CN106940829B - Personalized path recommendation method in Internet of vehicles environment - Google Patents
Personalized path recommendation method in Internet of vehicles environment Download PDFInfo
- Publication number
- CN106940829B CN106940829B CN201710319683.0A CN201710319683A CN106940829B CN 106940829 B CN106940829 B CN 106940829B CN 201710319683 A CN201710319683 A CN 201710319683A CN 106940829 B CN106940829 B CN 106940829B
- Authority
- CN
- China
- Prior art keywords
- path
- road
- road section
- convenience
- traveler
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096833—Systems involving transmission of navigation instructions to the vehicle where different aspects are considered when computing the route
- G08G1/096838—Systems involving transmission of navigation instructions to the vehicle where different aspects are considered when computing the route where the user preferences are taken into account or the user selects one route out of a plurality
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Abstract
The invention discloses a personalized path recommendation method under the environment of Internet of vehicles, which considers 4 travel factor sets, namely time, expense, convenience and distance, simultaneously in the personalized path recommendation process; in a traffic network, a backtracking algorithm for solving all feasible path sets from a starting point to a terminal point is realized by adopting a stack idea; and the path recommendation algorithm based on the Pearson correlation coefficient is realized according to the personality preference characteristic vector of the traveler and the feasible path characteristic vector. The invention fully utilizes real-time and accurate various traffic information provided by the Internet of vehicles, combines the individual travel requirements of travelers, provides path recommendation service meeting the individual preference requirements for the travelers, and realizes the private customization of individual travel paths; the system can respond to some unexpected traffic conditions in real time and provide personalized path recommendation when the unexpected traffic conditions occur.
Description
Technical Field
The invention belongs to the technical field of transportation and information science, and relates to a personalized path recommendation method in an internet of vehicles environment.
Background
Modern transportation based on information technology and communication technology has been receiving wide attention as an indispensable activity in people's daily life. The internet of vehicles is the centralized embodiment that the internet of things is applied to an intelligent traffic system, along with the continuous popularization and promotion of the internet of vehicles technology, the collection, transmission and storage of road network traffic information are more efficient and convenient, and a large amount of traffic information related to travel is gushed into the visual field of travelers. However, in the face of such complicated and complicated traffic information, how to extract information meeting the personalized preference of the user is always a difficult problem in the science of transportation and behavior science. The emerging recommendation system provides an effective way for coping with information overload and realizing information screening. By applying the recommendation technology, the most needed personalized information can be extracted from the massive information for the user.
Personalized article recommendation services (such as news pushing, private customization of movie books and periodicals and the like) of a plurality of manufacturers (such as electric business Amazon, movie and television Netflix, music Pandora, Last. fm, domestic bean sauce, Jingdong, Tianmao and the like) effectively meet the personalized requirements that people screen out information which accords with the interest preference of the people. Similarly, in the brand-new field of transportation travel for the recommendation system, the demand of people for traffic information personalization in the travel process is strong day by day. However, no public report (3 months as of 2017) exists at present for the personalized path recommendation field, especially for the personalized path recommendation method which is feasible by the system.
The existing path planning method does not consider personalized preference of travelers, has a single traffic information source, and most products are based on static information (no real-time change) or qualitatively graded road condition information (very congested, unobstructed and the like) instead of quantized traffic information. Mainstream path planning products, such as Baidu maps (navigation), Gauss maps (navigation), and the like, also only provide different types of path planning schemes (such as shortest time, minimum distance, preferential highway, and the like) according to situations, and do not consider different travel preferences of different individual travelers.
The main disadvantages of the existing path planning technology are the following 3 points:
(1) real-time dynamically changing path traffic information is not considered. One path in the traffic network is composed of a plurality of road sections, and information such as passing time, road conditions and the like on the road sections can change along with the change of time, so that the state information of the path presents nonlinear and random complex characteristics. That is, a path in the road network is a dynamic multi-attribute entity whose attributes change over time; the existing path planning system does not consider the real-time dynamically-changed path traffic information due to the limitation of the traditional traffic information acquisition, transmission, storage, calculation and other technologies.
(2) The quantitative influence of adverse weather factors (such as fog, rain, snow, haze and the like) on the path state and the individual travelers is not considered, and the influence of events such as road construction, large-scale events and the like on the traffic trip is also considered insufficiently. The influence degree of the variable weather factors on the travel of different roads is different; similarly, various types and properties of road construction, sports events and the like can also affect the road traffic conditions in local areas. While existing path planning techniques do not quantitatively take into account the actual impact of these factors on road capacity.
