CN117029847A - Personalized pedestrian travel path planning method based on crowd sensing - Google Patents
Personalized pedestrian travel path planning method based on crowd sensing Download PDFInfo
- Publication number
- CN117029847A CN117029847A CN202310792725.8A CN202310792725A CN117029847A CN 117029847 A CN117029847 A CN 117029847A CN 202310792725 A CN202310792725 A CN 202310792725A CN 117029847 A CN117029847 A CN 117029847A
- Authority
- CN
- China
- Prior art keywords
- user
- personalized
- task
- factors
- information
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000008447 perception Effects 0.000 claims abstract description 31
- 230000008901 benefit Effects 0.000 claims description 15
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical group [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 13
- 229910052799 carbon Inorganic materials 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 12
- 230000002776 aggregation Effects 0.000 claims description 9
- 238000004220 aggregation Methods 0.000 claims description 9
- 238000005286 illumination Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 3
- 238000013441 quality evaluation Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009131 signaling function Effects 0.000 claims description 3
- 238000001303 quality assessment method Methods 0.000 claims description 2
- 238000012790 confirmation Methods 0.000 claims 1
- 230000003746 surface roughness Effects 0.000 claims 1
- 230000006872 improvement Effects 0.000 abstract description 2
- 230000007246 mechanism Effects 0.000 abstract description 2
- 238000012360 testing method Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000011835 investigation Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000009435 building construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005251 gamma ray Effects 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000002352 surface water Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012795 verification 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
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3484—Personalized, e.g. from learned user behaviour or user-defined profiles
-
- 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
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a personalized pedestrian travel path planning method based on crowd sensing, which comprises the following steps: the personalized factors of the pedestrian walking travel demands are determined, and the accessibility and comfort are covered; the user responds to the personalized trip information acquisition task issued by the system; the system aggregates the user to upload information and load the information to the walking road network, and updates the user reputation level; designing an incentive mechanism based on contract theory, providing rewards for users with different credit grades, and encouraging the users to respond to system tasks actively; based on accessibility and comfort requirements of different crowd selection, individual travel impedance of the walking road network is calculated, and travel paths are optimized by using a constraint-containing path planning method. The method can fully utilize the information perception capability of pedestrian groups, provide diversified walking road condition information for the existing map navigation system, realize the improvement and updating of road network information, and plan a walking path scheme meeting the personalized requirements of pedestrians.
Description
Technical Field
The invention belongs to the technical field of pedestrian path planning and navigation, and particularly relates to a personalized pedestrian travel path planning method based on crowd sensing.
Background
Providing travel path planning for pedestrians in urban public spaces has become a fundamental business for navigation software developers. However, the existing navigation system has a limited walking path planning function due to a single walking road network information category. The following three points are specifically shown: firstly, the system lacks consideration about factors such as terrain height difference, urban facility configuration, weather influence and the like, so that pedestrians are difficult to grasp diversified road condition information; secondly, the intelligent degree of obtaining the walking road condition information is insufficient, so that the problems of too few users uploading effective information, untimely information feedback and the like are caused; thirdly, the system only carries out walking path planning according to the efficiency principles of shortest distance, shortest duration, least cost and the like, and the influence of the diversified factors and the travel demands of different people are not comprehensively considered.
The crowd sensing technology is used as a representation form of the Internet of things, an interactive and participatory sensing network is formed through existing mobile equipment, and sensing tasks are issued to user individuals in the network to complete information acquisition. Therefore, the crowd sensing technology is suitable for diversified information acquisition required by walking navigation function expansion based on the existing huge user population of the mobile equipment. However, how to guarantee the scale and data quality of participants in road condition acquisition tasks is a fundamental problem of the crowd sensing technology in the field. Meanwhile, the collected road condition information is generally heterogeneous, and how to reasonably quantify the influence level of various factors and provide a travel path matching with the user needs in walking navigation is also a key problem to be solved in the field.
Disclosure of Invention
The invention provides a personalized pedestrian travel path planning method based on crowd sensing, which is combined with the existing navigation software, so that large-scale and diversified walking road network information can be effectively obtained, and a personalized walking path planning scheme is provided for pedestrians.
