System, method, and computer program product for predicting operating behavior of vehicle during traffic jam

Abstract

The invention discloses a vehicle operation behavior prediction system during traffic jam, and the system comprises a traffic jam scene classification module which is used for inputting vehicle driving state data in a current traffic jam state into a traffic jam classification model, and obtaining a current traffic jam type; the association rule base establishment module establishes a traffic jam type and vehicle operation behavior association rule base through an association rule algorithm according to the traffic jam type and the historical vehicle operation behavior of the vehicle; and the vehicle operation behavior prediction module is used for predicting the vehicle operation behavior at the next moment in the current traffic jam type through the association rule base. According to the invention, based on traffic jam types and vehicle operation behavior data, an Apriori algorithm is adopted to construct an association rule base, and multi-level behavior prediction is realized.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, specifically to a vehicle operation behavior prediction system, method, and computer program product during traffic jams. Background Technology

[0002] With the acceleration of urbanization, traffic congestion has become a major factor affecting the driving experience.

[0003] Traditional methods primarily rely on GPS vehicle speed data to determine congestion, but they only provide a binary conclusion of "is there a traffic jam?", failing to delve into the specific behaviors of users during traffic jams. Furthermore, traditional methods underutilize vehicle operation data, making it impossible to analyze users' true needs during congestion. This makes it difficult for automakers to determine "what features should be provided during traffic jams" to improve user experience based on existing data. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a vehicle operation behavior prediction system during traffic jams, comprising: The traffic congestion scenario classification module is used to input vehicle driving status data under the current traffic congestion state into the traffic congestion classification model to obtain the current traffic congestion type; The association rule base building module establishes an association rule base between traffic jam type and vehicle operation behavior based on traffic jam type and vehicle historical operation behavior through association rule algorithms. The vehicle operation behavior prediction module is used to predict the vehicle operation behavior at the next moment under the current traffic jam type when the vehicle is in a traffic jam and there is no corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library; and when the vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, it uses the current traffic jam type and the current vehicle operation behavior to predict the vehicle operation behavior at the next moment under the current traffic jam type and vehicle operation behavior association rule library.

[0005] Furthermore, in the traffic jam scenario classification module, the vehicle driving status data under traffic jam conditions includes the following vehicle driving status characteristics: The average vehicle speed within the preset time window, the percentage of time within the preset time window when the vehicle speed is less than the preset speed, the braking frequency per unit time, the number of times the acceleration is less than the preset acceleration within the preset time window, and the number of vehicles in the preset area around the current vehicle.

[0006] Furthermore, in the traffic congestion scenario classification module, the traffic congestion classification model is obtained through the following method: Historical vehicle driving status dataset for all vehicles of the corresponding model during traffic jams , , For the corresponding model number Historical vehicle driving status data during traffic jams for each vehicle. For the vehicle Vehicle driving status characteristics; Data set of historical vehicle driving status during traffic jams for all vehicles of the corresponding model. The k-means clustering algorithm is used for classification, thereby classifying the historical traffic jam status of all vehicles into several traffic jam types. Cluster centers are determined based on these traffic jam types, and the number of classified traffic jam types is set as the k value. Based on the cluster centers and the k-value, a k-means clustering algorithm is set to obtain a traffic congestion classification model.

