CN110555476B - Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment - Google Patents
Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment Download PDFInfo
- Publication number
- CN110555476B CN110555476B CN201910809352.4A CN201910809352A CN110555476B CN 110555476 B CN110555476 B CN 110555476B CN 201910809352 A CN201910809352 A CN 201910809352A CN 110555476 B CN110555476 B CN 110555476B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- lane change
- track prediction
- change track
- optimal
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Traffic Control Systems (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The invention discloses an intelligent vehicle lane change track prediction method suitable for a man-machine hybrid driving environment, which comprises the following steps: s1, acquiring track data of an unmanned automobile which is traveling on a high-speed road section; s2, processing the acquired driving track data, acquiring the space information, the motion information and the data information of the relative states of the adjacent vehicles, screening lane change data according to the data information, and generating a sample database; s3, constructing a lane change track prediction model of the unmanned vehicle; s4, training a vehicle lane change track prediction model to obtain an optimal lane change track prediction model structure, an optimal training sample capacity and an optimal historical sequence length; s5, verifying the lane change track prediction model of the vehicle trained in the step S4.
Description
Technical Field
The invention relates to the technical field of unmanned, in particular to an intelligent vehicle lane change track prediction method suitable for a man-machine hybrid driving environment.
Background
The man-machine mixed driving traffic environment refers to a traffic environment in which a manual driving vehicle and an unmanned vehicle are mixed to run in a future road system. With the rapid development of automatic driving technology, in the foreseeable future, unmanned vehicles will enter an actual road traffic system and will run in a man-machine mixed driving traffic environment for a long time. The lane change behavior is a basic driving task of traffic, and has important significance for driving safety and traffic flow stabilization. However, the existing lane change model accuracy cannot meet the practical use. At present, no literature is available for predicting the lane change track of an unmanned vehicle based on deep learning in a man-machine hybrid driving environment.
Classical lane change behavior research can be mainly divided into two types, i.e. lane change intention prediction research and lane change track prediction research, and is typically represented by a Gipps model, a cellular automaton model, a Markov model and the like. In order to reduce the calculation difficulty, the models simplify the lane change behavior of a driver, so that the models cannot accurately reproduce the lane change process of the vehicle and cannot meet the requirements of the lane change behavior model of the unmanned vehicle in the man-machine mixed driving environment. In recent years, researchers have studied lane changing behavior using a machine learning method from a data-driven standpoint, and most of the research has focused on lane changing intention prediction, and some of the research has been directed to lane changing time prediction. Along with the development of automatic driving technology, lane change track prediction becomes a new research hotspot. However, the current unmanned vehicle channel-changing behavior model still faces the problems that the model is too complex and inapplicable or has insufficient precision.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the invention provides an intelligent vehicle lane change track prediction method suitable for a man-machine hybrid driving environment.
The invention adopts the following technical scheme:
a lane change track prediction method suitable for an intelligent vehicle in a man-machine hybrid driving environment comprises the following steps:
s1, acquiring track data of an unmanned automobile travelling on a high-speed road section;
s2, processing the acquired driving track data, acquiring the space information, the motion information and the data information of the relative states of the adjacent vehicles, screening out lane change data, and generating a sample database;
the data information of the relative states of the adjacent vehicles specifically refers to the relative state vectors of the lane change vehicle i and the adjacent vehicle jWherein Deltax i,j (t)，Δy i,j (t)，Δv i,j (t) represents the relative x-axis coordinates, the relative y-axis coordinates and the relative speed of vehicles i and j, respectively, at time t;
s3, constructing a lane change track prediction model of the unmanned vehicle, wherein the number of input variables of the model is 6N, and the number of output variables of the model is 2;
s4, training a vehicle lane change track prediction model to obtain an optimal lane change track prediction model structure, an optimal training sample capacity and an optimal historical sequence length;
s5, verifying the lane change track prediction model of the vehicle trained in the step S4.
