CN116592903A - Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment - Google Patents
Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment Download PDFInfo
- Publication number
- CN116592903A CN116592903A CN202310502145.0A CN202310502145A CN116592903A CN 116592903 A CN116592903 A CN 116592903A CN 202310502145 A CN202310502145 A CN 202310502145A CN 116592903 A CN116592903 A CN 116592903A
- Authority
- CN
- China
- Prior art keywords
- driving
- data
- behavior
- vehicle
- driving behavior
- 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 45
- 230000006399 behavior Effects 0.000 claims abstract description 369
- 238000004364 calculation method Methods 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 238000012946 outsourcing Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000010207 Bayesian analysis Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 10
- 238000005265 energy consumption Methods 0.000 description 9
- 238000013507 mapping Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3407—Route searching; Route guidance specially adapted for specific applications
- G01C21/3415—Dynamic re-routing, e.g. recalculating the route when the user deviates from calculated route or after detecting real-time traffic data or accidents
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a method for planning ecological driving paths with group preference in real time under a vehicle-road cooperative environment, which comprises the steps of pre-building a vehicle driving behavior library; the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in the cloud server; acquiring road data of a current vehicle, matching corresponding group preference driving behaviors in a vehicle driving behavior library based on the road data, and constructing a group preference crowdedness degree model based on vehicle-road cooperation; the group preference crowdedness model generates a dynamic behavior preference sequence through driving behavior data; and carrying out shortest path searching based on an AI algorithm and a preset destination, establishing a path dynamic planning model based on the group preference crowdedness degree model during the shortest path searching, and outputting an optimal driving strategy in real time by establishing the path dynamic planning model.
Description
Technical Field
The invention relates to the technical field of vehicle-road synchronization, in particular to a method for planning ecological driving paths with group preference in a vehicle-road cooperative environment in real time.
Background
The establishment of an intelligent traffic guidance system and a traffic management system is a key means for solving traffic jam at present. The intelligent traffic guidance system generates corresponding guidance information according to urban road traffic conditions, and feeds the guidance information back to traffic participants to influence the selection of the travelers on travel routes and travel modes so as to dredge traffic flow, maintain the optimal traffic capacity of roads and improve traffic safety, thereby achieving the purpose of efficiently utilizing road networks and realizing orderly smoothness of traffic. An important component of advanced traffic management systems is the traffic signal control system, which
The timing scheme of the traffic signal lamp on the road is optimally controlled by a specific control algorithm, so that the road traffic flow is scientifically organized and controlled, and the potential traffic capacity of the existing traffic network is fully exerted. Dynamic network traffic flow distribution is the most important core technology of a traffic guidance system and a traffic management system, the key point of research on traffic flow distribution in various countries is that a mechanism of path selection or a more effective solving algorithm is prone to be performed under different information conditions, but the research on important basis of vehicle driving road requirements is less, so that the progress of dynamic traffic distribution theory and application research is limited, the current navigation path planning method is difficult to reflect group dynamic decisions, and the demands of people in scientific travel, intelligent guidance and the like cannot be met.
Disclosure of Invention
The invention provides a real-time planning method for ecological driving paths with group preference under a vehicle-road cooperative environment, which fully considers the driving behavior preference mode of a driver, comprehensively considers driving transportation cost while providing personalized service for the driver, plans the driving paths of the vehicle in real time, and finally selects an optimal ecological driving path which meets the driving behavior preference of the driver and meets the green travel standard for the driver.
The invention provides a method for planning ecological driving paths with group preference in real time in a vehicle-road cooperative environment, which comprises the following steps:
pre-building a vehicle driving behavior library; wherein,,
the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in the cloud server;
acquiring road data of a current vehicle, matching corresponding group preference driving behaviors in a vehicle driving behavior library based on the road data, and constructing a group preference crowdedness degree model based on vehicle-road cooperation; wherein,,
the group preference crowdedness model generates a dynamic behavior preference sequence through driving behavior data;
and carrying out shortest path searching based on an AI algorithm and a preset destination, establishing a path dynamic planning model based on the group preference crowdedness degree model during the shortest path searching, and outputting an optimal driving strategy in real time by establishing the path dynamic planning model.
Preferably, the pre-building the vehicle driving behavior library includes:
presetting a driving behavior depth recognition algorithm, screening driving data captured by a cloud server, and determining driving behavior data; wherein,,
the driving behavior data includes: trajectory data, driver physiological data, and vehicle state data;
selecting behavior classification matching from a preset template library, and taking the behavior classification matching as a behavior classification template;
inputting the driving behavior data into the behavior classification template to generate a driving behavior data set;
calculating the occurrence frequency of different driving behaviors in a driving behavior data set based on a preset driving safety standard and a driving behavior specification, and taking the occurrence frequency as a group preference value;
generating according to the group preference value, and storing a scheduling sequence; wherein,,
in the stored schedule sequence, the higher the group preference value, the more convenient the rapid data scheduling.
Preferably, the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in a cloud server, including:
based on a crawler network of the cloud server, vehicle driving data are obtained from different approaches;
inputting the acquired vehicle driving data into a trained depth recognition algorithm to adaptively recognize driving behaviors and acquire different driving behaviors;
Inputting optimal representations of different driving behaviors to a multi-scale feature extraction unit to extract spatiotemporal features of driving behavior data;
the space-time characteristics are input into a multi-task learning unit to perform driving behavior learning and behavior prediction, and a behavior classification result and a behavior prediction result are obtained;
and carrying out fitting matching on the newly added driving data according to the classification result and the behavior prediction result, and updating the vehicle driving behavior library when the matching value accords with a preset threshold value.