[0007] Furthermore, in the association rule base module, the specific method for establishing an association rule base between traffic jam type and vehicle operation behavior based on traffic jam type and historical vehicle operation behavior using an association rule algorithm is as follows: The association rule algorithm uses the Apriori algorithm to construct a transaction item set based on traffic jam type i and the historical vehicle operation behavior k of all vehicles of the corresponding vehicle type under traffic jam type i: transaction item set = {traffic jam type i} or {traffic jam type i, historical vehicle operation behavior 1, historical vehicle operation behavior 2, ..., historical vehicle operation behavior j}; The transaction itemsets containing the same traffic jam type are grouped into a corresponding transaction database. The category of the transaction itemset is defined as a k-itemset, and the number of items in the k-itemset is equal to the total number of historical vehicle operation behaviors plus 1. In the Apriori algorithm, the support (A) of a transaction item set A is equal to the number of transaction item sets in a transaction database that contain a subset of transaction item set A, and the total number of transaction item sets in the transaction database. In the Apriori algorithm, the support (A,B) of a transaction item set containing both transaction item set A and transaction item set B is equal to the number of transaction item sets in a transaction database that contain both transaction item sets A and B, divided by the total number of transactions in the database. In the Apriori algorithm, the confidence of transaction itemset B is equal to the support (A, B) / support (A), where transaction itemset A and transaction itemset B have no intersection. For a given transaction database, calculate the support of all 1-itemsets in the transaction database, define transaction itemsets with support greater than or equal to a preset support as frequent itemsets, select frequent 1-itemsets from all 1-itemsets, generate candidate k-itemsets based on frequent 1-itemsets using join and branch pruning in the Apriori algorithm, and select frequent k-itemsets from the candidate k-itemsets. The support (A1) of transaction itemset A1, which is a frequent itemset, is calculated based on all frequent itemsets. The support (A1, B1) of transaction itemset B1, which contains both frequent itemsets, is calculated based on all frequent itemsets. The confidence of transaction itemset B1, which is a frequent itemset, is defined as support (A1, B1) / support (A1). The association rule is defined based on the confidence of transaction itemset B1, which is a frequent itemset: In the case of transaction itemset A1, which is a frequent itemset, there is a probability that transaction itemset B1, which is a frequent itemset, will have the same confidence value as transaction itemset B1. Based on the confidence threshold, the confidence of transaction itemset B1, which is a frequent itemset, is selected from all frequent itemsets. Based on the confidence of the selected frequent itemset B1, a rule base for the association between traffic jam type and vehicle operation behavior is generated.

[0008] Furthermore, when a vehicle is in a traffic jam and no corresponding vehicle operation behavior is currently performed in the traffic jam type and vehicle operation behavior association rule base, the specific method for predicting the vehicle operation behavior at the next moment under the current traffic jam type in the traffic jam type and vehicle operation behavior association rule base is as follows: When a vehicle is in a traffic jam and no vehicle operation behavior corresponding to the traffic jam type and vehicle operation behavior association rule base is performed, the association rule corresponding to the transaction item set A1 that only includes the current traffic jam type is found in the traffic jam type and vehicle operation behavior association rule base, and the transaction item set B2 in the association rule is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

[0009] Furthermore, when a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the specific method for predicting the vehicle operation behavior at the next moment in the current traffic jam type using the current traffic jam type and current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base is as follows: When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, the transaction item set A1, which consists of the current traffic jam type and the corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, is found. The transaction item B1 in the association rule corresponding to the transaction item set A1 is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

[0010] Furthermore, the traffic jam types include slow creeping, complete standstill, and frequent start-stop. The slow creeping type is defined as a vehicle speed between a first preset speed and a second preset speed and maintained for a period of more than a first time window. The complete standstill is defined as a vehicle speed of 0 and maintained for a period of more than a second time window. The frequent start-stop type is defined as a vehicle speed fluctuating between 0 and a third preset speed. Vehicle operation behaviors include air conditioning operation behaviors, in-vehicle entertainment system operation behaviors, window operation behaviors, seat operation behaviors, and driving operation behaviors.

[0011] Furthermore, it also includes a vehicle operation behavior execution module, which is used to predict the vehicle operation behavior based on association rules in the association rule base when the transaction item set A1 is traffic jam type i. The transaction item B1 is the transaction item set A1 of traffic jam type i and represents vehicle operation behavior. The set of transaction items B1 is called transaction item set C. Using transaction item set C, the vehicle operation behavior is predicted based on association rules in the association rule base. The transaction set D will contain vehicle operation behavior. and vehicle operation behavior The combination forms a preset mode, which controls the corresponding equipment to simultaneously execute all vehicle operation behaviors within the preset mode.

[0012] A method for predicting vehicle operating behavior during traffic jams includes: Input the vehicle driving status data under the current traffic jam condition into the traffic jam classification model to obtain the current traffic jam type; Based on traffic congestion types and historical vehicle operation behaviors, a rule base for associating traffic congestion types with vehicle operation behaviors is established using an association rule algorithm. When a vehicle is in a traffic jam and there is no corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, the vehicle operation behavior at the next moment in the current traffic jam type is predicted in the traffic jam type and vehicle operation behavior association rule library. When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the vehicle operation behavior at the next moment in the current traffic jam type is predicted by using the current traffic jam type and the current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base.

