A method and system for analyzing the activity area of a muck truck based on trajectory data

Abstract

The present invention provides a method and system for analyzing the activity area of a muck truck based on trajectory data, which relates to the technical field of traffic trajectories and includes the following steps: S1, obtaining a dwelling clustering point area with a dwelling point as the center point; S2, determining whether the distance difference between the center points of the dwelling clustering point areas exceeds a threshold. If not, updating the area of the dwelling clustering point area and proceeding to step S3; S3, obtaining the muck truck trajectory points located within the dwelling clustering point area; S4, performing a piecewise curve fitting of the average speed of the muck truck trajectory points within the dwelling clustering point area, and determining the scene type of the dwelling clustering point area according to the curve characteristics of the average speed trend curve, so as to detect whether there are new construction sites, dumping sites or other muck truck activity areas within the current area, and realize the supervision of the muck truck activity area.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic trajectory, and particularly relates to a method and system for analyzing the activity area of a muck truck based on trajectory data. Background Art

[0002] In order to ensure the safety of muck trucks in the activity areas of muck trucks such as construction sites and dumping sites, it is necessary to supervise the activity areas of muck trucks. Currently, the activity areas of muck trucks all have registration records, enabling staff to clarify the scene types of these areas. However, this registration record usually adopts the method of manual patrol at the suspected locations of muck truck activity areas to count the current activity areas of muck trucks. This method consumes a large amount of human resources, and when the area to be patrolled is large, the efficiency of manual statistics is very low. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and system for analyzing the activity area of a muck truck based on trajectory data. By clustering the residence points, the residence clustering point area where the muck truck permanently resides is obtained. Then, according to the change in the average speed of the muck truck in the residence clustering point area, the scene type of the current residence clustering point area can be determined, which can be used to detect whether there are new construction sites or dumping sites and other muck truck activity areas in the current area, and realize the function of supervising the muck truck activity area through the muck truck trajectory data.

[0004] To solve the above technical problems, the present invention adopts the following solutions:

[0005] A method for analyzing the activity area of a muck truck based on trajectory data, comprising the following steps:

[0006] S1. Obtain multiple groups of residence points screened from the muck truck trajectory data, match the residence points in the residence point groups to the divided grids, and perform clustering processing on each residence point group according to adjacent grids to obtain a residence clustering point area with the residence point as the center point;

[0007] S2. Judge whether the distance difference between the center points of the residence clustering point areas exceeds a threshold. If not, perform area update on the residence clustering point areas and go to step S3;

[0008] S3. Match the muck truck trajectory data corresponding to the residence point group to the residence clustering point area, and screen the muck truck trajectory points in the current muck truck trajectory data to obtain the muck truck trajectory points located in the residence clustering point area;

[0009] S4. Perform curve fitting on the average speed of the muck truck trajectory points in the residence clustering point area to obtain a corresponding average speed trend curve, and judge the scene type of the residence clustering point area according to the curve characteristics of the average speed trend curve.

[0010] Further, in S1, the multiple sets of staying points screened from the muck truck trajectory data refer to the grouping of the staying points screened by the matrix method and the adaptive jump method within a preset time and with a driving distance not exceeding a preset value according to time;

[0011] The matrix method can eliminate inactive muck truck trajectory points from the muck truck trajectory data;

[0012] The adaptive jump method can identify whether the muck truck trajectory point surrounded by multiple muck truck trajectory points in the muck truck trajectory data is a staying point. If so, match the staying point to the divided grid.

[0013] Further, in S1, the process of clustering each set of staying points according to adjacent grids is as follows:

[0014] Connect the staying points located in adjacent grids to obtain multiple sequentially connected staying points;

[0015] Calculate the distance differences between the staying points;

[0016] Respectively count the sum of the distance differences between each staying point and other staying points, compare the sums, take the staying point with the smallest sum as the center point, and form a staying cluster point area with the maximum distance difference of the center point as the radius.

[0017] Further, in S2, the process of updating the staying cluster point area includes the following steps:

[0018] SA1. Take the intermediate value of the distance difference between the center points of two staying cluster point areas, use the point where the intermediate value is located as the new center point, and judge whether there is an overlapping relationship between the two staying cluster point areas. If so, go to step SA2; if not, go to step SA3;

[0019] SA2. Select a radius value of one staying cluster point area according to the overlapping relationship and area size between the two staying cluster point areas, add it to the intermediate value, use it as the current radius value, form a new staying cluster point area with the current radius value based on the new center point, and delete the corresponding two staying cluster point areas;

[0020] SA3. Form a new staying cluster point area with the preset radius value as the radius based on the new center point, and delete the corresponding two staying cluster point areas.