Preferably, the matching the driving behavior of the corresponding group preference in the driving behavior library of the vehicle based on the road data includes:
classifying the road data to obtain a classification sequence; wherein,,
the classification process comprises the following steps: data type classification, road condition classification and driver status classification
Extracting behavior key points in each class of data of the classification sequence, and extracting behavior characteristics of each behavior key point;
asynchronous alignment is carried out on the classification sequence and similar data on a vehicle driving behavior library by utilizing a correlation algorithm;
screening target driving behavior data in the road data according to the asynchronous alignment, and generating a behavior type matching value;
carrying out weighted average calculation on target driving behavior data in a vehicle driving behavior library, determining the body preference crowdedness degree of the target driving behavior data, and taking the weighted average calculation as a second matching value;
Sorting the driving behavior day data in the road data according to the first matching value and the second matching value;
and determining group preference driving behaviors in the road data according to the sorting result.
Preferably, the building of the group preference crowdedness model based on vehicle-road cooperation includes:
presetting a model rule canvas and a control canvas; wherein,,
the model rule canvas is used for presetting the recognition rules of group preference;
the control canvas is used for connecting the data transmission control of the vehicle-road cooperation;
forming a rule model in a visual blueprint editing mode through a model rule canvas and a control canvas, and editing and supporting the rule model through a custom parameter table and a custom variable;
generating, by the rule model, a plurality of rule terms of driving behavior preferences; wherein,,
the rule items include: issuing an instruction preference rule item, a vehicle execution rule preference rule item, a trigger condition preference rule item, a vehicle track preference rule item and a vehicle execution instruction preference rule item;
and integrating a plurality of rule items into the depth calculation model to form a group preference crowdedness degree model.
Preferably, the generating the dynamic behavior preference sequence by the group preference popularity degree model through driving behavior data includes:
Pre-configuring a pre-trained complex model architecture, and generating a first model architecture through a structured dynamic data fusion interface;
configuring a dynamic training strategy in a first model framework, and setting repeated training test tasks according to the acquired road data and a vehicle driving behavior library;
based on the repeated training test task, executing a global iteration task when new driving behavior data are received;
and dynamically training the group preference crowdedness degree model through a global iteration task to generate a dynamic behavior preference sequence.
Preferably, the method further comprises:
according to the vehicle driving behavior library, a crowded driving behavior sequence and an ecological driving behavior sequence are generated; wherein,,
the crowd driving behavior sequence is determined by the implementation probability of driving behaviors;
the ecological driving behavior sequence is determined by the energy consumption cost of the driving behavior;
establishing a map mapping relation between a crowded driving behavior sequence and an ecological driving behavior sequence;
and screening the mass driving behavior sequence of the road data of the current vehicle according to the map mapping relation to determine the group preference driving behavior.
Preferably, the performing the shortest path search based on the AI algorithm and the preset destination includes:
Constructing an outsourcing rectangle by taking connecting lines of a starting destination and a destination as diagonal lines, and filtering the space of the outsourcing rectangle by an AI server to obtain road data in the outsourcing rectangle;
carrying out topological structure surface according to the connecting line and road data, searching a plurality of adjacent path polygons which are intersected with the connecting line space, merging the adjacent path polygons to obtain an initial path polygon, obtaining two paths of a connecting starting point and a connecting end point by taking the connecting line as a boundary according to the initial path polygon, and selecting the path with smaller length as an initial shortest path;
constructing a buffer rectangle by taking the connecting line as a central line and taking half of the length of the initial shortest path as a range, and spatially filtering out a second road central line in the buffer rectangle;
and carrying out topological structure according to the first road center line and the second road center line to obtain a path to be searched, and calculating the nodes of the path to be searched step by step from the starting point according to the path to be searched to obtain the shortest path.
Preferably, the path dynamic programming model is built when the shortest path is searched based on the group preference crowdedness model:
performing Bayesian analysis on the shortest path to obtain path risk factors, and establishing a path risk model based on the risk factors;
Based on two targets of path risk and path cost, a multi-target path dynamic planning model is established, and an objective function and constraint conditions are determined;
and integrating the objective function and the constraint condition into the population preference crowdedness model to generate a path dynamic programming model of the shortest path.
Preferably, the outputting the optimal driving strategy in real time by establishing a path dynamic programming model includes:
establishing a path dynamic programming model, specifically dividing the driving behavior of a user into general behavior, event subdivision behavior, crowd subdivision behavior and crowd preference behavior, generating corresponding strategy templates, and storing the strategy templates in a strategy template library in a database;
constructing a strategy tree, and determining an initial strategy node and sub-strategy nodes of each level of the strategy tree; the starting policy node serves as an entry for the entire policy tree;
all the strategy nodes are mutually related through connecting lines to form a strategy branch, and different driving behavior recommendation instructions are generated by strategy combination;
the current user enters a strategy tree through the starting node, and finally the corresponding optimal driving strategy is completed through crowded preference driving behavior matching recommendation of each level.
Preferably, the method further comprises:
setting monitoring equipment through a road side and taking the monitoring equipment as a patrol position;
After the optimal driving strategy is determined, when a user runs in the shortest path, collecting real-time running data of the vehicle;
screening driving behavior data in the real-time driving data, and judging whether the driving behavior data accords with an optimal driving strategy or not;
and when the optimal driving strategy is not met, vehicle patrol and alarm are carried out.
The invention has the beneficial effects that: the method comprises the steps of identifying the crowded preference behaviors in the driving process of the vehicle, establishing a path dynamic planning model, and obtaining the optimal driving path. The driving behavior preference mode of the driver is fully considered, the driving transportation cost is comprehensively considered while personalized service is provided for the driver, the vehicle driving path is planned in real time, and the optimal ecological driving path which meets the driving behavior preference of the driver and meets the green travel standard is selected for the driver.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of a method for planning an ecological driving path with group preference in a cooperative vehicle-road environment in real time in an embodiment of the invention;
FIG. 2 is a flowchart for constructing a vehicle driving behavior library in an embodiment of the invention;
fig. 3 is a monitoring flow chart of road side monitoring in an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
The invention provides a method for planning ecological driving paths with group preference in real time in a vehicle-road cooperative environment, which comprises the following steps:
pre-building a vehicle driving behavior library; wherein,,
the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in the cloud server;
acquiring road data of a current vehicle, matching corresponding group preference driving behaviors in a vehicle driving behavior library based on the road data, and constructing a group preference crowdedness degree model based on vehicle-road cooperation; wherein,,
The group preference crowdedness model generates a dynamic behavior preference sequence through driving behavior data;
and carrying out shortest path searching based on an AI algorithm and a preset destination, establishing a path dynamic planning model based on the group preference crowdedness degree model during the shortest path searching, and outputting an optimal driving strategy in real time by establishing the path dynamic planning model.