[0013] A computer program product includes a computer program / instructions that, when executed by a processor, implement the above-described method for predicting vehicle operation behavior during traffic jams.

[0014] The beneficial effects of this invention are as follows: 1. By combining multi-dimensional vehicle driving status features with a traffic jam scenario classification module, the k-means clustering algorithm is used to achieve dynamic classification of traffic jam types (slow crawling type / complete standstill type / frequent start-stop type).

[0015] 2. Based on traffic jam type and vehicle operation behavior data, an association rule base is constructed using the Apriori algorithm to achieve multi-level behavior prediction (such as predicting air conditioning operation based on traffic jam type, and predicting music playback operation from traffic jam type and air conditioning operation). It can also expand the preset pattern of behavior combination.

[0016] 3. Utilize existing vehicle network data, requiring no additional hardware investment. Attached Figure Description

[0017] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0018] To make the technical problems, technical solutions, and beneficial effects to be solved by this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and are not intended to limit the scope of this application.

[0019] Example 1 refer to Figure 1 A vehicle operation behavior prediction system during traffic jams, comprising: The traffic congestion scenario classification module is used to input vehicle driving status data under the current traffic congestion state into the traffic congestion classification model to obtain the current traffic congestion type; The association rule base building module establishes an association rule base between traffic jam type and vehicle operation behavior based on traffic jam type and vehicle historical operation behavior through association rule algorithms. The vehicle operation behavior prediction module is used to predict the vehicle operation behavior at the next moment under the current traffic jam type when the vehicle is in a traffic jam and there is no corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library; and when the vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, it uses the current traffic jam type and the current vehicle operation behavior to predict the vehicle operation behavior at the next moment under the current traffic jam type and vehicle operation behavior association rule library.

[0020] Traffic congestion is usually determined by the speed threshold method (the mainstream international standard). For urban expressways: speed ≤ 20 km / h (such as elevated roads in Beijing and Shanghai); for main roads: speed ≤ 15 km / h (six-lane dual carriageway in urban areas); for highways: speed ≤ 40 km / h (a congestion warning is triggered when the speed is 50% below the design speed).

[0021] Traffic jam vehicle status data includes the following characteristics: average vehicle speed within a preset time window (5 minutes), the percentage of time within the preset time window when the vehicle speed is lower than the preset speed (10 km / h), braking frequency per unit time, the number of times acceleration is lower than the preset acceleration (-0.5 m / s²) within the preset time window, and the number of vehicles within a preset range (50 meters). By quantifying traffic jam status through multi-dimensional features (average speed, braking frequency, number of surrounding vehicles, etc.), the limitations of a single speed indicator are avoided, improving the accuracy of traffic jam type classification. Preset time windows and thresholds (such as 10 km / h speed and -0.5 m / s² acceleration) enable the model to distinguish between scenarios such as "slow crawling" and "frequent starts and stops," providing differentiated input for subsequent behavior prediction. All features are derived from existing onboard sensors (such as vehicle speed sensors, radar, and accelerometers), requiring no additional hardware investment.

[0022] Vehicle operation behavior includes air conditioning usage records (temperature adjustment frequency, airflow changes), entertainment system operation (music switching, volume adjustment, voice assistant use), window / sunroof status (opening and closing frequency), seat adjustment records (such as lumbar support adjustment, massage function activation), and driving mode switching (such as switching from sport mode to comfort mode).

[0023] Traffic jams can be categorized into three types: slow crawling (e.g., speeds of 5-20 km / h lasting more than 10 minutes), complete standstill (e.g., speeds of 0 km / h lasting more than 2 minutes), and frequent stop-and-go traffic (e.g., speeds fluctuating repeatedly between 0-15 km / h).

[0024] (1) In the traffic congestion scenario classification module, the traffic congestion classification model is obtained through the following method: Historical vehicle driving status dataset for all vehicles of the corresponding model during traffic jams , , For the corresponding model number Historical vehicle driving status data during traffic jams for each vehicle. For the vehicle Vehicle driving status characteristics; Data set of historical vehicle driving status during traffic jams for all vehicles of the corresponding model. The k-means clustering algorithm is used for classification, thereby classifying the historical traffic jam status of all vehicles into several traffic jam types. Cluster centers are determined based on these traffic jam types (the cluster centers are ultimately determined through iteration during the classification process), and the number of classified traffic jam types is set as the k value. Based on the cluster centers and the k-value, a k-means clustering algorithm is set to obtain a traffic congestion classification model.