[0021] Further, in SA2, the process of selecting a radius value of one staying cluster point area according to the overlapping relationship and area size between the two staying cluster point areas and adding it to the intermediate value is as follows:

[0022] The overlapping relationship includes intersection and inclusion;

[0023] When the overlapping relationship between two areas of dwelling clustering points is intersection, add the radius value of the area of dwelling clustering points with a smaller area to the median value, and use the result as the current radius value;

[0024] When the overlapping relationship between two areas of dwelling clustering points is inclusion, add the radius value of the area of dwelling clustering points with a larger area to the median value, and use the result as the current radius value.

[0025] Furthermore, the step S4 includes the following steps:

[0026] S41. Connect the trajectory points of the muck trucks within the area of dwelling clustering points in chronological order, sequentially obtain the average speed values between the trajectory points of the muck trucks within the area of dwelling clustering points according to the chronological order, and sort the multiple average speed values in chronological order;

[0027] S42. Sequentially judge according to the chronological order whether the change values between the average speed values are continuously less than the minimum value. If so, perform screening according to the change values that are continuously less than the minimum value, select the average speed value at the central position as the segmentation point, and go to step S43;

[0028] S43. Split the multiple average speed values into two segments, the front and the back, according to the segmentation point, and perform linear fitting on the average speed values of the front and back segments respectively to obtain the corresponding average speed trend curves;

[0029] S44. Determine the scene type of the area of dwelling clustering points according to the slope magnitudes of the average speed trend curves of the front and back segments.

[0030] Furthermore, in S41, the process of sequentially obtaining the average speed values between the trajectory points of the muck trucks within the area of dwelling clustering points according to the chronological order is as follows:

[0031] Sequentially calculate the distance difference and time difference between the trajectory points of the muck trucks within the area of dwelling clustering points according to the chronological order, and through the calculation of the distance difference and time difference, the average speed value between two trajectory points of the muck trucks can be obtained.

[0032] Furthermore, in S44, the process of determining the scene type of the area of dwelling clustering points according to the slope magnitudes of the average speed trend curves of the front and back segments is as follows:

[0033] If the slope of the average speed trend curve of the front segment is greater than the slope of the average speed trend curve of the back segment, then determine that the scene type of the area of dwelling clustering points is a dumping site; if the slope of the average speed trend curve of the front segment is less than the slope of the average speed trend curve of the back segment, then determine that the scene type of the area of dwelling clustering points is a construction site.

[0034] A muck truck activity area analysis system based on trajectory data, which applies a muck truck activity area analysis method as described, includes:

[0035] Resident clustering point area division module: Obtain multiple groups of resident points screened from the muck truck trajectory data, match the resident points in the group of resident points to the divided grids, and perform clustering processing on each group of resident points according to adjacent grids to obtain a resident clustering point area with the resident point as the center point;

[0036] Resident clustering point area update module: Judge whether the distance difference between the center points of the resident clustering point areas exceeds the threshold. If not, perform area update on the resident clustering point areas;

[0037] Muck truck trajectory data matching module: Match the muck truck trajectory data corresponding to the group of resident points to the resident clustering point areas, and screen the muck truck trajectory points in the current muck truck trajectory data to obtain the muck truck trajectory points located in the resident clustering point areas;

[0038] Resident clustering point area determination module: Perform piecewise curve fitting on the average speed of the muck truck trajectory points in the resident clustering point areas to obtain the corresponding average speed trend curve, and determine the scene type of the resident clustering point area according to the curve characteristics of the average speed trend curve.

[0039] Advantages of the present invention:

[0040] The present invention provides a muck truck activity area analysis method and system based on trajectory data. The muck truck activity area analysis method mainly analyzes the resident clustering point areas through the trajectory data of the muck truck, can determine the scene type of the resident clustering point areas, does not need to statistically analyze the muck truck activity areas through manual inspections, can effectively improve the efficiency of statistically analyzing the muck truck activity areas, and reduce labor costs.

