Method, device and equipment for vehicle travel time prediction and vehicle

Abstract

This application provides a method, apparatus, device, and vehicle for predicting vehicle travel time, relating to the field of vehicle traffic technology. The method includes: acquiring the vehicle's historical travel times and historical travel trajectories; selecting multiple first moments where the dispersion of the vehicle's travel times meets preset conditions based on the historical travel times; selecting multiple second moments where there is overlap among the multiple travel trajectories based on the historical travel trajectories; calculating the similarity between the first and second moments and third moments in a preset dataset to obtain multiple fourth moments with a similarity greater than or equal to a preset value; and predicting the vehicle's target travel time based on the multiple fourth moments. This application can reduce the deviation between the predicted travel time and the user's actual travel time, making the predicted travel time closer to the user's actual travel time, and improving the accuracy of predicting vehicle travel time.

Description

Technical Field

[0001] This application relates to the field of vehicle transportation technology, and more specifically, to a method, apparatus, device, and vehicle for predicting vehicle travel time in the field of vehicle transportation technology. Background Technology

[0002] With the rapid development of vehicle control technology, it is now possible to predict vehicle travel time, thereby facilitating users' travel.

[0003] In related technologies, any of the following models—clustering models, neural network models, or decision trees—is used to predict vehicle travel time. For example, it may predict that a user might drive between 07:00 and 09:00 or between 17:00 and 19:00. However, the predicted travel time may deviate significantly from the user's actual travel time. Summary of the Invention

[0004] This application provides a method, apparatus, device, and vehicle for predicting vehicle travel time. The method can reduce the deviation between the predicted travel time and the user's actual travel time, making the predicted travel time more consistent with the user's actual travel time, and improving the accuracy of predicting vehicle travel time.

[0005] Firstly, a method for predicting vehicle travel time is provided. The method includes: acquiring historical travel data of the vehicle, including historical travel times and historical travel trajectories; filtering out multiple first moments where the dispersion of the vehicle's travel times meets preset conditions based on historical travel times; filtering out multiple second moments where there is overlap among the multiple travel trajectories of the vehicle based on historical travel trajectories; calculating the similarity between the first and second moments and a third moment in a preset dataset to obtain multiple fourth moments with a similarity greater than or equal to a preset value; and predicting the target travel time of the vehicle based on the multiple fourth moments.

[0006] In the embodiments of this application, multiple first moments that meet preset conditions in terms of the dispersion of vehicle travel times can be selected based on historical travel times; multiple second moments that have overlapping parts in multiple vehicle travel trajectories can be selected based on historical travel trajectories; by selecting two types of moments (i.e., first moments and second moments) that meet the two conditions to predict vehicle travel times, the error caused by the singularity of predicting vehicle travel times due to one type of moment is avoided. That is, using the travel times predicted by two types of moments to predict historical travel times can reduce the deviation between the predicted travel time and the user's actual travel time, so that the predicted travel time is closer to the user's actual travel time, thereby improving the accuracy of predicting vehicle travel times.

[0007] In conjunction with the first aspect, in some implementations of the first aspect, the target travel time of the vehicle is predicted based on multiple fourth moments, including: obtaining the similarity between the fourth moment and the third moment; and using the similarity as a weight to perform a weighted average of the multiple fourth moments to obtain the target travel time of the vehicle.

[0008] In the embodiments of this application, similarity can be used as a weight to perform a weighted average on multiple fourth time points to obtain the vehicle's target travel time. The weighted average process is a trend prediction method for the user's future driving vehicle travel time. Since the larger the weight, the greater the impact on the predicted target travel result, the similarity between the fourth time point and the third time point is used as a weight to perform a weighted average on multiple fourth time points. This increases the impact of the first time point with high dispersion and the second time point with the highest similarity to historical travel trajectories and overlapping trajectories on the final predicted target travel time, and reduces the impact of the filtered-out time points with low dispersion and the time points corresponding to low similarity to historical travel trajectories on the predicted target travel time. As a result, the predicted target travel time is more in line with the user's travel needs, further improving the accuracy of the predicted target travel time.

[0009] In combination with the first aspect and the above implementation methods, in some implementation methods of the first aspect, the historical trip data also includes user scheduling information for vehicle components. The method further includes: determining a target time period including the target travel time based on the target travel time; obtaining user scheduling information for vehicle components within the target time period based on the historical trip data; and scheduling the target components of the vehicle based on the scheduling information.

[0010] In the embodiments of this application, target components of the vehicle are scheduled according to scheduling information; by determining the scheduling information of target components in the vehicle during the target time period when the user drove the vehicle in the past, the target components in the vehicle are automatically scheduled in the future target time period, so that the target components in the vehicle can be put into operation in advance, thereby realizing the automatic adjustment of the vehicle's driving environment in advance and improving the user's driving experience.

[0011] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, multiple first moments that meet the preset conditions in terms of the dispersion of vehicle travel time are selected based on historical travel time, including: obtaining abnormal travel moments of vehicles based on the dispersion of historical travel time at each moment; removing abnormal travel moments from historical travel time to obtain multiple first moments.

[0012] In the embodiments of this application, abnormal travel times can be removed from historical travel times, and the historical travel times after removing abnormal travel times can be used as multiple first times. Since abnormal travel times are removed from historical travel times, the multiple first times used to predict vehicle travel times are more accurate. Therefore, based on the more accurate multiple first times, the predicted vehicle travel time can be more accurate.

[0013] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, multiple first moments whose dispersion of vehicle travel times meets preset conditions are selected based on historical travel times. This includes: obtaining multiple frequency values ​​based on the frequency of each travel moment in historical travel times; sorting the multiple frequency values ​​sequentially and selecting multiple target frequency values ​​that meet preset frequencies from among the multiple frequency values; obtaining the travel times corresponding to each of the multiple target frequency values ​​to obtain multiple fifth moments; and taking the moment whose dispersion meets preset conditions among the multiple fifth moments as the first moment.

[0014] In the embodiments of this application, the moment in which the dispersion of a plurality of fifth moments meets the preset conditions can be taken as the first moment. Since the frequency of travel of a large number of historical travel times is filtered first, the amount of data processing during the filtering of the dispersion of the fifth moments can be reduced, thereby improving the efficiency of filtering the first moment. Based on improving the efficiency of filtering the first moment, the efficiency of predicting the travel time of vehicles is further improved.

