CN113306558A - Lane changing decision method and system based on lane changing interaction intention - Google Patents
Lane changing decision method and system based on lane changing interaction intention Download PDFInfo
- Publication number
- CN113306558A CN113306558A CN202110867345.7A CN202110867345A CN113306558A CN 113306558 A CN113306558 A CN 113306558A CN 202110867345 A CN202110867345 A CN 202110867345A CN 113306558 A CN113306558 A CN 113306558A
- Authority
- CN
- China
- Prior art keywords
- lane
- change
- intention
- vehicle
- interaction
- 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
- 230000003993 interaction Effects 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000008859 change Effects 0.000 claims abstract description 136
- 238000012549 training Methods 0.000 claims abstract description 37
- 230000002452 interceptive effect Effects 0.000 claims abstract description 24
- 230000015654 memory Effects 0.000 claims abstract description 21
- 238000012795 verification Methods 0.000 claims abstract description 9
- 238000002372 labelling Methods 0.000 claims abstract description 8
- 230000001133 acceleration Effects 0.000 claims description 14
- 238000001914 filtration Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 9
- 230000006399 behavior Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000009499 grossing Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/18—Propelling the vehicle
- B60W30/18009—Propelling the vehicle related to particular drive situations
- B60W30/18163—Lane change; Overtaking manoeuvres
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
- B60W40/04—Traffic conditions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
- B60W40/06—Road conditions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
- B60W40/105—Speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a lane change decision-making method and system based on lane change interaction intents. The method comprises the following steps: collecting lane change interaction data; marking lane change interactive data through road information and lane scenes, and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set; labeling the driving characteristics and determining a transverse lane changing intention; training a long-time memory network and a short-time memory network through a training set and a transverse lane changing intention to obtain a transverse lane changing intention model; training a long-time memory network through a transverse lane changing intention and driving characteristics to obtain a longitudinal yielding intention model; judging the lane change interaction state of the lane change interaction vehicle according to the transverse lane change intention and the longitudinal lane giving intention; and determining a lane change decision according to the lane change interaction state. The invention helps to make a correct decision when an automatic driving vehicle changes lanes or vehicles around the vehicle change lanes in the lane changing scene and improve the traffic safety in the lane changing process by identifying the current lane changing interaction intention of the vehicle.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a lane change decision-making method and system based on lane change interaction intention.
Background
With the increasing number of automobiles and the increasing intelligence degree of automatic driving vehicles, higher requirements are put on understanding of interaction behaviors between vehicles, and research on the interaction behaviors between vehicles is more and more focused. The lane changing behavior is a common basic driving scene, and a lane changing model of an automatic driving vehicle is widely researched, however, the research on the vehicle interaction problem in the lane changing process at home and abroad is not systematic enough at present.
Improper driving operations such as illegal lane changing, cut-in and overtaking of vehicles can cause traffic conflicts and collisions, and the efficiency of traffic flow is greatly influenced. In a lane change scene, due to the reasons of high traffic density, multiple uncertain factors and the like, a reasonable passing decision is difficult to be given by the current automatic driving vehicle, and lane change interaction with surrounding traffic participants cannot be reasonably carried out.
On a traffic road, the decision of surrounding vehicles can be influenced after the vehicle generates the lane change intention, and the decision of lane change behavior of the vehicle can also be influenced by behaviors of the surrounding vehicles.
Disclosure of Invention
The invention aims to provide a lane change decision method and a lane change decision system based on lane change interaction intents, which help to make correct decisions and improve traffic safety in the lane change process when an automatic driving vehicle changes lanes or faces surrounding vehicles to change lanes in a lane change scene by identifying the current lane change interaction intents of the vehicles.
In order to achieve the purpose, the invention provides the following scheme:
a lane change decision-making method based on lane change interaction intents comprises the following steps:
collecting lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles comprise lane changing vehicles and surrounding vehicles;
marking the lane change interactive data through the road information and the lane scene, and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set;
labeling the driving characteristics and determining a transverse lane changing intention;
training a long-time memory network and a short-time memory network through the training set and the transverse lane changing intention to obtain a transverse lane changing intention model;
training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model;
judging a lane change interaction state of the lane change interaction vehicle according to a horizontal lane change intention output by a horizontal lane change intention model and a longitudinal lane giving intention output by a longitudinal lane giving intention model;
and determining a lane change decision according to the lane change interaction state.