[0041] Moreover, the present invention screens the trajectory data to obtain resident points, obtains the resident clustering point areas where the muck trucks often stay through resident point clustering. At the same time, by judging and updating the area size of the resident clustering point areas, the accuracy of muck truck activity area analysis can be improved when determining the muck truck activity areas. Description of the Drawings

[0042] Figure 1 It is a schematic flowchart of the muck truck activity area analysis method in Embodiment 1 of the present invention;

[0043] Figure 2 It is a schematic diagram of the resident clustering point area 1 in Embodiment 1 of the present invention;

[0044] Figure 3 Schematic diagram of the dwelling clustering point region 2 in Embodiment 1 of the present invention;

[0045] Figure 4 Schematic diagram of the intersecting dwelling clustering point regions 1 and 2 in Embodiment 1 of the present invention. Detailed implementation manners

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way restrictive of the present invention and its application or use. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] Unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present invention.

[0048] Meanwhile, it should be understood that, for the sake of description, the dimensions of the various parts shown in the drawings are not drawn in actual proportional relationships.

[0049] In addition, for the sake of clarity and conciseness, descriptions of well-known structures, functions, and configurations may be omitted. Those of ordinary skill in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0050] Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be regarded as part of the authorization specification.

[0051] In all the examples shown and discussed here, any specific value should be construed as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0052] The present invention will be described in detail below by referring to the accompanying drawings and in combination with embodiments:

[0053] Embodiment 1

[0054] In this embodiment, as Figure 1 shown, a scene type analysis method based on dwelling clustering point regions is proposed. The scene type analysis method includes the following steps:

[0055] S1. Obtain multiple groups of stay points filtered from the trajectory data of dump trucks, match the stay points in the stay point groups to the divided grids, and perform clustering processing on each stay point group according to adjacent grids to obtain a stay clustering point area centered on the stay points;

[0056] S2. Determine whether the distance difference between the central points of the stay clustering point areas exceeds a threshold. If not, update the area of the stay clustering point area and go to step S3;

[0057] S3. Match the trajectory data of the dump truck corresponding to the stay point group to the stay clustering point area, and screen the dump truck trajectory points in the current dump truck trajectory data to obtain the dump truck trajectory points located within the stay clustering point area;

[0058] S4. Perform piecewise curve fitting on the average speed of the dump truck trajectory points within the stay clustering point area to obtain the corresponding average speed trend curve, and determine the scene type of the stay clustering point area according to the curve characteristics of the average speed trend curve.

[0059] Based on the above scene type analysis method, it can be seen that the present invention mainly analyzes the scene type of the stay clustering point area by using the trajectory data generated by the dump truck in the stay clustering point area. The stay clustering point area is an area obtained by clustering stay points, and this area can represent the area where the dump truck often stays or the area where the dump truck often wanders. Then, the location of the suspected dump truck activity area can be obtained through the stay clustering point area, without the need for manual inspection, realizing the supervision of the dump truck activity area and reducing the labor cost.

[0060] Preferably, in S1, the multiple groups of stay points filtered from the dump truck trajectory data refer to the groups formed by grouping the stay points that do not exceed a preset value in driving distance within a preset time screened by the matrix method and the adaptive jump method according to time;

[0061] The matrix method can be used to eliminate non-active dump truck trajectory points from the dump truck trajectory data;

[0062] The adaptive jump method can identify whether the dump truck trajectory point surrounded by multiple dump truck trajectory points in the dump truck trajectory data is a stay point. If so, match the stay point to the divided grid.

[0063] The above method of screening the stay points that do not exceed a preset value in driving distance within a preset time by the matrix method and the adaptive jump method is a conventional technical means and will not be elaborated here.

[0064] Preferably, in S1, the process of clustering each stay point group in turn according to adjacent grids is as follows:

[0065] Connect the dwelling points located in adjacent grids to obtain multiple successively connected dwelling points;

[0066] Calculate the distance differences between the dwelling points;

[0067] Respectively count the sum of the distance differences between each dwelling point and other dwelling points, compare the sums, take the dwelling point with the smallest sum as the center point, and form a dwelling clustering point area with the maximum distance difference of the center point as the radius.