[0015] Combining the first aspect and the above implementation methods, in some implementation methods of the first aspect, multiple second moments with overlapping parts are selected from multiple travel trajectories of a vehicle based on historical travel trajectories, including: calculating the similarity between each trajectory in the historical travel trajectory; obtaining multiple travel trajectories with the highest similarity based on the similarity between the trajectories; performing overlap processing on multiple travel trajectories to obtain multiple target trajectories with overlapping parts; and determining the travel time corresponding to the multiple target trajectories as the second moment.

[0016] In the embodiments of this application, the time corresponding to the historical travel trajectory with the highest similarity and overlapping trajectory is determined as the second time, and historical travel trajectories with large differences are filtered out, that is, abnormal travel trajectories in the historical travel trajectories are filtered out. Since abnormal travel trajectories in the historical travel trajectories are filtered out, the multiple second times used to predict the travel time of the vehicle are more accurate. Therefore, based on the fact that the multiple second times are more accurate, the predicted travel time of the vehicle can also be more accurate.

[0017] In combination with the first aspect and the above implementation methods, in some implementation methods of the first aspect, obtaining the vehicle's historical trip data includes: collecting the driver's identity information; classifying the vehicle's historical trip data according to the driver's identity information to obtain the classification result; and obtaining the historical trip data corresponding to the target user according to the classification result.

[0018] In the embodiments of this application, the historical trip data of the vehicle is categorized based on the collected identity information of the driver user. Since the historical trip data corresponding to the same driver user is categorized together, it can avoid large deviations in the predicted travel time of the vehicle due to different driving habits of different users. Furthermore, by obtaining the historical trip data corresponding to the target user based on the categorization results, the accuracy of the predicted travel time of the vehicle can be improved.

[0019] Secondly, an apparatus for predicting vehicle travel time is provided. The apparatus includes: an acquisition module for acquiring historical travel data of a vehicle, including historical travel times and historical travel trajectories; a first filtering module for filtering multiple first moments where the dispersion of vehicle travel times meets preset conditions based on historical travel times; a second filtering module for filtering multiple second moments where there is overlap among multiple travel trajectories of the vehicle based on historical travel trajectories; a calculation module for calculating the similarity between the first and second moments and a third moment in a preset dataset to obtain multiple fourth moments with a similarity greater than or equal to a preset value; and a prediction module for predicting the target travel time of the vehicle based on the multiple fourth moments.

[0020] In conjunction with the second aspect, in some implementations of the second aspect, the target travel time of the vehicle is predicted based on multiple fourth moments. Specifically, the prediction module is used to: obtain the similarity between the fourth moment and the third moment; and use the similarity as a weight to perform a weighted average of the multiple fourth moments to obtain the target travel time of the vehicle.

[0021] In conjunction with the second aspect and the above-described implementation methods, in some implementation methods of the second aspect, the device further includes a scheduling module, which is used to determine a target time period including the target travel time based on the target travel time; obtain scheduling information of users for components in the vehicle within the target time period based on historical travel data; and schedule the target components of the vehicle based on the scheduling information.

[0022] Combining the second aspect and the above implementation methods, in some implementation methods of the second aspect, multiple first moments in which the dispersion of vehicle travel time meets preset conditions are filtered out based on historical travel time. The first filtering module is specifically used to: obtain abnormal travel moments of vehicles based on the dispersion of historical travel time at each moment; remove abnormal travel moments from historical travel time to obtain multiple first moments.

[0023] Combining the second aspect and the above implementation methods, in some implementation methods of the second aspect, multiple first moments whose dispersion of vehicle travel times meets preset conditions are selected based on historical travel times. The first filtering module is specifically used to: obtain multiple frequency values ​​based on the frequency of each travel moment in the historical travel time; sort the multiple frequency values ​​in sequence and select multiple target frequency values ​​that meet the preset frequency from the multiple frequency values; obtain the travel time corresponding to each of the multiple target frequency values ​​to obtain multiple fifth moments; and take the moment whose dispersion meets the preset conditions among the multiple fifth moments as the first moment.

[0024] Combining the second aspect and the above implementation methods, in some implementation methods of the second aspect, based on historical travel trajectories, multiple second moments with overlapping parts are selected from multiple travel trajectories of the vehicle. The second filtering module is specifically used to: calculate the similarity between each trajectory in the historical travel trajectory; obtain multiple travel trajectories with the highest similarity based on the similarity between the trajectories; perform overlap processing on multiple travel trajectories to obtain multiple target trajectories with overlapping parts; and determine the travel time corresponding to the multiple target trajectories as the second moment.

[0025] Combining the second aspect and the above implementation methods, in some implementation methods of the second aspect, the historical trip data of the vehicle is obtained. The acquisition module is specifically used to: collect the identity information of the driver user; classify the historical trip data of the vehicle according to the identity information of the driver user to obtain the classification result; and obtain the historical trip data corresponding to the target user according to the classification result.

[0026] In one implementation, the device for predicting vehicle travel time can be a cloud device (e.g., a cloud server), a vehicle, or a vehicle-mounted system.

[0027] Thirdly, a device for predicting vehicle travel time is provided, the device comprising: a memory for storing executable program code; and a processor for calling and running the executable program code from the memory, causing the device to perform the method described in the first aspect or any possible implementation thereof.

[0028] Fourthly, a vehicle is provided that includes a device for predicting vehicle travel time, causing the vehicle to perform the method described in the first aspect or any possible implementation thereof. Attached Figure Description

[0029] Figure 1 This is a schematic diagram illustrating a scenario of cloud server and vehicle interaction provided in an embodiment of this application;

[0030] Figure 2 This is a flowchart illustrating the method for predicting vehicle travel time provided in an embodiment of this application;

[0031] Figure 3 This is a schematic diagram showing the distribution of historical travel times at each moment, as provided in the embodiments of this application.