(3) Personalized preference difference among individuals on different trips is not considered. The traffic network involves a variety of factors, and the actual individual travelers are more in demand. The preference factors considered by different travelers are different, and even if the preference factors considered are approximately the same, the preference degrees are necessarily different. It must also be recognized that the level of preference of an individual traveler is a subjective, ambiguous, and difficult to measure quantitatively. Therefore, the traditional path planning technology cannot realize quantitative collection and analysis of preference factors of individual travelers.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides a personalized path recommendation method in an internet of vehicles environment, and is a whole set of solution for extracting dynamic personalized path information which accords with the preference of a user by applying a recommendation technology in traffic information provided by the internet of vehicles environment in order to meet the requirement of the user on personalized travel. In the invention, a road network model is established, a starting point and an end point are given, all feasible paths are searched in the road network, and the path most similar to the user preference is found by utilizing a recommendation algorithm according to the travel preference of a traveler.
In the personalized path recommendation process, 4 travel factor sets, namely { time, cost, convenience and distance } are considered at the same time; in a traffic network, a backtracking algorithm for solving all feasible path sets from a starting point to a terminal point is realized by adopting a stack idea; and the path recommendation algorithm based on the Pearson correlation coefficient is realized according to the personality preference characteristic vector of the traveler and the feasible path characteristic vector.
The specific technical scheme is as follows:
a personalized path recommendation method in a car networking environment comprises the following steps:
step 1: under the environment of the Internet of vehicles, a road network model conforming to personalized travel is established;
step 2: finding all feasible paths in the road network according to the starting point and the end point;
and step 3: obtaining a path feature vector of a feasible path;
and 4, step 4: obtaining a preference feature vector of the traveler according to the preference of the traveler;
and 5: sorting the feasible paths by using a recommendation algorithm;
step 6: and outputting the path ranked to the top as a personalized recommended path.
Further, the step 1 comprises:
step 1.1: definition G (V, a) represents a physical traffic network composed of a finite number of nodes, V representing a set of nodes, and a representing a set of directed road segments. a represents a directed road section in the road network, and a belongs to A; r represents a set of source points, ands represents a set of destination points, anR represents a source point and R ∈ R; s represents a destination point, and S belongs to S; krsRepresents the set of all feasible paths between OD and rs; k represents a feasible path between OD and rs,
step 1.2: and the path length attribute, wherein the path length can be obtained by superposing the physical distances of all road segments through which the actual path passes.
The path length attribute is defined as: physical travel distance l of k path between OD and rsrs,kCan be calculated from equation (1):
wherein laWhich represents the physical length of the road segment a,is a variable related to the link a and the path k between OD and rs, i.e. a variable of 0-1, if the link a is on the kth path connecting OD and rsOtherwise
Step 1.3: convenience is also an important attribute of a road section, main factors influencing convenience of the road section are road grade, weather conditions, traffic accident conditions and road control conditions, and path convenience is obtained by superposing convenience of all road sections forming the path.
Step 1.4: and improving the BPR function for calculating the road section passing time. Further, the route transit time may be obtained by superimposing transit times of all links constituting the route.
The BPR function is defined as: the BPR function is the federal highway administration function in the united states and is used for road section free-run time calculation.
The road section passing time is defined as: the actual travel time on the road segment can be calculated by equation (4):
wherein the content of the first and second substances,the time required for the vehicle to travel is zero flow impedance, namely the flow on the road section a is zero; x is the number ofaIs the traffic flow for road segment a; qa' is the actual traffic capacity of the section a; alpha is alpha1,α2The coefficient of retardation.
The actual capacity is defined as: the traffic capacity after being influenced by weather, traffic accidents and road control on the road section can be calculated by the following formula (5):
wherein Q isaIs the theoretical capacity of the road section a.
The roadThe diameter transit time is defined as: travel time of the kth path between the h time OD and the rsCan be represented by formula (6):
step 1.5: and the path cost attribute, wherein the path cost can be superposed by the actual charge of each road section through which the actual path passes.
The path cost attribute is defined as: travel cost e of OD to kth path between rsrs,kIt can be calculated by equation (7):
wherein e isaIndicating congestion charging on the road segment a.