The 6N input variables of the model are:
(1) lane change destination
The lane change destination of the lane change vehicle i at the time t- (N-1) tau is indicated, n=1, 2, …, N is the length of a history sequence, and tau is the time interval between the input time sequences;
(2) infrastructure information M
Taking the transverse x-axis coordinate of the median line of the lane with the nearest current position of the vehicle i as infrastructure information;
(3) vehicle parameter C i
The length and width of the vehicle i are used for representing the type of the vehicle, namely:
C i ＝{W i ,L i }
in which W is i 、L i The width and length of the vehicle i, respectively;
(4) state vector
The method comprises the following steps of the instant speed of the vehicle i, the direction and the amplitude of the motion state in the lane change process, namely:
wherein x is i (t-(n-1)τ)，y i (t-(n-1)τ)，v i (t- (n-1) τ) is the x-axis coordinate, y-axis coordinate and speed magnitude of the vehicle i at time t- (n-1) τ, respectively;
(5) vehicle parameter C j
The length, width of the neighboring vehicle j is used to characterize the vehicle type, namely:
C j ＝{W j ,L j }
in which W is j 、L j The width and length of vehicle j, respectively;
(6) relative state vector
The relative speed of the vehicles i and j is included, and the direction and the amplitude of the relative motion state of the vehicles i and j in the lane change process are as follows:
wherein Deltax is i,j (t-(n-1)τ)，Δy i,j (t-(n-1)τ)，Δv i,j (t- (n-1) τ) is the relative x-axis coordinate, relative y-axis coordinate, and relative speed magnitude of vehicles i and j at time t- (n-1) τ, respectively.
The output variables of the model are:
(1) a position prediction result x (t+t) of the model on the x-axis direction of the vehicle;
wherein T represents a prediction delay, and the value range is 0.1-1.5s;
(2) and predicting a result y (t+T) of the model on the position of the vehicle in the y-axis direction.
In the S4, RMSprop is adopted as an optimization method, and a back propagation algorithm is adopted as a basic method for calculating the gradient.
Preferably, the total number of training samples is not less than 50000, the total number of accuracy verification samples is not less than 10000, the track accuracy requirement is within 0.5m, and the sample sampling rate is between 1 and 10 Hz.
Preferably, the spatial information of the vehicle itself specifically refers to road information and a road position where the vehicle is located; the motion information refers to speed information.
The invention has the beneficial effects that:
1. compared with the traditional lane change track prediction method, the lane change track prediction method has the advantages that the lane change process of a driver is accurately reproduced, and in addition, the lane change track prediction method is based on a deep learning method, so that the lane change track prediction method has the advantage of simplicity and easiness in implementation.
2. The invention adopts a modeling method of deep learning, can process input parameters with higher dimensionality, and can restore the real driving behavior by acquiring more input information. At the same time, the invention integrates the predictive ability and memory effect of the driver into the model. Therefore, the method has higher precision than the traditional channel change track prediction method.
3. From the data driving point of view, the model training is carried out through the actual driving vehicle track data, and the obtained lane change track prediction model can be automatically adjusted according to the characteristics of the actual data, so that the method has good adaptability and robustness.
4. The method can not only predict the lane change track of the unmanned vehicle in the man-machine hybrid driving environment, but also can be used as the basis of the unmanned vehicle driving track prediction model in the completely unmanned vehicle environment in the future, and has great practical popularization value.
Drawings
FIG. 1 is a workflow diagram of the present invention;
fig. 2 (a) and 2 (b) are diagrams of the study route of the present invention;
FIG. 3 is a schematic diagram of a model structure of the present invention;
fig. 4 is a flow chart of a training sample of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but embodiments of the present invention are not limited thereto.
Examples
As shown in fig. 1-3, the lane change track prediction method suitable for the intelligent vehicle in the man-machine hybrid driving environment comprises the following steps:
s1, acquiring track data of an unmanned automobile which is traveling on a high-speed road section;
s2, processing the acquired driving track data, acquiring the space information, the motion information and the data information of the relative states of the adjacent vehicles, screening lane change data according to the data information, and generating a sample database required by the model.