[0068] Preferably, in S2, the process of updating the area of the dwelling clustering point area includes the following steps:

[0069] SA1. Take the median of the distance differences between the center points of two dwelling clustering point areas, use the point where the median is located as the new center point, and determine whether there is an overlapping relationship between the two dwelling clustering point areas. If so, go to step SA2; if not, go to step SA3;

[0070] SA2. According to the overlapping relationship and area size between the two dwelling clustering point areas, select the radius value of one dwelling clustering point area and add it to the median, use it as the current radius value, form a new dwelling clustering point area with the current radius value based on the new center point, and delete the corresponding two dwelling clustering point areas;

[0071] In SA2, the process of selecting the radius value of one dwelling clustering point area and adding it to the median according to the overlapping relationship and area size between the two dwelling clustering point areas is as follows:

[0072] The overlapping relationship includes intersection and inclusion;

[0073] When the overlapping relationship between the two dwelling clustering point areas is intersection, select the radius value of the dwelling clustering point area with a smaller area and add it to the median, use it as the current radius value;

[0074] When the overlapping relationship between the two dwelling clustering point areas is inclusion, select the radius value of the dwelling clustering point area with a larger area and add it to the median, use it as the current radius value;

[0075] SA3. Based on the new center point, form a new dwelling clustering point area with a preset radius value as the radius, and delete the corresponding two dwelling clustering point areas.

[0076] Specifically, through the above steps S1 - S2, when obtaining the staying points filtered from the trajectory data of the current muck truck within a preset time, the staying points with a long time interval are divided into two staying point groups A and B according to the time sequence. The time of the staying point group A is earlier than that of the staying point group B. The staying point group A is composed of multiple staying points sorted in time sequence, denoted as A1, A2, A3, A4, A5, A6, A7; the staying point group B is also composed of multiple staying points sorted in time sequence, denoted as B1, B2, B3, B4, B5, B6, B7. Match the two groups of staying points to the divided grids, and then cluster the staying point group A and the staying point group B in turn according to adjacent grids. Connect the staying points located in adjacent grids respectively to obtain a series of connected staying points. Suppose A3 - A4 - A5 - A6 - A7 is obtained currently, then calculate the distance difference between the staying points. It can be judged according to the grid area passed by the connection line between the staying points. Statistically calculate the sum of the distance differences between each staying point and other staying points respectively. At this time, the staying point with the smallest sum is A5, and the staying point with the largest distance difference from A5 is A7. Then, taking A5 as the center point and the distance difference r1 between A5 and A7 as the radius, form the staying clustering point area 1. The staying clustering point area 1 is as Figure 2 shown. Similarly, as described above, clustering the staying point group B can obtain the staying clustering point area 2 with B4 as the center point and the distance difference r2 between B4 and B6 as the radius. The staying clustering point area 2 is as Figure 3 shown. The location of the suspected muck truck activity area can be obtained according to the clustering of the staying points. At this time, the suspected muck truck activity area can be analyzed and judged.

[0077] In the present invention, in order to obtain the area size of the suspected muck truck activity area more accurately, further processing is also performed on the multiple staying clustering point areas respectively formed by multiple groups of staying points. Judge whether the distance difference between the center points of the staying clustering point areas exceeds the threshold. If the distance difference between the center point of the current staying clustering point area 1 and the center point of the staying clustering point area 2 does not exceed the threshold, then the center point of the staying clustering point area 1 and the staying clustering point area 2 need to be updated.

[0078] Among them, if the distance difference between the center points of the current two staying clustering point areas exceeds the threshold, then the area update of the current two staying clustering point areas is not required.

[0079] The distance difference between the center points of the staying clustering point region 1 and the staying clustering point region 2 does not exceed the threshold, which can indicate that the two regions where the muck truck stays or lingers more frequently are relatively close. By correcting the region size of the staying clustering point region according to the behavior of the muck truck staying or lingering in different time periods, the accuracy of subsequent analysis of the staying clustering point region through the muck truck trajectory data can be effectively improved.

[0080] Perform region updates on the above-mentioned staying clustering point region 1 and staying clustering point region 2. If the overlapping relationship between the current staying clustering point region 1 and staying clustering point region 2 is intersection, the intersecting staying clustering point region 1 and staying clustering point region 2 are as Figure 4 shown. Take the median value d / 2 of the distance difference d between the center point of the staying clustering point region 1 and the center point of the staying clustering point region 2. Use the position where the median value is located as the new center point, and select the radius value r1 of the staying clustering point region 1 with a smaller area and add it to the median value d / 2. Take r1 + d / 2 as the current radius value. Based on the new center point, form a new staying clustering point region 3 with the current radius value, and delete the center point of the staying clustering point region 1 and the staying clustering point region 2. The purpose is to obtain a more accurate suspected muck truck activity region.

