CN106599561A - Trajectory data cleaning method and device - Google Patents

Abstract

The embodiment of the invention discloses a trajectory data cleaning method and device. The method comprises following steps: obtaining trajectory data of a vehicle; recognizing mistaken trajectory points included in the trajectory data according to latitudes and longitudes of the trajectory points of the trajectory data and/or average speed corresponding to the trajectory points, wherein the average speed corresponding to each trajectory point refers to the average speed of a trajectory section composed of the trajectory point and a former trajectory point of the trajectory point. The embodiment of the trajectory data cleaning method and device has following beneficial effects: according to latitudes and longitudes of the trajectory points of the trajectory data and/or average speed corresponding to the trajectory points, the mistaken trajectory points included in the trajectory data are recognized; by comparing adjacent trajectory points or adjacent trajectory sections, the mistaken trajectory points are determined; trajectory data is fully utilized so that judgments over the mistaken trajectory points are made more accurate.

Description

A trajectory data cleaning method and device

technical field

Embodiments of the present invention relate to big data application technologies, and in particular to a method and device for cleaning trajectory data.

Background technique

The original trajectory data is the most basic and original data uploaded by the vehicle-mounted terminal to the server. The dynamic trajectory data processing is performed by the original trajectory data through data cleaning, vehicle information matching, road matching, administrative division matching, data analysis, stop point calculation, and missing points. Multi-step processing such as calculation, trajectory segment calculation, operation index calculation, and vehicle portrait finally obtains accurate vehicle data indicators. Dynamic data processing is the foundation of big data applications. The original trajectory data is the most basic and original data uploaded by the vehicle terminal to the server. Due to vehicle terminal failure, GPS positioning failure, SIM card traffic, network transmission errors and other reasons, the trajectory data is missing or the data field is abnormal. Therefore, the most important step in trajectory data processing is data cleaning, deleting data with serious errors in latitude and longitude and GPS time, and marking data with abnormal odometer and GPS speed, so as to ensure the accuracy and authenticity of data in the subsequent processing process.

Existing dirty data cleaning is only aimed at whether a single field is within a certain threshold range, the calculation method is simple, the trajectory data is not fully utilized, the judgment of the wrong trajectory point is not accurate enough, and there is still a large amount of dirty data in the final processed data, directly It affects the calculation of stops, operation index, frequent cities, frequent routes and other data calculations.

Contents of the invention

Embodiments of the present invention provide a track data cleaning method and device, which can make full use of the track data and make the judgment of wrong track points more accurate.

In the first aspect, the embodiment of the present invention provides a trajectory data cleaning method, including:

Obtain vehicle trajectory data;

According to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point, identify the wrong track point contained in the track data, wherein the average speed corresponding to each track point refers to the relationship between the track point and the track The average speed of the track segment composed of the previous track point of the point.

In the second aspect, the embodiment of the present invention also provides a trajectory data cleaning device, including:

A trajectory data acquisition module, configured to obtain vehicle trajectory data;

The wrong track point identification module is connected to the track data acquisition module, and is used to identify the wrong track points contained in the track data according to the latitude and longitude of the track points in the track data and/or the average speed corresponding to the track points, wherein The average speed corresponding to each track point refers to the average speed of the track segment composed of the track point and the previous track point of the track point.

In the embodiment of the present invention, the wrong track point contained in the track data is identified according to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point, and the error is judged by comparing with the adjacent track point or adjacent track segment Track points, making full use of track data, making the judgment of wrong track points more accurate.

Description of drawings

FIG. 1 is a flow chart of a trajectory data cleaning method provided in Embodiment 1 of the present invention;

FIG. 2 is a flow chart of a trajectory data cleaning method provided in Embodiment 2 of the present invention;

Fig. 3 is a flow chart of a trajectory data cleaning method provided in Embodiment 3 of the present invention;

FIG. 4 is a schematic structural diagram of a trajectory data cleaning device provided in Embodiment 4 of the present invention.

detailed description

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, only some structures related to the present invention are shown in the drawings but not all structures.

Embodiment one

Fig. 1 is a flow chart of the trajectory data cleaning method provided by Embodiment 1 of the present invention. This embodiment is applicable to the situation of cleaning trajectory data. The method can be executed by a trajectory data cleaning device, which can be implemented by software and/or Realized by means of hardware, the device can be integrated into any hardware device, such as a typical user terminal device or server. The method includes:

S101. Acquire trajectory data of the vehicle.

