Logistics transportation online monitoring method and device and storage medium

Abstract

The invention relates to the technical field of transportation monitoring, and discloses a logistics transportation online monitoring method and device and a storage medium, and the logistics transportation online monitoring method comprises the steps: constructing a basic judgment benchmark through a target reference track frame, and further combining historical position sequence information to position an initial deviation position after deviation is detected; accurate data support is provided for subsequent differentiation identification; meanwhile, geographic information of the energy complementing network and vehicle electric quantity information are introduced innovatively, reasonable deviation and abnormal malicious deviation caused by energy complementing requirements are effectively distinguished through energy complementing probability value calculation and energy complementing state judgment, and the early warning misjudgment rate is greatly reduced; and finally, the early warning is triggered only when the unreasonable deviation scene is confirmed, so that the timeliness and the accuracy of logistics transportation monitoring are ensured, the interference of invalid early warning on operation management is avoided, and reliable technical support is provided for improving the logistics transportation efficiency and reducing the operation risk.

Description

Technical Field

[0001] This invention relates to the field of transportation monitoring technology, specifically to an online monitoring method, device, and storage medium for logistics transportation. Background Technology

[0002] In modern logistics and transportation, online monitoring and safety management have become core requirements for ensuring transportation efficiency and reducing operational risks. Track deviation monitoring, as a key function of logistics monitoring systems, aims to promptly identify abnormal vehicle deviations from preset transportation routes, thereby preventing delays, losses, or safety hazards caused by these deviations.

[0003] In related technologies, trajectory deviation judgment often adopts a single location comparison approach. For example, the method for early warning of commercial vehicle dispatch routes based on vehicle-to-everything (V2X) technology disclosed in Chinese patent application CN202311794021.0 directly compares the real-time location of the vehicle with a preset fixed trajectory. Once the location exceeds the trajectory's defined range, it is directly judged as a route deviation and an early warning is triggered. This type of technical solution uses location information as the sole basis for deviation judgment, failing to establish a differentiated identification mechanism that incorporates the actual needs and dynamic factors in the transportation scenario. This results in an inability to effectively distinguish between reasonable deviations and abnormal malicious deviations, ultimately leading to a high false alarm rate. Summary of the Invention

[0004] The main objective of this invention is to provide an online monitoring method, device, and storage medium for logistics transportation, aiming to solve the technical problem of high false alarm rate for vehicle deviation warnings in the prior art.

[0005] To achieve the above objectives, in a first aspect, this application provides an online monitoring method for logistics transportation, the method comprising:

[0006] Obtain the real-time location and battery level information reported by the target transport vehicle;

[0007] The vehicle is determined to have deviated from the target reference trajectory frame based on its real-time location information, wherein the target reference trajectory frame is determined based on the logistics transportation order.

[0008] When a vehicle deviates from the target reference trajectory frame, the target deviation position of the target transport vehicle is predicted based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame;

[0009] The system retrieves all recharge stations within a preset range of the target's deviation location using the recharge network geographic information system, and obtains the real-time station status information of each recharge station.

[0010] The vehicle refueling probability value is determined based on the real-time site status information and the vehicle's current battery level information.

[0011] If the probability value of vehicle recharging is determined to be less than or equal to the probability threshold, and the vehicle is determined not to be recharging or not expected to be recharging based on the real-time battery level change information and vehicle movement status, an early warning of vehicle deviation from the transportation route is triggered.

[0012] In one possible implementation, before determining whether the vehicle has deviated from the target reference trajectory frame based on the real-time location information of the target transport vehicle, the method further includes:

[0013] Obtain the transportation order information and current transportation constraint information of the target transport vehicle, wherein the transportation order information includes origin information, destination information and route node information;

[0014] The initial reference trajectory frame is determined based on the starting point information, ending point information, and route node information.

[0015] The target reference trajectory frame is obtained by dynamically adjusting the initial reference trajectory frame based on the current transportation constraint information.

[0016] In one possible implementation, the current transportation constraint information includes current traffic congestion information and current traffic control information. The step of dynamically adjusting the initial reference trajectory frame based on the current transportation constraint information to obtain the target reference trajectory frame includes:

[0017] Obtain real-time traffic information of the path covered by the initial reference trajectory frame, and identify target congested road sections whose traffic congestion level exceeds a preset threshold;

[0018] Based on the congestion level and congestion length of the target congested road segment, the boundary width of the initial reference trajectory frame around the target congested road segment is dynamically expanded according to preset rules to obtain the candidate reference trajectory frame;

[0019] Obtain current traffic control information and identify target road sections and areas where passage is prohibited or restricted due to traffic control.

[0020] Based on the real-time road network, one or more detour routes are planned for the target road segment area, and the candidate reference trajectory boxes are adjusted according to the detour routes to obtain the target reference trajectory box.

[0021] In one possible implementation, predicting the target deviation position of the target transport vehicle based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame includes:

[0022] Obtain the historical position sequence of the target transport vehicle within a preset time period before it deviates from the target reference trajectory frame;

[0023] Based on the historical location sequence, a path prediction model is used to generate multiple candidate vehicle routes and estimate the probability of occurrence of each candidate route.