In order to achieve the above object, the present invention provides the following solutions:
a personalized pedestrian travel path planning method based on crowd sensing comprises the following steps:
step 1: determining personalized factors of pedestrian walking demands;
step 2: the system issues a perception task of walking road network information, and a user responds to the perception task;
the method comprises the following steps: based on the information data quality evaluation model, performing sensing data aggregation and user credit value updating, and loading the aggregated sensing data to a road network through a system for subsequent personalized impedance calculation;
step 4: establishing a contract incentive model based on a contract theory, and determining the basic profit form and amount of the user to motivate the user to respond to the perception task and ensure the quality of the uploaded information;
step 5: and according to the road condition information and the personalized factors selected by the pedestrians, weighting each factor, calculating personalized impedance of the walking road network, and finally providing an optimal travel path matched with the pedestrian demands by using a constraint path-containing planning algorithm.
Preferably, the individualization factors include reachability factors and comfort factors. The accessibility factors determine whether a user with inconvenient travel can reach a destination, including the traffic situation of a sidewalk, the number of steps, the occupied road surface of construction and the configuration of barrier-free facilities; the travelling experience and the comfort degree of the user are influenced by the comfort factors, and the travelling experience and the comfort degree comprise boulevard arrangement, crowd density, non-motor vehicle interference, crosswalk length and quantity, road surface ponding, gradient, road surface flatness and illumination conditions.
Preferably, the related content of the perception task includes: task type, task area, task rewards, task start-stop time and task description. The mode of the system for releasing the perception task is divided into event driving and time driving, wherein the event driving type task mainly triggers the information acquisition task of related factors through active feedback of a user or other ways and gives the specific condition of information to the user; and the time-driven task is periodically issued with a factor state update task by the system according to the characteristics of the states of the factors. The mode of the user responding to the perception task comprises the steps that the user uploads the state information of relevant personalized factors or the change condition of the state information actively fed back by the user in the task time.
Preferably, the information data quality assessment model includes:
user data aggregation is carried out through the user credit value, and the expression is as follows:
wherein,representing user u i The provided road section r j Kth dimension perception data,>for the aggregated road section r j Kth dimension data, ">For user u i Is a reputation value of (2);
and evaluating the user quality by adopting errors of the user data and the aggregate data, wherein the expression is as follows:
wherein e i,j Is u i The total deviation of the data is provided and,submitting the maximum value and the minimum value of the kth dimension data for the user respectively;
updating the user credit value according to the user quality, wherein the expression is as follows:
wherein,e, updating the reputation value for the user j Submitting road segments r for participants within a certain period of time j The median of the deviation value of road condition data, epsilon is a tolerable deviation value, alpha and beta are positive real numbers, sign (x) is a signal function, when x is more than 0, sign (x) =1, and when x is less than 0, sign (x) = -1;
according to the credit value of the user, four user grades are divided by taking the week as a settlement period: the corresponding reputation values of the high-grade, medium-grade, primary-grade and low-grade messages are specifically as follows:
a 90% frequency reputation value above 0.8 is a high-level user;
a frequency reputation value of 75% of users with intermediate levels above 0.7;
60% of the frequency reputation value is 0.5 or more and is a primary level user;
frequency reputation values of more than 40% are below 0.5 for low-confidence users.
Preferably, the user incentive mode based on the contract theory comprises a contract incentive model and a user rewarding system. The contract incentive model construction comprises the steps that a user completes a perception task, a platform obtains maximum benefit and ensures that the user benefit is not lower than expected benefit, and the method specifically comprises the following steps:
the optimal contract is established, and the expression is as follows:
wherein A is i For platform profit, R i The benefits obtained for each class of users, p i A probability that the user is an ith reputation level;
the excitation compatibility constraint IC and the participation constraint IR are introduced, and the expression is as follows:
wherein c is the cost required by the user to upload the data, f me Representing a desired threshold value established to reduce platform cost E [ u ] participate ]Representing the income expectation of a platform, wherein the income of the platform is a user perception result, the size of the income depends on the quality of data submitted by a user, and U is a preset income amount;
the method for rewarding the users of each level has the following expression:
R i =p i (σ+μA i ),i∈{1,...,4}
wherein σ is the basic reward and μ is the reward coefficient;
the basic income form of the user is a carbon coin, and the use mode of the carbon coin comprises the following steps:
(1) Redemption of the consumer ticket in the online supermarket;
(2) Exchanging public transportation comprises free travel times of subways, buses and shared bicycles;
(3) Directly exchanging the carbon coins into cash red bags according to a certain proportion.
