vehicle acquires its own position in real time by the vehicle-mounted GPS device, and obtains raw trajectory data expressed in longitude and latitude at each time.
4. The vehicle trajectory data compression method based on fuzzy prediction as claimed in claim 2, wherein fuzzy granularity is determined according to the accuracy of road type information or trajectory data, and r fuzzy sets are created.
5. The vehicle track data compression method based on fuzzy prediction as claimed in claim 1, wherein in step (2), the track fuzzy prediction is detailed as follows:
(2-1) prediction of driving intention: inputting information of a target vehicle and information of surrounding vehicles into a network overlapped by a GRU unit, and outputting driving intentions of three target vehicles, namely left lane changing, straight lane changing and right lane changing;
(2-2) track position prediction: and selecting a corresponding multi-order fusion Markov model according to the driving intention of the target vehicle, and predicting a future fuzzy displacement increment sequence according to the historical fuzzy displacement increment sequence.
6. The vehicle track data compression method based on the fuzzy prediction as claimed in claim 5, wherein the information of the target vehicle comprises historical track position, speed and number of lanes; the information of the surrounding vehicles includes the time of collision of the target vehicle with the surrounding vehicles and the number of lanes in which the surrounding vehicles are located.
7. The vehicle track data compression method based on fuzzy prediction according to claim 1, wherein in the step (3), the detailed steps of residual calculation and track point deletion are as follows:
(3-1) residual calculation based on fuzzy characters: the model predicted fuzzy character needs to be compared with the real fuzzy character, and the residual error of the fuzzy character is calculated as follows:
wherein i is the serial number of the fuzzy character set to which the predicted fuzzy character belongs, j is the serial number of the fuzzy character set to which the original fuzzy character belongs, and r is the number of the fuzzy characters;
(3-2) removing track points according to the set threshold: setting a threshold value
，
If the prediction result is not deviated from the real track position, namely the residual error is smaller than the threshold value, the track position is removed, and if the prediction result is deviated from the real track position, namely the residual error is larger than the threshold value, the track position needs to be reserved, so that the purpose of compression is achieved.
8. The vehicle track data compression method based on the fuzzy prediction as claimed in claim 7, wherein when the threshold value is set to 0, it represents that the prediction is not allowed to have an error, and if the predicted value and the original value have an error, the original value is added to the compressed track; when the threshold is setA value of 1 indicates that any prediction is within the error tolerance, and no original value is added to the compressed trace; thus, the threshold value
The larger the error range allowed for prediction.
9. A vehicle trajectory data compression device based on fuzzy prediction, comprising a memory and one or more processors, wherein the memory stores executable code, and the processors are configured to implement the steps of the vehicle trajectory data compression method based on fuzzy prediction according to any one of claims 1 to 8 when executing the executable code.
10. A computer-readable storage medium, on which a program is stored, which, when being executed by a processor, carries out the steps of a fuzzy prediction based vehicle trajectory data compression method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211706702.2A CN115688682B (en) | 2022-12-29 | 2022-12-29 | Vehicle track data compression method and device based on fuzzy prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211706702.2A CN115688682B (en) | 2022-12-29 | 2022-12-29 | Vehicle track data compression method and device based on fuzzy prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115688682A true CN115688682A (en) | 2023-02-03 |
| CN115688682B CN115688682B (en) | 2023-11-14 |
Family
ID=85055170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211706702.2A Active CN115688682B (en) | 2022-12-29 | 2022-12-29 | Vehicle track data compression method and device based on fuzzy prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115688682B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247961A (en) * | 2017-05-10 | 2017-10-13 | 西安交通大学 | A kind of trajectory predictions method of application blurring trajectorie sequence |
| CN111654290A (en) * | 2020-04-24 | 2020-09-11 | 北京中交兴路车联网科技有限公司 | Method, device and equipment for compressing vehicle track data and storage medium |
| US20200385014A1 (en) * | 2017-04-24 | 2020-12-10 | Mobileye Vision Technologies Ltd. | Systems and methods for compression of lane data |
