Target similar trajectory matching method considering motion elements

Abstract

The invention discloses a target similar track matching method considering motion elements, and relates to the field of geographic information processing. The method comprises the following steps: firstly, constructing a candidate matching track data set with motion elements, then obtaining a to-be-matched track, obtaining an element maximum difference value set according to the to-be-matched track and the candidate matching track data set, and obtaining a distance matrix according to the to-be-matched track, the candidate matching track data set and the element maximum difference value set; and then obtaining a path matrix according to the distance matrix, obtaining an element similar distance between the to-be-matched track and any track in the candidate matching track data set according to the path matrix, and recording the track corresponding to the minimum element similar distance as a matching track. According to the invention, through comprehensive consideration of the to-be-matched target and the motion state of each track in the candidate matching track data set, accurate matching can be carried out on similar tracks of the to-be-matched target.

Description

Technical Field

[0001] The invention relates to the field of geographic information processing, and in particular to a target similarity trajectory matching method taking into account motion elements. Background Art

[0002] In today's rapidly developing social and economic activities, target similar trajectory matching technology plays an increasingly important role. With the advent of the big data era, the popularity of various mobile devices and sensors has made it easy for people to obtain and record the movement trajectories of individuals or objects. These trajectory data are not only widely used in traffic management, logistics optimization, urban planning and other fields, but also play a key role in security monitoring, personal health monitoring and other aspects. For example, in the field of transportation, by analyzing the driving trajectory of vehicles, traffic flow can be optimized and congestion can be reduced; in the field of logistics, similar trajectory matching can help companies optimize delivery routes and reduce costs. However, existing trajectory matching methods often ignore the diversity of motion elements and the time differences of trajectories, resulting in limited accuracy and practicality of matching results.

[0003] Target motion elements, such as speed and direction, are key factors affecting trajectory similarity. Traditional trajectory matching methods often rely on simple geometric distance or time series similarity, while ignoring the combined influence of these motion elements. This makes it difficult for matching results to achieve ideal accuracy and robustness when facing complex and changing actual situations. Therefore, how to combine the motion element information of different targets to improve the matching accuracy and anti-interference of target similar trajectories has become a research focus. Summary of the invention

[0004] In view of this, the present invention proposes a target similar trajectory matching method taking into account motion elements. The present invention can accurately match the similar trajectory of the target to be matched by comprehensively considering the motion state of the target to be matched and each trajectory in the candidate matching trajectory data set.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for matching target similar trajectories taking into account motion elements comprises the following steps:

[0007] Step 1: Build the initial candidate matching trajectory dataset

[0008]

[0009] Among them, l i Represents the initial candidate matching trajectory dataset The i-th trajectory in, i = 1, 2, 3, ..., n, n is the initial candidate matching trajectory dataset The number of trajectories in Respectively represent the trajectory l i The time, longitude and latitude of the kth trajectory point in , k = 1, 2, 3, ..., m i , m i Represents the trajectory l i The number of trajectory points included;

[0010] Step 2: Match the initial candidate trajectory dataset For each trajectory in the , calculate the motion element of each trajectory point in the trajectory, and construct a candidate matching trajectory dataset with motion elements

[0011] Step 3: Get the trajectory to be matched, calculate the motion element of each trajectory point in the trajectory to be matched, and construct the trajectory to be matched L with motion elements re ;

[0012] Step 4: Calculate the trajectory L to be matched re and candidate matching trajectory dataset The maximum difference set B of all trajectories in max ;

[0013] Step 5: Calculate the trajectory L to be matched re and candidate matching trajectory dataset Any trajectory l' i The distance matrix D(L re ,l' i );

[0014] Step 6: According to the distance matrix D(L re ,l' i ) Calculate the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l' i The path matrix W(L re ,l' i );

[0015] Step 7: According to the path matrix W(L re ,l' i ) Get the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l' i The element similarity distance T(L re ,l' i );

[0016] Step 8: Select the candidate matching trajectory dataset corresponding to the minimum feature similarity distance The trajectory in is recorded as the trajectory to be matched Lre The trajectory with the highest matching degree completes the target similarity trajectory matching taking into account the motion elements.