[0032] Figure 4 This is a schematic diagram showing the comparison of historical travel trajectories provided in the embodiments of this application;

[0033] Figure 5 This is a schematic diagram of the structure for predicting vehicle travel time provided in an embodiment of this application;

[0034] Figure 6 This is a schematic diagram of the structure of the vehicle travel time prediction device provided in the embodiments of this application;

[0035] Figure 7 This is a schematic diagram of the vehicle structure provided in the embodiments of this application. Detailed Implementation

[0036] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0037] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0038] Figure 1 This is a schematic diagram illustrating the interaction between a cloud server and a vehicle, as provided in an embodiment of this application.

[0039] For example, such as Figure 1 As shown, Figure 1 (a) includes vehicle 110 and cloud server 120. Vehicle 110 and cloud server 120 can exchange information.

[0040] For example, vehicle 110 can transmit user driving data to cloud server 120. After receiving the driving data sent by the vehicle, cloud server 120 can process it through methods such as... Figure 1 The model shown in (b) can be any of the clustering model, neural network model, or decision tree model to predict vehicle travel time. For example, it might predict that the vehicle will be driven to work during the morning rush hour (07:00-09:00) and the evening rush hour (17:00-19:00) from Monday to Friday. However, when predicting vehicle travel time using a single model such as a clustering model, neural network model, or decision tree, the predicted travel time may deviate significantly from the user's actual travel time.

[0041] To address the aforementioned issues, this application proposes a method for predicting vehicle travel time, which will be explained in detail below.

[0042] Figure 2 This is a flowchart illustrating the vehicle travel time prediction method provided in this application embodiment. The method can be... Figure 1 Vehicle 110 was executed.

[0043] For example, such as Figure 2 As shown, the method 200 includes the following implementation process:

[0044] S210, obtain the vehicle's historical trip data.

[0045] Historical trip data is used to represent the vehicle's historical travel history, which may include the vehicle's historical travel time and historical travel trajectory. It should be understood that, to avoid significant deviations in predicting vehicle travel times, driving data from a period preceding the current moment can be used, such as three months or two months. This avoids a large gap between the historical trip data acquisition period and the current moment, which could lead to excessive discrepancies between the predicted travel time and the actual travel time.

[0046] For example, a vehicle can send its trip data up to the current moment to a cloud server, so that the cloud server can obtain the vehicle's historical travel time and historical travel trajectory based on the historical trip data.

[0047] For example, the historical travel data obtained by the cloud server includes the vehicle's historical travel time and historical travel trajectory.

[0048] For example, when acquiring historical trip data of a vehicle, the identity information of the driver can be collected; based on the identity information of the driver, the historical trip data of the vehicle can be categorized to obtain the categorization results; based on the categorization results, the historical trip data corresponding to the target user can be obtained.

[0049] Among them, the driver user refers to the user who has authenticated their identity with the vehicle and has a legal driving identity for the vehicle; and the identity information may include one or more of the driver user's image information, fingerprint information, iris information and voice information.

[0050] The classification results may include the historical trip data of the driving user.

[0051] For example, driver image information can be used to identify drivers. After collecting driver image information, intelligent recognition can be performed on the image information to identify the driver's identity, for example, driver 1 and driver 2, indicating that the vehicle is used by two drivers. Furthermore, the historical trip data of driver 1 and driver 2 can be categorized to obtain their respective categorization results. For example, driver 1's historical trip data can be categorized in categorization result 1, and driver 2's historical trip data can be categorized in categorization result 2.

[0052] Optionally, when predicting the travel time of a vehicle, the driver who last used the vehicle before the current time can be used as the target user (e.g., driver 1), and the historical trip data corresponding to driver 1 in classification result 1 can be obtained.

[0053] Optionally, the vehicle can also receive prediction instructions sent by the driver in advance. When sending the prediction instructions, the driver can select the target user, for example, driver 2 as the target user. When predicting the vehicle's travel time, the historical trip data corresponding to driver 2 in classification result 2 can be obtained.

[0054] In the embodiments of this application, the historical trip data of the vehicle is categorized based on the collected identity information of the driver user. Since the historical trip data corresponding to the same driver user is categorized together, it can avoid large deviations in the predicted travel time of the vehicle due to different driving habits of different users. Furthermore, by obtaining the historical trip data corresponding to the target user based on the categorization results, the accuracy of the predicted travel time of the vehicle can be improved.

[0055] S220 filters out multiple first moments based on historical travel times, where the dispersion of vehicle travel times meets preset conditions.

[0056] The degree of dispersion can be represented by the number of times a historical travel time exists at each moment. The preset condition can refer to historical travel times with a high degree of dispersion.

[0057] Optionally, drivers can set the dispersion level based on the vehicle's historical trip data over a period of time (e.g., three months); or, the vehicle can set the dispersion level based on the acquired historical trip data. See Table 1:

[0058] Table 1

[0059] Number of trips at the same time Dispersion Less than or equal to 5 times Low dispersion More than 5 times and less than 20 times The degree of dispersion is generally 20 times or more High degree of dispersion

[0060] Referring to Table 1, if a vehicle makes 5 or fewer trips at the same time, the dispersion level at that time is low; if a vehicle makes more than 5 but less than 20 trips at the same time, the dispersion level is moderate; and if a vehicle makes more than 20 trips at the same time, the dispersion level is high. In other words, there is a correlation between the number of trips a vehicle makes at the same time and the dispersion level; the more trips a vehicle makes at the same time, the higher the dispersion level at that time. It should be understood that the number of trips and the dispersion level at the same time in Table 1 can be set based on the start and end times of the obtained historical trip data, and this embodiment does not limit this.

[0061] Figure 3 This is a schematic diagram showing the distribution of historical travel times at each moment, as provided in the embodiments of this application.

[0062] For example, such as Figure 3 As shown, Figure 3 In (a), the black dots 10, 11 and 12 indicate that the vehicle departed at 01:10 and the number of times the vehicle departed at 01:10 was 3. Compared with Table 1, the dispersion of the vehicle's departure at 01:10 is low. Figure 3In (a), the black dots 20, 21, 22, 23, 24, 25 and 26 can indicate that the vehicle departed at 03:15 and the vehicle departed 6 times at 01:10. Compared with Table 1, the dispersion of the vehicles departing at 03:15 is generally moderate. Figure 3 In (a), black dots 30 to 51 indicate that the vehicle departed at 08:35 and the number of times the vehicle departed at 08:35 was 21. Compared with Table 1, the degree of dispersion of the vehicle's departure at 08:35 is high. Figure 3 The dispersion of 10:00, 12:05, 15:11, 18:27 and 23:55 in (a) is the same as that analyzed above.