Firstly, primarily screening the obtained data, providing vehicle data with missing information, obtaining the space information, the motion information and the data information of the relative states of adjacent vehicles, screening lane change data according to the information, and generating a sample database with the sequence length of N.
The vehicle space information includes a lane in which the vehicle is traveling and a position of the vehicle within the lane.
The vehicle motion information includes information such as vehicle speed, vehicle length, vehicle width, and the like.
The relative states of the adjacent vehicles mainly refer to relative state vectors of the lane vehicle i and the adjacent vehicle jWherein Deltax i,j (t)，Δy i,j (t)，Δv i,j (t) represents the relative x-axis coordinates, the relative y-axis coordinates and the relative speed of vehicles i and j, respectively, at time t.
Screening vehicle historical data deletion from the lane change data obtained by processing, and eliminating lane change data with abnormal data in the data; and generating a sample library according to the required historical sequence length N.
S3, constructing a lane change track prediction model of the unmanned vehicle, wherein the number of input variables of the model is 6N, the number of output variables is 2, and N is the length of a sample sequence;
and (3) processing the data obtained in the step (S2) to obtain a training sample set and an accuracy verification sample set, wherein more than 110000 training samples and 37440 accuracy verification samples are obtained.
The 6 input variables of the model are:
(1) lane change destination
The lane change destination of the lane change vehicle i at the time t- (N-1) tau is indicated, n=1, 2, …, N and N are the lengths of historical sequences, tau is the time interval between the input time sequences, and the range of the value is 0.05-0.4s in consideration of the development of the current intelligent network vehicle (Intelligent Connected Vehicle, ICV) data acquisition technology and the speed data precision requirement.
(2) Infrastructure information M
Taking the transverse x-axis coordinate of the median line of the lane with the nearest current position of the vehicle i as infrastructure information;
(3) vehicle parameter C i
The length and width of the vehicle i are used for representing the type of the vehicle, namely:
C i ＝{W i ,L i }
in which W is i 、L i The width and length of the vehicle i, respectively;
(4) state vector
The method comprises the following steps of the instant speed of the vehicle i, the direction and the amplitude of the motion state in the lane change process, namely:
wherein x is i (t-(n-1)τ)，y i (t-(n-1)τ)，v i (t- (n-1) τ) is the x-axis coordinate, y-axis coordinate and speed magnitude of the vehicle i at time t- (n-1) τ, respectively;
(5) vehicle parameter C j
The length, width of the neighboring vehicle j is used to characterize the vehicle type, namely:
C j ＝{W j ,L j }
in which W is j 、L j The width and length of vehicle j, respectively;
(6) relative state vector
The relative speed of the vehicles i and j is included, and the direction and the amplitude of the relative motion state of the vehicles i and j in the lane change process are as follows:
wherein Deltax is i,j (t-(n-1)τ)，Δy i,j (t-(n-1)τ)，Δv i,j (t- (n-1) τ) is the relative x-axis coordinate, the relative y-axis coordinate and the relative speed of the vehicles i and j at the time t- (n-1) τ, and the calculation formulas are respectively:
Δx i,j (t-(n-1)τ)＝x j (t-(n-1)τ)-x i (t-(n-1)τ)
wherein x is i (t-(n-1)τ)，x j (t- (n-1) τ) is the x-axis coordinates of vehicles i and j at time t- (n-1) τ, respectively;
Δy i,j (t-(n-1)τ)＝y j (t-(n-1)τ)-y i (t-(n-1)τ)
wherein y is i (t-(n-1)τ，y j (t- (n-1) τ) is the y-axis coordinates of vehicles i and j at time t- (n-1) τ, respectively;
Δv i,j (t-(n-1)τ)＝v j (t-(n-1)τ)-v i (t-(n-1)τ)
in the formula, v i (t-(n-1)τ，v j (t- (n-1) τ) is the speed of vehicles i and j at time t- (n-1) τ, respectively.