[0017] Furthermore, in step 2, a candidate matching trajectory dataset with motion elements is constructed Specifically:

[0018]

[0019] Among them, l' i Represents the candidate matching trajectory dataset The i-th trajectory in Respectively represent the trajectory l' i The heading and speed of the kth trajectory point in ;

[0020] When k = 1:

[0021]

[0022] When k=2,3,......,m i hour:

[0023]

[0024] Among them, arctan() means to find the inverse tangent.

[0025] Furthermore, the trajectory L to be matched with motion elements in step 3 re Specifically:

[0026]

[0027] in, Respectively represent the trajectory to be matched L re The time, longitude, latitude, heading and speed corresponding to the jth track point in , j = 1, 2, 3, ..., m re , m re Represents the trajectory to be matched L re The number of trajectory points included;

[0028] When j = 1:

[0029]

[0030] When j=2,3,......,m re hour:

[0031]

[0032] Furthermore, the maximum difference set B of the elements in step 4 max Specifically:

[0033] B max ={x max ,y max ,d max ,s max}

[0034]

[0035] Among them, x max ,y max ,d max ,s max They represent the maximum differences among longitude, latitude, heading and speed respectively. max() indicates the maximum value operation and |*| indicates the absolute value operation.

[0036] Furthermore, the distance matrix D(L re ,l' i ) is calculated as follows:

[0037]

[0038] Among them, d(j,k) is the distance matrix D(L re ,l' i )’s jth row and kth column data, dis(L re (j),l' i (k)) represents the calculation of the trajectory to be matched L re The j-th trajectory point and candidate matching trajectory dataset Middle track l' i The multidimensional distance between the k-th trajectory points.

[0039] Furthermore, in step 6, the path matrix W(L re ,l' i ) is calculated as follows:

[0040]

[0041] Among them, w(j,k) is the path matrix W(L re ,l' i ), min{} means to find the minimum value. To make the above formula meaningful, set the following initial conditions:

[0042] w(0,0)=0, w(j,0)=∞, w(0,k)=∞.

[0043] Furthermore, in step 7, the trajectory L to be matched re and candidate matching trajectory dataset Any trajectory l' i The element similarity distance T(L re ,l' i) is the path matrix W(L re ,l' i ) in w(m re ,m i ).

[0044] Due to the adoption of the above technical solution, the beneficial effects of the present invention compared with the prior art are:

[0045] 1. The present invention can accurately match similar trajectories of the target to be matched by comprehensively considering the motion state of the target to be matched and each trajectory in the candidate matching trajectory data set.

[0046] 2. The present invention can not only significantly improve the accuracy of trajectory similarity analysis, but also effectively deal with complex noise and interference in trajectory data, providing more reliable and efficient technical support for multiple data fusion. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 4 is an overall flow chart of a target similar trajectory matching method taking motion elements into consideration in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0049] A target similar trajectory matching method taking into account motion elements, such as Figure 1 As shown, the following steps are included:

[0050] Step 1: Build the initial candidate matching trajectory dataset

[0051]

[0052] Among them, l i Represents the initial candidate matching trajectory dataset The i-th trajectory in, i = 1, 2, 3, ..., n, n is the initial candidate matching trajectory dataset The number of trajectories in the dataset, and the initial candidate matching trajectory Each trajectory in is a currently known historical trajectory. Respectively represent the trajectory l i The time, longitude and latitude of the kth trajectory point in , k = 1, 2, 3, ..., m i , m i Represents the trajectory l i The number of trajectory points included;

[0053] Step 2: Match the initial candidate trajectory dataset For each trajectory in the , calculate the motion element of each trajectory point in the trajectory, and construct a candidate matching trajectory dataset with motion elements