Vehicle trajectory data includes latitude and longitude, average speed, mileage and time. The mileage can be obtained from the odometer of the vehicle, and the latitude and longitude, average speed and time can be obtained from the vehicle positioning device or the mobile terminal on the vehicle.

S102. According to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point, identify the wrong track point contained in the track data, wherein the average speed corresponding to each track point refers to the relationship between the track point and the The average speed of the track segment composed of the previous track point of this track point.

Identifying whether a track point is an error track point requires not only analyzing the latitude and longitude and/or average speed of the track point itself, but also analyzing the latitude and longitude and/or average speed of adjacent track points or adjacent track segments. When the latitude and longitude and/or average speed of the track point are greatly different from those of adjacent track points or adjacent track segments, the track point is likely to be a wrong track point.

In this embodiment, the wrong track points contained in the track data are identified according to the latitude and longitude of the track points in the track data and/or the average speed corresponding to the track points, and the wrong track is judged by comparing with adjacent track points or adjacent track segments Points, making full use of trajectory data, making the judgment of wrong trajectory points more accurate.

Embodiment two

Fig. 2 is a flow chart of the track data cleaning method provided by the second embodiment of the present invention. The second embodiment is based on the above-mentioned embodiment, and before identifying the wrong track points contained in the track data, the proportion of judging wrong track points is increased.

The trajectory data cleaning method provided in this embodiment includes the following steps: step S201, step S202 and step S203. Wherein, step S201 is the same as step S101 in the first embodiment, step S203 is the same as step S102 in the first embodiment, and the same steps will not be repeated here.

S201. Acquire trajectory data of the vehicle.

S202. If the average velocity of two adjacent trajectory points in the trajectory data is greater than the fifth average velocity threshold or the reporting time of the two adjacent trajectory points is the same, then determine that the sequence of the adjacent trajectory points is an error trajectory point ; If the ratio of the number of wrong track points to the total number of track points in the track data is greater than a ratio threshold, abandon cleaning the track data.

For example, the average speed of two adjacent track points in the track data is greater than 50m/s, and it is impossible for the vehicle to reach this speed during normal driving, so it is determined that the next sequence in the adjacent track points is the wrong track point . Any time may only correspond to one track point, and if the reporting time of two adjacent track points is the same, it is determined that the next-ranked track point among the adjacent track points is a wrong track point. If the ratio of the number of wrong track points to the total number of track points in the track data is greater than the ratio threshold, for example, greater than 20%, it means that there are many wrong track points in the track segment, and the track data is unavailable, so give up cleaning the track data , that is, discard all trajectory data.

S203. According to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point, identify the wrong track point contained in the track data, wherein the average speed corresponding to each track point refers to the relationship between the track point and the The average speed of the track segment composed of the previous track point of this track point.

In this embodiment, by judging whether the ratio of the number of wrong trajectory points in the trajectory data to the total number of trajectory points is greater than a ratio threshold, it is generally determined whether the entire segment of trajectory data is available.

Embodiment three

Fig. 3 is a flow chart of the trajectory data cleaning method provided by Embodiment 3 of the present invention. Embodiment 3 is based on the above-mentioned embodiments, and identifies the above-mentioned points according to the latitude and longitude of the trajectory points in the trajectory data and/or the average speed corresponding to the trajectory points. Wrong trajectory points included in trajectory data were optimized.

The trajectory data cleaning method provided in this embodiment includes the following steps: step S301, step S302, step S303, step S304 and step S305. Wherein, step S302 , step S303 , step S304 and step S305 are not sequential, and may not all be executed, and step S301 is the same as step S101 in Embodiment 1, and the same steps will not be repeated.

S301. Acquire trajectory data of the vehicle.

S302. If the latitude and longitude of at least one previous trajectory point before any trajectory point in the trajectory data is the same as the latitude and longitude of the trajectory point and at least one subsequent trajectory point, and the trajectory point is the same as the at least one previous trajectory point If the latitude and longitude of the track are different, then the track point is determined to be the docking drift point.

For example, if the latitude and longitude sequences of five adjacent track points are similar to AABAA, then point B is considered to be the docking drift point.