Still further, in step 1.3, the road grade is defined as: the urban road grade is divided into four grades of an express way, a main road, a secondary road and a branch;
aiming at the convenience of the road section, setting a first-level road section to be 1, a second-level road section to be 2, a third-level road section to be 3 and a fourth-level road section to be 4;
the weather conditions are defined as: degree of influence of weather conditions on road segment a;
the traffic accident situation is defined as: the degree of influence of the traffic accident situation on the road section a;
the road regulation condition is defined as: the degree of influence of the road control condition on the road section a;
the convenience of the road section is defined as: convenience b of road section aaCalculated from equation (2):
wherein, gaIs the road grade for the road segment a,the influence degree of the weather condition on the road section a is expressed, such as fog, rainy days, snowy days and the like;representing the degree of influence of the traffic accident situation on the road section a;expressed as the degree of influence of the road control situation on the road section a, such as repairing roads, repairing subways, etc., and
the path convenience is defined as: convenience of OD to kth Path between rs brs,kCalculated from equation (3):
further, the step 2 comprises:
step 2.1: input starting point vrAnd endpoint vs;
Step 2.2: v is to bejThe adjacent nodes (j belongs to {1, 2, 3,.., n }) are sequentially stored into the array from small to large according to the numbering sequencePerforming the following steps; initializing an array of pointersNewly building a Stack; m is 1;
step 2.3: push (v)r);
Step 2.4: if Stack is not empty, vkPeek (); otherwise, turning to step 2.8;
step 2.6: if peek () > vsTurning to step 2.4;
2.7: printing all elements in the Stack in reverse order (from the bottom to the top of the Stack) and recording to Krs,mM +1, pop (), go to step 2.4;
2.8: generating a new set Krs={Krs,m|m∈N*};
2.9: the algorithm ends.
The peek () is defined as: the top-of-stack element is read and its value returned.
The pop () is defined as: and deleting the stack top element.
Further, the step 3 comprises:
Step 3.2: in order to make the path recommendation have a unified and feasible standard, the data in the path feature vector needs to be normalized, as shown in equation (8):
wherein x ismaxIs the maximum value of each attribute in the quadruple of the alternative path.
Step 3.3: path feature vectorAfter normalization processing, converting the normalized path feature vector into a normalized path feature vector
Further, the step 4 comprises:
step 4.1: different travelers have different preference degrees for travel time, cost, convenience and path distance. Thus, a path preference feature vector for actor i may be definedAnd set 0. ltoreq. ulIs less than or equal to 1, and
further, the step 5 comprises:
step 5.1: the Pearson correlation coefficient is used to measure the degree of correlation between the preference feature vector and the path feature vector. Calculating K from Pearson correlation coefficientrsAnd recommending the path with the highest similarity to the actor i according to the similarity of each feasible path and the assigned actor i. The similarity between the traveler i and the alternative path k at the h-th momentCan be calculated from equation (9):
further, the step 6 includes:
step 6.1:
aiming at the traveler i, the path with the maximum similarity at the h momentRepresents a recommended path for the traveler i that best meets the traveler's preference, and
compared with the prior art, the invention has the beneficial effects that:
1. the real-time and accurate various kinds of traffic information provided by the Internet of vehicles are fully utilized, and the personalized travel requirements of travelers are combined, so that the path recommendation service meeting the individual preference requirements is provided for the travelers, and the individual travel paths are customized;
2. the system can respond to some unexpected traffic conditions in real time, such as traffic accidents, road construction maintenance, large-scale events and the like, and provides personalized path recommendation when the conditions occur;
3. from the perspective of the whole traffic network, due to the fact that the preferences of travelers are different, personalized path recommendation can effectively achieve distribution of traffic flow according to individual preferences, and objectively avoid aggregation of all travelers on individual paths, so that traffic flow distribution of the whole traffic network is more reasonable and balanced.
Drawings
FIG. 1 is a flow diagram of a method for personalized path recommendation in a vehicle networking environment;
fig. 2 is a simulated road network.
Detailed Description
The technical solution of the present invention will be described in further detail with reference to the specific drawings and examples.