Preferably, the personalized factor weight determining method comprises the steps of confirming the identity of a special requirement of a user, selecting the personalized requirement factor, and combining an entropy weight method with a G1 method to give weight to determine the weight of each factor;
after the weights of all factors are obtained, the personalized trip impedance of the walking road network is determined, and the expression is as follows:
wherein F is i (x i ,x i1 ,...,x in ) Representing the personalized impedance of road section i, x i Representing the length of the road section i, dist (x i ) Represents a distance function, w j Weights, x, representing personality factors j ij Normalized value of personality j for road segment i, (x) i1 ,...,x in ) The standardized result of each individual factor state in the road section i is represented, and comprises the traffic situation of the sidewalk, the number of steps, the construction encroaching on the road surface, the configuration of barrier-free facilities and the setting of the boulevardCrowd density, non-motor vehicle interference, crosswalk length and number, road surface water accumulation, gradient, road surface flatness and illumination condition, coefficient gamma 1 、γ 2 Reflecting distance function Dist (x i ) And the extent to which the user preference affects the impedance;
the path planning method considers the individual requirements of two layers of user reachability and comfortableness, the reachability requirement realizes feasible path set screening by setting constraint, the comfortableness requirement is reflected in weight determination of individual factors, and finally, paths are matched by using a multi-constraint path planning method, and a planning model is as follows:
wherein F is i (x i ,x i1 ,...,x in ) Represents the personalized impedance of the road segment i,for the reachability requirement factor selected by the user, i=1, 2,..m, j=1, 2,..n.
The beneficial effects are that: compared with the existing map navigation system, the invention has the remarkable advantages that: compared with the single walking road network information and walking path planning target of the existing system, the method fully plays the main role of pedestrians, and utilizes the perception capability of people such as idle users, users with rewards and users using navigation users to acquire road condition information, so that accurate and rapid acquisition and updating of walking road condition diversified information are realized. Meanwhile, the accessibility and comfort level requirements of pedestrian travel are emphasized, the planned walking path can effectively avoid travel obstacles possibly encountered by special crowds such as the old, and experience and comfort level of common crowds in travel are remarkably improved, so that diversified slow-going requirements of resident safety, health, comfort and the like are met. With the arrival of an aging society and the improvement of the attention of the society to the disabled, a huge special group brings wide application prospect to the invention. In practical application, the system function of the invention can be embedded into the existing map navigation APP, and the scale of information perception groups is ensured through a large amount of user basis of the APP, so that the perfection and optimization of system data and path planning are realized.
Drawings
FIG. 1 is a flow chart of the overall process of the present invention;
FIG. 2 is a personalized factor state update flow of the present invention;
FIG. 3 is a diagram of a reputation value update system according to the present invention;
FIG. 4 is a diagram illustrating reputation level user classification in accordance with the present invention;
FIG. 5 is a schematic diagram of an incentive contract model of a crowd sensing task of the present invention;
fig. 6 is a diagram comparing a planned route of a conventional map with a planned route of the present invention.
Detailed Description
The invention is described in detail below with reference to the attached drawings and the specific embodiments:
example 1
As shown in fig. 1, a personalized pedestrian travel path planning method based on crowd sensing comprises the following steps:
step 1: determining personalized factors of pedestrian walking travel;
in the initial stage, the personalized factor data of pedestrian walking travel is obtained by consulting investigation statistical modes such as literature, field interviews and the like and by a rolling collection mode in the application stage. For people with inconvenient travel (handicapped people, old people and the like), the accessibility requirement of the people for completing travel is mainly considered; for the general population, comfort requirements are further considered in addition to accessibility requirements.