S303. For the trajectory sequence composed of multiple continuous trajectory points in the trajectory data; if the average speed corresponding to the continuous multiple trajectory points in any trajectory sequence is the first average speed threshold, and the previous bit of the trajectory sequence If the average speed of the starting point corresponding to the track point and the average speed of the end point corresponding to the first track point after the track sequence are both greater than the second average speed threshold, it is determined that all the track points in the track sequence are wrong track points, wherein the first Two average velocity thresholds are greater than the first average velocity threshold; if the average velocity of a plurality of continuous trajectory segments in any trajectory sequence is the first average velocity threshold, the corresponding starting point average velocity is less than the second average velocity threshold, corresponding The average velocity at the end of the trajectory sequence is greater than the second average velocity threshold, and the average velocity corresponding to the second trajectory point after the trajectory sequence is greater than the second average velocity threshold, then it is determined that the first trajectory point after the trajectory sequence is Wrong track point; if the average speed of multiple consecutive track segments in any track sequence is the first average speed threshold, the corresponding starting point average speed is less than the second average speed threshold, and the corresponding end point average speed is greater than the The second average speed threshold, and the average speed corresponding to the second track point after the track sequence is less than or equal to the second average speed threshold, then according to the average speed of the first and last track points in the track sequence and the preset standard speed value, which determines the number of error points included in this trajectory sequence.

In this embodiment, it is assumed that the first average speed threshold is 0, and the second average speed threshold is assumed to be 25 m/s. For example, the average velocity corresponding to multiple consecutive trajectory points in a trajectory sequence is 0, and the average velocity of the starting point corresponding to the previous trajectory point of the trajectory sequence and the average velocity of the end point corresponding to the first trajectory point after the trajectory sequence are greater than 25m/s, it means that the track sequence and the previous track point mistakenly uploaded the latitude and longitude of the previous track point repeatedly, so that the average speed calculated by the first track point before and after the track sequence is very large, then It is determined that all the track points in the track sequence are wrong track points; if the average speed of the corresponding starting point is less than 25m/s, it means that the vehicle may really stop so that the average speed of the track sequence is 0, and the corresponding end point and the track The average speed corresponding to the second track point after the sequence is greater than 25m/s, indicating that the latitude and longitude drift makes the average speed of the corresponding end point very large, then it is determined that the first track point after the track sequence is a wrong track point; if The corresponding end point average speed is greater than 25m/s, and the average speed corresponding to the second track point after the track sequence is less than or equal to 25m/s, indicating that the track sequence does not have a latitude and longitude drift error but has repeatedly uploaded the previous one by mistake. The latitude and longitude of the corresponding starting point is less than 25m/s. It is impossible to know whether the starting point corresponding to the track sequence is a wrong track point, that is, it cannot be determined that the longitude and latitude of a previous track point that was wrongly uploaded repeatedly is from this track point. Which point of the trajectory sequence starts, assuming that the vehicle advances at a constant speed at a preset standard speed value, the average speed of the end point corresponding to the trajectory sequence is proportional to the number of error points, divided by the average speed of the end point corresponding to the trajectory sequence divided by The value obtained from the preset standard speed value is used to estimate the number of error points, and so many trajectory points from the corresponding end point are error trajectory points. The preset standard speed value can be 5m/s, and the purpose of taking this smaller value is to include as many error trajectory points as possible.

S304. For the three consecutive track points in the track data, determine the corresponding gps speed threshold according to the gps speed of each track point; if the second track point and the third track point in the three continuous track points The average speed of the rear track segment composed of track points is greater than the third average speed threshold, the average speed of the first and last track points is less than or equal to the sum of the gps speed threshold and the first correction speed, and the average speed of the rear track segment minus the first and last If the difference between the average speeds of the track points is greater than the second corrected speed, it is determined that the middle track point is an error point.