[0054] Step 3: Get the trajectory to be matched, calculate the motion element of each trajectory point in the trajectory to be matched, and construct the trajectory to be matched L with motion elements re ;

[0055] Step 4: Calculate the trajectory L to be matched re and candidate matching trajectory dataset The maximum difference set B of all trajectories in max ;

[0056] Step 5: Calculate the trajectory L to be matched re and candidate matching trajectory dataset Any trajectory l' i The distance matrix D(L re ,l' i );

[0057] Step 6: According to the distance matrix D(L re ,l' i ) Calculate the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l' i The path matrix W(L re ,l' i );

[0058] Step 7: According to the path matrix W(L re ,l' i ) Get the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l' i The element similarity distance T(L re ,l' i );

[0059] Step 8: Select the candidate matching trajectory dataset corresponding to the minimum feature similarity distance The trajectory in is recorded as the trajectory to be matched L re The trajectory with the highest matching degree completes the target similarity trajectory matching taking into account the motion elements.

[0060] Furthermore, in step 2, a candidate matching trajectory dataset with motion elements is constructed Specifically:

[0061]

[0062] Among them, l'i Represents the candidate matching trajectory dataset The i-th trajectory in Respectively represent the trajectory l' i The heading and speed of the kth trajectory point in ;

[0063] When k = 1:

[0064]

[0065] When k=2,3,......,m i hour:

[0066]

[0067] Among them, arctan() means to find the inverse tangent.

[0068] Furthermore, the trajectory L to be matched with motion elements in step 3 re Specifically:

[0069]

[0070] in, Respectively represent the trajectory to be matched L re The time, longitude, latitude, heading and speed corresponding to the jth track point in , j = 1, 2, 3, ..., m re , m re Represents the trajectory to be matched L re The number of trajectory points included;

[0071] When j = 1:

[0072]

[0073] When j=2,3,......,m re hour:

[0074]

[0075] Furthermore, the maximum difference set B of the elements in step 4 max Specifically:

[0076] B max ={x max ,y max ,d max ,s max}

[0077]

[0078] Among them, x max ,y max ,dmax ,s max They represent the maximum differences among longitude, latitude, heading and speed respectively. max() indicates the maximum value operation and |*| indicates the absolute value operation.

[0079] Furthermore, the distance matrix D(L re ,l' i ) is calculated as follows:

[0080]

[0081] Among them, d(j,k) is the distance matrix D(L re ,l' i )’s jth row and kth column data, dis(L re (j),l' i (k)) represents the calculation of the trajectory to be matched L re The j-th trajectory point and candidate matching trajectory dataset Middle track l' i The multidimensional distance between the k-th trajectory points.

[0082] Furthermore, in step 6, the path matrix W(L re ,l' i ) is calculated as follows:

[0083]

[0084] Among them, w(j,k) is the path matrix W(L re ,l' i ), min{} means to find the minimum value. To make the above formula meaningful, set the following initial conditions:

[0085] w(0,0)=0, w(j,0)=∞, w(0,k)=∞.

[0086] Furthermore, in step 7, the trajectory L to be matched re and candidate matching trajectory dataset Any trajectory l' i The element similarity distance T(L re ,l' i ) is the path matrix W(L re ,l' i ) in w(m re ,m i ).

[0087] In summary, the present invention can accurately match similar trajectories of the target to be matched by comprehensively considering the motion state of the target to be matched and each trajectory in the candidate matching trajectory data set.