Step 2: the system issues a perception task of personalized trip information, and a user responds to the perception task;
the actual information of personalized factors is acquired through the perception task, and the tasks issued by the system mainly comprise the following two sources: firstly, an event driven type triggers an information acquisition task of related factors mainly through active feedback of a user or other ways, and the information acquisition task is handed to the user to realize specific situations of information; secondly, a time driving type system regularly issues a factor state update task according to the characteristics of each factor state;
as shown in fig. 2, relevant content of a perceived task is determined according to the status information requirement condition of the personalized factors, including task type, task area, task rewards, start and stop time of the task, task description and the like, and the system distributes road network travel information perceived task to surrounding idle users, users with reward requirements and mobile devices of users using navigation to complete task invitation. The user receiving the invitation may complete the response by: firstly, actively feeding back the change condition of state information by a user; secondly, the system determines the period and time of task release, and the user uploads the state information of relevant personalized factors in the task time.
Step 3: based on the information data quality evaluation model, performing sensing data aggregation and user credit value updating, and loading the aggregated sensing data to a road network through a system for personalized impedance calculation of the walking road network;
(1) User data aggregation is carried out through the user credit value, and the expression is as follows:
wherein,representing user u i The provided road section r j Kth dimension perception data,>for the aggregated road section r j Kth dimension data, ">For user u i Is a reputation value of (2);
(2) And evaluating the user quality by adopting errors of the user data and the aggregate data, wherein the expression is as follows:
wherein e i,j Is u i The total deviation of the data is provided and,submitting the maximum value and the minimum value of the kth dimension data for the user respectively;
as shown in fig. 3, the user reputation value is updated according to the user quality, and the expression is:
wherein,e, updating the reputation value for the user j Submitting road segments r for participants within a certain period of time j The median of the deviation value of the road condition data, epsilon is a tolerable deviation value and is used for counteracting deviation of participants caused by individual perception difference, perception time, environmental change and other factors; alpha and beta are positive real numbers for controlling +.>Is a convergence speed of (2); sign (x) is a signal function, sign (x) =1 when x > 0, sign (x) = -1 when x < 0;
(3) As shown in FIG. 4, users are classified into four classes according to the reputation value updating mechanism: in each settlement period, when the frequency reputation value of the user is more than 0.8, the user is a high-level user with the reputation value; when the frequency credit value of 75% is above 0.7, the user is a medium-grade user; when the frequency credit value of 60% is 0.5 or above, the user is a primary level user; if the frequency credit value exceeding 40% is below 0.5, the credit value is a low credit user, and the credit value class settlement period is one week; low-credit to high-ranking users correspond to i=1, 2,3,4, respectively.
Step 4: establishing a contract incentive model based on a contract theory, and determining the basic profit form and amount of the user to motivate the user to respond to the perception task and ensure the quality of the uploaded information; the user incentive mode based on contract theory comprises a contract incentive model and a user rewarding system:
(1) As shown in fig. 5, a contractual incentive model is built: on the premise that a user completes a perception task, the income of a platform is maximized, an optimal contract is sought, and an expression of an objective function is:
wherein A is i For platform profit, R i The benefits obtained for each class of users, p i A probability that the user is an ith reputation level;
in the optimal contract, in order to avoid the 'agent problem', an incentive compatibility constraint IC and a participation constraint IR are introduced, and in the road network information sensing task, even if the sensing data is continuously acquired after the acquired data reaches a certain quantity, the discovery and correction effect on a true value is lower. Thus, in order to reduce the cost of the platform as much as possible, a reasonable expected threshold f is established me At this time, the constraint conditions are:
wherein c is the cost required by the user to upload the data, f me Representing a desired threshold value established to reduce platform cost E [ u ] participate ]Representing the income expectation of a platform, wherein the income of the platform is a user perception result, the size of the income depends on the quality of data submitted by a user, and U is a preset income amount; the incentive compatibility constraint IC effectively binds benefits of the platform and the user, the maximization of the benefits of the platform is realized through the maximization behavior of the utility of the user, and the participation constraint IR means that the obtained benefits cannot be lower than a certain preset benefit U after the user fulfills contract;
the specific benefits of the user rewards of each level are calculated according to the following formula:
R i =p i (σ+μA i ),i∈{1,...,4}
wherein σ is the basic reward and μ is the reward coefficient;
(2) The user basic income is issued in the form of carbon coins, and the carbon coins can be converted into cash and consumption coupons according to a certain proportion;
there are three options for carbon coin exchange: 1) exchanging consumption coupons in online supermarkets, 2) exchanging free travel times of subways, buses and shared bicycles, 3) directly exchanging carbon coins into cash red bags according to a certain proportion. The personal carbon credit is deduced by cooperation with the bank and corresponding policy incentives are strived for, and the user carbon coin exchange mode is increased, wherein the user carbon coin exchange mode comprises certain government public service exchange, such as enjoyment of preferential in aspects of book borrowing and the like of a public library.