[0063] It should be understood that Figure 3 (a) in this example is merely illustrative. This application embodiment can obtain the dispersion of a vehicle's travel time at any time of day, and there is no limit to the number of times it can be obtained within a fixed duration. For example, in the period from 00:00 to 03:00, besides... Figure 3 The historical trip count of the vehicle shown in (a) at 01:10 can also be obtained at any time such as 00:33, 01:55 and 02:45. This application embodiment does not limit this.

[0064] For example, based on the obtained historical travel times, multiple first moments that meet preset conditions regarding the dispersion of vehicle travel times can be selected; such as... Figure 3 As shown in (a), the times with high dispersion can be filtered out as 08:35, 12:05 and 18:27.

[0065] Optionally, if the only time with a high degree of dispersion is 08:35, then 08:35 will be taken as the first time.

[0066] In one possible implementation, to ensure the accuracy of vehicle travel time prediction, abnormal travel times can be obtained based on the dispersion of historical travel times at each moment; abnormal travel times are then removed from historical travel times to obtain multiple first moments.

[0067] Abnormal travel times can refer to the travel times of vehicles with low dispersion.

[0068] refer to Figure 3In step (a), the dispersion of historical travel times at each moment can be filtered using an anomaly detection algorithm (Locally Selective Combination In Parallel Outlier Ensembles, LSCP) to identify abnormal travel times. These abnormal travel times are then removed from the historical travel time list. An adaptive density clustering model is then used to stabilize these historical travel times (e.g., 01:10, 03:15, 08:55, and 12:05), resulting in a more stable time (which can be denoted as "first moment"). Because the adaptive density clustering model learns continuously during the prediction of vehicle travel times, it provides personalized predictions. Therefore, predicting vehicle travel times using this model can better meet the travel needs of different users.

[0069] Optionally, abnormal travel times can be filtered multiple times. For example, LSCP can be used to... Figure 3 (a) The dispersion of multiple times was used for the first screening, and 23:55 was found to be an abnormal travel time. Removing 23:55 from the historical travel times yielded the following results: Figure 3 Multiple first moments are shown in (b); LSCP can also be used to... Figure 3 (b) The dispersion of multiple times was used for a second screening, and 01:10 was identified as an abnormal travel time. 01:10 was removed from the historical travel times, and adaptive density clustering was used to stabilize the historical travel times after removing the abnormal travel time. This yielded the following results: Figure 3 Multiple first moments are shown in (c).

[0070] Alternatively, LSCP can be used to... Figure 3 When filtering for the dispersion of multiple time points in (a), multiple anomalous time points can be obtained through a single filtering; for example, using LSCP to analyze the dispersion of multiple time points. Figure 3 (a) By filtering the dispersion of multiple times, 01:10 and 23:55 were identified as abnormal travel times. These two times were removed from the historical travel times. Adaptive density clustering was then used to stabilize the historical travel times after removing the abnormal travel times, resulting in the following... Figure 3 Multiple first moments are shown in (c).

[0071] In the embodiments of this application, abnormal travel times can be removed from historical travel times, and the historical travel times after removing abnormal travel times can be used as multiple first times. Since abnormal travel times are removed from historical travel times, the multiple first times used to predict vehicle travel times are more accurate. Therefore, based on the more accurate multiple first times, the predicted vehicle travel time can be more accurate.

[0072] In another possible implementation, multiple frequency values ​​can be obtained based on the frequency of each travel time in the historical travel time; the multiple frequency values ​​are sorted in sequence, and multiple target frequency values ​​that meet the preset frequency are selected from the multiple frequency values; the travel time corresponding to each of the multiple target frequency values ​​is obtained to obtain multiple fifth times; the time when the dispersion of the multiple fifth times meets the preset condition is taken as the first time.

[0073] The preset frequency can refer to a specific moment when the number of trips at multiple times falls within the top 25%; or a specific moment when the number of trips at multiple times falls within the top 10%. It should be understood that the order of the times with the most trips among the multiple times can be set according to needs, and this application embodiment does not limit this.

[0074] For example, the frequency of each travel time in historical travel time can be counted to obtain multiple frequency values; the multiple frequency values ​​can be sorted in sequence, and multiple target frequency values ​​that meet the preset frequency can be selected; the travel time corresponding to each of the multiple target frequency values ​​can be obtained to obtain multiple fifth times.

[0075] refer to Figure 3 (a) in the table provides the frequency values ​​for multiple travel times, as detailed in Table 2.

[0076] Table 2

[0077] Historical travel time Frequency value 01:10 3 03:15 6 08:55 21 10:00 9 12:05 34 15:11 7 18:27 63 23:55 1

[0078] Referring to Table 2, the frequency values ​​corresponding to different travel times can be obtained. For example, the frequency value corresponding to a travel time of 01:10 is 3, the frequency value corresponding to a travel time of 03:15 is 6, the frequency value corresponding to a travel time of 08:55 is 21, and so on. The frequency values ​​corresponding to other travel times in Table 2 are shown in Table 2 and will not be introduced one by one here.

[0079] Furthermore, after obtaining the frequency values ​​corresponding to different travel times, these frequency values ​​are sorted in descending order, as detailed in Table 3:

[0080] Table 3

[0081] Historical travel time Frequency value Sort 18:27 63 1 12:05 34 2 08:55 21 3 10:00 9 4 15:11 7 5 03:15 6 6 01:10 3 7 23:55 1 8

[0082] Referring to Table 2, the frequency values ​​corresponding to different travel times can be ranked. For example, the frequency value 3 corresponding to travel time 01:10 is ranked as 1, the frequency value 6 corresponding to travel time 03:15 is ranked as 6, the frequency value 21 corresponding to travel time 08:55 is ranked as 3, and so on. The frequency values ​​corresponding to other travel times are shown in Table 3, and will not be described in detail here. It should be understood that the ranking of the frequency values ​​corresponding to travel times can vary depending on vehicle usage, and this application embodiment does not limit this.