Assume that the third average speed threshold is 30m/s, the first corrected speed is 3m/s, and the second corrected speed is 2m/s. Assuming there are three consecutive track points P1, P2 and P3, the second track point P2 and the third track point P3 form the rear track segment S23, and the first and last track points form the entire track segment S13. gps_speed1, gps_speed2 and gps_speed3 are the gps speeds of track points P1, P2 and P3 respectively, gps_speed_limit is the corresponding gps speed threshold, the mode of determining the gps speed threshold can be gps_speed_limit=min(max(gps_speed1, gps_speed2, gps_speed3,15), for example, 27), among which 15m/s and 27m/s can also be changed to other values according to needs. If the average speed gps_speed_ave23 of S23 is greater than 30m/s, and the average speed gps_speed ave13 of S13 is less than or equal to gps_speed_limit+3, and gps_speed_ave23-gps_speed_ave13>2, it is determined that the middle track point P2 is an error point. The error of P2 is caused by the time latitude and longitude matching error, that is, the latitude and longitude of point P2 is delayed in uploading, so that the difference between the average speed of the latter segment and the average speed of the entire segment exceeds the threshold. If the average speed change is less than 2m/s, it may be true that the entire section is overspeeded, and at this time it cannot be determined that P2 is the wrong trajectory point.

S305. Select at least three consecutive points whose speed is less than the fourth average speed threshold as the reference trajectory sequence; determine whether the average speed corresponding to the trajectory point before or after the reference trajectory sequence is greater than the A fourth average speed threshold, and according to the determination result, identify the wrong track points included in the track points before or after the reference track sequence.

Assuming that the fourth average speed threshold is 35m/s, select at least three consecutive points with a speed less than 35m/s as the reference trajectory sequence; if the average speed between the end point of the reference trajectory sequence and the first point behind is less than 35m/s , then the first point behind is not an error track point and the average speed of the second point behind is calculated based on the first point behind, if the average speed between the end point of the reference track sequence and the first point behind is greater than 35m/ s, then the first point behind is the wrong track point and the average speed of the second point behind is still calculated based on the end point of the reference track sequence, and so on, it is also akin.

In this embodiment, wrong track points are identified by comparing the longitudes and latitudes or average speeds of adjacent track points or track segments, and full use of track data makes the judgment of wrong track points more accurate.

Embodiment four

FIG. 4 is a schematic structural diagram of a trajectory data cleaning device provided in Embodiment 4 of the present invention, and the device is used to implement the trajectory data cleaning method in the above-mentioned embodiment. The device includes: a trajectory data acquisition module 401 and a wrong trajectory point identification module 402 .

The track data acquisition module 401 is used to acquire track data of the vehicle.

The wrong track point identification module 402 is connected to the track data acquisition module 401, and is used to identify the wrong track point contained in the track data according to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point , wherein the average speed corresponding to each track point refers to the average speed of the track segment composed of the track point and the previous track point of the track point.

In this embodiment, the wrong track points contained in the track data are identified according to the latitude and longitude of the track points in the track data and/or the average speed corresponding to the track points, and the wrong track is judged by comparing with adjacent track points or adjacent track segments Points, making full use of trajectory data, making the judgment of wrong trajectory points more accurate.

Further, the wrong track point identification module is specifically configured to: if the latitude and longitude of at least one front track point before any track point in the track data is the same as the longitude and latitude of the track point and at least one back track point after that, And the track point is different from the latitude and longitude of the at least one previous track point, then it is determined that the track point is a docking drift point.

Further, the wrong track point identification module is specifically used to: aim at the track sequence composed of multiple continuous track points in the track data; if the average speed corresponding to the multiple continuous track points in any track sequence is the first The average speed threshold, and the average speed of the starting point corresponding to the previous track point of the track sequence and the average speed of the end point corresponding to the first track point after the track sequence are both greater than the second average speed threshold, then determine that all in the track sequence The trajectory points are all wrong trajectory points, wherein the second average velocity threshold is greater than the first average velocity threshold; if the average velocity of a plurality of consecutive trajectory segments in any trajectory sequence is the first average velocity threshold, the corresponding The average velocity at the starting point is less than the second average velocity threshold, the corresponding average velocity at the end point is greater than the second average velocity threshold, and the average velocity corresponding to the second track point after the trajectory sequence is greater than the second average velocity threshold , then it is determined that the first track point after the track sequence is an error track point; if the average speed of a plurality of consecutive track segments in any track sequence is the first average speed threshold, the corresponding starting point average speed is less than the described The second average speed threshold, the corresponding terminal average speed is greater than the second average speed threshold, and the average speed corresponding to the second track point after the track sequence is less than or equal to the second average speed threshold, then according to the track The average speed of the first and last track points in the sequence and the preset standard speed value determine the number of error points contained in the track sequence.