[0088] Those skilled in the art will appreciate that the embodiments described are intended to help readers understand the principles of the present invention, and should be understood that the scope of protection of the present invention is not limited to the embodiments described. For those skilled in the art, the present invention may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for matching target similar trajectories taking into account motion elements, characterized in that: The following steps are involved: Step 1: Build the initial candidate matching trajectory dataset Among them, l i Represents the initial candidate matching trajectory dataset The i-th trajectory in, i = 1, 2, 3, ..., n, n is the initial candidate matching trajectory dataset The number of trajectories in Respectively represent the trajectory l i The time, longitude and latitude of the kth trajectory point in , k = 1, 2, 3, ..., m i , m i Represents the trajectory l i The number of trajectory points included; Step 2: Match the initial candidate trajectory dataset For each trajectory in the , calculate the motion element of each trajectory point in the trajectory, and construct a candidate matching trajectory dataset with motion elements Step 3: Get the trajectory to be matched, calculate the motion element of each trajectory point in the trajectory to be matched, and construct the trajectory to be matched L with motion elements re ; Step 4: Calculate the trajectory L to be matched re and candidate matching trajectory dataset The maximum difference set B of all trajectories in max ; Step 5: Calculate the trajectory L to be matched re and candidate matching trajectory dataset Any trajectory in i 'The distance matrix D(L re ,l i '); Step 6: According to the distance matrix D(L re ,l′ i ) Calculate the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l′ i The path matrix W(L re ,l′ i ); Step 7: According to the path matrix W(L re ,l′ i ) Get the trajectory to be matched L re and candidate matching trajectory dataset Any trajectory l′ i The element similarity distance T(L re ,l′ i ); Step 8: Select the candidate matching trajectory dataset corresponding to the minimum feature similarity distance The trajectory in is recorded as the trajectory to be matched L re The trajectory with the highest matching degree completes the target similarity trajectory matching taking into account the motion elements.

2. The target similarity trajectory matching method taking into account motion elements according to claim 1, characterized in that: In step 2, a candidate matching trajectory dataset with motion elements is constructed Specifically: Among them, l i 'Represents the candidate matching trajectory dataset The i-th trajectory in Respectively represent the trajectory l i 'The heading and speed of the kth track point; When k = 1: When k=2,3,......,m i hour: Among them, arctan() means to find the inverse tangent.

3. The target similarity trajectory matching method taking into account motion elements according to claim 2, characterized in that: The trajectory L to be matched with motion elements in step 3 re Specifically: in, Respectively represent the trajectory to be matched L re The time, longitude, latitude, heading and speed corresponding to the jth track point in , j = 1, 2, 3, ..., m re , m re Represents the trajectory to be matched L re The number of trajectory points included; When j = 1: When j=2,3,......,m re hour:

4. The target similarity trajectory matching method taking into account motion elements according to claim 3, characterized in that: The maximum difference set B of the elements in step 4 max Specifically: Among them, x max ,y max ,d max ,s max They represent the maximum differences among longitude, latitude, heading and speed respectively. max() indicates the maximum value operation and * indicates the absolute value operation.

5. The target similarity trajectory matching method taking into account motion elements according to claim 4, characterized in that: In step 5, the distance matrix D(L re ,l i ') is calculated as follows: Among them, d(j,k) is the distance matrix D(L re ,l′ i )’s jth row and kth column data, dis(L re (j),l′ i (k)) represents the calculation of the trajectory to be matched L re The j-th trajectory point and candidate matching trajectory dataset Middle track i 'The multidimensional distance between the k-th trajectory points.

6. The target similarity trajectory matching method considering motion elements according to claim 5, characterized in that: In step 6, the path matrix W(L re ,l i ') is calculated as follows: Among them, w(j,k) is the path matrix W(L re ,l′ i ), min{} means to find the minimum value. To make the above formula meaningful, set the following initial conditions: w(0,0)=0, w(j,0)=∞, w(0,k)=∞.

7. The target similarity trajectory matching method considering motion elements according to claim 6, characterized in that: The trajectory L to be matched in step 7 re and candidate matching trajectory dataset Any trajectory in i 'The element similarity distance T(L re ,l′ i ) is the path matrix W(L re ,l′ i ) in w(m re ,m i ).

Images