[0083] Optionally, the frequency values ​​ranked in the top 25% of Table 3 (which can be denoted as "target frequency values") can be filtered out using the interquartile range detection model, i.e., the frequency values ​​ranked 1 and 2. The frequency values ​​ranked in the top 25% of Table 3 are 63 and 34, respectively. Then, the travel times corresponding to frequency values ​​63 and 34 are obtained. The travel time corresponding to frequency value 63 is 18:27, and the travel time corresponding to frequency value 34 is 12:05. The travel times 18:27 and 12:05 can be used as the fifth time.

[0084] Optionally, if the frequency value obtained in the top 25% of Table 3 is only 63, then the departure time 18:27 corresponding to the frequency value 63 is taken as the fifth time.

[0085] It should be understood that the order can also be sorted from smallest to largest, and the corresponding filter is the travel time corresponding to the bottom 25%.

[0086] Alternatively, frequency values ​​can be filtered by specific sorting values. For example, the top four frequency values ​​in Table 3 can be filtered out, and the travel times corresponding to the top four frequency values ​​can be used as the fifth time.

[0087] Furthermore, after selecting multiple fifth moments, the moment that meets the preset conditions among the fifth moments is taken as the first moment.

[0088] Optionally, if the selected fifth time slots include 18:27, 12:05, 08:55 and 10:00, where 18:27, 12:05 and 08:55 have a high degree of dispersion, while 10:00 has a moderate degree of dispersion, then 18:27, 12:05 and 08:55 that meet the preset conditions can be taken as the first time slot, and 10:00 that does not meet the preset conditions can be filtered out.

[0089] Optionally, if the selected fifth time includes 18:27 and 12:05, and the dispersion of 18:27 and 12:05 is high, both meeting the preset conditions, then 18:27 and 12:05 can be used as the first time.

[0090] In the embodiments of this application, the moment in which the dispersion of a plurality of fifth moments meets the preset conditions can be taken as the first moment. Since the frequency of travel of a large number of historical travel times is filtered first, the amount of data processing during the filtering of the dispersion of the fifth moments can be reduced, thereby improving the efficiency of filtering the first moment. Based on improving the efficiency of filtering the first moment, the efficiency of predicting the travel time of vehicles is further improved.

[0091] S230, based on historical travel trajectories, filters out multiple second moments where there is overlap in multiple travel trajectories of the vehicle.

[0092] For example, multiple historical travel trajectories can be obtained from the vehicle's historical travel data. The multiple historical travel trajectories can be compared using an overlapping trajectory calculation model to filter out the time corresponding to the travel trajectories that have overlapping parts (which can be referred to as "second time").

[0093] For example, the similarity between each trajectory in the historical travel trajectory is calculated; based on the similarity between the trajectories, the multiple travel trajectories with the highest similarity are obtained; the multiple travel trajectories are overlapped to obtain multiple target trajectories with overlapping parts; the travel time corresponding to the multiple target trajectories is determined as the second time.

[0094] Figure 4 This is a schematic diagram showing the comparison of historical travel trajectories provided in the embodiments of this application.

[0095] For example, such as Figure 4 As shown, Figure 4 This includes historical travel trajectory 1, historical travel trajectory 2... historical travel trajectory 8, and includes the corresponding travel time for each historical travel trajectory. For example, the travel time corresponding to historical travel trajectory 1 is 06:10, the travel time corresponding to historical travel trajectory 2 is 08:00... the travel time corresponding to historical travel trajectory 8 is 03:30, and so on. They will not be described in detail here.

[0096] Optionally, the similarity of multiple historical travel trajectories can be calculated using the similarity calculation of Global Positioning System (GPS) trajectory points in the overlapping trajectory calculation model to obtain the multiple travel trajectories with the highest similarity, such as... Figure 4 The similarity between historical travel trajectories 1, 5, and 7 is 90%, which is greater than the similarity between historical travel trajectories 2, 3, 4, 6, and 8.

[0097] Furthermore, the overlapping portion between historical travel trajectories 1, 5, and 7 is calculated using variance and key GPS point distance constraints; that is, the overlapping trajectory of historical travel trajectories 1, 5, and 7. When an overlapping trajectory is identified between historical travel trajectories 1, 5, and 7, the departure time corresponding to each of these trajectories can be designated as the second time. Specifically, the departure time 06:10 for historical travel trajectory 1 is designated as the second time, the departure time 21:00 for historical travel trajectory 5 is designated as the second time, and the departure time 20:42 for historical travel trajectory 7 is designated as the second time.

[0098] In the embodiments of this application, the time corresponding to the historical travel trajectory with the highest similarity and overlapping trajectory is determined as the second time, and historical travel trajectories with large differences are filtered out, that is, abnormal travel trajectories in the historical travel trajectories are filtered out. Since abnormal travel trajectories in the historical travel trajectories are filtered out, the multiple second times used to predict the travel time of the vehicle are more accurate. Therefore, based on the fact that the multiple second times are more accurate, the predicted travel time of the vehicle can also be more accurate.

[0099] It should be understood that Figure 4 The historical travel trajectories shown are merely illustrative examples and are not intended to limit the scope of this application.

[0100] It should also be understood that the execution order of S220 and S230 can be as follows: Figure 2 The execution of S220 and S230 can be carried out simultaneously; or, S220 can be executed first, followed by S230; or S230 can be executed first, followed by S220. That is, the execution order of S220 and S230 will not affect the implementation of the scheme in this application.

[0101] S240, calculate the similarity between the first and second moments and the third moment in the preset dataset to obtain multiple fourth moments with a similarity greater than or equal to a preset value.

[0102] The preset third time in the dataset can represent the user's high probability of traveling at that time in the future. For example, 08:10 can be the third time, or 06:10, 08:10, 12:09, and 18:27 can be the third time. The preset value can represent the user's probability of traveling at that time in the future, for example, 95% or 98%. The third time can be set by the user according to their needs, and this application embodiment does not limit it.

[0103] Optionally, the similarity calculation model can be used to calculate the similarity between the first time point with a high degree of dispersion, such as 18:27 and 12:05, and the third time point in the preset dataset. The similarity between two times points that are exactly the same is defined as 100%, the similarity between two times points that are less than or equal to 5 minutes is defined as 95%, and the similarity between two times points that are greater than 5 minutes and less than or equal to 10 minutes is defined as 80%.