The principle of the counting scheme is as follows:
as shown in fig. 1, in this embodiment:
the vehicle driving behavior library is used for identifying all driving behaviors selectable when the vehicle runs, and the database can be continuously updated based on real-time crawling of the cloud server. In this embodiment, the body preference crowdedness degree model may determine that driving behaviors are better in different road conditions, and that driving behaviors are more adopted by people, so as to achieve optimal driving behavior matching and preference recommendation.
In the embodiment, the most frequent preferred driving behavior of the user can be judged by constructing the group preference crowdedness model based on the vehicle-road cooperation, so that the preferred driving behavior is recommended to the user.
In this embodiment, the path dynamic planning model may recommend the crowd preference behavior in the driving process under the condition that the path dynamic planning model is already the shortest planning path, so that the user can drive more safely and stably, and a corresponding optimal recommendation policy is generated.
The beneficial effects of the embodiment are that: the method comprises the steps of identifying the crowded preference behaviors in the driving process of the vehicle, establishing a path dynamic planning model, and obtaining the optimal driving path. The driving behavior preference mode of the driver is fully considered, the driving transportation cost is comprehensively considered while personalized service is provided for the driver, the vehicle driving path is planned in real time, and the optimal ecological driving path which meets the driving behavior preference of the driver and meets the green travel standard is selected for the driver.
Preferably, the pre-building the vehicle driving behavior library includes:
presetting a driving behavior depth recognition algorithm, screening driving data captured by a cloud server, and determining driving behavior data; wherein,,
the driving behavior data includes: trajectory data, driver physiological data, and vehicle state data;
selecting behavior classification matching from a preset template library, and taking the behavior classification matching as a behavior classification template;
Inputting the driving behavior data into the behavior classification template to generate a driving behavior data set;
calculating the occurrence frequency of different driving behaviors in a driving behavior data set based on a preset driving safety standard and a driving behavior specification, and taking the occurrence frequency as a group preference value;
generating according to the group preference value, and storing a scheduling sequence; wherein,,
in the stored schedule sequence, the higher the group preference value, the more convenient the rapid data scheduling.
The principle of the technical scheme is as follows:
as shown in fig. 2:
in this embodiment, in order to screen driving behavior data during the depth recognition algorithm, driving behavior data is continuously collected in real time through the cloud.
In this embodiment, the classification template may classify driving data, so that under the condition that the driving safety standard and the driving behavior specification are met, data scheduling, that is, instruction scheduling of driving behaviors with crowdedness preference, is performed according to the data scheduling mode of the maximum number of times adopted by people.
In this embodiment, the driving safety standard and the driving behavior specification are preset by using chinese, and the specific setting standard and the behavior specification conform to the safety requirements of actual driving.
In this embodiment, by calculating the occurrence frequency of different driving behaviors in the driving behavior data set, the group preference can be determined, so as to realize rapid group preference judgment and rapid data scheduling.
The beneficial effects of the technical scheme are that:
the driving behavior preference of the masses can be rapidly judged, so that a corresponding preference sequence is generated, the preference value of each driving behavior is calculated, and more rapid recommendation and scheduling are performed.
Preferably, the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in a cloud server, including:
based on a crawler network of the cloud server, vehicle driving data are obtained from different approaches;
inputting the acquired vehicle driving data into a trained depth recognition algorithm to adaptively recognize driving behaviors and acquire different driving behaviors;
inputting optimal representations of different driving behaviors to a multi-scale feature extraction unit to extract spatiotemporal features of driving behavior data;
the space-time characteristics are input into a multi-task learning unit to perform driving behavior learning and behavior prediction, and a behavior classification result and a behavior prediction result are obtained;
and carrying out fitting matching on the newly added driving data according to the classification result and the behavior prediction result, and updating the vehicle driving behavior library when the matching value accords with a preset threshold value.
The principle of the technical scheme is as follows:
in this embodiment, the crawler network of the cloud server may collect vehicle driving data as a data sample, then train to obtain screened vehicle driving behaviors, and finally generate a vehicle driving behavior library through all the vehicle driving behaviors, and continuously update the vehicle driving behavior library.
The multi-scale feature extraction unit can realize multi-scale division of driving behavior data based on scale analysis, so that space-time features in the driving behavior data can be identified and extracted; the space-time feature represents a change feature point of the driving behavior of the user in the time dimension;
the multi-task learning unit is used for learning the driving behavior of the user through the space-time characteristics and predicting the behavior; the behavior prediction and the driving behavior are based on the change feature points of the driving behavior of the user in the time dimension; and then classifying driving behaviors and specifically predicting the behaviors in a behavior prediction mode.
According to the invention, the newly-added driving data is matched in a fitting way, the behavior characteristics of the newly-added driving data are extracted, and the behavior of the newly-added driving data can be represented to have similar behavior in the previous driving behavior when the matching value reaches the preset threshold in a matching way.
The beneficial effects of the technical scheme are that:
the invention can realize the real-time update of the vehicle driving behavior library based on the crawler network; compared with the database updating mode in the prior art, the method fuses a plurality of modes such as training identification, space-time characteristics, behavior classification, behavior prediction, fitting matching and the like to realize database updating, and has the updating function and the classification division and classification prediction capability of data; finally, by means of fitting and matching, identification errors can be prevented.