[0024] Calculate the intersection point between each candidate path and the boundary line of the target reference trajectory box;

[0025] The intersection points of each candidate path are weighted and fused based on their probability of occurrence, and the calculated weighted average position is determined as the target deviation position.

[0026] In one possible implementation, the step of generating multiple candidate vehicle travel paths based on the historical location sequence using a path prediction model and estimating the probability of occurrence of each candidate path includes:

[0027] The historical location sequence is mapped to the digital road network to determine the historical travel segment sequence of the vehicle in the road network;

[0028] Based on the historical travel route sequence, the vehicle's direction of travel and initial route are determined;

[0029] Starting from the initial road segment of the vehicle in the road network and guided by the direction of travel, multiple paths are explored in the digital road network to generate multiple candidate paths that start from the initial road segment and conform to the direction of vehicle travel.

[0030] For each candidate path, calculate its directional matching degree with the historical location sequence, and obtain the average road grade of the path;

[0031] The probability of occurrence of the candidate path is estimated based on the matching degree and the average road grade; wherein, the candidate path with a higher matching degree with the historical location sequence and a higher average road grade is assigned a higher probability of occurrence.

[0032] In one possible implementation, determining the vehicle recharging probability value based on the site status information and the vehicle's current battery level includes:

[0033] Based on the real-time site status information, candidate power replenishment sites that are idle or about to become idle are selected.

[0034] Calculate the estimated energy consumption of the target transport vehicle as it travels from the target deviation position to each of the candidate refueling stations;

[0035] The vehicle's current battery level is compared with the estimated energy consumption to obtain a first probability sub-value for each candidate charging station.

[0036] Based on the path distance between the target deviation position and each of the candidate refueling stations, a second probability sub-value is obtained for each candidate refueling station;

[0037] Based on the first probability sub-value and the second probability sub-value, the vehicle refueling probability value for each candidate refueling station is obtained;

[0038] The maximum value among all candidate refueling station vehicle refueling probability values ​​is determined as the final vehicle refueling probability value.

[0039] In one possible implementation, after triggering the warning that the vehicle has deviated from the transport route, the method further includes:

[0040] Generate an early warning report that includes the target deviation location, real-time vehicle battery level, a list of nearby charging stations, and a recommended return route;

[0041] The warning report is pushed to the cloud monitoring platform and the vehicle terminal of the target transport vehicle;

[0042] If the driver confirmation information is received from the vehicle terminal and the deviation is reasonable, the warning is lifted and the reason for the deviation is recorded; if no confirmation information is received or the deviation is unreasonable, the cloud monitoring platform initiates a manual intervention process.

[0043] In one possible implementation, determining that the vehicle does not require recharging based on real-time battery level changes and the vehicle's motion status includes:

[0044] The real-time trajectory of the vehicle after deviation is obtained, and the trajectory feature vector is extracted. The trajectory feature vector includes the average speed change rate, the direction angle change frequency, and the distance change rate to the nearest refueling station.

[0045] The trajectory feature vector is input into a pre-trained power replenishment intention recognition model, which is trained based on the correlation between vehicle motion patterns and final power replenishment behavior in historical deviation cases.

[0046] Obtain the expected energy replenishment score output by the model. If the score is lower than a preset threshold, it is determined that the vehicle does not have an expected energy replenishment.

[0047] Secondly, embodiments of this application also provide an online monitoring device for logistics transportation, comprising:

[0048] The status acquisition unit is used to acquire the real-time location information and real-time power information reported by the target transport vehicle.

[0049] The deviation judgment unit is used to determine whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transport order;

[0050] A position prediction unit is used to predict the target deviation position of the target transport vehicle based on historical position sequence information before the vehicle deviates from the target reference trajectory frame when the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame;

[0051] The site information acquisition unit is used to retrieve all replenishment sites within a preset range of the target deviation location through the replenishment network geographic information system, and to acquire the real-time site status information of each replenishment site.

[0052] The probability determination unit is used to determine the vehicle recharging probability value based on the real-time site status information and the vehicle's current battery level information.

[0053] The warning triggering unit is used to determine that the probability value of the vehicle recharging is less than or equal to the probability threshold, and to determine that the vehicle has not recharged based on the real-time power change information of the vehicle, thereby triggering a warning that the vehicle has deviated from the transportation route.

[0054] Thirdly, embodiments of this application also provide a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as described in the first aspect.

[0055] Unlike existing technologies, the online monitoring method for logistics transportation provided in this application first acquires the real-time location information and real-time battery information reported by the target transport vehicle; determines whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transportation order; if the vehicle deviates from the target reference trajectory frame, predicts the target deviation position of the target transport vehicle based on the historical position sequence information of the vehicle before deviating from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame; retrieves all refueling stations within a preset range of the target deviation position through a refueling network geographic information system, and acquires the real-time station status information of each refueling station; determines the vehicle refueling probability value based on the real-time station status information and the vehicle's current battery information; determines that the vehicle refueling probability value is less than or equal to a probability threshold, and determines that the vehicle is not refueling or there is no expected refueling based on the vehicle's real-time battery change information and vehicle movement status, triggering an early warning of vehicle deviation from the transportation route. Thus, this application constructs a basic judgment benchmark through a target reference trajectory frame, and further combines historical location sequence information to locate the initial deviation position after detecting a deviation, providing accurate data support for subsequent differentiated identification. At the same time, it innovatively introduces the geographic information of the energy replenishment network and vehicle battery information, and effectively distinguishes between reasonable deviations caused by energy replenishment needs and abnormal malicious deviations through the calculation of energy replenishment probability values ​​and the determination of energy replenishment status, significantly reducing the false alarm rate. Finally, the alarm is triggered only when an unreasonable deviation is confirmed, which not only ensures the timeliness and accuracy of logistics transportation monitoring, but also avoids the interference of invalid alarms on operation management, providing reliable technical support for improving logistics transportation efficiency and reducing operational risks. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0057] Figure 1 This is a flowchart illustrating the online monitoring method for logistics transportation in some embodiments of this application;