Referring to fig. 1, a personalized path recommendation method in an internet of vehicles environment includes the following steps:
step 1: under the environment of the Internet of vehicles, a road network model conforming to personalized travel is established;
step 2: finding all feasible paths in the road network according to the starting point and the end point;
and step 3: obtaining a path feature vector of a feasible path;
and 4, step 4: obtaining a preference feature vector of the traveler according to the preference of the traveler;
and 5: sorting the feasible paths by using a recommendation algorithm;
step 6: and outputting the path ranked to the top as a personalized recommended path.
Further, the step 1 comprises:
step 1.1: definition G (V, a) represents a physical traffic network composed of a finite number of nodes, V representing a set of nodes, and a representing a set of directed road segments. a represents a directed road section in the road network, and a belongs to A; r represents a set of source points, ands represents a set of destination points, anR represents a source point and R ∈ R; s represents a destination point, and S belongs to S; krsRepresents the set of all feasible paths between OD and rs; k represents a feasible path between OD and rs,
step 1.2: and the path length attribute, wherein the path length can be obtained by superposing the physical distances of all road segments through which the actual path passes.
The path length attribute is defined as: physical travel distance l of k path between OD and rsrs,kCan be calculated from equation (1):
wherein laWhich represents the physical length of the road segment a,is a variable related to the link a and the path k between OD and rs, i.e. a variable of 0-1, if the link a is on the kth path connecting OD and rsOtherwise
Step 1.3: convenience is also an important attribute of a road section, and the main factors influencing the convenience of the road section are road grade, weather conditions, traffic accident conditions and road control conditions. Further, the convenience of the route can be obtained by superposing the convenience of all the road segments constituting the route.
The road grade is defined as: the grade of the urban road is divided into four grades of an express way, a main road, a secondary road and a branch. According to the relevant regulations of the national temporary regulations of urban planning quota indexes, roads can be divided into four levels, as shown in table 1:
TABLE 1 road grade demarcation Table
For convenience of quantitative research, aiming at convenience of road sections, the first-level road section is set to be 1, the second-level road section is set to be 2, the third-level road section is set to be 3, and the fourth-level road section is set to be 4.
The weather conditions are defined as: the degree of influence of the weather conditions on the road section a, such as fog, rainy days, snowy days, etc.
The traffic accident situation is defined as: the degree of influence of the traffic accident situation on the road section a.
The road regulation condition is defined as: the influence degree of the road control condition on the road section a, such as road repair, subway repair and the like.
The convenience of the road section is defined as: convenience b of road section aaCalculated from equation (2):
wherein, gaIs the road grade for the road segment a,the influence degree of the weather condition on the road section a is expressed, such as fog, rainy days, snowy days and the like;representing the degree of influence of the traffic accident situation on the road section a;expressed as the degree of influence of the road control situation on the road section a, such as repairing roads, repairing subways, etc., and
the path convenience is defined as: convenience of OD to kth Path between rs brs,kCan be calculated from equation (3):
step 1.4: and improving the BPR function for calculating the road section passing time. Further, the route transit time may be obtained by superimposing transit times of all links constituting the route.
The BPR function is defined as: the BPR function is the federal highway administration function in the united states and is used for road section free-run time calculation.
The road section passing time is defined as: the actual travel time on the road segment can be calculated by equation (4):
wherein the content of the first and second substances,the time required for the vehicle to travel is zero flow impedance, namely the flow on the road section a is zero; x is the number ofaIs the traffic flow for road segment a; qa' is the actual traffic capacity of the section a; alpha is alpha1,α2The coefficient of retardation.
The actual capacity is defined as: the traffic capacity after being influenced by weather, traffic accidents and road control on the road section can be calculated by the following formula (5):
wherein Q isaIs the theoretical capacity of the road section a.
The path transit time is defined as: travel time of the kth path between the h time OD and the rsCan be represented by formula (6):
step 1.5: and the path cost attribute, wherein the path cost can be superposed by the actual charge of each road section through which the actual path passes.
The path cost attribute is defined as: travel cost e of OD to kth path between rsrs,kIt can be calculated by equation (7):
the step 2 comprises the following steps:
step 2.1: input starting point vrAnd endpoint vs;
Step 2.2: v is to bejThe adjacent nodes (j belongs to {1, 2, 3,.., n }) are sequentially stored into the array from small to large according to the numbering sequencePerforming the following steps; initializing an array of pointersNewly building a Stack; m is 1;
step 2.3: push (v)r);
Step 2.4: if Stack is not empty, vkPeek (); otherwise, turning to step 2.8;
step 2.6: if peek () > vsTurning to step 2.4;
2.7: printing all elements in the Stack in reverse order (from the bottom to the top of the Stack) and recording to Krs,mM +1, pop (), go to step 2.4;
2.8: generating a new set Krs={Krs,m|m∈N*};
2.9: the algorithm ends.