[0081] Preferably, the step S4 includes the following steps:

[0082] S41. Connect the muck truck trajectory points in the staying clustering point region in chronological order, sequentially obtain the average speed values between the muck truck trajectory points in the staying clustering point region according to the chronological order, and sort the multiple average speed values in chronological order;

[0083] In S41, the process of sequentially obtaining the average speed values between the muck truck trajectory points in the staying clustering point region according to the chronological order is as follows:

[0084] Sequentially calculate the distance difference and time difference between the muck truck trajectory points in the staying clustering point region according to the chronological order. Through the calculation of the distance difference and time difference, the average speed value between two muck truck trajectory points can be obtained;

[0085] S42. Sequentially judge whether the change values between the average speed values are continuously less than the minimum value according to the chronological order. If so, screen according to the change values that are continuously less than the minimum value, and select the average speed value in the central position as the segmentation point, and then go to step S43;

[0086] S43. Split the multiple average speed values into two segments before and after according to the segmentation point, and perform linear fitting on the average speed values of the two segments before and after respectively to obtain the corresponding average speed trend curves;

[0087] S44. Determine the scene type of the dwelling clustering point area according to the slope magnitudes of the average speed trend curves of the front and rear sections;

[0088] In S44, the process of determining the scene type of the dwelling clustering point area according to the slope magnitudes of the average speed trend curves of the front and rear sections is as follows:

[0089] If the slope of the average speed trend curve of the front section is greater than the slope magnitude of the average speed trend curve of the rear section, then determine that the scene type of the dwelling clustering point area is a dumping site; if the slope of the average speed trend curve of the front section is less than the slope magnitude of the average speed trend curve of the rear section, then determine that the scene type of the dwelling clustering point area is a construction site.

[0090] Specifically, when the suspected muck truck activity area is obtained, the scene type of the muck truck activity area can be judged according to the speed change between the muck truck trajectory points in the muck truck trajectory data, and the suspected muck truck activity area refers to the newly generated dwelling clustering point area after area update or the dwelling clustering point area that has not been updated. At this time, the suspected muck truck activity area refers to the dwelling clustering point area 3.

[0091] When the dwelling clustering point area is obtained, one set of dwelling point groups corresponding to a section of muck truck trajectory data can be optionally or separately matched to the dwelling clustering point area, the muck truck trajectory points in the muck truck trajectory data are screened to obtain multiple muck truck trajectory points located in the dwelling clustering point area, and the multiple muck truck trajectory points are connected in chronological order. Since the muck truck trajectory points are uploaded by the GPS device carried by the muck truck itself, the muck truck trajectory points have time information and coordinate information. According to the time information of the muck truck trajectory points, the time difference between two muck truck trajectory points can be calculated, and according to the coordinate information of the muck truck trajectory points, the distance difference between two muck truck trajectory points can be calculated. The distance difference can be calculated using the Haversine formula. Then, according to the distance difference and time difference between two muck truck trajectory points, the average speed value between two muck truck trajectory points can be obtained.

[0092] Since the time interval between two GPS data of the muck truck is generally less than 30 seconds, that is, the time difference between two muck truck trajectory points is generally less than 30 seconds, so the time interval between two muck truck trajectory points is short. Then the current average speed value can be regarded as the instantaneous speed at a time point. Then, the driving behavior of the muck truck in the trajectory data can be analyzed according to the change value of the average speed, and combined with the behavior characteristics of the muck truck carrying out different activities in the muck truck activity areas of different scene types, the scene type of the current dwelling clustering point area can be judged.

[0093] In this embodiment, the judgment of the scene type of the current resident clustering point area mainly determines whether the current resident clustering point area is a dumping site or a construction site. The dumping site and the construction site are the two most common activity areas in the activity area of the muck truck. In these two activity areas, the muck truck needs to take certain measures, such as loading and unloading behaviors, and these behaviors need to be supervised, so that the inspection personnel need to inspect these two activity areas.

[0094] For a construction site, when a muck truck drives into the construction site, since the muck truck is generally empty, its front section speed is relatively fast; when the muck truck is inside the construction site, the muck truck maintains a low speed according to the needs inside the construction site or stops due to loading earthwork and stonework, resulting in a relatively low or zero middle speed; when the muck truck drives out of the construction site, since the muck truck is generally full of earthwork and stonework, its rear section speed is slower than the front section speed.