Step 5: according to the information of the perception data aggregation road network, the user demand is divided into two layers of reachability and comfortableness, the personalized impedance of the walking road network is calculated by determining the weight of each impedance factor, and a constraint path-containing planning algorithm is used for providing personalized optimal travel paths for the user;
(1) Personalized impedance factor weight determination
And combining an entropy weight method with G1 to give weights, and analyzing weights of different personalized factors in a pedestrian demand path. The specific method comprises the following steps:
the system determines objective weights according to attribute differences of individual factors of the road network by utilizing an entropy weight method. The entropy value represents the information quantity contained in the index, and the smaller the information entropy is, the larger the information quantity contained in the index is, the larger the variation degree of different road sections on the attribute is, the larger the function which can be played in the decision process is, and the larger the corresponding weight is. The specific calculation formula is as follows:
wherein x is ij Normalized value of index j of road section i, P ij For the specific gravity of the index j of the road section i, H j Is the entropy value of index j, w 1j Is entropy weight;
according to the importance degree of the user on the personality factors, analyzing the user preference, determining the subjective weight through the G1 method, firstly, selecting the personality factors according to the importance degree by the user, and sequencing the personality factors as x' 1 >x′ 2 >…x n And then, calculating weight coefficients according to the sequence:
wherein w is 2j Absolute subjective weights, w, for personality factors 2j ' scoring coefficients for user weights, R j For the user to index x j And x j-1 The ratio scale value of the importance level comparison,
in order to fully reflect the difference and the importance degree among the attributes and reflect the subjective intention of a decision maker, the combination weight of each factor is finally determined by using linear weighting, and the specific formula is as follows:
wherein w is j Weights, delta, representing personality factors j combining objective and subjective factors 1 、δ 2 For combining coefficients, the decision maker is heavy to subjective and objectiveWhen the degrees of vision are different, the selected combination weights are different;
after determining the combining weights of the factors, the personalized impedance of the road segment can be calculated by the following formula:
wherein F is i (x i ,x i1 ,…,x in ) Representing the personalized impedance of road section i, x i Representing the length of the road section i, dist (x i ) Representing distance function, x ij Normalized value for personality factor j for road segment i, (x) i1 ,...,x in ) The standardized results of the individual factor states determined in the road section i are represented, and comprise the traffic situation of the sidewalk, the number of steps, the construction encroaching road surface, the accessible facility configuration, the boulevard arrangement, the crowd density, the non-motor vehicle interference, the length and the number of crosswalk, the accumulated water on the road surface, the gradient, the road surface flatness and the illumination condition. Coefficient gamma 1 、γ 2 Reflects the distance function Dist (x i ) And the extent to which the user prefers the magnitude of the effect on impedance. Gamma ray 1 The larger the distance function, the greater the influence of the distance function on the impedance; gamma ray 2 The larger the contribution of the user preference to the impedance, and the earlier case may be set to γ 1 =γ 2 The weight can be regarded as a product of the distance and the individualization factor prolonged on the basis of the road distance;
(2) Constrained path planning
The basic idea of the determination of the personalized path of the pedestrian is as follows: firstly, setting constraint of corresponding factors for accessibility requirements of users, and further screening out a feasible node set; secondly, introducing user preference factors into user comfort factors, determining weights of the factors and personalized impedance values of road sections according to importance sequences of users, and finally searching walking paths with minimum impedance in a feasible road network by adopting Dijkstra algorithm. The specific model is as follows:
wherein,for the reachability requirement factor selected by the user, i=1, 2,..m, j=1, 2,..n.