The peek () is defined as: the top-of-stack element is read and its value returned.
The pop () is defined as: and deleting the stack top element.
The step 3 comprises the following steps:
Step 3.2: in order to make the path recommendation have a unified and feasible standard, the data in the path feature vector needs to be normalized, as shown in equation (8):
wherein x ismaxIs the maximum value of each attribute in the quadruple of the alternative path.
Step 3.3: path feature vectorAfter normalization processing, converting the normalized path feature vector into a normalized path feature vector
The step 4 comprises the following steps:
step 4.1: different travelers have different preference degrees for travel time, cost, convenience and path distance. Thus, a path preference feature vector for actor i may be definedAnd set 0. ltoreq. ulIs less than or equal to 1, and
the step 5 comprises the following steps:
step 5.1: the Pearson correlation coefficient is used to measure the degree of correlation between the preference feature vector and the path feature vector. Calculating K from Pearson correlation coefficientrsAnd recommending the path with the highest similarity to the actor i according to the similarity of each feasible path and the assigned actor i. The similarity between the traveler i and the alternative path k at the h-th momentCan be calculated from equation (9):
the step 6 comprises the following steps:
step 6.1:
aiming at the traveler i, the path with the maximum similarity at the h momentRepresents a recommended path for the traveler i that best meets the traveler's preference, and
examples
First, a real-time traffic network model, such as a simulated traffic network as shown in fig. 2, is established. The number sequence on the side of the figure is the road section number, length (unit: km), traffic capacity (unit: veh/h), road section cost (unit: yuan), road section grade.
Next, all the feasible routes are calculated from the starting point number 1 to the end point number 9 according to the travel purpose of the traveler. The path base attribute is obtained from the link attributes on the feasible path, as shown in table 2.
TABLE 2 Path basis Attribute
The path attribute data is normalized to obtain a path feature vector, and the path features are analyzed according to the path feature vector, as shown in table 3.
TABLE 3 Path feature vectors
And finally, calculating Pearson similarity coefficients between all feasible paths and the travelers with different preferences, and sequencing the Pearson similarity coefficients, wherein the larger the Pearson similarity coefficient, the more the Pearson similarity coefficient accords with the preference of the travelers, so that a recommended path result is obtained, as shown in Table 4.
TABLE 4 Pearson similarity coefficient between traveler and path
For traveler 1, travel prefers time and convenience, so route 1 is most similar; for the traveler 2, the travel preference is mainly paid, and the travel preference is also better for time and convenience, so the route 2 is most similar to the travel preference; for the traveler 3, the travel preference is focused on convenience and is more preferred to time, so the path 1 is most similar to the travel preference; for the traveler 4, the trip prefers to be longer, more preferred for time and cost, so the route 4 is most similar to it; for the traveler 5, the travel prefers to be more time-critical, and also prefers to be more cost-effective and convenient, so the route 2 is most similar to the travel; for the traveler 6, the travel preference is focused on convenience, and also on time and cost, so the route 1 is most similar to it.
The above description is only a preferred embodiment of the present invention, and the scope of the present invention is not limited thereto, and any simple modifications or equivalent substitutions of the technical solutions that can be obviously obtained by those skilled in the art within the technical scope of the present invention are within the scope of the present invention.