| US20210237779A1 (en) * | 2018-09-27 | 2021-08-05 | Kpit Technologies Limited | System for grid-based merge cut-in prediction and method thereof |
| WO2021249020A1 (en) * | 2020-06-10 | 2021-12-16 | 华为技术有限公司 | Method and apparatus for predicting driving state, and terminal device |
| CN113911129A (en) * | 2021-11-23 | 2022-01-11 | 吉林大学 | Traffic vehicle intention identification method based on driving behavior generation mechanism |
| CN114368387A (en) * | 2021-12-21 | 2022-04-19 | 吉林大学 | Attention mechanism-based driver intention identification and vehicle track prediction method |
| CN115027484A (en) * | 2022-05-23 | 2022-09-09 | 吉林大学 | Human-computer fusion perception method for high-degree automatic driving |
-
2022
- 2022-12-29 CN CN202211706702.2A patent/CN115688682B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200385014A1 (en) * | 2017-04-24 | 2020-12-10 | Mobileye Vision Technologies Ltd. | Systems and methods for compression of lane data |
| CN107247961A (en) * | 2017-05-10 | 2017-10-13 | 西安交通大学 | A kind of trajectory predictions method of application blurring trajectorie sequence |
| US20210237779A1 (en) * | 2018-09-27 | 2021-08-05 | Kpit Technologies Limited | System for grid-based merge cut-in prediction and method thereof |
| CN111654290A (en) * | 2020-04-24 | 2020-09-11 | 北京中交兴路车联网科技有限公司 | Method, device and equipment for compressing vehicle track data and storage medium |
| WO2021249020A1 (en) * | 2020-06-10 | 2021-12-16 | 华为技术有限公司 | Method and apparatus for predicting driving state, and terminal device |
| CN113911129A (en) * | 2021-11-23 | 2022-01-11 | 吉林大学 | Traffic vehicle intention identification method based on driving behavior generation mechanism |
| CN114368387A (en) * | 2021-12-21 | 2022-04-19 | 吉林大学 | Attention mechanism-based driver intention identification and vehicle track prediction method |
| CN115027484A (en) * | 2022-05-23 | 2022-09-09 | 吉林大学 | Human-computer fusion perception method for high-degree automatic driving |
Non-Patent Citations (6)
| Title |
|---|
| ARLEI SILVA, ET AL.: "Prediction-based Online Trajectory Compression", 《ARXIV》 * |
| QI DENG, ET AL.: "Improved driving behaviors prediction based on Fuzzy Logic-Hidden Markov Model (FL-HMM)", 《IEEE》 * |
| YINGLONG LI,ET AL.: "fuzzyCom: Privacy-Aware Trajectory Data Compression Using Fuzzy Sets in Edge Vehicular Networks", pages 613 - 619 * |
| 李立 等: "面向群体行驶场景的时空信息融合车辆轨迹预测", 《交通运输工程学报》 * |
| 王献伟: "基于自然驾驶数据的驾驶行为辨识和建模方法研究", no. 2, pages 035 - 454 * |
| 陈煜 等: "基于预测模型的轨迹数据压缩方法", vol. 38, no. 1, pages 171 - 175 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115688682B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220105961A1 (en) | Method for automatic control of vehicle and method for training lane change intention prediction network | |
| CN113261035B (en) | Trajectory prediction method and related equipment | |
| Yang et al. | MF-CNN: traffic flow prediction using convolutional neural network and multi-features fusion | |
| EP4152204A1 (en) | Lane line detection method, and related apparatus | |
| US11132211B1 (en) | Neural finite state machines | |
| KR102539942B1 (en) | Method and apparatus for training trajectory planning model, electronic device, storage medium and program | |
| CN112382099B (en) | Traffic road condition prediction method and device, electronic equipment and storage medium | |
| US11620492B2 (en) | Flexible edge-empowered graph convolutional networks with node-edge enhancement | |
| JP7168708B2 (en) | MAP DATA UPDATE METHOD, APPARATUS, DEVICE, AND READABLE STORAGE MEDIUM | |
| CN112163446B (en) | Obstacle detection method and device, electronic equipment and storage medium | |
| Varghese et al. | Deep learning in transport studies: A meta-analysis on the prediction accuracy | |
| CN110363098B (en) | Violent behavior early warning method and device, readable storage medium and terminal equipment | |
| US20220383075A1 (en) | System and method for conditional marginal distributions at flexible evaluation horizons | |
| Chen et al. | CAE‐GAN: A hybrid model for vehicle trajectory prediction | |
| Zhao et al. | Traffic inflow and outflow forecasting by modeling intra-and inter-relationship between flows | |
| Chen et al. | Follownet: A comprehensive benchmark for car-following behavior modeling | |
| CN112233419B (en) | Data processing method, device, equipment and storage medium | |
| CN115688682A (en) | Vehicle track data compression method and device based on fuzzy prediction | |
| CN111898199A (en) | Vehicle dynamics data driving modeling method | |
| CN113119996B (en) | Trajectory prediction method and apparatus, electronic device and storage medium | |
| CN115773744A (en) | Model training and road network processing method, device, equipment, medium and product | |
| CN113094422B (en) | Urban road traffic flow map generation method, system and equipment | |
| CN113192315B (en) | Traffic flow distribution prediction method, prediction device and terminal equipment | |
| CN114548572A (en) | Method, device, equipment and medium for predicting urban road network traffic state | |
| CN115272738A (en) | Data processing method, model training method and device |
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 |