[0058] Figure 2 This is a flowchart illustrating step S300 of the online monitoring method for logistics transportation in some embodiments of this application;

[0059] Figure 3 This is a flowchart illustrating step S500 of the online monitoring method for logistics transportation in some embodiments of this application;

[0060] Figure 4 This is a schematic diagram of the module structure of the online monitoring device for logistics transportation in some embodiments of this application;

[0061] Figure 5 This is a schematic diagram of the hardware structure of the online monitoring system for logistics transportation in some embodiments of this application.

[0062] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0064] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0065] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the term "and / or" throughout the text includes three solutions; taking A and / or B as an example, it includes technical solution A, technical solution B, and a technical solution that simultaneously satisfies A and B. Furthermore, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of a person skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0066] In modern logistics and transportation, online monitoring and safety management have become core requirements for ensuring transportation efficiency and reducing operational risks. Track deviation monitoring, as a key function of logistics monitoring systems, aims to promptly identify abnormal vehicle deviations from preset transportation routes, thereby preventing delays, losses, or safety hazards caused by these deviations.

[0067] In related technologies, trajectory deviation judgment often adopts a single location comparison implementation mode. For example, the method for early warning of commercial vehicle dispatch routes based on vehicle-to-everything (V2X) disclosed in Chinese patent application CN202311794021.0 directly compares the real-time location of the vehicle with a preset fixed trajectory. Once the location exceeds the trajectory's limit range, it is directly judged as a route deviation and an early warning is triggered. However, such technical solutions rely solely on location information as the sole basis for deviation judgment, failing to establish a differentiated identification mechanism that incorporates the actual needs and dynamic factors in the transportation scenario. Especially with the gradual popularization of new energy logistics vehicles, there is an unavoidable need for refueling during vehicle operation (such as charging and battery swapping), and the refueling process usually takes a certain amount of time. Vehicles may need to temporarily deviate from the preset route to reach a refueling station and stop. Existing technologies do not consider such reasonable scenarios and still judge them as abnormal deviations, resulting in a high false alarm rate. This not only interferes with the normal work of logistics monitoring personnel but may also affect the rationality of logistics transportation scheduling.

[0068] To address the aforementioned technical problems, this application provides an online monitoring method for logistics transportation. This method can be applied to a monitoring system, which may include vehicles, a cloud platform, and monitoring terminals. The following description uses the execution of this online monitoring method by a monitoring system as an example. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order. Please refer to the appendix. Figure 1 The method includes the following steps S100-S600:

[0069] Step S100: Obtain the real-time location information and real-time battery information reported by the target transport vehicle;

[0070] The logistics monitoring system can receive real-time data reported by target transport vehicles through vehicle-to-everything (V2X) communication modules (such as 4G / 5G and BeiDou / GPS dual-mode positioning modules). Real-time location information includes key parameters such as the vehicle's latitude and longitude coordinates, positioning timestamp, and positioning accuracy, used to accurately pinpoint the vehicle's current location. Real-time battery information includes data such as the vehicle's remaining battery percentage, current voltage / current, and estimated driving range, reflecting the vehicle's current energy replenishment status. Vehicles can proactively report data at preset reporting intervals (e.g., every 10 seconds) or passively respond when the monitoring system issues a data request, ensuring the monitoring system promptly acquires key vehicle operating data, providing fundamental data support for subsequent trajectory deviation assessments and energy replenishment demand identification. This step is the initial stage of the entire monitoring process, achieving real-time collection and synchronization of key vehicle status data, ensuring the accuracy and timeliness of subsequent judgments.

[0071] Step S200: Determine whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transport order;

[0072] After obtaining the vehicle's real-time location information, the monitoring system can retrieve the logistics transportation order information corresponding to the vehicle. Based on parameters such as the origin, destination, preset waypoints, and transportation route priority in the order, a target reference trajectory frame is constructed using a trajectory frame generation algorithm. Subsequently, the monitoring system spatially compares the real-time location information obtained in step S100 with the geographical boundary of the target reference trajectory frame. If the real-time location coordinates fall within the boundary range of the trajectory frame, the vehicle is determined to be driving normally; if the real-time location coordinates exceed the boundary range of the trajectory frame, the vehicle is determined to be deviating from its course.

[0073] The target reference trajectory frame refers to a virtual geographic area with specific boundaries formed by geospatial modeling, with the preset transportation route corresponding to the logistics transportation order as the core, used to define the spatial threshold for normal vehicle driving; the target reference trajectory frame can be a static reference trajectory frame determined based on the logistics transportation order information, or a dynamic reference trajectory frame dynamically adjusted based on the logistics transportation order information and actual road conditions.