Claims (1)
1. A personalized path recommendation method in a car networking environment is characterized by comprising the following steps:
step 1: under the environment of the Internet of vehicles, a road network model conforming to personalized travel is established;
the step 1 comprises the following steps:
step 1.1: defining G (V, A) to represent a physical traffic network consisting of a limited number of nodes, wherein V represents a node set, and A represents a directed road section set; a represents a directed road section in the road network, and a belongs to A; r represents a set of source points, ands represents a set of destination points, anR represents a source point and R ∈ R; s represents a destination point, and S belongs to S; krsRepresents the set of all feasible paths between OD and rs; k represents a feasible path between OD and rs,
step 1.2: the path length attribute is obtained by superposing the physical distances of all road sections through which the actual path passes;
the path length attribute is defined as: physical travel distance l of k path between OD and rsrs,kCalculated from equation (1):
wherein laWhich represents the physical length of the road segment a,is a variable related to the link a and the path k between OD and rs, i.e. a variable of 0-1, if the link a is on the kth path connecting OD and rsOtherwise
Step 1.3: the convenience is also an important attribute of the road section, the main factors influencing the convenience of the road section are road grade, weather condition, traffic accident condition and road control condition, and the convenience of the path is obtained by superposing the convenience of all the road sections forming the path;
the road grade is defined as: the urban road grade is divided into four grades of an express way, a main road, a secondary road and a branch;
aiming at the convenience of the road section, setting a first-level road section to be 1, a second-level road section to be 2, a third-level road section to be 3 and a fourth-level road section to be 4;
the weather conditions are defined as: degree of influence of weather conditions on road segment a;
the traffic accident situation is defined as: the degree of influence of the traffic accident situation on the road section a;
the road regulation condition is defined as: the degree of influence of the road control condition on the road section a;
the convenience of the road section is defined as: convenience b of road section aaCalculated from equation (2):
wherein, gaIs the road grade for the road segment a,representing the influence degree of the weather condition on the road section a;representing the degree of influence of the traffic accident situation on the road section a;is expressed as the degree of influence of the road control situation on the road section a, and
the path convenience is defined as: convenience of OD to kth Path between rs brs,kCalculated from equation (3):
step 1.4: improving the BPR function and calculating the road section passing time; further, the path passing time is obtained by superposing the passing time of all the road sections forming the path;
the BPR function is defined as: the BPR function is a function of the U.S. Federal road administration and is used for calculating free running time of a road section;
the road section passing time is defined as: the actual travel time on the road segment is calculated by equation (4):
wherein the content of the first and second substances,the time required for the vehicle to travel is zero flow impedance, namely the flow on the road section a is zero; x is the number ofaIs the traffic flow for road segment a; qa' is the actual traffic capacity of the section a; alpha is alpha1,α2Is the retardation coefficient;
the actual capacity is defined as: the traffic capacity after being influenced by weather, traffic accidents and road control on the road section is calculated by the formula (5):
wherein Q isaIs the theoretical traffic capacity of the section a;
passage of said pathThe time is defined as: travel time of the kth path between the h time OD and the rsRepresented by formula (6):
step 1.5: the route cost attribute, the route cost can be obtained by superposing the actual charges of all road sections through which the actual route passes;
the path cost attribute is defined as: travel cost e of OD to kth path between rsrs,kCalculated from equation (7):
wherein e isaIndicating congestion charging on road segment a;
step 2: finding all feasible paths in the road network according to the starting point and the end point;
the step 2 comprises the following steps:
step 2.1: input starting point vrAnd endpoint vs;
Step 2.2: v is to bej(j is belonged to {1, 2, 3, K, n }) adjacent nodes are sequentially stored in an array from small to large according to the numbering sequencePerforming the following steps; initializing an array of pointersNewly building a Stack; m is 1;
step 2.3: push (v)r);
Step 2.4: if Stack is not empty, vkPeek (); otherwise, turning to step 2.8;
step 2.6: if peek () > vsTurning to step 2.4;