Preferably, the matching the driving behavior of the corresponding group preference in the driving behavior library of the vehicle based on the road data includes:
classifying the road data to obtain a classification sequence; wherein,,
the classification process comprises the following steps: data type classification, road condition classification and driver status classification
Extracting behavior key points in each class of data of the classification sequence, and extracting behavior characteristics of each behavior key point;
asynchronous alignment is carried out on the classification sequence and similar data on a vehicle driving behavior library by utilizing a correlation algorithm;
screening target driving behavior data in the road data according to the asynchronous alignment, and generating a behavior type matching value;
carrying out weighted average calculation on target driving behavior data in a vehicle driving behavior library, determining the body preference crowdedness degree of the target driving behavior data, and taking the weighted average calculation as a second matching value;
sorting the driving behavior day data in the road data according to the first matching value and the second matching value;
and determining group preference driving behaviors in the road data according to the sorting result.
The principle of the technical scheme is as follows:
in this embodiment, for the crowd preferred driving behavior, when the user drives in real time, real-time road data of the user is collected, driving behavior data in the road data is identified, and then matching calculation is performed with the crowd preferred driving behavior in the vehicle driving behavior library, so as to determine the crowd preferred driving behavior in real-time driving.
In this embodiment, through the classification processing of the road data, the driving behaviors of the driver may be classified, so as to facilitate extraction of key points of different driving behaviors and determine corresponding behavior features. The key points are identification key points describing different behavior characteristics and are behavior key factors;
in this embodiment, asynchronous alignment is to align the same kind of driving behavior, but the same kind of data formed in different time periods; thus, for the same driving behavior, only one data acquisition is needed; without the need to sample multiple data because of the difference in driving behavior time.
In this embodiment, the type matching value is a matching value of the target driving behavior data and the similar data in the vehicle driving behavior library, and after weighted average calculation, the matching value can more accurately determine the crowd preference, so that the driving behaviors of the crowd preference are ranked and recommended to the driver, and the driving behaviors are not limited to the control behaviors of the vehicle and the selection behaviors of the road.
The beneficial effects of the technical scheme are that:
in the process of determining the group preference driving behaviors, the method adopts the data classification based on key points and combines an asynchronous alignment mode, so that the similar data can be obtained in the vehicle driving behavior library in the road data more quickly, and further, the single driving behavior and all the driving behaviors of the masses are subjected to weighted average calculation through the matching value, so that the group preference degree is determined, the group preference driving behaviors in the road data are further determined, and the accuracy of the data is ensured.
Preferably, the building of the group preference crowdedness model based on vehicle-road cooperation includes:
presetting a model rule canvas and a control canvas; wherein,,
the model rule canvas is used for presetting the recognition rules of group preference;
the control canvas is used for connecting the data transmission control of the vehicle-road cooperation;
forming a rule model in a visual blueprint editing mode through a model rule canvas and a control canvas, and editing and supporting the rule model through a custom parameter table and a custom variable;
generating, by the rule model, a plurality of rule terms of driving behavior preferences; wherein,,
the rule items include: issuing an instruction preference rule item, a vehicle execution rule preference rule item, a trigger condition preference rule item, a vehicle track preference rule item and a vehicle execution instruction preference rule item;
and integrating a plurality of rule items into the depth calculation model to form a group preference crowdedness degree model.
The principle of the technical scheme is as follows:
in this embodiment, the present invention may set an identification rule of a preset group preference, and. When the vehicles and roads are cooperated, the data instruction transmission control of the vehicles and roads are cooperated, the identification rules and the transmission control are used as rule items, the crowd preference degree of different driving behaviors can be judged through different rule items, and then the crowd preference degree model is formed by being integrated into a depth calculation model.
In the embodiment, the model rule canvas and the control canvas can convert the current road data into a visual blueprint editing form, and further realize the editing of the visual blueprint through a custom parameter table and a custom variable; the custom parameters represent parameters for blueprint planning; the custom variable is a data variable which is changed correspondingly according to the real-time change of the visual blueprint and the road data;
in this embodiment, the rule term and the depth calculation model may be used to manufacture the preference directivity when the depth calculation model performs the driving behavior preference calculation.
The beneficial effects of the technical scheme are that:
the method can generate the editable road rule blueprint, the road rule blueprint can realize automatic editing, and the driving behaviors of the group preference are guided to the user in an automatic editing mode, including road guiding behaviors, so that the establishment of the crowdedness degree model of the group preference is realized by fusing a depth calculation model through rule items.
Preferably, the generating the dynamic behavior preference sequence by the group preference popularity degree model through driving behavior data includes:
pre-configuring a pre-trained complex model architecture, and generating a first model architecture through a structured dynamic data fusion interface;
Configuring a dynamic training strategy in a first model framework, and setting repeated training test tasks according to the acquired road data and a vehicle driving behavior library;
based on the repeated training test task, executing a global iteration task when new driving behavior data are received;
and dynamically training the group preference crowdedness degree model through a global iteration task to generate a dynamic behavior preference sequence.
The principle of the technical scheme is as follows:
in this embodiment, the driving behavior of the group preference may be changed in real time, and in this process, by setting a test task for repeated training, the new driving behavior data is subjected to continuous iterative global calculation, so as to generate a dynamic behavior preference sequence.
In this embodiment, the complex model architecture is a nonlinear, multiple-input and multiple-output neural network model, and may be fused with structured dynamic data, that is, structured change data of different behaviors in driving behavior data of a driver, to generate a first model architecture;
in this embodiment, the dynamic training strategy is a training strategy that performs repeated dynamic changes according to road data and vehicle driving behavior data; the newly acquired data is used as an identification task to carry out test identification, and different driving behaviors are converted into a sequence mode in the preference condition of the group body weight in a global iteration mode, so that the group-preferred driving behaviors can be recommended to the driver, namely, a path is planned in real time.
The beneficial effects of the technical scheme are that:
the invention converts the neural network into the complicated neural model, so that the structured data with different behaviors can be fused, and the dynamic fusion of the structured data can be realized by combining a plurality of different data conversion and identification models because the neural network is a complicated model.
Because the invention sets the test task of repeated training in the first model framework, the iteration processing can be carried out on the newly received driving behavior data at any time when the preference identification is carried out, so that the group preference behavior is continuously updated to be identified.