[0025] Employing the k-means clustering algorithm, traffic congestion patterns (such as three-category cluster centers) are automatically discovered from historical data, eliminating the need for manual labeling and reducing model development costs. Training is based on data from the same vehicle model, ensuring that the model's output congestion type (such as "complete standstill") matches the actual driving scenarios of users of that vehicle type. The classification granularity can be flexibly increased or decreased by adjusting the k-value (e.g., adding "intermittent congestion").

[0026] (2) In the vehicle operation behavior prediction module, the association rule base is obtained through the following method: Data preprocessing: Discretize historical vehicle operation behavior, that is, divide continuous data (such as air conditioning temperature) into bins (e.g., "low temperature: <22°C", "medium temperature: 22-26°C", "high temperature: >26°C").

[0027] The specific method for establishing a rule base for association between traffic congestion type and vehicle operation behavior based on traffic congestion type and historical vehicle operation behavior using an association rule algorithm is as follows: The association rule algorithm uses the Apriori algorithm to construct a transaction item set based on traffic congestion type i and the historical vehicle operation behavior k of all vehicles of the corresponding vehicle type under traffic congestion type i (obtained through the vehicle network): Transaction item set = {traffic congestion type i} or {traffic congestion type i, historical vehicle operation behavior 1, historical vehicle operation behavior 2, ..., historical vehicle operation behavior j}. The transaction itemsets containing the same traffic jam type are grouped into a corresponding transaction database. The category of the transaction itemset is defined as a k-itemset, and the number of items in the k-itemset is equal to the total number of historical vehicle operation behaviors plus 1. In the Apriori algorithm, the support (A) of a transaction item set A is equal to the number of transaction item sets in a transaction database that contain a subset of transaction item set A, and the total number of transaction item sets in the transaction database. In the Apriori algorithm, the support (A,B) of a transaction item set containing both transaction item set A and transaction item set B is equal to the number of transaction item sets in a transaction database that contain both transaction item sets A and B, divided by the total number of transactions in the database. In the Apriori algorithm, the confidence of transaction itemset B is equal to the support (A, B) / support (A), where transaction itemset A and transaction itemset B have no intersection. For a given transaction database, calculate the support of all 1-itemsets in the transaction database, define transaction itemsets with support greater than or equal to a preset support as frequent itemsets, select frequent 1-itemsets from all 1-itemsets, generate candidate k-itemsets based on frequent 1-itemsets using join and branch pruning in the Apriori algorithm, and select frequent k-itemsets from the candidate k-itemsets. The support (A1) of transaction itemset A1, which is a frequent itemset, is calculated based on all frequent itemsets. The support (A1, B1) of transaction itemset B1, which contains both frequent itemsets, is calculated based on all frequent itemsets. The confidence of transaction itemset B1, which is a frequent itemset, is defined as support (A1, B1) / support (A1). The association rule is defined based on the confidence of transaction itemset B1, which is a frequent itemset: In the case of transaction itemset A1, which is a frequent itemset, there is a probability that transaction itemset B1, which is a frequent itemset, will have the same confidence value as transaction itemset B1. Based on the confidence threshold, the confidence of transaction itemset B1 that is a frequent itemset is selected from all frequent itemsets (the confidence of transaction itemset B1 that is a frequent itemset is greater than or equal to the preset confidence). Based on the confidence of the selected transaction itemset B1 that is a frequent itemset, a rule base for the association between traffic jam type and vehicle operation behavior is generated.