The 2 output variables of the model are respectively:
(1) a position prediction result x (t+t) of the model on the x-axis direction of the vehicle;
wherein T represents a prediction delay, and the value range is 0.1-1.5s;
(2) and predicting a result y (t+T) of the model on the position of the vehicle in the y-axis direction.
S4, training a vehicle lane change track prediction model to obtain an optimal lane change track prediction model structure, an optimal training sample capacity and an optimal historical sequence length;
modeling a lane change track by using different neural network model structures and different sample volumes, selecting a neural network structure with a representative hidden layer number and node number of each layer, and training each network structure by using different training data volumes to obtain an optimal structure scheme and an optimal training sample volume;
under the conditions of the optimal network structure and the optimal training sample size, the lane change track prediction models with different historical sequence lengths N are trained, and the optimal historical sequence length N is selected.
S5, verifying the lane change track prediction model of the vehicle trained in the step S4.
Verifying the trained optimal lane change track prediction model by using verification sample data, and determining and calculating corresponding input variables, wherein the method comprises the following steps of:M，C i ，/>C j ，/>wherein n=1, 2, …, N; inputting the variables into a lane change track prediction model, and outputting position prediction results x (t+T) and y (t+T) of the vehicle in the x-axis direction and the y-axis direction by the model;
the error analysis index is the Mean Square Error (MSE).
The invention is specifically illustrated by the method according to one embodiment:
firstly, selecting actual driving vehicle track data on an American I-101 expressway and an I-80 expressway section in an NGSIM data set as original data for constructing a lane change track sample library, wherein the original data comprises manual driving vehicle track data in a 7:50-8:35a.m. period. The study route is shown in fig. 2 (a) and fig. 2 (b).
Then, the acquired driving track data are processed, the space information and the motion information of the vehicle are acquired, and data containing the relative states of the adjacent vehicles are calculated; preliminary screening is carried out on the original data, and vehicle data of the missing information is removed; judging the obtained data to be adjacent to the vehicle, and storing related information (including information such as vehicle length, vehicle width, speed, position and the like); calculating the relative state of the acquired data with the adjacent vehicles, and screening out lane change data; screening vehicle historical data deletion from the lane change data obtained by processing, and eliminating lane change data with abnormal data in the data; and generating a sample library according to the required historical sequence length N.
And then constructing a lane change track prediction model of the unmanned vehicle based on the long-short-term memory neural network, wherein the model structure is shown in figure 3.
And performing data processing on the lane change track original database according to the input and output variables of the model to form a lane change track sample library of the manual driving vehicle, wherein the lane change track sample library comprises training sample sets and precision verification sample sets, and more than 110000 training samples and 37440 precision verification samples are obtained.
Randomly extracting 10,000 models under the condition of different historical sequence lengths N to obtain a test historical sequence length N=5;
modeling a lane change track by using different neural network model structures and different sample sizes, selecting a neural network structure with a representative hidden layer number and node number of each layer, and training each network structure by using different training data sizes to obtain an optimal structural scheme 1LS1 and an optimal training sample size 110000;
under the conditions of an optimal network structure 1LS1 and an optimal training sample size of 110000, training LSTM-LC models with different historical sequence lengths N, and selecting an optimal historical sequence length N=5;
and finally, performing accuracy verification on the trained LSTM-LC model by using 37440 accuracy verification samples.
The model verification result of the embodiment is relatively goodIdeally, the mean square error of the constructed LSTM-based unmanned vehicle lane change track prediction model is 3.18m 2 This result is satisfactory in view of the complex concrete operations and diversification of situations in the lane changing behavior of the vehicle.
In conclusion, the method for predicting the lane change track of the unmanned vehicle based on deep learning is suitable for the man-machine hybrid driving environment, can rapidly and effectively predict the lane change track of the unmanned vehicle by using measured data, lays a foundation for developing a driving track prediction model of the unmanned vehicle, has practical popularization value, and is worthy of popularization.