Further, the wrong track point identification module is specifically used to: for three consecutive track points in the track data, determine the corresponding gps speed threshold according to the gps speed of each track point; if the three continuous tracks The average speed of the rear track segment composed of the second track point and the third track point in the points is greater than the third average speed threshold, and the average speed of the first and last track points is less than or equal to the sum of the GPS speed threshold and the first corrected speed , and the difference between the average velocity of the rear trajectory segment minus the average velocity of the first and last trajectory points is greater than the second corrected velocity, then it is determined that the middle trajectory point is an error point.

Further, the wrong track point identification module is specifically used to: select at least three consecutive points whose speed is less than the fourth average speed threshold as a reference track sequence; determine the track before the reference track sequence according to the reference track sequence Whether the average speed corresponding to the first or later track points is greater than the fourth average speed threshold, and identify the wrong track points contained in the previous track points or the last track points of the reference track sequence according to the determination result.

In this embodiment, wrong track points are identified by comparing the longitudes and latitudes or average speeds of adjacent track points or track segments, and full use of track data makes the judgment of wrong track points more accurate.

Further, the track data cleaning device also includes a wrong track point ratio judging module, which is respectively connected to the track data acquisition module and the wrong track point identification module, and is used to execute before identifying the wrong track points contained in the track data. : if the average speed of two adjacent track points in the track data is greater than the fifth average speed threshold or the reporting time of two adjacent track points is the same, then it is determined that the next-ranked track point in the adjacent track points is an error track point; If the ratio of the number of wrong track points to the total number of track points in the track data is greater than a ratio threshold, then abandon cleaning the track data.

In this embodiment, by judging whether the ratio of the number of wrong trajectory points in the trajectory data to the total number of trajectory points is greater than a ratio threshold, it is generally determined whether the entire segment of trajectory data is available.

The trajectory data cleaning device provided in the embodiment of the present invention can be used to execute the trajectory data cleaning method provided in any embodiment of the present invention, and has corresponding functions and beneficial effects for performing the method.

Note that the above are only preferred embodiments of the present invention and applied technical principles. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and can also include more other equivalent embodiments without departing from the concept of the present invention, and the present invention The scope is determined by the scope of the appended claims.

Claims (10)