[0074] Step S300: When the vehicle deviates from the target reference trajectory frame, predict the target deviation position of the target transport vehicle based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame;

[0075] When step S200 determines that the vehicle has deviated from the target reference trajectory frame, directly using the last location point within the trajectory frame as the deviation point would introduce significant errors due to vehicle location reporting delays or intervals. Therefore, the monitoring system automatically extracts historical location sequence information within a preset time period before the vehicle's deviation (e.g., all location data within 5 minutes prior to the deviation). This sequence information includes key parameters such as latitude and longitude, timestamp, driving speed, and heading angle for each historical location. Based on this historical location sequence, trajectory prediction algorithms (such as Kalman filter-based trajectory fitting algorithms or LSTM time-series prediction models) are used to reverse-engineer and fit the vehicle's deviation trajectory, accurately locating the geographic coordinates of the first point where the vehicle exceeds the trajectory frame boundary—the target deviation location. This step aims to eliminate errors caused by reporting delays and intervals through algorithmic optimization, clearly defining the starting point of the vehicle's deviation behavior, providing a precise spatial reference for subsequent retrieval of nearby charging stations, and avoiding deviations in the charging station retrieval range due to continued driving after the deviation.

[0076] Step S400: Retrieve all replenishment stations within a preset range of the target deviation location through the replenishment network geographic information system, and obtain the real-time station status information of each replenishment station;

[0077] After determining the target deviation location, the monitoring system can call the Geographic Information System (GIS) of the Energy Supplement Network via an interface and set a preset search range (such as a circular area with a radius of 3 kilometers centered on the target deviation location; this range can be dynamically adjusted according to vehicle type refueling needs and road conditions). Based on spatial indexing technology, the Energy Supplement Network GIS system quickly filters out all energy supplement sites (including charging stations, battery swapping stations, or hydrogen refueling stations) within the search range and returns basic information for each site (such as site name, detailed address, latitude and longitude coordinates, energy supplement type, and number of charging piles / battery swapping stations). Simultaneously, the monitoring system obtains real-time status information for each site through data integration with the energy supplement site operation system, including the number of available energy supplement equipment, the number of vehicles currently in the queue, the estimated refueling wait time, the refueling price, and whether the site is operating normally (e.g., whether it is malfunctioning or closed). This step achieves a comprehensive survey of energy supplement resources around the deviation location, providing data support for subsequent judgments on the probability of refueling.

[0078] Step S500: Determine the vehicle recharge probability value based on the real-time station status information and the vehicle's current battery level information;

[0079] Among them, the vehicle refueling probability value refers to the likelihood that a vehicle will refuel at a refueling station within a preset range after deviating from the target deviation position and deviating from the normal trajectory. The embodiments of this application achieve accurate quantitative assessment of the refueling probability through the fusion of two-dimensional data of "vehicle energy demand - surrounding refueling supply", providing key quantitative basis for distinguishing between reasonable refueling deviation and abnormal malicious deviation.

[0080] For example, if a vehicle's current battery level is only 5% (with a remaining range of approximately 15 kilometers), and there are three operational fast-charging stations within a 2-kilometer radius of the target location, with two of these stations having available charging equipment and an average waiting time of only 10 minutes, and calculations show that the vehicle's remaining battery level is sufficient to reach any of these stations, it indicates that the vehicle's energy demand is urgent and the surrounding charging supply is sufficient. In this case, the charging probability value calculated by the monitoring system will be in a relatively high range (e.g., 0.7-0.9). Conversely, if the vehicle's current battery level is 60%, and there is only one charging station within a 5-kilometer radius of the target location that is out of service, then the urgency of the charging demand is low and the supply is insufficient, and the charging probability value will be in a relatively low range (e.g., 0.1-0.3).

[0081] Step S600: Determine that the vehicle's recharge probability value is less than or equal to the probability threshold, and determine that the vehicle is not recharging or there is no expectation of recharging based on the vehicle's real-time battery level change information and vehicle movement status, triggering an early warning that the vehicle is deviating from the transportation route.

[0082] The monitoring system first compares the replenishment probability value calculated in step S500 with a preset probability threshold. If the replenishment probability value is greater than the probability threshold, it is determined that the vehicle deviation is likely a reasonable replenishment behavior, and no warning is triggered temporarily, while the vehicle status is continuously monitored. If the replenishment probability value is less than or equal to the probability threshold, a second verification is performed by combining the vehicle's real-time battery level change information and movement status: the real-time battery level change information is used to determine whether the vehicle is in a replenishment state (if the battery level is continuously rising and the rate of increase matches the replenishment characteristics, it is determined that it is replenishing; if the battery level is continuously decreasing or remains stable and there is no upward trend, it is determined that it is not replenishing); the movement status information is used to determine whether the vehicle has a replenishment expectation. A replenishment expectation means that the vehicle is on its way to a replenishment station and expects to replenish. If the vehicle speed is 0 and the continuous dwell time exceeds 5 minutes, and the location coincides with the coordinates of the replenishment station, then there is a replenishment expectation; if the vehicle speed is greater than 0 and the vehicle is continuously driving, and the location is far from the replenishment station, then there is no replenishment expectation.