Preferably, the method further comprises:
according to the vehicle driving behavior library, a crowded driving behavior sequence and an ecological driving behavior sequence are generated; wherein,,
the crowd driving behavior sequence is determined by the implementation probability of driving behaviors;
the ecological driving behavior sequence is determined by the energy consumption cost of the driving behavior;
establishing a map mapping relation between a crowded driving behavior sequence and an ecological driving behavior sequence;
and screening the mass driving behavior sequence of the road data of the current vehicle according to the map mapping relation to determine the group preference driving behavior.
The principle of the technical scheme is as follows:
In this embodiment, the present invention calculates implementation probabilities and cost energy consumption of different driving behaviors, and then screens the optimum energy consumption cost and the optimum crowded preference driving behavior by means of map mapping through the corresponding relationship between the implementation probabilities and cost energy consumption, so that the implemented driving behavior accords with group preference and also accords with the minimum cost.
In this embodiment, the crowded driving behavior sequence is used to describe the occurrence probability of different driving behaviors, and the probability of occurrence in the driving process is greater as the behavior before the sequence is shot;
in this embodiment, the ecological driving behavior sequence is used to describe different driving behaviors, or driving choices, so that the energy consumption degree is higher or lower, and the more the behavior before the sequence is taken, the less energy is consumed in the driving process.
The map mapping relationship is a relationship which is mutually connected in two sequences, namely, the mathematical relationship of the same behavior between energy consumption and occurrence probability;
and screening the weight of the occurrence probability and the weight of the energy consumption, and calculating the priority of the same driving behaviors so as to determine the driving behaviors with group preference.
The beneficial effects of the technical scheme are that:
the invention can further realize preference screening of the driving behaviors through the occurrence probability or the energy consumption cost of different driving behaviors, and simultaneously, when new driving behaviors are continuously increased, new driving behavior statistics is carried out, and corresponding group preference calculation is carried out.
In the implementation process of the invention, the crowd-preferred driving behavior is determined by screening the crowd-preferred driving behavior sequence of the road data of the current vehicle according to the map mapping relation,
step 1: according to the map mapping relation, determining the association function of the same driving behavior in the crowded driving behavior sequence and the ecological driving behavior sequence by the following formula:
F(i)＝K(fp(x i )，fp(y i ))
wherein F (i) represents an association function of the ith driving behavior; p (x) i ) A crowdedness parameter indicating that the ith driving behavior is crowdedness driving behavior x; p (y) i ) The ith driving behavior is shown as an ecological parameter of the ecological driving behavior y; f represents a characteristic parameter of the ith driving behavior; k (fp (x) i )，fp(y i ) A correlation coefficient corresponding to the crowdedness parameter and the ecology parameter corresponding to the ith driving behavior; i epsilon n, n represents the total number of driving behaviors;
in this embodiment, through the image mapping relationship, a correlation coefficient between the crowd driving behavior sequence and the ecological driving behavior can be determined, through the correlation coefficient, whether the same driving behavior has correlation in group judgment or not can be determined, and if the correlation exists, the corresponding driving behavior is indicated to be easily identified by a behavior identification model based on a behavior identification mechanism.
Step 2: according to the association function, carrying out group preference calculation on each driving behavior, and judging whether the driving behavior is the group preference behavior or not:
wherein H is i A detection amount indicating the ith driving behavior; g (F (i) represents the recognition function of the association function of the ith driving behavior in a behavior recognition model based on a behavior recognition mechanism, for determining the recognition coefficient which can be detected by each driving behavior, for determining H under the condition of having recognition errors i * G (F (i)) may determine the number of i-th driving behaviors that are not identified; when P > 1 indicates that the ith driving behavior is a group-preferred driving behavior.
In this embodiment, step 2 is a comparison between the actual value of each behavior and the total behavior, and determines whether the compared value is greater than 1; when the driving behavior is more than 1, the corresponding driving behavior is represented, and the average value of all behaviors is exceeded under the condition of recognition loss; thus, it belongs to group-preferred driving behavior.
Preferably, the performing the shortest path search based on the AI algorithm and the preset destination includes:
constructing an outsourcing rectangle by taking connecting lines of a starting destination and a destination as diagonal lines, and filtering the space of the outsourcing rectangle by an AI server to obtain road data in the outsourcing rectangle;
Carrying out topological structure surface according to the connecting line and road data, searching a plurality of adjacent path polygons which are intersected with the connecting line space, merging the adjacent path polygons to obtain an initial path polygon, obtaining two paths of a connecting starting point and a connecting end point by taking the connecting line as a boundary according to the initial path polygon, and selecting the path with smaller length as an initial shortest path;
constructing a buffer rectangle by taking the connecting line as a central line and taking half of the length of the initial shortest path as a range, and spatially filtering out a second road central line in the buffer rectangle;
and carrying out topological structure according to the first road center line and the second road center line to obtain a path to be searched, and calculating the nodes of the path to be searched step by step from the starting point according to the path to be searched to obtain the shortest path.
The principle of the technical scheme is as follows:
in this embodiment, in the process of searching the shortest path through the AI algorithm, a global rectangular area is established, all road data in the area are determined, and then the shortest path is determined by comparing the road data.
In this embodiment, the connection lines of the starting destination and the destination are diagonal lines to construct an outsourcing rectangle, that is, under the direct distance, all road data from the starting point to the destination which can be planned are determined;
In this embodiment, by constructing a topological structure plane of the connection line and the road data, that is, an overall path plane having a plurality of paths on a plane, the shortest path among all the programmable paths from the start point to the destination can be determined through the overall path plane; the initial shortest path may have multiple paths, so the invention needs to screen again to further determine the shortest path;
in this embodiment, the second shortest path planned by the second buffer rectangle is determined and calculated by means of path search.
The beneficial effects of the technical scheme are that:
in the process of determining the shortest path, the method determines all roads capable of realizing planning based on a mode of wrapping a rectangle; and in all the planned roads, the shortest path is screened twice, so that the shortest path can be accurately positioned.