The embodiments described above are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the embodiments described above, and any other changes, modifications, substitutions, combinations, and simplifications that do not depart from the spirit and principles of the present invention should be made in the equivalent manner, and are included in the scope of the present invention.
Claims (2)
1. The intelligent vehicle lane change track prediction method suitable for the man-machine hybrid driving environment is characterized by comprising the following steps of:
s1, acquiring track data of an unmanned automobile travelling on a high-speed road section;
s2, processing the acquired driving track data, acquiring the space information, the motion information and the data information of the relative states of the adjacent vehicles, screening lane change data, and generating a sample database, wherein the sample database comprises a sample data set and an accuracy verification sample set;
the spatial information comprises the position of a lane where the vehicle is traveling in the lane;
the motion information comprises vehicle speed, vehicle length and vehicle width;
the data information of the relative states of the adjacent vehicles specifically refers to the relative state vectors of the lane change vehicle i and the adjacent vehicle jWherein Deltax i,j (t)，Δy i,j (t)，Δv i,j (t) represents the relative x-axis coordinates, the relative y-axis coordinates and the relative speed of vehicles i and j, respectively, at time t;
s3, constructing a lane change track prediction model of the unmanned vehicle, wherein the number of input variables of the model is 6N, the number of output variables is 2, and N is the length of a historical sample sequence;
the lane change track prediction model of the unmanned vehicle is constructed based on a long-short-period memory neural network;
the 6N input variables of the lane change track prediction model of the unmanned vehicle are as follows:
(1) lane change destination
The lane change destination of the lane change vehicle i at the time t- (N-1) tau is indicated, n=1, 2, …, N are the lengths of historical sample sequences, and tau is the time interval between the input time sequences;
(2) infrastructure information M
Taking the transverse x-axis coordinate of the median line of the lane with the nearest current position of the vehicle i as infrastructure information;
(3) vehicle parameter C i
The length and width of the vehicle i are used for representing the type of the vehicle, namely:
C i ＝{W i ,L i }
in which W is i 、L i The width and length of the vehicle i, respectively;
(4) state vector
The method comprises the following steps of the instant speed of the vehicle i, the direction and the amplitude of the motion state in the lane change process, namely:
wherein x is i (t-(n-1)τ)，y i (t-(n-1)τ)，v i (t- (n-1) τ) is the vehicle respectivelyi is the x-axis coordinate, the y-axis coordinate and the speed of t- (n-1) tau moment;
(5) vehicle parameter C j
The length, width of the neighboring vehicle j is used to characterize the vehicle type, namely:
C j ＝{W j ,L j }
in which W is j 、L j The width and length of vehicle j, respectively;
(6) relative state vector
The relative speed of the vehicles i and j is included, and the direction and the amplitude of the relative motion state of the vehicles i and j in the lane change process are as follows:
wherein Deltax is i,j (t-(n-1)τ)，Δy i,j (t-(n-1)τ)，Δv i,j (t- (n-1) τ) is the relative x-axis coordinates, relative y-axis coordinates, and relative speed magnitudes of vehicles i and j, respectively, at time t- (n-1) τ;
the output variables of the lane change track prediction model of the unmanned vehicle are as follows:
(1) a position prediction result x (t+t) of the model on the x-axis direction of the vehicle;
wherein T represents a prediction delay, and the value range is 0.1-1.5s;
(2) a position prediction result y (t+T) of the model in the y-axis direction of the vehicle;
s4, training a vehicle lane change track prediction model, training by using a RMSprop optimization method with a model track MSE error as a conditional function for training model convergence, adopting a back propagation algorithm as a basic gradient calculation method, obtaining an optimal vehicle lane change track prediction model structure and an optimal training sample size when the model track MSE error is minimum, and further selecting an optimal historical sequence length;
10000 samples of different sequence length N conditions are randomly extracted, and an LSTM model is tested to obtain a test history sequence length;
training the vehicle lane change track prediction model by using different training sample amounts to obtain an optimal training sample amount and an optimal lane change track prediction model structure;
on the premise of knowing the optimal training sample size and the optimal lane change track prediction model structure, training the vehicle lane change track prediction models with different historical sequence lengths to obtain the optimal historical sequence length;
and S5, verifying the structure of the optimal lane change track prediction model by utilizing data in the accuracy verification sample set, and performing lane change track prediction by using the verified optimal lane change track prediction model.