[0028] In the Apriori algorithm, a join means that if two (k-1)-itemsets are identical in their first (k-2) items and differ only in their last item, then joining them can generate a k-itemset. The goal is to generate candidate k-itemsets from frequent (k-1)-itemsets. Example (k=3): For frequent 2-itemsets {A,B}, {A,C}, {B,C}, {B,D}, {C,D}, we can join them to generate candidate 3-itemsets: {A,B} + {A,C} joins to {A,B,C} (because the first item is the same, both are {A}); {B,C} + {B,D} joins to {B,C,D}; {A,C} + {B,C} cannot be joined (because the first item is different). In the Apriori algorithm, pruning refers to checking whether all (k-1)-item subsets of a candidate k-itemset are frequent (k-1)-itemsets. If a subset is not a frequent (k-1)-itemset, then the candidate k-itemset cannot be frequent (Apriori property: all subsets of a frequent itemset must also be frequent). The goal is to reduce unnecessary computation by removing candidates that cannot be frequent itemsets. Example (k=3): For a candidate 3-itemset {A,B,C}, check all its 2-item subsets {A,B}, {A,C}, {B,C}. If {A,B}, {A,C}, and {B,C} are frequent 2-itemsets, then keep {A,B,C}. If {A,B} is not a frequent 2-itemset, then prune (remove {A,B,C}).

[0029] The above methods support multi-level behavioral reasoning, such as single-step prediction: directly inferring behavior from traffic jam type, e.g., predicting to activate seat massage based on a complete standstill; and multi-step prediction: combining the current behavior sequence, e.g., predicting to play music based on traffic jam type + turning down the air conditioning, achieving a chain reaction and covering users' continuous operation needs. The Apriori algorithm mines high-frequency behavior combinations (e.g., support ≥ 10%) and filters noisy data. A confidence threshold (e.g., 60%) ensures that only strongly associated rules are retained, improving prediction reliability. The rule base is pre-calculated and filtered, so during actual prediction, it only needs to match the current scenario with the rules, resulting in fast response speed.

[0030] (3) In the vehicle operation behavior prediction module, when the vehicle is in a traffic jam and no vehicle operation behavior corresponding to the traffic jam type and vehicle operation behavior association rule library is performed, the specific method for predicting the vehicle operation behavior at the next moment in the traffic jam type using the current traffic jam type and vehicle operation behavior association rule library is as follows: When a vehicle is in a traffic jam and no vehicle operation behavior corresponding to the traffic jam type and vehicle operation behavior association rule base is performed, the association rule corresponding to the transaction item set A1 that only includes the current traffic jam type is found in the traffic jam type and vehicle operation behavior association rule base, and the transaction item set B2 in the association rule is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

[0031] When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the specific method for predicting the vehicle operation behavior at the next moment under the current traffic jam type using the current traffic jam type and current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base is as follows: When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, the transaction item set A1, which consists of the current traffic jam type and the corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, is found. The transaction item B1 in the association rule corresponding to the transaction item set A1 is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

[0032] (4) It also includes a vehicle operation behavior execution module, which is used to predict the vehicle operation behavior based on the association rules in the association rule base when the transaction item set A1 is traffic jam type i. The transaction item B1 is the transaction item set A1 of traffic jam type i and represents vehicle operation behavior. The set of transaction items B1 is called transaction item set C. Using transaction item set C, the vehicle operation behavior is predicted based on association rules in the association rule base. The transaction set D will contain vehicle operation behavior. and vehicle operation behavior The system combines preset modes, and the number of vehicle operation behaviors within these modes can be expanded based on association rules in a rule base. This allows the corresponding devices to simultaneously execute all vehicle operation behaviors within the preset mode. Specific control methods include vehicle bus control systems that send control commands via CAN bus or AutoSAR architecture. Devices such as the Infineon AURIX series MCU + MOST150 multimedia bus interface module generate APPD (Application Programming Messages) based on probability weights to control the corresponding devices.

[0033] For example: Automatic Comfort Mode: When "frequent stop-and-go" traffic jams are detected, it automatically switches to a smooth driving mode. Intelligent Entertainment Mode: When "completely stationary" traffic jams are detected, it turns on the music. Automatic Air Conditioning Mode: When slow-moving traffic jams are detected, it adjusts the air conditioning temperature to a specific temperature and turns on the music.

[0034] By generating preset patterns (such as "air conditioning set to low temperature + music on") through association rule chains (such as "traffic jam type → behavior A → behavior B"), multiple user needs can be met at once, improving the smoothness of the experience. The preset patterns can be automatically expanded as the rule base is updated (such as adding the behavior of "opening the window"), supporting function iteration or user-selected functions, and can directly link with devices such as air conditioning, audio, and seats to achieve fully automated scenario-based services.