[0104] By comparison, we can see that if the preset dataset includes 18:27, then the similarity corresponding to the first time point 18:27 is 100%; if the preset dataset includes 12:09, then the first time points 12:05 and 12:09 differ by 4 minutes, which is less than 5 minutes, so the similarity corresponding to 12:05 is 95%. Therefore, 18:27 and 12:05 can be determined as the fourth time point.

[0105] Optionally, the similarity calculation model can be used to calculate the similarity between the second time point with the highest similarity and overlapping trajectory, such as 06:10 and 20:42, and the third time point in the preset dataset. By comparison, it can be seen that the preset dataset includes 06:10, so the similarity corresponding to the second time point 06:10 is 100%. However, the preset dataset does not include 20:42, and the deviation does not meet the screening criteria. Therefore, 06:10 can be identified as the fourth time point, but 20:42 cannot be identified as the fourth time point.

[0106] Optionally, the first, second, and fifth moments can be compared with the third moment in the preset dataset to obtain multiple fourth moments with similarity greater than or equal to a preset value.

[0107] S250 predicts the vehicle's target travel time based on multiple fourth moments.

[0108] For example, the similarity between the fourth time point and the third time point is obtained; the similarity is used as a weight to perform a weighted average on multiple fourth time points to obtain the vehicle's target travel time.

[0109] For example, the fourth time point can include 06:10, 18:27, and 12:05. The similarity between 06:10 and 18:27 and the third time point in the preset dataset is 100%, and the similarity between 12:05 and the third time point in the preset dataset is 95%. 100%, 100%, and 95% can be used as weights to perform a weighted average on 06:10, 18:27, and 12:05 to finally obtain the vehicle's target travel time.

[0110] Alternatively, if the determined fourth time is only 06:10, then 06:10 can be directly used as the predicted target travel time.

[0111] Alternatively, the fourth times 06:10, 18:27 and 12:05 can all be determined as the target travel time of the vehicle, without performing a weighted average on the times 06:10, 18:27 and 12:05.

[0112] In the embodiments of this application, similarity can be used as a weight to perform a weighted average on multiple fourth time points to obtain the vehicle's target travel time. The weighted average process is a trend prediction method for the user's future driving vehicle travel time. Since the larger the weight, the greater the impact on the predicted target travel result, the similarity between the fourth time point and the third time point is used as a weight to perform a weighted average on multiple fourth time points. This increases the impact of the first time point with high dispersion and the second time point with the highest similarity to historical travel trajectories and overlapping trajectories on the final predicted target travel time, and reduces the impact of the filtered-out time points with low dispersion and the time points corresponding to low similarity to historical travel trajectories on the predicted target travel time. As a result, the predicted target travel time is more in line with the user's travel needs, further improving the accuracy of the predicted target travel time.

[0113] exist Figure 2 In the method 200 shown, multiple first moments that meet preset conditions in terms of the dispersion of vehicle travel times can be selected based on historical travel times; multiple second moments that have overlapping parts in multiple vehicle travel trajectories can be selected based on historical travel trajectories; by selecting two types of moments (i.e., first moments and second moments) that meet the two conditions, the vehicle travel time is predicted, avoiding the error caused by the singularity of predicting vehicle travel time due to the single type of moment. That is, using the travel time predicted by the two types of moments to predict historical travel time can reduce the deviation between the predicted travel time and the user's actual travel time, so that the predicted travel time is closer to the user's actual travel time, thus improving the accuracy of predicting vehicle travel time.

[0114] After predicting the vehicle's target travel time, a target time period can be determined based on the target travel time; based on historical trip data, user scheduling information for vehicle components within the target time period can be obtained; and based on the scheduling information, the target components of the vehicle can be scheduled.

[0115] The historical trip data also includes user scheduling information for vehicle components. The target time period can be any time of day, and can be a period of 5 or 10 minutes, for example, 12:00 to 12:10. The interval can be set according to user needs, and this application embodiment does not limit it.

[0116] The target component may refer to one or more of the following in a vehicle: air conditioning, audio system, and seat heating pad.

[0117] For example, based on the predicted target travel time (e.g., 12:07), a time period that can include that time (e.g., 12:00 to 12:10) can be determined; then, the scheduling information of the user's use of the vehicle's components between 12:00 and 12:10 can be obtained through the vehicle's historical trip data. For example, the user used the air conditioner to cool down the vehicle between 12:00 and 12:10.

[0118] Furthermore, based on the obtained scheduling information of the vehicle's components, the vehicle's air conditioning is automatically scheduled at the predicted target travel time or a preset time before the target travel time, so that the air conditioning is turned on to cool the passenger compartment.

[0119] In the embodiments of this application, target components of the vehicle are scheduled according to scheduling information; by determining the scheduling information of target components in the vehicle during the target time period when the user drove the vehicle in the past, the target components in the vehicle are automatically scheduled in the future target time period, so that the target components in the vehicle can be put into operation in advance, thereby realizing the automatic adjustment of the vehicle's driving environment in advance, thereby ensuring the comfort of the vehicle's driving environment and improving the user's driving experience.

[0120] Figure 5 This is a schematic diagram of the structure for predicting vehicle travel time provided in an embodiment of this application.

[0121] For example, such as Figure 5 As shown, Figure 5 It includes a data preprocessing unit 510, a travel time prediction unit 520, a similarity calculation unit 530, and a preset processing unit 540.

[0122] The data preprocessing unit 510 can be used to acquire historical trip data and extract historical travel time, historical travel trajectory and historical scheduling information of vehicle components from the historical trip data. The extracted historical travel time, historical travel trajectory and historical scheduling information of vehicle components are sent to the travel time prediction unit 520 so that the travel time prediction unit 520 can predict the travel time of the vehicle.

[0123] The travel time prediction unit 520 can be used to predict the travel time of vehicles and send the predicted travel time to the similarity calculation unit 550. The travel time prediction unit 520 may include one or more of an adaptive density clustering model 521, an overlapping trajectory calculation model 522, and an interquartile range detection model 523. It should be understood that the travel time prediction unit 520 can include any model capable of predicting the travel time of vehicles at every moment of the day, or it can be other models used to predict the travel time of vehicles; this application embodiment does not limit this. Specifically, the adaptive density clustering model 521 can be used to use a clustering algorithm to filter out the first moment with high dispersion; the overlapping trajectory calculation model 522 is used to filter out the second moment with the highest similarity and overlapping trajectories; and the interquartile range detection model 523 is used to filter out the fifth moment where the historical travel time ranks in the top 25%.