step 2.7: printing all elements in Stack in reverse order, and recording to Krs,mM +1, pop (), go to step 2.4;
step 2.8: generating a new set Krs={Krs,m|m∈N*};
Step 2.9: finishing the algorithm;
The peek () is defined as: reading the top element of the stack and returning the value of the top element of the stack;
the pop () is defined as: deleting the elements at the top of the stack;
and step 3: obtaining a path feature vector of a feasible path;
the step 3 comprises the following steps:
step 3.2: in order to make the path recommendation have a unified and feasible standard, the data in the path feature vector needs to be normalized, as shown in equation (8):
wherein x ismaxThe maximum value of each attribute in the quadruple of the alternative path is obtained;
step 3.3: path feature vectorAfter normalization processing, converting the normalized path feature vector into a normalized path feature vector
And 4, step 4: obtaining a preference feature vector of the traveler according to the preference of the traveler;
the step 4 comprises the following steps:
step 4.1: different travelers have different preference degrees on travel time, cost, convenience and path distance; defining path preference feature vector for traveler iAnd set 0. ltoreq. uiIs less than or equal to 1, and
and 5: sorting the feasible paths by using a recommendation algorithm;
the step 5 comprises the following steps:
step 5.1: using Pearson correlation coefficient to measure the degree of correlation between the preference feature vector and the path feature vector; calculating K from Pearson correlation coefficientrsThe similarity degree of each feasible path and the designated traveler i is obtained, and then the path with the highest similarity degree is recommended to the traveler i; the similarity between the traveler i and the alternative path k at the h-th momentCalculated from equation (9):
step 6: outputting the path ranked to the top as the personalized recommended path
The step 6 comprises the following steps: step 6.1:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710319683.0A CN106940829B (en) | 2017-04-28 | 2017-04-28 | Personalized path recommendation method in Internet of vehicles environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710319683.0A CN106940829B (en) | 2017-04-28 | 2017-04-28 | Personalized path recommendation method in Internet of vehicles environment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106940829A CN106940829A (en) | 2017-07-11 |
CN106940829B true CN106940829B (en) | 2021-06-18 |
Family
ID=59464388
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710319683.0A Expired - Fee Related CN106940829B (en) | 2017-04-28 | 2017-04-28 | Personalized path recommendation method in Internet of vehicles environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106940829B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11846519B2 (en) | 2018-07-23 | 2023-12-19 | Tencent Technology (Shenzhen) Company Limited | Travel time determining method and apparatus, computer device, and storage medium |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107146416B (en) * | 2017-07-18 | 2020-01-21 | 深圳市锦粤达科技有限公司 | Intelligent traffic management system |
CN107368931B (en) * | 2017-08-09 | 2020-10-09 | 西南交通大学 | Logistics distribution path dynamic planning method and system based on big data analysis technology |
CN108663062B (en) * | 2018-03-27 | 2022-07-22 | 斑马网络技术有限公司 | Path planning method and system |
CN110415546A (en) * | 2018-04-26 | 2019-11-05 | 中移(苏州)软件技术有限公司 | It parks abductive approach, device and medium |
CN110569450B (en) * | 2018-05-18 | 2024-03-26 | 北京搜狗科技发展有限公司 | Path recommendation method and device |
CN110930747B (en) * | 2018-09-20 | 2021-11-19 | 上海丰豹商务咨询有限公司 | Intelligent internet traffic service system based on cloud computing technology |
CN109272169A (en) * | 2018-10-10 | 2019-01-25 | 深圳市赛为智能股份有限公司 | Traffic flow forecasting method, device, computer equipment and storage medium |
CN109598367B (en) * | 2018-11-14 | 2023-05-09 | 创新先进技术有限公司 | Multipath processing method and device |
CN109724616B (en) * | 2019-02-28 | 2021-02-05 | 广州大学 | Electric vehicle navigation method, system, readable storage medium and terminal equipment |
CN110095134B (en) * | 2019-05-16 | 2023-02-10 | 李培嵩 | Method and system for planning and navigating path by taking preference of user as core |
CN110647693B (en) * | 2019-09-23 | 2021-03-05 | 北京京东智能城市大数据研究院 | Path recommendation method and device |
CN110992724B (en) * | 2019-11-05 | 2021-11-19 | 华为技术有限公司 | Vehicle navigation method and terminal |
CN110853375B (en) * | 2019-11-21 | 2020-12-01 | 东南大学 | Random user balanced day-by-day dynamic traffic flow prediction method considering overlapping paths |
CN112991740B (en) * | 2021-03-23 | 2022-03-11 | 苏州蓝图智慧城市科技有限公司 | Vehicle guiding method, medium and equipment based on travel dispersion and clustering |
CN113380064B (en) * | 2021-05-21 | 2022-07-05 | 徐州工程学院 | Efficient highway passing system and method |
CN113032444A (en) * | 2021-05-24 | 2021-06-25 | 车巴达(苏州)网络科技有限公司 | Travel planning method and device, electronic equipment and medium |