[0083] When the probability value of refueling is less than or equal to the probability threshold, and refueling is neither expected nor anticipated, the vehicle deviation is determined to be an abnormal and malicious deviation, and the monitoring system immediately triggers an early warning mechanism. This step, through dual verification logic, achieves accurate differentiation between reasonable and abnormal deviations, effectively reducing the false alarm rate.

[0084] Based on this, this application constructs a basic judgment benchmark through a target reference trajectory frame, and further locates the initial deviation position by combining historical location sequence information after detecting deviation, providing accurate data support for subsequent differentiated identification. At the same time, it innovatively introduces the geographic information of the energy replenishment network and the vehicle's battery information, and effectively distinguishes between reasonable deviations caused by energy replenishment needs and abnormal malicious deviations by calculating the energy replenishment probability value and determining the energy replenishment status, significantly reducing the false alarm rate. Finally, the alarm is triggered only when an unreasonable deviation is confirmed, which not only ensures the timeliness and accuracy of logistics transportation monitoring, but also avoids the interference of invalid alarms on operation management, providing reliable technical support for improving logistics transportation efficiency and reducing operational risks.

[0085] In one embodiment, before determining whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, the method further includes: obtaining transport order information and current transport constraint information of the target transport vehicle, wherein the transport order information includes origin information, destination information and route node information; determining an initial reference trajectory frame based on the origin information, destination information and route node information; and dynamically adjusting the initial reference trajectory frame based on the current transport constraint information to obtain the target reference trajectory frame.

[0086] Specifically, the process begins by acquiring the transport order information and current transport constraints of the target vehicle. The transport order information serves as the foundation for trajectory planning. It includes origin information defining the vehicle's initial departure location, destination information defining the final destination of the transport task, and route node information identifying key locations encountered during transport (such as transfer stations and designated checkpoints). These elements collectively constitute the core node elements of trajectory planning. Current transport constraints are crucial variables for dynamic trajectory adaptation. They can include real-time traffic conditions (such as congested or under construction), weather conditions (such as heavy rain or fog affecting traffic), road control information (such as temporary road closures and traffic restrictions), and vehicle-specific constraints (such as remaining battery power and load limits), providing real-time dynamic data for subsequent trajectory adjustments.

[0087] Subsequently, an initial reference trajectory frame is determined based on the starting point information, the ending point information, and the information of the nodes along the way. This process can be implemented using trajectory planning algorithms (such as Dijkstra's algorithm, A* algorithm, etc.). Taking the starting point as the starting node and the ending point as the target node, and the nodes along the way as intermediate constraint nodes that must be passed, the preliminary optimal path connecting each node is calculated. Then, based on this preliminary path and combined with parameters such as the safe width of vehicle driving and steering redundancy, an initial reference trajectory frame with a certain range of boundaries is constructed to provide preliminary trajectory guidance range for vehicle driving.

[0088] Next, the initial reference trajectory frame is dynamically adjusted based on the current transportation constraints to obtain the target reference trajectory frame. This adjustment process requires targeted adaptation to the different types of constraints: for example, if the current transportation constraints indicate congested sections within the initial trajectory frame, the algorithm can replan a route around the congested area and adjust the trajectory frame boundaries; if road closures prevent passage on certain sections, the trajectory portion corresponding to the closure section is directly removed, and a new, compliant trajectory frame is generated; if the vehicle's remaining battery power is insufficient, charging stations can be added within the trajectory frame as temporary transit nodes, and the path and boundary range adjusted. This dynamic adjustment method allows the initial reference trajectory frame to adapt to real-time changes in the transportation environment, ultimately resulting in an accurate and feasible target reference trajectory frame.

[0089] Thus, by first obtaining basic order information and real-time constraint information, and then sequentially determining the initial trajectory frame and dynamically adjusting it to the target trajectory frame, this embodiment of the application can provide accurate reference for subsequent vehicle deviation judgment, thereby improving the rationality of the transportation trajectory and the execution efficiency of the transportation task.

[0090] In one embodiment, the current transportation constraint information includes current traffic congestion information and current traffic control information. The step of dynamically adjusting the initial reference trajectory frame according to the current transportation constraint information to obtain the target reference trajectory frame includes:

[0091] Obtain real-time traffic information of the path covered by the initial reference trajectory frame, and identify target congested road sections whose traffic congestion level exceeds a preset threshold;

[0092] Based on the congestion level and congestion length of the target congested road segment, the boundary width of the initial reference trajectory frame around the target congested road segment is dynamically expanded according to preset rules to obtain the candidate reference trajectory frame;

[0093] Obtain current traffic control information and identify target road sections and areas where passage is prohibited or restricted due to traffic control.

[0094] Based on the real-time road network, one or more detour routes are planned for the target road segment area, and the candidate reference trajectory boxes are adjusted according to the detour routes to obtain the target reference trajectory box.

[0095] Specifically, the system first acquires real-time traffic information for the path covered by the initial reference trajectory frame, identifying target congested road segments whose congestion levels exceed preset thresholds. Real-time traffic information can be obtained through public data interfaces of traffic management departments, real-time data acquisition modules of in-vehicle navigation systems, or third-party traffic information service platforms. This information includes key parameters such as traffic speed, vehicle density, and queue length. Congestion levels can be defined according to preset evaluation criteria (e.g., traffic speeds below 20 km / h are considered severe congestion, and 20-40 km / h are considered moderate congestion). Preset thresholds can be flexibly configured based on the timeliness requirements of transportation tasks (e.g., thresholds can be set to moderate congestion for emergency transportation tasks and severe congestion for ordinary tasks). By comparing the congestion levels of each road segment with preset thresholds, target congested road segments requiring priority handling are accurately identified.