[0124] The similarity calculation model 531 in the similarity calculation unit 530 can be used to calculate the similarity between the first time, the second time, and the fifth time and the third time in the preset dataset to obtain multiple fourth times, and send the calculated multiple fourth times to the preset processing unit 540.

[0125] The preset processing unit 540 can be used to perform weighted average processing on multiple received fourth time points.

[0126] pass Figure 5 The method for predicting vehicle travel time provided in this application can be illustrated by example. Figure 5 In this system, vehicle travel time can be predicted using any one of the adaptive density clustering model 521, overlapping trajectory calculation model 522, and interquartile range detection model 523; alternatively, vehicle travel time can be predicted using any two of these models; furthermore, vehicle travel time can be predicted simultaneously using all three models; and additional models can be added or removed based on the required vehicle travel time prediction. Figure 5 The model in this application is not limited to this embodiment.

[0127] Optionally, the interquartile range detection model 523 can be used to first select the fifth time point with a frequency in the top 25%, and then the first time point with a high degree of dispersion can be selected based on the fifth time point; alternatively, the adaptive density clustering model 521 can be used to first select the first time point with a high degree of dispersion, and then the fifth time point with a frequency in the top 25% can be selected based on the first time point; alternatively, the interquartile range detection model 523 can be used to select the fifth time point with a frequency in the top 25%, and the adaptive density clustering model 521 can be used to select the first time point with a high degree of dispersion.

[0128] For example, three models—adaptive density clustering model 521, overlapping trajectory calculation model 522, and interquartile range detection model 523—are used simultaneously to predict vehicle travel time.

[0129] For example, the data preprocessing unit 510 can extract the historical travel time and historical travel trajectory required to predict the vehicle's travel time from the acquired historical travel data. The extracted historical travel time is sent to the adaptive density clustering model 521 and the interquartile range detection model 523 in the travel time prediction unit 520. Multiple first moments are selected by the adaptive density clustering model 521, and multiple fifth moments are selected by the interquartile range detection model 523. At the same time, the historical travel trajectory is sent to the overlapping trajectory calculation model 522 in the travel time prediction unit 520, and a second moment is selected by the overlapping trajectory calculation model 522. The selected first moments, second moments, and fifth moments are sent to the similarity calculation model 531 in the similarity calculation unit 530 for similarity calculation to obtain a fourth moment with a similarity greater than or equal to 95%. The calculated fourth moment is sent to the preset processing unit 540 for weighted average processing to predict the vehicle's target travel time.

[0130] Figure 6 This is a schematic diagram of the vehicle travel time prediction device provided in the embodiments of this application.

[0131] For example, such as Figure 6 As shown, the device 600 includes:

[0132] Module 610: Used to acquire historical trip data of the vehicle, including the vehicle's historical travel time and historical travel trajectory;

[0133] First filtering module 620: used to filter out multiple first moments when the dispersion of vehicle travel times meets preset conditions based on historical travel times;

[0134] The second filtering module 630 is used to filter out multiple second moments with overlapping parts among multiple travel trajectories of a vehicle based on historical travel trajectories.

[0135] Calculation module 640: used to calculate the similarity between the first time point and the second time point and the third time point in the preset dataset, and obtain multiple fourth time points with a similarity greater than or equal to a preset value;

[0136] Prediction module 650: Used to predict the vehicle's target travel time based on multiple fourth moments.

[0137] In one possible implementation, the target travel time of the vehicle is predicted based on multiple fourth moments. Specifically, the prediction module 650 is used to: obtain the similarity between the fourth moment and the third moment; and use the similarity as a weight to perform a weighted average of the multiple fourth moments to obtain the target travel time of the vehicle.

[0138] Optionally, the device 600 further includes a scheduling module 660, which is used to determine a target time period including the target travel time based on the target travel time; to obtain scheduling information of user's use of vehicle components within the target time period based on historical travel data; and to schedule the target components of the vehicle based on the scheduling information.

[0139] In one possible implementation, based on historical travel times, multiple first moments where the dispersion of vehicle travel times meets preset conditions are selected. The first filtering module 620 is specifically used to: obtain abnormal travel moments of vehicles based on the dispersion of historical travel times at each moment; and remove abnormal travel moments from historical travel times to obtain multiple first moments.

[0140] In one possible implementation, based on historical travel times, multiple first moments whose travel time dispersion meets preset conditions are selected. The first filtering module 620 is specifically used to: obtain multiple frequency values ​​based on the frequency of each travel moment in the historical travel time; sort the multiple frequency values ​​in sequence and select multiple target frequency values ​​that meet the preset frequency; obtain the travel time corresponding to each of the multiple target frequency values ​​to obtain multiple fifth moments; and take the moment whose dispersion meets the preset conditions among the multiple fifth moments as the first moment.

[0141] In one possible implementation, based on historical travel trajectories, multiple second moments with overlapping portions are selected from multiple travel trajectories of the vehicle. The second filtering module 630 is specifically used to: calculate the similarity between each trajectory in the historical travel trajectory; obtain multiple travel trajectories with the highest similarity based on the similarity between the trajectories; perform overlap processing on the multiple travel trajectories to obtain multiple target trajectories with overlapping portions; and determine the travel time corresponding to the multiple target trajectories as the second moment.

[0142] In one possible implementation, the historical trip data of the vehicle is obtained. Specifically, the acquisition module 610 is used to: collect the identity information of the driver; classify the historical trip data of the vehicle according to the identity information of the driver to obtain the classification result; and obtain the historical trip data corresponding to the target user according to the classification result.

[0143] It should be noted that the vehicle travel time prediction device provided in the above embodiments is only illustrated by the division of the above functional modules when performing the vehicle travel time prediction method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0144] Furthermore, the vehicle travel time prediction device and the vehicle travel time prediction method provided in the above embodiments belong to the same concept. Therefore, for details not disclosed in the device embodiments of this specification, please refer to the vehicle travel time prediction embodiments described above in this specification, which will not be repeated here.

[0145] For example, this application also provides a device for predicting vehicle travel time, the device comprising: a memory for storing executable program code; and a processor for calling and running the executable program code from the memory, causing the device to perform a method for predicting vehicle travel time.