Preferably, the path dynamic programming model is built when the shortest path is searched based on the group preference crowdedness model:
performing Bayesian analysis on the shortest path to obtain path risk factors, and establishing a path risk model based on the risk factors;
Based on two targets of path risk and path cost, a multi-target path dynamic planning model is established, and an objective function and constraint conditions are determined;
and integrating the objective function and the constraint condition into the population preference crowdedness model to generate a path dynamic programming model of the shortest path.
The principle of the technical scheme is as follows:
in this embodiment, risks possibly encountered in the shortest path process are determined through bayesian analysis, then a multi-objective path dynamic planning model is generated according to risk factors and path cost, a driving objective function and driving constraint conditions of the shortest path are determined, and under the condition of determining the two points, corresponding group preference driving behaviors are configured through a group preference crowdedness model.
In this embodiment, the bayesian analysis is to analyze risk factors in the shortest path, identify risk factors possibly existing in paths such as traffic congestion and uneven roads in the shortest path through a bayesian network, and further generate a path risk model, where the path risk model is based on conversion of the bayesian network, and the bayesian network adopts a bayesian network with risk identification.
In this embodiment, a multi-objective path dynamic planning model is determined by constructing constraint conditions, wherein the constraint conditions are the risk degree of path risk and the cost of path cost, so that the risk and cost of different planning paths are ordered, and the path dynamic planning model capable of identifying the shortest path is determined in a mode of integrating a group preference crowdedness degree model
The beneficial effects of the technical scheme are that:
the invention can ensure that the shortest path is ensured to be shortest and the risk is ensured to be minimum when the shortest path is planned by integrating the risk factors and the constraint conditions when the shortest path is identified.
Preferably, the outputting the optimal driving strategy in real time by establishing a path dynamic programming model includes:
establishing a path dynamic programming model, specifically dividing the driving behavior of a user into general behavior, event subdivision behavior, crowd subdivision behavior and crowd preference behavior, generating corresponding strategy templates, and storing the strategy templates in a strategy template library in a database;
constructing a strategy tree, and determining an initial strategy node and sub-strategy nodes of each level of the strategy tree; the starting policy node serves as an entry for the entire policy tree;
all the strategy nodes are mutually related through connecting lines to form a strategy branch, and different driving behavior recommendation instructions are generated by strategy combination;
the current user enters a strategy tree through the starting node, and finally the corresponding optimal driving strategy is completed through crowded preference driving behavior matching recommendation of each level.
The principle of the technical scheme is as follows:
in this embodiment, the path dynamic programming model established in the present invention may divide driving behaviors into general behaviors, event subdivision behaviors, crowd subdivision behaviors and crowd preference behaviors, which may all be recommended to a driving user, but these behaviors are not necessarily crowd preference behaviors, so that the present invention determines an optimal driving strategy through node policy matching in different levels.
In this embodiment, the path dynamic planning model may subdivide the driving behavior, and generate a corresponding policy model after subdivision, where the policy model may perform different path planning weights according to different behaviors.
In this embodiment, the policy tree is constructed based on different policy models, generates recommendation instructions corresponding to driving behaviors, and recommends different driving behaviors, thereby recommending an optimal implementation policy for the user.
Preferably, the method further comprises:
setting monitoring equipment through a road side and taking the monitoring equipment as a patrol position;
after the optimal driving strategy is determined, when a user runs in the shortest path, collecting real-time running data of the vehicle;
screening driving behavior data in the real-time driving data, and judging whether the driving behavior data accords with an optimal driving strategy or not;
and when the optimal driving strategy is not met, vehicle patrol and alarm are carried out.
The principle of the technical scheme is as follows:
in this embodiment, as shown in fig. 3, the present invention may further determine, through the path monitoring device, whether the vehicle runs strictly according to the optimal driving policy after executing the optimal driving policy, so as to prevent risk of occurrence of hardware of the vehicle and risk of undefined instruction.
In the embodiment, the road side is provided with the road side detection equipment, and the road side detection equipment comprises traffic lights, electronic signboards, electronic snapshots and other equipment on the road side, so that the near field communication with the vehicle is realized;
in this embodiment, the electronic device on the road side is used to determine whether the existing driving behavior accords with the optimal driving strategy according to the real-time driving data of the vehicle, and the electronic device on the road side is used to determine specific road conditions except for driving preference, and the optimal driving strategy is updated again through the specific robust geometry existing driving strategy, so as to achieve the aim of optimal driving.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. The ecological driving path real-time planning method for group preference under the vehicle-road cooperative environment is characterized by comprising the following steps of:
pre-building a vehicle driving behavior library; wherein,,
the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in the cloud server;
Acquiring road data of a current vehicle, matching corresponding group preference driving behaviors in a vehicle driving behavior library based on the road data, and constructing a group preference crowdedness degree model based on vehicle-road cooperation; wherein,,
the group preference crowdedness model generates a dynamic behavior preference sequence through driving behavior data;
and carrying out shortest path searching based on an AI algorithm and a preset destination, establishing a path dynamic planning model based on the group preference crowdedness degree model during the shortest path searching, and outputting an optimal driving strategy in real time by establishing the path dynamic planning model.
2. The method for planning the ecological driving path of the group preference in real time in the cooperative vehicle-road environment according to claim 1, wherein the pre-constructing the vehicle driving behavior library comprises the following steps:
presetting a driving behavior depth recognition algorithm, screening driving data captured by a cloud server, and determining driving behavior data; wherein,,
the driving behavior data includes: trajectory data, driver physiological data, and vehicle state data;
selecting behavior classification matching from a preset template library, and taking the behavior classification matching as a behavior classification template;
inputting the driving behavior data into the behavior classification template to generate a driving behavior data set;
Calculating the occurrence frequency of different driving behaviors in a driving behavior data set based on a preset driving safety standard and a driving behavior specification, and taking the occurrence frequency as a group preference value;
generating according to the group preference value, and storing a scheduling sequence; wherein,,
in the stored schedule sequence, the higher the group preference value, the more convenient the rapid data scheduling.