Example 2
(1) Implementation area
In this embodiment, a specific area of a city is selected as a research area of a demonstration case. The area is 1.92 square kilometers, and public service facilities such as hospitals, schools, park scenic spots, stadiums and the like are arranged, so that various demand crowds can be covered. The road condition attribute such as non-mixing of people, building construction, accumulated water on the road surface, crowding degree, boulevard, terrain slope difference and the like is contained, and the influence of different personalized impedance on walking path planning can be effectively embodied.
(2) Test user
In order to understand the consideration factors of pedestrians in different groups and different age groups on walking roads, in this embodiment, 50 persons of an experimental object are selected to perform walking personalized factor investigation, basic road network information filling and path planning test, and the basic conditions of the users are shown in the following table 1:
table 1 test subscriber base case
(3) Process of implementation
Test 1: the testers participated in the personalized factor survey. The tester fills in the requirement factors for the walking path, the embodiment performs statistics on the investigation result, and the personalized factors and the quantification standard of each factor of the embodiment are determined;
test 2: the test was performed under daytime, good weather conditions. Distributing paths to testers in a research area, feeding back road condition information of a road, and sorting the collected information to form impedance information;
test 3: the test does not limit weather conditions, covering peak and peaked periods. The tester determines the starting and ending point of the path according to the requirement, selects the personalized factors which are expected to be considered, and uses the method in the travel process.
(4) Results of the implementation
The personalized factors and the quantification standards are determined through investigation statistics and literature review by experiments, and the following table 2 is shown in the specification:
table 2 personalized factor quantization table
Examples of common starting and ending points in volunteers were selected. Test subjects A, B have jointly selected to travel from their home cell to stadium:
college student user a walks to the stadium as a general user, and selects the options of "no motor vehicle is less disturbed", "there is a boulevard" according to his needs, while the desired path is relatively short. The system recommends a path I as shown in FIG. 6 for the user A after comprehensively considering the requirements of the user A, and sends a task invitation to the user A to inform the user A that feedback is rewarded. Under the excitation of the system, the user A actively feeds back the condition that the construction part encroaches on the road surface at the road section (2) in fig. 6 through the navigation popup window. The information is identified as a new information state by the system, the acquisition task of the construction factors is triggered, and after receiving the task, users needing compensation and nearby leisure go to the road section for information verification. The system carries out perception data aggregation on the construction condition of the road section (2) to obtain an aggregate data value of 0.548, and loads the information to the road network to update the construction state of the road section. The deviation between the sensing data and the aggregation data on the user A is only 0.048, the information fidelity is high, and then the system updates the reputation value, and the updating process is as follows in the table 3:
table 3 reputation value update table
The old user B selects the same travel origin-destination as the old user A, the old user B is older and inconvenient in legs and feet, the old user B is confirmed to be a special-demand user, and the old user B expects to acquire a accessibility path without building construction occupying the road surface and steps, and selects other demands with low crowd density. The system comprehensively considers the individual requirements of the user B and the walking road network information, avoids the construction and the step road section fed back by the user A, recommends the user B to avoid the temporary congestion road section and the intersection caused by school learning and pass through the park as shown in the II paths shown in fig. 6, and the recommended paths have good road conditions, low crowd density and high comfort level. After the trip is completed, the test object B makes a good comment on the trip.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (6)
1. The personalized pedestrian travel path planning method based on crowd sensing is characterized by comprising the following steps of:
step 1: determining personalized factors of pedestrian walking travel demands;
step 2: the system issues a perception task of personalized trip information, and a user responds to the perception task;
step 3: based on the information data quality evaluation model, performing sensing data aggregation and user credit value updating, and loading the aggregated sensing data to a road network through a system for subsequent personalized impedance calculation;
step 4: establishing a contract incentive model based on a contract theory, and determining the basic profit form and amount of the user to motivate the user to respond to the perception task and ensure the quality of the uploaded information;
step 5: and according to the road condition information and the personalized factors selected by the pedestrians, weighting each factor, calculating personalized impedance of the walking road network, and finally providing an optimal walking path matched with the pedestrian demands by using a constraint path-containing planning algorithm.