Optionally, after the acquiring lane change interaction data, the method further includes:
and carrying out noise reduction and filtering processing on the channel change interactive data.
Optionally, the driving characteristics include lane change vehicle absolute status informationI E Absolute state information of surrounding vehiclesI O Interaction status information with two vehiclesI R ;
Wherein,x,yrespectively vehicle longitudinal and transverse position, ΔxIndicating the relative lateral position, Δ, between the lane-change vehicle and the surrounding vehiclesyIndicating the relative longitudinal position, Δ, between the lane-change vehicle and the surrounding vehiclev x Representing the relative lateral velocity component, Δ, between the lane-change vehicle and the surrounding vehiclev y Indicating between a lane-change vehicle and a surrounding vehicleRelative longitudinal velocity component, Δa x Representing the relative lateral acceleration component, Δ, between the lane-change vehicle and the surrounding vehiclea y Representing the relative longitudinal acceleration component between the lane-change vehicle and the surrounding vehicle,Ein order to change the lane of the vehicle,Oare the vehicles around the vehicle, and are,Ris an interactive vehicle.
Optionally, the lateral lane change intent comprises a left-passing lane change intent, a right-passing lane change intent, a left-following lane change intent, a right-following lane change intent, and a lane keeping intent.
Optionally, the formula for longitudinal yield intent is as follows:
whereinLCGFor the purpose of vertical yielding, an active yielding is indicated when the value is-1, a passive yielding is indicated when the value is 1,a x o is the acceleration component of the surrounding vehicle in the x-direction.
Optionally, the rule for determining the lane change interaction state of the lane change interaction vehicle is as follows:
wherein,LCITin order to determine the rules, the rules are determined,LCIfor the purpose of the transverse lane change,LCGfor the purpose of giving longitudinal consent.
Optionally, the method further comprises:
and verifying the accuracy of the transverse lane change intention model through the verification set.
The invention also provides a lane change decision system based on the lane change interaction intention, which comprises the following steps:
the data acquisition module is used for acquiring lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles comprise lane changing vehicles and surrounding vehicles;
the marking and extracting module is used for marking the lane change interactive data through the road information and the lane scene and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set;
the label module is used for labeling the driving characteristics and determining a transverse lane changing intention;
the first training module is used for training a long-time memory network through the training set and the transverse lane change intention to obtain a transverse lane change intention model;
the second training module is used for training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model;
the lane changing interaction state determination module is used for determining a lane changing interaction state of the lane changing interaction vehicle according to a transverse lane changing intention output by the transverse lane changing intention model and a longitudinal lane giving intention output by the longitudinal lane giving intention model;
and the channel changing decision determining module is used for determining a channel changing decision according to the channel changing interaction state.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention firstly identifies the transverse lane changing intention of the lane changing vehicles, identifies the yielding intention of the surrounding vehicles, and further judges the lane changing interaction state between the vehicles on the basis.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flowchart of a lane change decision method based on lane change interaction intention according to an embodiment of the present invention;
FIG. 2 is a schematic view of a left lane-change overtaking scene;
FIG. 3 is a schematic diagram of a right lane change overtaking scene;
FIG. 4 is a schematic view of a left lane-changing and following scene;
FIG. 5 is a schematic view of a right lane-change car-following scene;
fig. 6 is a view of a car following scene.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a lane change decision method and a lane change decision system based on lane change interaction intents, which help to make correct decisions and improve traffic safety in the lane change process when an automatic driving vehicle changes lanes or faces surrounding vehicles to change lanes in a lane change scene by identifying the current lane change interaction intents of the vehicles.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the lane change decision method based on the lane change interaction intention disclosed by the present invention comprises the following steps:
step 101: collecting lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles include lane change vehicles and surrounding vehicles.