[0146] Figure 7 This is a schematic diagram of the vehicle structure provided in the embodiments of this application.

[0147] For example, such as Figure 7 As shown, the vehicle 110 includes a device for predicting vehicle travel time, specifically including: a memory 710 and a processor 720, wherein the memory 710 stores executable program code 7101, and the processor 720 is used to call and execute the executable program code 7101 to perform a method for predicting vehicle travel time.

[0148] This application can divide the vehicle into functional modules based on the above method example. For example, each module can correspond to a separate function module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0149] When each functional module is divided according to its corresponding function, the vehicle may include: a first acquisition module, a second acquisition module, and a preset processing module, etc. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0150] The vehicle provided in this application is used to execute the above-described method for predicting vehicle travel time, and thus can achieve the same effect as the above-described implementation method.

[0151] When using integrated units, the vehicle may include a processing module and a storage module. The processing module is used to control and manage the vehicle's actions. The storage module supports the vehicle in executing program code and data.

[0152] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as disclosed in this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0153] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described in the foregoing embodiments. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs (Digital Video Discs), CD-ROMs (Compact Disc Read-Only Memory), microdrives, magneto-optical disks, ROMs (Read-Only Memory), RAMs (Random Access Memory), EPROMs (Erasable Programmable Read-Only Memory), EEPROMs (Electrically Erasable Programmable Read-Only Memory), DRAMs (Dynamic Random Access Memory), VRAMs (Video Random Access Memory), flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0154] This application also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement a method for predicting vehicle travel time in the above embodiments.

[0155] In addition, the vehicle provided in the embodiments of this application may specifically be a chip, component or module. The vehicle may include a connected processor and a memory. The memory is used to store instructions. When the vehicle is running, the processor may call and execute the instructions to make the chip execute a vehicle travel time prediction method in the above embodiments.

[0156] The vehicle, computer-readable storage medium, computer program product or chip provided in this application are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods provided above, and will not be repeated here.

[0157] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0158] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0159] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of vehicle travel time prediction, the method comprising: The method includes: Obtain historical trip data of the vehicle, including the vehicle's historical travel time, historical travel trajectory, and user scheduling information for components in the vehicle; Based on the historical travel times, select multiple first moments where the dispersion of the vehicle's travel times meets preset conditions; Based on the historical travel trajectory, multiple second moments with overlapping parts are selected from the multiple travel trajectories of the vehicle; The similarity between the first time point and the second time point and the third time point in the preset dataset is calculated to obtain multiple fourth time points with a similarity greater than or equal to a preset value. Based on the multiple fourth moments, predict the target travel time of the vehicle; Based on the target travel time, determine the target time period that includes the target travel time; Based on the historical travel data, obtain the user's scheduling information for components in the vehicle within the target time period; Based on the scheduling information, the target components of the vehicle are scheduled; The prediction of the vehicle's target travel time based on a plurality of the fourth times includes: Obtain the similarity between the fourth time point and the third time point; Using the similarity as a weight, a weighted average is performed on multiple fourth time points to obtain the target travel time of the vehicle.

2. The method of claim 1, wherein, The step of filtering out multiple first moments where the dispersion of the vehicle's travel time meets preset conditions based on the historical travel time includes: The abnormal travel times of the vehicle are obtained based on the dispersion of the historical travel times at each moment. By removing the abnormal travel times from the historical travel times, multiple first times are obtained.

3. The method of claim 1, wherein, The step of filtering out multiple first moments where the dispersion of the vehicle's travel time meets preset conditions based on the historical travel time includes: Based on the frequency of each travel time in the historical travel time, multiple frequency values ​​are obtained; The multiple frequency values ​​are sorted sequentially, and multiple target frequency values ​​that satisfy the preset frequency are selected from the multiple frequency values. By obtaining the travel time corresponding to each of the multiple target frequency values, multiple fifth times are obtained; The moment in which the dispersion of the plurality of fifth moments meets the preset condition is taken as the first moment.

4. The method of claim 1, wherein, The step of filtering out multiple second moments with overlapping portions among the vehicle's multiple travel trajectories based on the historical travel trajectory includes: Calculate the similarity between the various trajectories in the historical travel trajectory; Based on the similarity between the trajectories, the multiple travel trajectories with the highest similarity are obtained; Multiple travel trajectories are overlapped to obtain multiple target trajectories with overlapping portions; The travel times corresponding to the multiple target trajectories are determined as the second time.

5. The method of claim 1, wherein, The acquisition of historical travel data of vehicles includes: Collect driver user identity information; Based on the driver's identity information, the vehicle's historical trip data is categorized to obtain the categorization results; Based on the classification results, obtain the historical travel data corresponding to the target user.

6. An apparatus for vehicle travel time prediction, the apparatus comprising: The device includes: The first acquisition module is used to acquire the vehicle's historical travel data, which includes the vehicle's historical travel time, historical travel trajectory, and user scheduling information for components in the vehicle. The first filtering module is used to filter out multiple first moments in which the dispersion of the vehicle's travel time meets preset conditions based on the historical travel time. The second filtering module is used to filter out multiple second moments with overlapping parts among the multiple travel trajectories of the vehicle based on the historical travel trajectory. The calculation module is used to calculate the similarity between the first time point and the second time point and the third time point in the preset dataset, and obtain multiple fourth time points with a similarity greater than or equal to a preset value; The prediction module is used to predict the target travel time of the vehicle based on multiple fourth time points; The scheduling module is used to determine a target time period including the target travel time based on the target travel time; to obtain the user's scheduling information for the vehicle's components within the target time period based on the historical travel data; and to schedule the target components of the vehicle based on the scheduling information. The prediction module is specifically used to: obtain the similarity between the fourth time point and the third time point; and use the similarity as a weight to perform a weighted average on multiple fourth time points to obtain the target travel time of the vehicle.

7. An apparatus for vehicle travel time prediction, the apparatus comprising: The device includes: Memory, used to store executable program code; A processor for calling and running the executable program code from the memory, causing the device to perform the method as described in any one of claims 1 to 5.

8. A vehicle characterized by comprising: The vehicle includes the vehicle travel time prediction device as described in claim 7.

Images