3. The method for planning the ecological driving path in real time according to the group preference in the vehicle-road cooperative environment as claimed in claim 1, wherein the vehicle driving behavior database updates driving behavior data in real time through a driving behavior depth recognition algorithm in a cloud server, and the method comprises the following steps:
based on a crawler network of the cloud server, vehicle driving data are obtained from different approaches;
inputting the acquired vehicle driving data into a trained depth recognition algorithm to adaptively recognize driving behaviors and acquire different driving behaviors;
inputting optimal representations of different driving behaviors to a multi-scale feature extraction unit to extract spatiotemporal features of driving behavior data;
the space-time characteristics are input into a multi-task learning unit to perform driving behavior learning and behavior prediction, and a behavior classification result and a behavior prediction result are obtained;
and carrying out fitting matching on the newly added driving data according to the classification result and the behavior prediction result, and updating the vehicle driving behavior library when the matching value accords with a preset threshold value.
4. The method for planning ecological driving paths with group preference in real time in a vehicle-road cooperative environment according to claim 3, wherein the matching of the driving behavior with the group preference in the vehicle driving behavior library based on the road data comprises the following steps:
classifying the road data to obtain a classification sequence; wherein,,
the classification process comprises the following steps: data type classification, road condition classification and driver status classification;
extracting behavior key points in each class of data of the classification sequence, and extracting behavior characteristics of each behavior key point;
asynchronous alignment is carried out on the classification sequence and similar data on a vehicle driving behavior library by utilizing a correlation algorithm;
screening target driving behavior data in the road data according to the asynchronous alignment, and generating a behavior type matching value;
carrying out weighted average calculation on target driving behavior data in a vehicle driving behavior library, determining the body preference crowdedness degree of the target driving behavior data, and taking the weighted average calculation as a second matching value;
sorting the driving behavior day data in the road data according to the first matching value and the second matching value;
and determining group preference driving behaviors in the road data according to the sorting result.
5. The method for planning the ecological driving path of the group preference in real time in the vehicle-road cooperation environment according to claim 1, wherein the constructing the group preference crowdedness model based on the vehicle-road cooperation comprises the following steps:
Presetting a model rule canvas and a control canvas; wherein,,
the model rule canvas is used for presetting the recognition rules of group preference;
the control canvas is used for connecting the data transmission control of the vehicle-road cooperation;
forming a rule model in a visual blueprint editing mode through a model rule canvas and a control canvas, and editing and supporting the rule model through a custom parameter table and a custom variable;
generating, by the rule model, a plurality of rule terms of driving behavior preferences; wherein,,
the rule items include: issuing an instruction preference rule item, a vehicle execution rule preference rule item, a trigger condition preference rule item, a vehicle track preference rule item and a vehicle execution instruction preference rule item;
and integrating a plurality of rule items into the depth calculation model to form a group preference crowdedness degree model.
6. The method for planning an ecological driving path with group preference in real time in a vehicle-road cooperative environment according to claim 1, wherein the generating a dynamic behavior preference sequence by the group preference crowdedness model through driving behavior data comprises:
pre-configuring a pre-trained complex model architecture, and generating a first model architecture through a structured dynamic data fusion interface;
Configuring a dynamic training strategy in a first model framework, and setting repeated training test tasks according to the acquired road data and a vehicle driving behavior library;
based on the repeated training test task, executing a global iteration task when new driving behavior data are received;
and dynamically training the group preference crowdedness degree model through a global iteration task to generate a dynamic behavior preference sequence.
7. The method for planning the ecological driving path in real time according to the group preference in the cooperative vehicle-road environment of claim 1, wherein the shortest path searching based on the AI algorithm and the preset destination comprises the following steps:
constructing an outsourcing rectangle by taking connecting lines of a starting destination and a destination as diagonal lines, and filtering the space of the outsourcing rectangle by an AI server to obtain road data in the outsourcing rectangle;
carrying out topological structure surface according to the connecting line and road data, searching a plurality of adjacent path polygons which are intersected with the connecting line space, merging the adjacent path polygons to obtain an initial path polygon, obtaining two paths of a connecting starting point and a connecting end point by taking the connecting line as a boundary according to the initial path polygon, and selecting the path with smaller length as an initial shortest path;
constructing a buffer rectangle by taking the connecting line as a central line and taking half of the length of the initial shortest path as a range, and spatially filtering out a second road central line in the buffer rectangle;
And carrying out topological structure according to the first road center line and the second road center line to obtain a path to be searched, and calculating the nodes of the path to be searched step by step from the starting point according to the path to be searched to obtain the shortest path.
8. The method for planning the ecological driving path with group preference in real time in the vehicle-road cooperative environment according to claim 7, wherein the step of establishing the path dynamic planning model based on the group preference crowdedness model during the shortest path search comprises the following steps:
performing Bayesian analysis on the shortest path to obtain path risk factors, and establishing a path risk model based on the risk factors;
based on two targets of path risk and path cost, a multi-target path dynamic planning model is established, and an objective function and constraint conditions are determined;
and integrating the objective function and the constraint condition into the population preference crowdedness model to generate a path dynamic programming model of the shortest path.
9. The method for planning the ecological driving path in real time according to the group preference in the vehicle-road cooperative environment of claim 7, wherein the outputting the optimal driving strategy in real time by establishing the path dynamic planning model comprises the following steps:
Establishing a path dynamic programming model, specifically dividing the driving behavior of a user into general behavior, event subdivision behavior, crowd subdivision behavior and crowd preference behavior, generating corresponding strategy templates, and storing the strategy templates in a strategy template library in a database;
constructing a strategy tree, and determining an initial strategy node and sub-strategy nodes of each level of the strategy tree; the starting policy node serves as an entry for the entire policy tree;
all the strategy nodes are mutually related through connecting lines to form a strategy branch, and different driving behavior recommendation instructions are generated by strategy combination;
the current user enters a strategy tree through the starting node, and finally the corresponding optimal driving strategy is completed through crowded preference driving behavior matching recommendation of each level.