The lane-changing interactive data are collected through a vehicle-mounted sensor, and the sensor comprises a camera, a laser radar, a millimeter wave radar, a navigation system, an electronic control system and the like. And ensure the time consistency of the collected data.
The scene may be a three-lane scene, where the left lane is a fast lane and the right lane is a slow lane, but is not limited thereto, and the number of lanes may be determined according to actual situations. The vehicle is located in a middle lane, called a lane change vehicle, a lane change left/right, or lane keeping. Lane changing vehicles and surrounding vehicles perform lane changing interaction, and focus on five sub-lane changing scenes as shown in fig. 2-6. The traffic flow comprises low-speed and high-speed traffic conditions, and complex interactive behaviors such as overtaking exist in the driving process of the vehicle, and the traffic flow belongs to a combined derivative form in the invention.
Step 102: marking the lane change interactive data through the road information and the lane scene, and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set.
And preprocessing the acquired data in the step 101, wherein the data preprocessing method comprises noise reduction, filtering and the like, and the driving data is subjected to smoothing processing by selecting an interpolation, fitting or filtering method so as to reduce noise interference of the data and improve the performance of the intention identification model. For example, the smoothing method may be a moving average filtering method, and the formula is as follows:
wherein,F(k) Is shown inkFiltered data of each filtering window, and, ,frepresenting collected driving data prior to filtering;Nrepresents the total length of the sample;Lrepresents the filter window length; max (-) and min (-) represent the maximum and minimum values of the array, respectively. And obtaining the result of filtering all the lane change interaction data according to a formula.
Marking lane change data according to scenes shot by the camera and road information, and extracting three driving characteristics including lane change vehicle absolute state information from collected informationI E Absolute state information of surrounding vehiclesI O Interaction status information with two vehiclesI R Three types of driving information:
wherein,x,yrespectively the longitudinal and transverse position of the vehicle,v x ,v y are respectively vehiclesx,yThe component of the velocity in the direction of the,a x ,a y are respectively vehiclesx,yAcceleration component in the direction. DeltaxIndicating the relative lateral position, Δ, between the lane-change vehicle and the surrounding vehiclesyIndicating the relative longitudinal position, Δ, between the lane-change vehicle and the surrounding vehiclev x Representing the relative lateral velocity component, Δ, between the lane-change vehicle and the surrounding vehiclev y Representing the relative longitudinal speed component, Δ, between the lane-change vehicle and the surrounding vehiclea x Representing the relative lateral acceleration component, Δ, between the lane-change vehicle and the surrounding vehiclea y Representing the relative longitudinal acceleration component between the lane-change vehicle and the surrounding vehicle,Efor changing lanes,OAre the vehicles around the vehicle, and are,Ris an interactive vehicle.
In addition, based on a cross validation theory, the preprocessed driving feature data are divided into a training data set and a validation data set
Step 103: and labeling the driving characteristics and determining a transverse lane changing intention.
And labeling different driving characteristics to express 5 transverse lane changing driving intents of left overtaking lane changing, right overtaking lane changing, left following lane changing, right following lane changing or lane keeping of the lane changing vehicle.
Step 104: and training a long-time memory network and a short-time memory network through the training set and the transverse lane changing intention to obtain a transverse lane changing intention model.
And further identifying the transverse lane-changing intention of the lane-changing vehicle according to the driving characteristics processed in the step 102. The data input format for the lateral intent recognition model is as follows:
selecting long and short time memory network (LSTM) to drive data under different transverse intention labelsIRespectively learning, training and feature extraction, and then establishing a transverse lane change intention model corresponding to the lane change vehicle. The LSTM network layer consists of and point operations that act as gates for data input, output and forgetting, three gates providing information for the LSTM cell state, while the information state remains long-term memory throughout the network learning process as well as input and output. The model training process specifically comprises: based on the preprocessed driving data, firstly, a forgetting gate of the network determines [0,1 ] by using a sigmoid activation function]Represents the probability of forgetting the state of the network element; then, the input gate controls to store and update information by using a sigmoid activation function, and obtains new candidate state information through a tanh function so as to obtain new internal state information; and finally, inputting the state information and the hidden state of the previous sequence through a sigmoid activation function of an output gate, and transmitting the internal state information to the external state by combining the output gate, namely the recognized driving intention.