[0096] Subsequently, based on the congestion level and length of the target congested road segment, the boundary width of the initial reference trajectory frame around the target congested road segment is dynamically expanded according to preset rules to obtain candidate reference trajectory frames. The preset rules must fully consider the impact of congestion level on vehicle maneuverability. For example, for heavily congested road segments exceeding 500 meters in length, the trajectory frame boundary can be expanded by 3-5 meters on each side to provide vehicles with more space for lane changing and avoidance; for moderately congested road segments between 200-500 meters in length, it can be expanded by 1-3 meters on each side; for short-distance congested road segments less than 200 meters in length, it can be expanded only slightly by 0.5-1 meter to avoid excessive expansion leading to redundancy in the trajectory frame range. This differentiated boundary expansion method ensures both vehicle safety in congested road segments and maintains the overall accuracy of the trajectory frame, forming candidate reference trajectory frames.

[0097] Next, current traffic control information is obtained to identify target road sections or areas where passage is prohibited or restricted due to traffic control measures. Traffic control information can be obtained through traffic radio broadcasts, real-time notifications from traffic management departments, or the control information push function of navigation systems. These types of information include temporary road closures (such as road construction or security for large-scale events), one-way traffic, and time-limited passage. During the identification process, the location information of the road section, the control time range, and the expected travel time of transport vehicles must be matched. If a vehicle is expected to pass through a road section that is under control (prohibited or restricted), then that road section is designated as the target road section area to ensure that no key traffic restriction factors are overlooked.

[0098] Finally, based on the real-time road network, one or more detour routes for the target road segment area are planned, and the candidate reference trajectory boxes are adjusted according to the detour routes to obtain the target reference trajectory box. Detour route planning can use path optimization algorithms (such as A* algorithm, genetic algorithm, etc.), with "shortest detour distance", "minimum travel time" or "lowest travel cost" as optimization objectives. Combining information such as road segment traffic conditions and intersection turning restrictions in the real-time road network, 1-3 alternative detour routes are generated. Then, by evaluating the feasibility of each alternative route (such as whether it complies with vehicle traffic restrictions and whether there are potential congestion points), the optimal detour route is selected. The portion of the candidate reference trajectory box corresponding to the target road segment area is replaced with the trajectory segment corresponding to the optimal detour route, and the boundary width of the new trajectory segment is adjusted simultaneously to finally form a target reference trajectory box that adapts to the dual constraints of congestion and traffic control.

[0099] Thus, this embodiment of the application processes the two types of constraint information, congestion and control, in steps. First, it optimizes the trajectory boundary of the congested road segment, and then plans the detour path of the controlled road segment. This achieves precise dynamic adjustment of the initial reference trajectory frame, thereby providing an accurate and reliable reference for subsequent vehicle trajectory deviation judgment.

[0100] In one embodiment, such as Figure 2 As shown, step S300: predicting the target deviation position of the target transport vehicle based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame, including:

[0101] S310. Obtain the historical position sequence of the target transport vehicle within a preset time period before it deviates from the target reference trajectory frame;

[0102] S320. Based on the historical location sequence, a path prediction model is used to generate multiple candidate vehicle operation paths and estimate the probability of occurrence of each candidate path.

[0103] S330. Calculate the intersection point of each candidate path with the boundary line of the target reference trajectory box;

[0104] S340. Based on the probability of occurrence of each candidate path, the corresponding intersection points are weighted and fused, and the calculated weighted average position is determined as the target deviation position.

[0105] Specifically, firstly, the historical position sequence of the target transport vehicle within a preset time period (5 minutes) before deviating from the target reference trajectory frame is obtained; then, based on the historical position sequence, multiple candidate vehicle operation paths are generated using a path prediction model, and the probability of occurrence of each candidate path is estimated; next, the intersection point of each candidate path with the boundary line of the target reference trajectory frame is calculated; finally, the intersection points corresponding to each candidate path are weighted and fused based on the probability of occurrence of each candidate path, and the calculated weighted average position is determined as the target deviation position.

[0106] Thus, by acquiring historical location sequences, generating candidate paths with probabilities, calculating boundary intersections, and weighted fusion, this embodiment of the application can accurately predict the target deviation position of the vehicle, thereby providing an accurate spatial reference for subsequent retrieval of surrounding refueling stations.

[0107] It is understandable that the historical location sequence is only a sequence of changes in vehicle coordinates during passage and does not directly correspond to the actual operating trajectory in the digital road network. Therefore, to improve the accuracy of target deviation location acquisition, in one embodiment, the steps include: based on the historical location sequence, using a path prediction model to generate multiple candidate vehicle operating paths and estimating the probability of occurrence of each candidate path, including:

[0108] The historical location sequence is mapped to the digital road network to determine the historical travel segment sequence of the vehicle in the road network;

[0109] Based on the historical travel route sequence, the vehicle's direction of travel and initial route are determined;

[0110] Starting from the initial road segment of the vehicle in the road network and guided by the direction of travel, multiple paths are explored in the digital road network to generate multiple candidate paths that start from the initial road segment and conform to the direction of vehicle travel.