[0095] For a dumping site, when a muck truck drives into the dumping site, since the muck truck is generally full, its front section speed is relatively slow; when the muck truck is inside the dumping site, the muck truck will maintain a low speed according to the needs inside the dumping site or stop due to dumping, resulting in a relatively low or zero middle speed; when the muck truck drives out of the dumping site, since the muck truck is generally empty, its rear section speed is faster than the front section speed.

[0096] Combining the above situations, the present invention selects the lowest change value of the average speed value to split the average speed value into the front and rear sections, and linearly fits the average speed values of the front and rear sections respectively to obtain the corresponding average speed trend curves. If the slope of the front-section average speed trend curve is greater than the slope of the rear-section average speed trend curve, it is determined that the scene type of the resident clustering point area is a dumping site; if the slope of the front-section average speed trend curve is less than the slope of the rear-section average speed trend curve, it is determined that the scene type of the resident clustering point area is a construction site. The scene type of the current resident clustering point area can be accurately analyzed. In actual situations, by comparing the current resident clustering point area with the registered construction sites or dumping sites, it is possible to detect whether there are new construction sites or dumping sites and other muck truck activity areas in the current area, realizing the role of supervising the muck truck activity area through the muck truck trajectory data. There is no need to statistically analyze the muck truck activity area by manual inspection, which can effectively improve the efficiency of statistically analyzing the muck truck activity area and reduce labor costs.

[0097] Embodiment 2

[0098] A muck truck activity area analysis system based on trajectory data, which applies the muck truck activity area analysis method as described above, includes:

[0099] Resident clustering point area division module: Obtain multiple groups of resident points screened from the trajectory data of muck trucks, match the resident points in the group of resident points to the divided grids, and perform clustering processing on each group of resident points according to adjacent grids to obtain a resident clustering point area centered on the resident points;

[0100] Resident clustering point area update module: Judge whether the distance difference between the central points of the resident clustering point areas exceeds the threshold. If not, perform area update on the resident clustering point areas;

[0101] Muck truck trajectory data matching module: Match the trajectory data of the muck truck corresponding to the group of resident points to the resident clustering point area, and screen the muck truck trajectory points in the current muck truck trajectory data to obtain the muck truck trajectory points located in the resident clustering point area;

[0102] Resident clustering point area determination module: Perform piecewise curve fitting on the average speed of the muck truck trajectory points in the resident clustering point area to obtain the corresponding average speed trend curve, and judge the scene type of the resident clustering point area according to the curve characteristics of the average speed trend curve.

[0103] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Based on the technical essence of the present invention, any simple modifications, equivalent replacements, and improvements made to the above embodiments within the spirit and principles of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for analyzing the activity area of a muck truck based on trajectory data, characterized in that, The analysis method for the activity area of the muck truck includes the following steps: S1. Obtain multiple groups of staying points screened from the muck truck trajectory data, match the staying points in the staying point groups to the divided grids, and perform clustering processing on each staying point group according to adjacent grids to obtain a staying clustering point area with the staying points as the center points; S2. Judge whether the distance difference between the center points of the staying clustering point areas exceeds the threshold. If not, update the area of the staying clustering point area and go to step S3; S3. Match the muck truck trajectory data corresponding to the staying point groups to the staying clustering point areas, and screen the muck truck trajectory points in the current muck truck trajectory data to obtain the muck truck trajectory points located within the staying clustering point areas; S4. Perform a piecewise curve fitting of the average speeds of the muck truck trajectory points within the staying clustering point areas to obtain the corresponding average speed trend curves, and judge the scene type of the staying clustering point area according to the curve characteristics of the average speed trend curves; Specifically, step S4 includes the following steps: S41. Connect the muck truck trajectory points within the staying clustering point areas in chronological order, sequentially obtain the average speed values between the muck truck trajectory points within the staying clustering point areas according to the chronological order, and sort the multiple average speed values in chronological order; S42. Sequentially judge whether the change values between the average speed values are continuously less than the minimum value according to the chronological order. If so, screen according to the change values that are continuously less than the minimum value, select the average speed value at the central position as the segmentation point, and go to step S43; S43. Split the multiple average speed values into two segments before and after according to the segmentation point, and perform linear fitting on the average speed values of the two segments before and after respectively to obtain the corresponding average speed trend curves; S44. Judge the scene type of the staying clustering point area according to the slope magnitudes of the average speed trend curves of the two segments before and after; In S44, the process of judging the scene type of the staying clustering point area according to the slope magnitudes of the average speed trend curves of the two segments before and after is as follows: If the slope of the average speed trend curve of the previous segment is greater than the slope of the average speed trend curve of the latter segment, it is judged that the scene type of the staying clustering point area is a dumping site; if the slope of the average speed trend curve of the previous segment is less than the slope of the average speed trend curve of the latter segment, it is judged that the scene type of the staying clustering point area is a construction site.