And finally, extracting five transverse intention characteristic parameters including left overtaking lane changing, right overtaking lane changing, left following lane changing, right following lane changing and lane keeping from the driving characteristics based on the selected long-time memory network. And respectively establishing corresponding horizontal lane change intention recognition models which respectively correspond to the 5 lane change scenes in the figures 2 to 6. The 5 driving intentions are: intention to change lane for left-hand overtaking (LCI= 1), right overtaking lane change intention (LCI= 2), left car-following lane-changing intention (LCI= 3), right car-following lane-changing intention (LCI= 4) and lane keeping intention (lane keeping intention: (lane keeping intention) ((b))LCI= 5). And then, aiming at different lane changing intentions, the model outputs corresponding driving intentionsLCI=1,2,3,4,5。
Step 105: and training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model.
In recognizing transverse lane change intentionLCIAnd on the basis of the =1,2,3,4 and 5, identifying the yielding intention of the surrounding vehicles on the target lane. The data input of the longitudinal yielding intention recognition model is transverse lane changing intention and driving dataI。
The giving way intention of the surrounding vehicle means: and on the basis of the identified transverse lane changing intention of the lane changing vehicles, the concessional intention of the surrounding vehicles is identified so as to analyze and judge lane changing interactive behaviors among the vehicles. Specifically, a long-time and short-time memory network is selected to learn, train and extract characteristics of driving data, longitudinal acceleration parameters of surrounding vehicles are predicted, and then a longitudinal giving intention model corresponding to the surrounding vehicles is established. The process of establishing the longitudinal yielding intention model is similar to the process of establishing the transverse driving intention model, and then the yielding intention is judged. When the intention of transverse lane change is lane keepingLCIIf =5, the surrounding vehicle is not allowed to passLCG= 0; when the transverse lane change is intended to change the lane of the left following carLCI=3 or right car-following lane changeLCI=4, since surrounding vehicles are located in front of the lane change vehicle target lane, not paying attention to their concessional intention LCG; when the transverse lane changing intention is left overtaking lane changing or right overtaking lane changing, the line giving intention formula is as follows:
whereinLCGFor the purpose of vertical yielding, an active yielding is indicated when the value is-1, a passive yielding is indicated when the value is 1,a x o is the acceleration component of the surrounding vehicle in the x-direction.
Step 106: and judging the lane change interaction state of the lane change interaction vehicle according to the lateral lane change intention output by the lateral lane change intention model and the longitudinal lane giving intention output by the longitudinal lane giving intention model.
And judging the lane change interaction state according to the following judgment rule:
taking the two identified intention results as input of lane changing interaction state judgment, wherein when the transverse lane changing intention is lane keeping, the lane changing interaction state between the vehicles is a weak interaction state; when the transverse lane changing intention is that a left lane is changed or a right lane is changed, the lane changing interaction state is a one-way interaction state, mainly used as a lane changing vehicle of a rear vehicle to interact with surrounding front vehicles in a one-way mode, and the longitudinal acceleration of the lane changing vehicle is obtained specificallya x E And lateral accelerationa y E To embody; when the transverse lane changing intention is left-hand overtaking lane changing or right-hand overtaking lane changing, the lane changing interaction state comprises active lane-giving interaction and passive lane-giving interaction, a lane changing interaction degree/mode is obtained, if forced lane changing interaction is carried out, the longitudinal acceleration of the lane changing vehicle is obtaineda x E And lateral accelerationa y E And longitudinal acceleration of surrounding vehiclea x O To be embodied.
Step 107: and determining a lane change decision according to the lane change interaction state.
Based on the interaction state determined in step 106, when the interaction state is a weak interaction state or a one-way interaction state, performing a lane change decision according to a set lane change decision model; if the lane changing is in the active yielding interactive state, the lane changing is considered to have small influence on surrounding vehicles, and a set lane changing decision can be made; if the vehicle is in a passive yielding interactive state, the influence on surrounding rear vehicles is considered to be large, and in order to reduce the collision risk and ensure the traffic safety, the lane change decision needs to be appropriately changed, for example, the lane change is accelerated, and even the lane change is abandoned.