[0035] Example 2 A method for predicting vehicle operating behavior during traffic jams includes: Input the vehicle driving status data during the current traffic jam into the traffic jam classification model to obtain the current traffic jam type; Based on the association rule base, the user's behavior under the current traffic jam type is predicted based on the current traffic jam type or the combination of the current traffic jam type and the current vehicle operation behavior.

[0036] Example 3 A computer program product includes a computer program / instructions that, when executed by a processor, implement the vehicle operation behavior prediction method during traffic jams as described in Embodiment 2.

[0037] The contents not described in detail in this specification are prior art known to those skilled in the art. Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0038] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0039] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0040] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the invention.

Claims

1. A vehicle operation behavior prediction system during traffic jams, characterized in that, include: The traffic congestion scenario classification module is used to input vehicle driving status data under the current traffic congestion state into the traffic congestion classification model to obtain the current traffic congestion type; The association rule base building module establishes an association rule base between traffic jam type and vehicle operation behavior based on traffic jam type and vehicle historical operation behavior through association rule algorithms. The vehicle operation behavior prediction module is used to predict the vehicle operation behavior at the next moment in the current traffic jam situation when the vehicle is in a traffic jam and there is no corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library. When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the vehicle operation behavior at the next moment in the current traffic jam type is predicted by using the current traffic jam type and the current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base.

2. The vehicle operation behavior prediction system during traffic jams according to claim 1, characterized in that, In the traffic jam scenario classification module, the vehicle driving status data under traffic jam conditions includes the following vehicle driving status characteristics: The average vehicle speed within the preset time window, the percentage of time within the preset time window when the vehicle speed is less than the preset speed, the braking frequency per unit time, the number of times the acceleration is less than the preset acceleration within the preset time window, and the number of vehicles in the preset area around the current vehicle.

3. The vehicle operation behavior prediction system during traffic jams according to claim 2, characterized in that, In the traffic congestion scenario classification module, the traffic congestion classification model is obtained through the following method: Historical vehicle driving status dataset for all vehicles of the corresponding model during traffic jams , , For the corresponding model number Historical vehicle driving status data during traffic jams for each vehicle. For the vehicle Vehicle driving status characteristics; Data set of historical vehicle driving status during traffic jams for all vehicles of the corresponding model. The k-means clustering algorithm is used for classification, thereby classifying the historical traffic jam status of all vehicles into several traffic jam types. Cluster centers are determined based on these traffic jam types, and the number of classified traffic jam types is set as the k value. Based on the cluster centers and the k-value, a k-means clustering algorithm is set to obtain a traffic congestion classification model.

4. The vehicle operation behavior prediction system during traffic jams according to claim 3, characterized in that, In the association rule base module, the specific method for establishing an association rule base between traffic jam type and vehicle operation behavior based on traffic jam type and historical vehicle operation behavior using an association rule algorithm is as follows: The association rule algorithm uses the Apriori algorithm to construct a transaction item set based on traffic jam type i and the historical vehicle operation behavior k of all vehicles of the corresponding vehicle type under traffic jam type i: transaction item set = {traffic jam type i} or {traffic jam type i, historical vehicle operation behavior 1, historical vehicle operation behavior 2, ..., historical vehicle operation behavior j}; The transaction itemsets containing the same traffic jam type are grouped into a corresponding transaction database. The category of the transaction itemset is defined as a k-itemset, and the number of items in the k-itemset is equal to the total number of historical vehicle operation behaviors plus 1. In the Apriori algorithm, the support (A) of a transaction item set A is equal to the number of transaction item sets in a transaction database that contain a subset of transaction item set A, and the total number of transaction item sets in the transaction database. In the Apriori algorithm, the support (A,B) of a transaction item set containing both transaction item set A and transaction item set B is equal to the number of transaction item sets in a transaction database that contain both transaction item sets A and B, divided by the total number of transactions in the database. In the Apriori algorithm, the confidence of transaction itemset B is equal to the support (A, B) / support (A), where transaction itemset A and transaction itemset B have no intersection. For a given transaction database, calculate the support of all 1-itemsets in the transaction database, define transaction itemsets with support greater than or equal to a preset support as frequent itemsets, select frequent 1-itemsets from all 1-itemsets, generate candidate k-itemsets based on frequent 1-itemsets using join and branch pruning in the Apriori algorithm, and select frequent k-itemsets from the candidate k-itemsets. The support (A1) of transaction itemset A1, which is a frequent itemset, is calculated based on all frequent itemsets. The support (A1, B1) of transaction itemset B1, which contains both frequent itemsets, is calculated based on all frequent itemsets. The confidence of transaction itemset B1, which is a frequent itemset, is defined as support (A1, B1) / support (A1). The association rule is defined based on the confidence of transaction itemset B1, which is a frequent itemset: In the case of transaction itemset A1, which is a frequent itemset, there is a probability that transaction itemset B1, which is a frequent itemset, will have the same confidence value as transaction itemset B1. Based on the confidence threshold, the confidence of transaction itemset B1, which is a frequent itemset, is selected from all frequent itemsets. Based on the confidence of the selected frequent itemset B1, a rule base for the association between traffic jam type and vehicle operation behavior is generated.