[0111] For each candidate path, calculate its directional matching degree with the historical location sequence, and obtain the average road grade of the path;

[0112] The probability of occurrence of the candidate path is estimated based on the matching degree and the average road grade; wherein, the candidate path with a higher matching degree with the historical location sequence and a higher average road grade is assigned a higher probability of occurrence.

[0113] Specifically, the historical location sequence (composed of coordinate data) is first mapped to a preset digital road network, and the historical driving segment sequence of the vehicle in the digital road network is determined by the road network matching algorithm. The historical driving segment sequence includes the starting segment of the vehicle's driving trajectory, the intermediate continuous segments, and the ending segment of the deviation from the trajectory. The determination of this segment sequence provides a basic road network reference for subsequent driving direction extraction and path expansion.

[0114] Then, based on the historical driving segment sequence, the vehicle's running direction is extracted through a trajectory direction analysis algorithm, and the starting segment of the historical driving segment sequence is identified as the initial segment for path prediction.

[0115] Next, taking the initial road segment as the starting point and the extracted running direction as the main guide, the path is expanded in the digital road network through a multi-branch path exploration algorithm. During the expansion process, the topological structure of the digital road network (such as intersection connection relationship and lane traffic rules) is strictly combined to constrain the process and avoid generating invalid paths that do not conform to the actual road network traffic logic. Finally, multiple candidate vehicle running paths starting from the initial road segment and consistent with the vehicle running direction are constructed.

[0116] Then, for each constructed candidate path, a directional similarity calculation model (such as the vector angle cosine algorithm) is used to calculate its directional matching degree with the historical location sequence. This matching degree can quantify the degree of fit between the candidate path and the vehicle's historical driving direction. At the same time, based on the preset levels of each road segment in the digital road network (such as highways, arterial roads, secondary arterial roads, and branch roads), the average road level of the candidate path is statistically obtained. The level of the average road level directly reflects the traffic efficiency and priority of the path.

[0117] Finally, the direction matching degree and average road grade are used as input parameters, and the probability of occurrence of each candidate path is estimated by weighted calculation (such as setting matching degree weight coefficient and road grade weight coefficient). The output of the probability estimation model satisfies the following: the candidate path with higher direction matching degree and higher average road grade is assigned a higher probability value. This design conforms to the actual law that vehicles prioritize paths with consistent direction and high traffic efficiency during driving, ensuring the rationality and accuracy of probability estimation.

[0118] Thus, the embodiments of this application can construct multiple candidate paths with actual passability based on historical location sequences, effectively making up for the deficiency that simply relying on coordinate sequences cannot reflect the actual trajectory of the road network, thereby providing reliable data support for the accurate positioning and calibration of the target's deviation position.

[0119] In one embodiment, such as Figure 3 As shown, step S500: determining the vehicle recharging probability value based on the station status information and the vehicle's current battery level information, including:

[0120] S510. Based on the real-time site status information, filter out candidate power replenishment sites that are idle or about to become idle.

[0121] S520. Calculate the estimated energy consumption of the target transport vehicle as it travels from the target deviation position to each of the candidate refueling stations;

[0122] S530. Compare the vehicle's current battery level with the estimated energy consumption to obtain a first probability sub-value for each candidate charging station.

[0123] S540. Based on the path distance between the target deviation position and each of the candidate refueling stations, obtain a second probability sub-value for each candidate refueling station.

[0124] S550. Based on the first probability sub-value and the second probability sub-value, obtain the vehicle refueling probability value for each candidate refueling station.

[0125] S560. The maximum value among all candidate refueling station vehicle refueling probability values ​​is determined as the final vehicle refueling probability value.

[0126] Specifically, firstly, candidate charging stations that are idle or about to become idle are selected based on real-time station status information; then, the vehicle's current battery level is compared with the estimated energy consumption to obtain a first probability sub-value for each candidate charging station; next, a second probability sub-value is obtained for each candidate charging station based on the path distance between the target's deviation position and each candidate charging station; then, the charging probability value for each candidate charging station is calculated based on the first and second probability sub-values, which can be obtained by weighted summation; finally, the maximum value among the vehicle charging probability values ​​corresponding to all candidate charging stations is determined as the final vehicle charging probability value.

[0127] The logic for determining the first probability sub-value can be as follows: if the current remaining power is greater than or equal to the estimated energy consumption plus the preset safe power (e.g., 5 kWh), then the first probability sub-value corresponds to a higher probability (e.g., 0.8-1.0); if the current remaining power is greater than or equal to the estimated energy consumption and less than the estimated energy consumption plus the preset safe power, then the first probability sub-value corresponds to a medium probability (e.g., 0.5-0.7); if the current remaining power is less than the estimated energy consumption, then the first probability sub-value corresponds to a lower probability (e.g., 0-0.4). This comparison method quantifies the feasibility of the vehicle reaching the candidate charging station based on its power, ensuring that the basic conditions for charging are met. The logic for determining the second probability sub-value can be based on the distance; the greater the path distance, the lower the corresponding probability value.