The invention also provides a lane change decision system based on the lane change interaction intention, which comprises the following steps:
the data acquisition module is used for acquiring lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles comprise lane changing vehicles and surrounding vehicles;
the marking and extracting module is used for marking the lane change interactive data through the road information and the lane scene and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set;
the label module is used for labeling the driving characteristics and determining a transverse lane changing intention;
the first training module is used for training a long-time memory network through the training set and the transverse lane change intention to obtain a transverse lane change intention model;
the second training module is used for training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model;
the lane changing interaction state determination module is used for determining a lane changing interaction state of the lane changing interaction vehicle according to a transverse lane changing intention output by the transverse lane changing intention model and a longitudinal lane giving intention output by the longitudinal lane giving intention model;
and the channel changing decision determining module is used for determining a channel changing decision according to the channel changing interaction state.
The invention firstly identifies the transverse lane changing intention of the lane changing vehicles, identifies the yielding intention of the surrounding vehicles, and further judges the lane changing interaction state between the vehicles on the basis.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. A lane change decision method based on a lane change interaction intention is characterized by comprising the following steps:
collecting lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles comprise lane changing vehicles and surrounding vehicles;
marking the lane change interactive data through the road information and the lane scene, and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set;
labeling the driving characteristics and determining a transverse lane changing intention;
training a long-time memory network and a short-time memory network through the training set and the transverse lane changing intention to obtain a transverse lane changing intention model;
training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model;
judging a lane change interaction state of the lane change interaction vehicle according to a horizontal lane change intention output by a horizontal lane change intention model and a longitudinal lane giving intention output by a longitudinal lane giving intention model;
and determining a lane change decision according to the lane change interaction state.
2. The lane-change decision method based on lane-change interaction intention as claimed in claim 1, further comprising, after said collecting lane-change interaction data:
and carrying out noise reduction and filtering processing on the channel change interactive data.
3. The lane-change interaction intent-based lane-change decision making method according to claim 1, wherein the driving characteristics comprise lane-change vehicle absoluteStatus informationI E Absolute state information of surrounding vehiclesI O Interaction status information with two vehiclesI R ;
Wherein,x,yrespectively the longitudinal and transverse position of the vehicle,v x ,v y are respectively vehiclesx,yThe component of the velocity in the direction of the,a x ,a y are respectively vehiclesx,yComponent of acceleration in the direction, ΔxIndicating the relative lateral position, Δ, between the lane-change vehicle and the surrounding vehiclesyIndicating the relative longitudinal position, Δ, between the lane-change vehicle and the surrounding vehiclev x Representing the relative lateral velocity component, Δ, between the lane-change vehicle and the surrounding vehiclev y Representing the relative longitudinal speed component, Δ, between the lane-change vehicle and the surrounding vehiclea x Representing the relative lateral acceleration component, Δ, between the lane-change vehicle and the surrounding vehiclea y Representing the relative longitudinal acceleration component between the lane-change vehicle and the surrounding vehicle,Ein order to change the lane of the vehicle,Oare the vehicles around the vehicle, and are,Ris an interactive vehicle.
4. The lane-change decision-making method based on the lane-change interaction intention, according to claim 1, wherein the lateral lane-change intention comprises a left-passing lane-change intention, a right-passing lane-change intention, a left-following lane-change intention, a right-following lane-change intention and a lane-keeping intention.
5. The lane-change decision making method based on lane-change interaction intention of claim 1, wherein the formula of the longitudinal yielding intention is as follows:
whereinLCGFor the purpose of vertical yielding, an active yielding is indicated when the value is-1, a passive yielding is indicated when the value is 1,a x o is the acceleration component of the surrounding vehicle in the x-direction.
6. The lane-change decision method based on the lane-change interaction intention as claimed in claim 1, wherein the rule for determining the lane-change interaction state of the lane-change interaction vehicle is as follows:
wherein,LCITin order to determine the rules, the rules are determined,LCIfor the purpose of the transverse lane change,LCGfor the purpose of giving longitudinal consent.