2. The intelligent vehicle lane change track prediction method suitable for the man-machine hybrid driving environment according to claim 1, wherein the total number of training samples is not less than 50000, the total number of accuracy verification samples is not less than 10000, the track accuracy requirement is within 0.5m, and the sample sampling rate is between 1 and 10 Hz.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910809352.4A CN110555476B (en) | 2019-08-29 | 2019-08-29 | Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910809352.4A CN110555476B (en) | 2019-08-29 | 2019-08-29 | Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110555476A CN110555476A (en) | 2019-12-10 |
| CN110555476B true CN110555476B (en) | 2023-09-26 |
Family
ID=68738437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910809352.4A Active CN110555476B (en) | 2019-08-29 | 2019-08-29 | Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110555476B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111114556A (en) * | 2019-12-24 | 2020-05-08 | 北京工业大学 | Lane change intention identification method based on LSTM under multi-source exponential weighting loss |
| CN111243309A (en) * | 2020-01-10 | 2020-06-05 | 北京航空航天大学 | Expressway traffic flow full-sample trajectory reconstruction method based on automatic driving vehicle movement detection |
| CN111310735A (en) * | 2020-03-18 | 2020-06-19 | 桂林电子科技大学 | Automobile track prediction method based on LSTM technology |
| US11433923B2 (en) | 2020-06-10 | 2022-09-06 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for long-term prediction of lane change maneuver |
| CN112965489B (en) * | 2021-02-05 | 2022-04-08 | 北京理工大学 | Intelligent vehicle high-speed lane change planning method based on collision detection |
| CN113033899B (en) * | 2021-03-29 | 2023-03-17 | 同济大学 | Unmanned adjacent vehicle track prediction method |
| CN115071704B (en) * | 2022-07-19 | 2022-11-11 | 小米汽车科技有限公司 | Trajectory prediction method, apparatus, medium, device, chip and vehicle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106926844A (en) * | 2017-03-27 | 2017-07-07 | 西南交通大学 | A kind of dynamic auto driving lane-change method for planning track based on real time environment information |
| CN107315411A (en) * | 2017-07-04 | 2017-11-03 | 合肥工业大学 | A kind of lane-change method for planning track based on automatic driving vehicle under collaborative truck |
| CN109460023A (en) * | 2018-11-09 | 2019-03-12 | 上海理工大学 | Driver's lane-changing intention recognition methods based on Hidden Markov Model |
| CN109471436A (en) * | 2018-11-09 | 2019-03-15 | 上海理工大学 | Based on mixed Gaussian-Hidden Markov Model lane-change Model Parameter Optimization method |
| CN109711557A (en) * | 2018-12-28 | 2019-05-03 | 西安航空学院 | A kind of wheelpath prediction technique, system, computer equipment and storage medium |
| CN110085056A (en) * | 2019-04-24 | 2019-08-02 | 华南理工大学 | Vehicle lane-changing instantaneous risk recognition methods under a kind of highway bus or train route cooperative surroundings |
| CN110146100A (en) * | 2018-02-13 | 2019-08-20 | 华为技术有限公司 | Trajectory predictions method, apparatus and storage medium |
-
2019
- 2019-08-29 CN CN201910809352.4A patent/CN110555476B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106926844A (en) * | 2017-03-27 | 2017-07-07 | 西南交通大学 | A kind of dynamic auto driving lane-change method for planning track based on real time environment information |
| CN107315411A (en) * | 2017-07-04 | 2017-11-03 | 合肥工业大学 | A kind of lane-change method for planning track based on automatic driving vehicle under collaborative truck |
| CN110146100A (en) * | 2018-02-13 | 2019-08-20 | 华为技术有限公司 | Trajectory predictions method, apparatus and storage medium |