2. The method for analyzing the activity area of a muck truck based on trajectory data according to claim 1, wherein In S1, the multiple groups of staying points screened from the muck truck trajectory data refer to the groups formed by grouping the staying points that do not exceed a preset value in driving distance within a preset time screened by the matrix method and the adaptive jump method according to time; The matrix method can be used to eliminate the inactive muck truck trajectory points from the muck truck trajectory data; The adaptive jump method can identify whether the muck truck trajectory point around which multiple muck truck trajectory points in the muck truck trajectory data are located is a staying point. If so, match the staying point to the divided grid; 3. The method for analyzing the activity area of a muck truck based on trajectory data according to claim 1, wherein In S1, the process of performing clustering processing on each staying point group according to adjacent grids is as follows: Connect the staying points located in adjacent grids to obtain multiple sequentially connected staying points; Calculate the distance difference between the residence points; Statistically calculate the sum of the distance differences between each residence point and other residence points respectively, compare the sums, take the residence point with the smallest sum as the center point, and form a residence clustering point area with the maximum distance difference of the center point as the radius.

4. The method for analyzing the activity area of a muck truck based on trajectory data according to claim 3, wherein In S2, the process of region updating for the residence clustering point area includes the following steps: SA1. Take the intermediate value of the distance difference between the center points of two residence clustering point areas, use the point where the intermediate value is located as the new center point, and determine whether there is an overlapping relationship between the two residence clustering point areas. If so, go to step SA2; if not, go to step SA3; SA2. Select the radius value of one of the residence clustering point areas according to the overlapping relationship and area size between the two residence clustering point areas, add it to the intermediate value, take it as the current radius value, form a new residence clustering point area with the current radius value based on the new center point, and delete the corresponding two residence clustering point areas; SA3. Form a new residence clustering point area with the preset radius value as the radius based on the new center point, and delete the corresponding two residence clustering point areas.

5. The method for analyzing the activity area of a muck truck based on trajectory data according to claim 4, wherein In SA2, the process of selecting the radius value of one of the residence clustering point areas according to the overlapping relationship and area size between the two residence clustering point areas and adding it to the intermediate value is as follows: The overlapping relationships include intersection and inclusion; When the overlapping relationship between two residence clustering point areas is intersection, select the radius value of the residence clustering point area with the smaller area, add it to the intermediate value, and take it as the current radius value; When the overlapping relationship between two residence clustering point areas is inclusion, select the radius value of the residence clustering point area with the larger area, add it to the intermediate value, and take it as the current radius value.

6. The method for analyzing the activity area of a muck truck based on trajectory data according to claim 1, wherein, In S41, the process of sequentially obtaining the average speed values between the trajectory points of the dump trucks in the residence clustering point area according to the time sequence is as follows: Sequentially calculate the distance difference and time difference between the trajectory points of the dump trucks in the residence clustering point area according to the time sequence, and calculate through the distance difference and time difference to obtain the average speed value between two dump truck trajectory points.

7. A muck truck activity area analysis system based on trajectory data, characterized in that, Applying a method for analyzing the activity area of dump trucks based on trajectory data according to any one of claims 1-6, including: Residence clustering point area division module: Obtain multiple groups of residence points screened from the dump truck trajectory data, match the residence points in the residence point group to the divided grids, and perform clustering processing on each residence point group according to adjacent grids to obtain a residence clustering point area with the residence point as the center point; Residence clustering point area update module: Determine whether the distance difference between the center points of the residence clustering point areas exceeds the threshold. If not, perform region updating on the residence clustering point area; Dump truck trajectory data matching module: Match the dump truck trajectory data corresponding to the residence point group to the residence clustering point area, and screen the dump truck trajectory points in the current dump truck trajectory data to obtain the dump truck trajectory points located in the residence clustering point area; Resident clustering point area determination module: Perform piecewise curve fitting on the trajectory points of the muck trucks in the resident clustering point area to obtain the corresponding average speed trend curve, and determine the scene type of the resident clustering point area according to the curve characteristics of the average speed trend curve.

Images