7. The lane change decision method based on lane change interaction intention of claim 1, further comprising:
and verifying the accuracy of the transverse lane change intention model through the verification set.
8. A lane change decision system based on a lane change interaction intention, comprising:
the data acquisition module is used for acquiring lane change interaction data; the lane change interaction data comprises: position information, speed, road information and lane scene of the vehicle; the vehicles comprise lane changing vehicles and surrounding vehicles;
the marking and extracting module is used for marking the lane change interactive data through the road information and the lane scene and extracting driving characteristics from the lane change interactive data to construct a training set and a verification set;
the label module is used for labeling the driving characteristics and determining a transverse lane changing intention;
the first training module is used for training a long-time memory network through the training set and the transverse lane change intention to obtain a transverse lane change intention model;
the second training module is used for training a long-time memory network through the transverse lane changing intention and the driving characteristics to obtain a longitudinal yielding intention model;
the lane changing interaction state determination module is used for determining a lane changing interaction state of the lane changing interaction vehicle according to a transverse lane changing intention output by the transverse lane changing intention model and a longitudinal lane giving intention output by the longitudinal lane giving intention model;
and the channel changing decision determining module is used for determining a channel changing decision according to the channel changing interaction state.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110867345.7A CN113306558B (en) | 2021-07-30 | 2021-07-30 | Lane changing decision method and system based on lane changing interaction intention |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110867345.7A CN113306558B (en) | 2021-07-30 | 2021-07-30 | Lane changing decision method and system based on lane changing interaction intention |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113306558A true CN113306558A (en) | 2021-08-27 |
CN113306558B CN113306558B (en) | 2021-11-09 |
Family
ID=77382476
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110867345.7A Active CN113306558B (en) | 2021-07-30 | 2021-07-30 | Lane changing decision method and system based on lane changing interaction intention |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113306558B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113753049A (en) * | 2021-11-10 | 2021-12-07 | 北京理工大学 | Social preference-based automatic driving overtaking decision determination method and system |
CN115294546A (en) * | 2022-07-22 | 2022-11-04 | 中兵智能创新研究院有限公司 | Lane change primitive-based lane change decision-making behavior method and system |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105912814A (en) * | 2016-05-05 | 2016-08-31 | 苏州京坤达汽车电子科技有限公司 | Lane change decision model of intelligent drive vehicle |
CN106874597A (en) * | 2017-02-16 | 2017-06-20 | 北理慧动(常熟)车辆科技有限公司 | A kind of highway passing behavior decision-making technique for being applied to automatic driving vehicle |
WO2018007262A1 (en) * | 2016-07-08 | 2018-01-11 | Volkswagen Aktiengesellschaft | Turned-wheel detection for yielding during low-speed lane changes |
CN109991987A (en) * | 2019-04-29 | 2019-07-09 | 北京智行者科技有限公司 | Automatic Pilot decision-making technique and device |
CN110723146A (en) * | 2019-10-31 | 2020-01-24 | 哈尔滨工业大学 | Multi-vehicle exchange lane decision method considering driving benefit increment |
CN111081065A (en) * | 2019-12-13 | 2020-04-28 | 北京理工大学 | Intelligent vehicle collaborative lane change decision model under road section mixed traveling condition |
CN111845787A (en) * | 2020-08-03 | 2020-10-30 | 北京理工大学 | Lane change intention prediction method based on LSTM |
CN112373485A (en) * | 2020-11-03 | 2021-02-19 | 南京航空航天大学 | Decision planning method for automatic driving vehicle considering interactive game |
CN112406867A (en) * | 2020-11-19 | 2021-02-26 | 清华大学 | Emergency vehicle hybrid lane change decision method based on reinforcement learning and avoidance strategy |
-
2021
- 2021-07-30 CN CN202110867345.7A patent/CN113306558B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105912814A (en) * | 2016-05-05 | 2016-08-31 | 苏州京坤达汽车电子科技有限公司 | Lane change decision model of intelligent drive vehicle |