5. The vehicle operation behavior prediction system during traffic jams according to claim 4, characterized in that, When a vehicle is in a traffic jam and no corresponding vehicle operation behavior is currently performed in the traffic jam type and vehicle operation behavior association rule base, the specific method for predicting the vehicle operation behavior at the next moment under the current traffic jam type in the traffic jam type and vehicle operation behavior association rule base is as follows: When a vehicle is in a traffic jam and no vehicle operation behavior corresponding to the traffic jam type and vehicle operation behavior association rule base is performed, the association rule corresponding to the transaction item set A1 that only includes the current traffic jam type is found in the traffic jam type and vehicle operation behavior association rule base, and the transaction item set B2 in the association rule is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

6. The vehicle operation behavior prediction system during traffic jams according to claim 4, characterized in that, When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the specific method for predicting the vehicle operation behavior at the next moment under the current traffic jam type using the current traffic jam type and current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base is as follows: When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, the transaction item set A1, which consists of the current traffic jam type and the corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, is found. The transaction item B1 in the association rule corresponding to the transaction item set A1 is used as the predicted vehicle operation behavior for the next moment in the current traffic jam type.

7. The vehicle operation behavior prediction system during traffic jams according to claim 6, characterized in that: The traffic jam types include slow creep, complete standstill, and frequent stop-and-go. The slow creep is defined as a vehicle speed between a first preset speed and a second preset speed and maintained for a period of more than a first time window. The complete standstill is defined as a vehicle speed of 0 and maintained for a period of more than a second time window. The frequent stop-and-go is defined as a vehicle speed fluctuating between 0 and a third preset speed. Vehicle operation behaviors include air conditioning operation behaviors, in-vehicle entertainment system operation behaviors, window operation behaviors, seat operation behaviors, and driving operation behaviors.

8. The vehicle operation behavior prediction system during traffic jams according to claim 6, characterized in that, It also includes a vehicle operation behavior execution module, which is used to predict vehicle operation behavior based on association rules in the association rule base when the transaction item set A1 is traffic jam type i. The transaction item B1 is the transaction item set A1 of traffic jam type i and represents vehicle operation behavior. The set of transaction items B1 is called transaction item set C. Using transaction item set C, the vehicle operation behavior is predicted based on association rules in the association rule base. The transaction set D will contain vehicle operation behavior. and vehicle operation behavior The combination forms a preset mode, which controls the corresponding equipment to simultaneously execute all vehicle operation behaviors within the preset mode.

9. A method for predicting vehicle operating behavior during traffic jams, characterized in that, include: Input the vehicle driving status data under the current traffic jam condition into the traffic jam classification model to obtain the current traffic jam type; Based on traffic congestion types and historical vehicle operation behaviors, a rule base for associating traffic congestion types with vehicle operation behaviors is established using an association rule algorithm. When a vehicle is in a traffic jam and there is no corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule library, the vehicle operation behavior at the next moment in the current traffic jam type is predicted in the traffic jam type and vehicle operation behavior association rule library. When a vehicle is in a traffic jam and there is a corresponding vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base, the vehicle operation behavior at the next moment in the current traffic jam type is predicted by using the current traffic jam type and the current vehicle operation behavior in the traffic jam type and vehicle operation behavior association rule base.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the vehicle operation behavior prediction method during traffic jams as described in claim 9.

Images