CN113538920B (en) * | 2021-09-14 | 2021-12-10 | 腾讯科技(深圳)有限公司 | Method and device for determining target driving route, electronic equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008180577A (en) * | 2007-01-24 | 2008-08-07 | Xanavi Informatics Corp | Navigation device |
CN102798394A (en) * | 2011-05-26 | 2012-11-28 | 上海博泰悦臻电子设备制造有限公司 | Route planning method and system |
CN102916929A (en) * | 2011-08-01 | 2013-02-06 | 杭州信雅达数码科技有限公司 | Trust evaluating method based on fuzzy Petri net |
CN104331743A (en) * | 2014-10-11 | 2015-02-04 | 清华大学 | Electric vehicle travel planning method based on multi-target optimization |
CN104899195A (en) * | 2014-01-26 | 2015-09-09 | 武汉联影医疗科技有限公司 | Customized educational resource recommending method and apparatus |
CN105547306A (en) * | 2015-08-11 | 2016-05-04 | 深圳大学 | Route pushing method and system thereof |
CN106096756A (en) * | 2016-05-31 | 2016-11-09 | 武汉大学 | A kind of urban road network dynamic realtime Multiple Intersections routing resource |
-
2017
- 2017-04-28 CN CN201710319683.0A patent/CN106940829B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008180577A (en) * | 2007-01-24 | 2008-08-07 | Xanavi Informatics Corp | Navigation device |
CN102798394A (en) * | 2011-05-26 | 2012-11-28 | 上海博泰悦臻电子设备制造有限公司 | Route planning method and system |
CN102916929A (en) * | 2011-08-01 | 2013-02-06 | 杭州信雅达数码科技有限公司 | Trust evaluating method based on fuzzy Petri net |
CN104899195A (en) * | 2014-01-26 | 2015-09-09 | 武汉联影医疗科技有限公司 | Customized educational resource recommending method and apparatus |
CN104331743A (en) * | 2014-10-11 | 2015-02-04 | 清华大学 | Electric vehicle travel planning method based on multi-target optimization |
CN105547306A (en) * | 2015-08-11 | 2016-05-04 | 深圳大学 | Route pushing method and system thereof |
CN106096756A (en) * | 2016-05-31 | 2016-11-09 | 武汉大学 | A kind of urban road network dynamic realtime Multiple Intersections routing resource |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11846519B2 (en) | 2018-07-23 | 2023-12-19 | Tencent Technology (Shenzhen) Company Limited | Travel time determining method and apparatus, computer device, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106940829A (en) | 2017-07-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106940829B (en) | Personalized path recommendation method in Internet of vehicles environment | |
Haustein et al. | European mobility cultures: A survey-based cluster analysis across 28 European countries | |
Faghih-Imani et al. | Analysing bicycle-sharing system user destination choice preferences: Chicago’s Divvy system | |
Klobucar et al. | Network evaluation tool to improve real and perceived bicycle safety | |
CN107024217B (en) | The method, apparatus and system of the route planning of Intercity Transportation | |
CN107134142A (en) | A kind of urban road method for predicting based on multisource data fusion | |
CN106203725A (en) | Door-to-door trip route scheme personalized recommendation method based on heuristic search | |
Murphy et al. | Implementing low-stress bicycle routing in national accessibility evaluation | |
CN110413886A (en) | A kind of point of interest methods of exhibiting and device | |
Turner et al. | Bicycle and pedestrian travel demand forecasting: Literature review | |
Winters et al. | The Canadian Bikeway Comfort and Safety metrics (Can-BICS): National measures of the bicycling environment for use in research and policy | |
CN108090220A (en) | Point of interest search sort method and system | |
Huber et al. | Disaggregation of aggregate GPS-based cycling data–How to enrich commercial cycling data sets for detailed cycling behaviour analysis | |
CN107918895A (en) | One kind is called a taxi order matching process and device | |
CN113095570B (en) | Bicycle riding path recommending method based on demand difference | |
Masli et al. | Geowiki+ route analysis= improved transportation planning | |
CN114595962A (en) | Method for evaluating influence of university campus entrance on urban road traffic based on cloud matter element model | |
Celiński et al. | Sustainable development of the transport system through rationalization of transport tasks using a specialised travel planner | |
Chen et al. | Classification criteria and application of level of service for bicycle lanes in China | |
Fields | The correlates of congestion: Investigating the links between congestion and urban area characteristics | |
Cui | Influencing factors of public participation willingness in shared bicycles and intervention strategies | |
CN114896523B (en) | Road planning method and device based on country tourism line | |
Shan et al. | A novel personalized dynamic route recommendation approach based on pearson similarity coefficient in cooperative vehicle-infrastructure systems | |
Naumov et al. | Using Clustering Algorithms to Identify Recreational Trips within a Bike-Sharing System | |
Li | Cycling in Toronto: route choice behavior and implications to infrastructure planning |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210618 |
|
CF01 | Termination of patent right due to non-payment of annual fee |