2. The personalized pedestrian travel path planning method based on crowd sensing according to claim 1, wherein the personalized factors comprise accessibility factors and comfort factors, and the accessibility factors determine whether users with inconvenient travel can reach a destination, including sidewalk traffic conditions, step number, construction occupation road surface conditions and barrier-free facility configuration; the travelling experience of user is influenced to travelling factor, including boulevard setting, crowd density, non-motor vehicle interference, cross street crosswalk length and quantity, road surface ponding, slope, road surface roughness and illumination condition.
3. The personalized pedestrian travel path planning method based on crowd sensing according to claim 1, wherein the content of the sensing task comprises task type, task area, task rewards, task start-stop time and task description. The mode of the system for releasing the perception task is divided into event driving and time driving, wherein the event driving task triggers the information acquisition task of related factors through active feedback of a user or other ways and gives the specific condition of implementing information to the user; the time-driven task is periodically issued with a factor state update task by the system according to the characteristics of the states of the factors. The mode of the user responding to the perception task comprises the steps that the user uploads the state information of relevant personalized factors or the change condition of the state information actively fed back by the user in the task time.
4. The personalized pedestrian travel path planning method based on crowd sensing according to claim 1, wherein the information data quality assessment model comprises:
user data aggregation is carried out through the user credit value, and the expression is as follows:
wherein,representing user u i The provided road section r j Kth dimension perception data,>for the aggregated road section r j Kth dimension data, ">For user u i Is a reputation value of (2);
and evaluating the user quality by adopting errors of the user data and the aggregate data, wherein the expression is as follows:
wherein e i,j Is u i The total deviation of the data is provided and,submitting the maximum value and the minimum value of the kth dimension data for the user respectively;
updating the user credit value according to the user quality, wherein the expression is as follows:
wherein,e, updating the reputation value for the user j Submitting road segments r for participants within a certain period of time j The median of the deviation value of road condition data, epsilon is a tolerable deviation value, alpha and beta are positive real numbers, sign(s) is a signal function, when x is more than 0, sign (x) =1, and when x is less than 0, sign (x) = -1;
according to the credit value of the user, four user grades are divided by taking the week as a settlement period: the corresponding reputation values of the high-grade, medium-grade, primary-grade and low-grade messages are specifically as follows:
a 90% frequency reputation value above 0.8 is a high-level user;
a frequency reputation value of 75% of users with intermediate levels above 0.7;
60% of the frequency reputation value is 0.5 or more and is a primary level user;
frequency reputation values of more than 40% are below 0.5 for low-confidence users.
5. The personalized pedestrian travel path planning method based on crowd sensing as claimed in claim 1, wherein the method comprises the following steps: the user incentive mode based on the contract theory comprises a contract incentive model and a user rewarding system. The contract incentive model construction comprises the steps that a user completes a perception task, a platform obtains maximum benefit and ensures that the user benefit is not lower than expected benefit, and the method specifically comprises the following steps:
the optimal contract is established, and the expression is as follows:
wherein A is i For platform profit, R i The benefits obtained for each class of users, p i A probability that the user is an ith reputation level;
the excitation compatibility constraint IC and the participation constraint IR are introduced, and the expression is as follows:
wherein c is the cost required by the user to upload the data, f me Representing a desired threshold value established to reduce platform cost E [ u ] participate ]Representing the income expectation of a platform, wherein the income of the platform is a user perception result, the size of the income depends on the quality of data submitted by a user, and U is a preset income amount;
the method for rewarding the users of each level has the following expression:
R i =p i (σ+μA i ),i∈{1,...,4}
wherein σ is the basic reward and μ is the reward coefficient;
the basic income form of the user is a carbon coin, and the use mode of the carbon coin comprises the following steps:
(1) Redemption of the consumer ticket in the online supermarket;
(2) Exchanging public transportation comprises free travel times of subways, buses and shared bicycles;
(3) Directly exchanging the carbon coins into cash red bags according to a certain proportion.