1. A trajectory data cleaning method, characterized in that, comprising: Obtain vehicle trajectory data; According to the latitude and longitude of the track point in the track data and/or the average speed corresponding to the track point, identify the wrong track point contained in the track data, wherein the average speed corresponding to each track point refers to the relationship between the track point and the track The average speed of the track segment composed of the previous track point of the point. 2. The method according to claim 1, wherein, according to the latitude and longitude of the track point in the track data, identifying the wrong track point contained in the track data comprises: If the latitude and longitude of at least one previous trajectory point before any trajectory point in the trajectory data is the same as the latitude and longitude of the trajectory point and at least one subsequent trajectory point, and the latitude and longitude of the trajectory point and the at least one previous trajectory point different, it is determined that the track point is a docking drift point. 3. The method according to claim 1, wherein, according to the average velocity corresponding to the track point in the track data, identifying the wrong track point contained in the track data comprises: A trajectory sequence composed of a plurality of continuous trajectory points in the trajectory data; If the average velocity corresponding to multiple consecutive trajectory points in any trajectory sequence is the first average velocity threshold, and the average velocity of the starting point corresponding to the previous trajectory point of the trajectory sequence corresponds to the first trajectory point after the trajectory sequence If the average speed at the end point is greater than the second average speed threshold, then it is determined that all the trajectory points in the trajectory sequence are error trajectory points, wherein the second average speed threshold is greater than the first average speed threshold; If the average speed of multiple consecutive trajectory segments in any trajectory sequence is the first average speed threshold, the corresponding starting point average speed is less than the second average speed threshold, and the corresponding end point average speed is greater than the second average speed threshold, and the average speed corresponding to the second track point after the track sequence is greater than the second average speed threshold, then it is determined that the first track point after the track sequence is an error track point; If the average speed of multiple consecutive trajectory segments in any trajectory sequence is the first average speed threshold, the corresponding starting point average speed is less than the second average speed threshold, and the corresponding end point average speed is greater than the second average speed threshold, and the average speed corresponding to the second track point after the track sequence is less than or equal to the second average speed threshold, then according to the average speed of the first and last track points in the track sequence and the preset standard speed value, determine the The number of error points included in the trajectory sequence. 4. The method according to claim 1, wherein, according to the average velocity corresponding to the track point in the track data, identifying the wrong track point contained in the track data comprises: For three consecutive track points in the track data, determine the corresponding gps speed threshold according to the gps speed of each track point; If the average speed of the rear track segment consisting of the second track point and the third track point in the three consecutive track points is greater than the third average speed threshold, the average speed of the first and last track points is less than or equal to the GPS speed The sum of the threshold and the first corrected speed, and the difference between the average speed of the rear track segment minus the average speed of the first and last track points is greater than the second corrected speed, then it is determined that the middle track point is an error point. 5. The method according to claim 1, wherein, according to the average speed corresponding to the track points in the track data, identifying the wrong track points contained in the track data comprises: Selecting at least three consecutive points whose speed is less than the fourth average speed threshold as a reference trajectory sequence; according to the reference trajectory sequence, determine whether the average speed corresponding to the trajectory point before or after the reference trajectory sequence is greater than the A fourth average speed threshold, and according to the determination result, identify the wrong track points included in the track points before or after the reference track sequence. 6. The method according to claim 1, wherein, before identifying the wrong track points contained in the track data, further comprising: If the average velocity of two adjacent trajectory points in the trajectory data is greater than the fifth average velocity threshold or the reporting time of the two adjacent trajectory points is the same, it is determined that the sequence of the adjacent trajectory points is an error trajectory point; If the ratio of the number of wrong track points to the total number of track points in the track data is greater than a ratio threshold, then abandon cleaning the track data. 7. A trajectory data cleaning device, characterized in that, comprising: A trajectory data acquisition module, configured to obtain vehicle trajectory data; The wrong track point identification module is connected to the track data acquisition module, and is used to identify the wrong track points contained in the track data according to the latitude and longitude of the track points in the track data and/or the average speed corresponding to the track points, wherein The average speed corresponding to each track point refers to the average speed of the track segment composed of the track point and the previous track point of the track point. 8. The track data cleaning device according to claim 7, wherein the wrong track point identification module is specifically used for: If the latitude and longitude of at least one previous trajectory point before any trajectory point in the trajectory data is the same as the latitude and longitude of the trajectory point and at least one subsequent trajectory point, and the latitude and longitude of the trajectory point and the at least one previous trajectory point different, it is determined that the track point is a docking drift point. 9. The track data cleaning device according to claim 7, wherein the wrong track point identification module is specifically used for: A trajectory sequence composed of a plurality of continuous trajectory points in the trajectory data; If the average velocity corresponding to multiple consecutive trajectory points in any trajectory sequence is the first average velocity threshold, and the average velocity of the starting point corresponding to the previous trajectory point of the trajectory sequence corresponds to the first trajectory point after the trajectory sequence If the average speed at the end point is greater than the second average speed threshold, then it is determined that all the trajectory points in the trajectory sequence are error trajectory points, wherein the second average speed threshold is greater than the first average speed threshold; If the average speed of multiple consecutive trajectory segments in any trajectory sequence is the first average speed threshold, the corresponding starting point average speed is less than the second average speed threshold, and the corresponding end point average speed is greater than the second average speed threshold, and the average speed corresponding to the second track point after the track sequence is greater than the second average speed threshold, then it is determined that the first track point after the track sequence is an error track point; If the average speed of multiple consecutive trajectory segments in any trajectory sequence is the first average speed threshold, the corresponding starting point average speed is less than the second average speed threshold, and the corresponding end point average speed is greater than the second average speed threshold, and the average speed corresponding to the second track point after the track sequence is less than or equal to the second average speed threshold, then according to the average speed of the first and last track points in the track sequence and the preset standard speed value, determine the The number of error points included in the trajectory sequence. 10. The track data cleaning device according to claim 7, wherein the wrong track point identification module is specifically used for: For three consecutive track points in the track data, determine the corresponding gps speed threshold according to the gps speed of each track point; If the average speed of the rear track segment consisting of the second track point and the third track point in the three consecutive track points is greater than the third average speed threshold, the average speed of the first and last track points is less than or equal to the GPS speed The sum of the threshold and the first corrected speed, and the difference between the average speed of the rear track segment minus the average speed of the first and last track points is greater than the second corrected speed, then it is determined that the middle track point is an error point.