10. The method for planning in real time an ecological driving path with group preference in a cooperative vehicle-road environment according to claim 7, wherein the method further comprises:
setting monitoring equipment through a road side and taking the monitoring equipment as a patrol position;
after the optimal driving strategy is determined, when a user runs in the shortest path, collecting real-time running data of the vehicle;
screening driving behavior data in the real-time driving data, and judging whether the driving behavior data accords with an optimal driving strategy or not;
And when the optimal driving strategy is not met, vehicle patrol and alarm are carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310502145.0A CN116592903B (en) | 2023-05-06 | 2023-05-06 | Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310502145.0A CN116592903B (en) | 2023-05-06 | 2023-05-06 | Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116592903A true CN116592903A (en) | 2023-08-15 |
| CN116592903B CN116592903B (en) | 2024-02-23 |
Family
ID=87600034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310502145.0A Active CN116592903B (en) | 2023-05-06 | 2023-05-06 | Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116592903B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117238131A (en) * | 2023-09-14 | 2023-12-15 | 中国民航大学 | Traffic flow characteristic analysis method in Internet of vehicles environment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3309640A2 (en) * | 2016-10-13 | 2018-04-18 | Baidu USA LLC | Group driving style learning framework for autonomous vehicles |
| CN108891422A (en) * | 2018-07-09 | 2018-11-27 | 深圳市易成自动驾驶技术有限公司 | Control method, device and the computer readable storage medium of intelligent vehicle |
| CN110466534A (en) * | 2018-05-10 | 2019-11-19 | 通用汽车环球科技运作有限责任公司 | For customizing the method and system of the driving behavior of autonomous vehicle |
| CN112896185A (en) * | 2021-01-25 | 2021-06-04 | 北京理工大学 | Intelligent driving behavior decision planning method and system for vehicle-road cooperation |
| CN114559959A (en) * | 2022-03-18 | 2022-05-31 | 重庆长安汽车股份有限公司 | Automatic driving behavior decision system and method based on driving preference |
| CN115384539A (en) * | 2022-08-11 | 2022-11-25 | 重庆长安汽车股份有限公司 | Method and device for generating intelligent driving strategy of vehicle, vehicle and storage medium |
-
2023
- 2023-05-06 CN CN202310502145.0A patent/CN116592903B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3309640A2 (en) * | 2016-10-13 | 2018-04-18 | Baidu USA LLC | Group driving style learning framework for autonomous vehicles |
| CN110466534A (en) * | 2018-05-10 | 2019-11-19 | 通用汽车环球科技运作有限责任公司 | For customizing the method and system of the driving behavior of autonomous vehicle |
| CN108891422A (en) * | 2018-07-09 | 2018-11-27 | 深圳市易成自动驾驶技术有限公司 | Control method, device and the computer readable storage medium of intelligent vehicle |
| CN112896185A (en) * | 2021-01-25 | 2021-06-04 | 北京理工大学 | Intelligent driving behavior decision planning method and system for vehicle-road cooperation |
| CN114559959A (en) * | 2022-03-18 | 2022-05-31 | 重庆长安汽车股份有限公司 | Automatic driving behavior decision system and method based on driving preference |
| CN115384539A (en) * | 2022-08-11 | 2022-11-25 | 重庆长安汽车股份有限公司 | Method and device for generating intelligent driving strategy of vehicle, vehicle and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 周元峰: "基于信息的驾驶员路径选择行为及动态诱导模型研究", 中国博士学位论文全文数据库 工程科技Ⅱ辑, no. 03 * |
| 张晓燕: "融合用户习惯的车辆导航路径规划策略研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, no. 12 * |
| 程 潇，王 凯，张 硕: "基于群体智能的目标搜索及路径规划机制", 软件工程与应用, vol. 11, no. 2 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117238131A (en) * | 2023-09-14 | 2023-12-15 | 中国民航大学 | Traffic flow characteristic analysis method in Internet of vehicles environment |
| CN117238131B (en) * | 2023-09-14 | 2024-05-07 | 中国民航大学 | Traffic flow characteristic analysis method in Internet of vehicles environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116592903B (en) | 2024-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109191896A (en) | Personalized parking stall recommended method and system | |
| EP3035314A1 (en) | A traffic data fusion system and the related method for providing a traffic state for a network of roads | |
| CN111582559B (en) | Arrival time estimation method and device | |
| CN106714110A (en) | Auto building method and system of Wi-Fi position fingerprint map | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| CN113704956A (en) | Urban road online microscopic simulation method and system based on digital twin technology | |
| CN116504076A (en) | Expressway traffic flow prediction method based on ETC portal data | |
| CN116592903B (en) | Ecological driving path real-time planning method for group preference under vehicle-road cooperative environment | |
| JP5070574B2 (en) | Local traffic prediction program generation device, local traffic prediction device, local traffic prediction program generation method, local traffic prediction method and program | |
| CN110021161B (en) | Traffic flow direction prediction method and system | |
| CN110738366B (en) | Outdoor blind zone behavior prediction method | |
| KR20230057558A (en) | Short-term traffic flow prediction method and device using deep learning | |
| CN110459050B (en) | Short-term bus passenger flow prediction method based on mixed decision tree | |
| Tilg et al. | Simulation-based design of urban bi-modal transport systems | |
| CN117094534B (en) | Intelligent control method and system for Internet of things | |
| Thu et al. | Multi-source data analysis for bike sharing systems | |
| KR102359902B1 (en) | Crossroads LOS Prediction Method Based on Big Data and AI, and Storage Medium Having the Same | |
| CN117114220A (en) | Vehicle scheduling algorithm based on graph neural network | |
| Wang et al. | Dynamic traffic prediction based on traffic flow mining | |
| CN113688200B (en) | Decision tree-based special population action track collection method and system | |
| Ajani et al. | Dynamic path planning approaches based on artificial intelligence and machine learning | |
| Salamanis et al. | Evaluating the effect of time series segmentation on STARIMA-based traffic prediction model | |
| CN115691140A (en) | Analysis and prediction method for space-time distribution of automobile charging demand | |
| Zhou et al. | Time related class association rule mining and its application to traffic prediction | |
| CN115565388A (en) | Traffic light control method based on multi-channel vehicle detection and three-dimensional feature labeling |
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 |