WO2018007262A1 (en) * | 2016-07-08 | 2018-01-11 | Volkswagen Aktiengesellschaft | Turned-wheel detection for yielding during low-speed lane changes |
CN106874597A (en) * | 2017-02-16 | 2017-06-20 | 北理慧动(常熟)车辆科技有限公司 | A kind of highway passing behavior decision-making technique for being applied to automatic driving vehicle |
CN109991987A (en) * | 2019-04-29 | 2019-07-09 | 北京智行者科技有限公司 | Automatic Pilot decision-making technique and device |
CN110723146A (en) * | 2019-10-31 | 2020-01-24 | 哈尔滨工业大学 | Multi-vehicle exchange lane decision method considering driving benefit increment |
CN111081065A (en) * | 2019-12-13 | 2020-04-28 | 北京理工大学 | Intelligent vehicle collaborative lane change decision model under road section mixed traveling condition |
CN111845787A (en) * | 2020-08-03 | 2020-10-30 | 北京理工大学 | Lane change intention prediction method based on LSTM |
CN112373485A (en) * | 2020-11-03 | 2021-02-19 | 南京航空航天大学 | Decision planning method for automatic driving vehicle considering interactive game |
CN112406867A (en) * | 2020-11-19 | 2021-02-26 | 清华大学 | Emergency vehicle hybrid lane change decision method based on reinforcement learning and avoidance strategy |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113753049A (en) * | 2021-11-10 | 2021-12-07 | 北京理工大学 | Social preference-based automatic driving overtaking decision determination method and system |
CN115294546A (en) * | 2022-07-22 | 2022-11-04 | 中兵智能创新研究院有限公司 | Lane change primitive-based lane change decision-making behavior method and system |
Also Published As
Publication number | Publication date |
---|---|
CN113306558B (en) | 2021-11-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10068171B2 (en) | Multi-layer fusion in a convolutional neural network for image classification | |
CN113486822B (en) | Surrounding vehicle track prediction method and system based on driving intention | |
CN110843789B (en) | Vehicle lane change intention prediction method based on time sequence convolution network | |
CN112242059B (en) | Intelligent decision-making method for unmanned vehicle based on motivation and risk assessment | |
CN113306558B (en) | Lane changing decision method and system based on lane changing interaction intention | |
CN110751847B (en) | Decision-making method and system for automatically driving vehicle behaviors | |
CN110027553A (en) | A kind of anti-collision control method based on deeply study | |
CN112793576B (en) | Lane change decision method and system based on rule and machine learning fusion | |
CN112706785B (en) | Method and device for selecting cognitive target of driving environment of automatic driving vehicle and storage medium | |
CN114043989B (en) | Driving style recognition model, lane change decision model and decision method based on recursion diagram and convolutional neural network | |
CN113548054B (en) | Vehicle lane change intention prediction method and system based on time sequence | |
Zhang et al. | A framework for turning behavior classification at intersections using 3D LIDAR | |
CN106569214A (en) | Method and system for processing vehicle-mounted radar data of adaptive cruise vehicle in conjunction with navigation information | |
CN115451987A (en) | Path planning learning method for automatic driving automobile | |
CN112896166A (en) | Vehicle lane changing method and device and electronic equipment | |
CN116341288A (en) | Heterogeneous traffic epidemic car security field modeling method | |
CN114371015B (en) | Automatic driving test method, automatic driving test device, computer equipment and storage medium | |
CN112810604B (en) | Intelligent vehicle behavior decision method and system based on parking lot scene | |
CN114248780A (en) | IDM-LSTM combined following model establishing method considering driver style | |
Sanberg et al. | Asteroids: A stixel tracking extrapolation-based relevant obstacle impact detection system | |
CN112590792A (en) | Vehicle convergence control method based on deep reinforcement learning algorithm | |
CN118238847B (en) | Autonomous lane change decision planning method and system adaptive to different driving styles and road surface environments | |
CN111126338A (en) | Intelligent vehicle environment perception method integrating visual attention mechanism | |
US20220405517A1 (en) | System, method, and vehicle for recognition of traffic signs | |
CN113807009B (en) | Segmentation extraction method for microscopic lane change track |
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 |