6. The personalized pedestrian travel path planning method based on crowd sensing as claimed in claim 1, wherein the method comprises the following steps: the personalized factor weight determining method comprises the steps of user special requirement identity confirmation, personalized requirement factor selection, combination weighting of an entropy weight method and a G1 method, and determination of various factor weights;
after the weights of all factors are obtained, the personalized trip impedance of the walking road network is determined, and the expression is as follows:
wherein F is i (x i ,x i1 ,...,x in ) Representing the personalized impedance of road section i, x i Representing the length of the road section i, dist (x i ) Represents a distance function, w j Weights, x, representing personality factors j ij Normalized value for personality factor j for road segment i, (x) i1 ,...,x in ) The standardized result for representing each individual factor state in road section i comprises pavement traffic situation, step number, construction encroaching road surface, no-obstacle facility configuration, boulevard setting, crowd density, non-motor vehicle interference, crosswalk length and number, road surface ponding, gradient, road surface flatness and illumination condition, coefficient gamma 1 、γ 2 Reflecting distance function Dist (x i ) And the extent to which the user preference affects the impedance;
the path planning method considers the individual requirements of two layers of user reachability and comfortableness, the reachability requirement realizes feasible path set screening by setting constraint, the comfortableness requirement is reflected in weight determination of individual factors, and finally, paths are matched by using a multi-constraint path planning method, and a planning model is as follows:
wherein F is i (x i ,x i1 ,...,x in ) Represents the personalized impedance of the road segment i,for the reachability requirement factor selected by the user, i=1, 2,..m, j=1, 2,..n.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310792725.8A CN117029847A (en) | 2023-06-30 | 2023-06-30 | Personalized pedestrian travel path planning method based on crowd sensing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310792725.8A CN117029847A (en) | 2023-06-30 | 2023-06-30 | Personalized pedestrian travel path planning method based on crowd sensing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117029847A true CN117029847A (en) | 2023-11-10 |
Family
ID=88630658
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310792725.8A Pending CN117029847A (en) | 2023-06-30 | 2023-06-30 | Personalized pedestrian travel path planning method based on crowd sensing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117029847A (en) |
-
2023
- 2023-06-30 CN CN202310792725.8A patent/CN117029847A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yan et al. | The effectiveness of parking policies to reduce parking demand pressure and car use | |
Cao et al. | The relationship between the built environment and nonwork travel: A case study of Northern California | |
Guo et al. | Assessing the cost of transfer inconvenience in public transport systems: A case study of the London Underground | |
Lin et al. | Built environment effects on leisure travel for children: Trip generation and travel mode | |
Davidson et al. | CT-RAMP family of activity-based models | |
Kaplan et al. | Exploring en-route parking type and parking-search route choice: Decision making framework and survey design | |
de Jong et al. | The driving factors of passenger transport | |
Alabed et al. | The characteristics of livable streets: A study of physical aspects of two streets in Riyadh | |
Davidson et al. | New advancements in activity-based models | |
Jiageng et al. | A study on the trip behavior of shared bicycles and shared electric bikes in Chinese universities based on NL model—Henan Polytechnic University as an example | |
Tran et al. | Effects of perceived neighborhood walkability on mode choice of short-distance trips in Hanoi city | |
CN117029847A (en) | Personalized pedestrian travel path planning method based on crowd sensing | |
Sammer | Non-negligible side effects of traffic demand management | |
Lukman | Development and implementation of a transit oriented development (TOD) index around the current transit nodes | |
Obi Lawrence | Application of gravity model for trip analysis in transportation and traffic engineering for Owerri Metropolis Nigeria | |
Pöhler et al. | Urban land use policies for efficient autonomous on-demand transportation–a case study on the japanese island of izu oshima | |
CN108417040B (en) | Characteristic town traffic distribution prediction method | |
Harmatuck | Revealed parking choices and the value of time | |
Turki Haykal | Influence of street design characteristics on walkability: Case studies of two neighborhoods in Erbil | |
Jung | Enhanced Transit Ridership Forecasting Using Automatic Passenger Counting Data | |
Jehn et al. | A GIS Performance Measure-based Methodology for Prioritizing Bike Share System Expansion | |
Truong et al. | Analysis of parking demand and impacts of parking pricing on commuter mode choice in Hanoi, Vietnam | |
Ulrich et al. | Modeling Private and Business Travelers in an Agent-Based Travel Demand Model | |
Moua et al. | Ride On: Strategies for promoting transit ridership at the University of Minnesota-Twin Cities | |
Jou et al. | Parking preferences of tourists in Sun Moon Lake scenic area |
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 |