| CN109460023A (en) * | 2018-11-09 | 2019-03-12 | 上海理工大学 | Driver's lane-changing intention recognition methods based on Hidden Markov Model |
| CN109471436A (en) * | 2018-11-09 | 2019-03-15 | 上海理工大学 | Based on mixed Gaussian-Hidden Markov Model lane-change Model Parameter Optimization method |
| CN109711557A (en) * | 2018-12-28 | 2019-05-03 | 西安航空学院 | A kind of wheelpath prediction technique, system, computer equipment and storage medium |
| CN110085056A (en) * | 2019-04-24 | 2019-08-02 | 华南理工大学 | Vehicle lane-changing instantaneous risk recognition methods under a kind of highway bus or train route cooperative surroundings |
Non-Patent Citations (2)
| Title |
|---|
| Capturing Drivers" Lane Changing Behaviors on Operational Level by Data Driven Methods;LING HUANG et al.;《IEEE Access：SPECIAL SECTION ON ADVANCED BIG DATA ANALYSIS FOR VEHICULAR SOCIAL NETWORKS》;20181029;第6卷;57497-57506 * |
| 基于LSTM网络的驾驶意图识别及车辆轨迹预测;季学武 等;《中国公路学报》;20190630;第32卷(第6期);34-42 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110555476A (en) | 2019-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110555476B (en) | Intelligent vehicle lane change track prediction method suitable for man-machine hybrid driving environment | |
| CN111832225B (en) | Method for constructing driving condition of automobile | |
| EP2610836B1 (en) | Device and method for the on-line prediction of the driving cycle in an automotive vehicle | |
| CN107330217B (en) | Mesoscopic oil consumption prediction method based on RBFNN | |
| CN109285348A (en) | A kind of vehicle behavior recognition methods and system based on two-way length memory network in short-term | |
| CN113704956A (en) | Urban road online microscopic simulation method and system based on digital twin technology | |
| CN107564290A (en) | A kind of urban road intersection saturation volume rate computational methods | |
| CN113239986B (en) | Training method and device for vehicle track evaluation network model and storage medium | |
| CN112633602B (en) | Traffic congestion index prediction method and device based on GIS map information | |
| CN111695737A (en) | Group target advancing trend prediction method based on LSTM neural network | |
| CN114881339A (en) | Vehicle trajectory prediction method, system, computer device, and storage medium | |
| CN116913088A (en) | Intelligent flow prediction method for expressway | |
| CN112749508B (en) | Road feel simulation method based on GMM and BP neural network | |
| CN112632706B (en) | Road feel simulation method based on GMM and CART regression tree | |
| CN106781508B (en) | Short-time Traffic Flow Forecasting Methods based on multiple phase space under a kind of Spark environment | |
| CN115691140B (en) | Analysis and prediction method for space-time distribution of automobile charging demand | |
| CN116611864A (en) | Urban rail transit train speed and passenger flow prediction method based on RNN | |
| CN116495007A (en) | Vehicle track deep learning prediction method considering physical constraint | |
| EP4194249A1 (en) | Method for predicting energy consumption-recovery ratio of new energy vehicle, and energy saving control method and system | |
| CN114912719A (en) | Heterogeneous traffic individual trajectory collaborative prediction method based on graph neural network | |
| CN114004406A (en) | Vehicle track prediction method and device, storage medium and electronic equipment | |
| CN112949939A (en) | Taxi passenger carrying hotspot prediction method based on random forest model | |
| Ni et al. | Research on Port truck trajectory completion based on Long Short-Term Memory model and Speed distribution characteristics | |
| Bohan et al. | Short-term traffic flow prediction based on bus floating car | |
| Li et al. | Modeling of merging decision during execution period based on random forest |
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 |