[0128] In other words, the technical solution of this application can select the optimal charging station and determine the vehicle charging probability value corresponding to the station based on the station status information and the vehicle's current battery level information. If the charging probability value corresponding to the optimal charging station is still at a low level, it means that the target transport vehicle is less likely to be able to charge reasonably and there is a greater possibility of abnormal deviation.

[0129] In one embodiment, the step of determining that the vehicle does not require recharging based on the vehicle's real-time battery level changes and vehicle motion status includes:

[0130] The real-time trajectory of the vehicle after deviation is obtained, and the trajectory feature vector is extracted. The trajectory feature vector includes the average speed change rate, the direction angle change frequency, and the distance change rate to the nearest refueling station.

[0131] The trajectory feature vector is input into a pre-trained power replenishment intention recognition model, which is trained based on the correlation between vehicle motion patterns and final power replenishment behavior in historical deviation cases.

[0132] Obtain the expected energy replenishment score output by the model. If the score is lower than a preset threshold, it is determined that the vehicle does not have an expected energy replenishment.

[0133] Specifically, the system first acquires the vehicle's real-time trajectory after deviating from the preset driving route. A trajectory acquisition module (such as a GPS positioning unit or inertial navigation sensor) continuously records the vehicle's position coordinates, speed, and direction information. Trajectory feature vectors are then extracted from the acquired trajectory data. These extracted feature vectors are then input into a pre-trained refueling intention recognition model. This model is trained based on a historical deviation case dataset, constructing a mapping model by mining the intrinsic correlation between vehicle movement patterns and the final refueling behavior. Finally, the refueling intention recognition model outputs a refueling expectation score, calculated based on the similarity between real-time trajectory features and historical refueling case features (the score range can be set to 0-100 points, with higher scores indicating a greater probability of refueling). A preset threshold is a judgment threshold determined according to the actual application scenario (e.g., by statistically analyzing the lowest score distribution of historical non-refueling cases, the threshold is set to 30 points). If the model's output refueling expectation score is lower than this preset threshold, it indicates that the current vehicle's movement pattern highly matches the movement characteristics of historical non-refueling cases, thus determining that the vehicle does not have a refueling expectation.

[0134] Thus, this embodiment of the application extracts multi-dimensional trajectory features after the vehicle deviates and combines them with a refueling intention recognition model trained based on historical data to achieve accurate determination of the vehicle's refueling expectation.

[0135] In one embodiment, after triggering a warning about a vehicle deviating from its transport route, the method further includes:

[0136] Generate an early warning report that includes the target deviation location, real-time vehicle battery level, a list of nearby charging stations, and a recommended return route;

[0137] The warning report is pushed to the cloud monitoring platform and the vehicle terminal of the target transport vehicle;

[0138] If the driver confirmation information is received from the vehicle terminal and the deviation is reasonable, the warning is lifted and the reason for the deviation is recorded; if no confirmation information is received or the deviation is unreasonable, the cloud monitoring platform initiates a manual intervention process.

[0139] Specifically, after triggering an alert for a vehicle deviating from its transport route, a warning report is generated, including the target deviation location, the vehicle's real-time battery level, a list of nearby charging stations, and a recommended return route. This report is then simultaneously pushed to the cloud monitoring platform and the onboard terminal of the target transport vehicle, enabling multi-terminal information sharing. When pushed to the onboard terminal, the driver is notified prominently via pop-ups, voice prompts, or other eye-catching methods, ensuring the driver is promptly aware of the deviation and related auxiliary information, facilitating quick judgment and response.

[0140] The system continuously receives driver confirmation messages from the in-vehicle terminal. After reviewing the warning report, drivers can provide confirmation messages through the in-vehicle terminal based on the actual transportation scenario (such as sudden road conditions or temporary transportation needs). If the confirmation message indicates a reasonable deviation (e.g., the driver submits a temporary transportation task notification or proof of sudden malfunction), the system automatically lifts the warning and records the specific reason for the deviation (e.g., "temporary delivery need" or "road construction detour") through the in-vehicle terminal or cloud platform, creating a deviation processing file to provide data for subsequent route optimization and management strategy adjustments. If no driver confirmation message is received within a preset time (e.g., 5 minutes), or if the driver's confirmation message indicates an unreasonable deviation (e.g., arbitrary deviation without any reasonable supporting materials), the cloud monitoring platform automatically triggers a manual intervention process. Monitoring personnel contact the driver via telephone, in-vehicle voice, etc., to verify the specific reason for the deviation and guide the driver on subsequent actions (e.g., immediately return to base or adjust the route) to ensure the standardization and safety of the transportation process.

[0141] Thus, in this embodiment of the application, after a vehicle deviates from the transportation route, a comprehensive warning report is generated, multi-terminal information is pushed, and a handling mechanism is built that allows for driver self-confirmation and manual intervention as a fallback. This ensures that deviations are handled in a timely and reasonable manner and forms a complete deviation handling closed loop, thereby improving the efficiency and safety of monitoring and management of the transportation process.

[0142] like Figure 4 As shown, Figure 4 This is a schematic diagram of the module structure of the online monitoring device for logistics transportation in some embodiments of this application. This application also provides an online monitoring device for logistics transportation, including:

[0143] The status acquisition unit 100 is used to acquire the real-time location information and real-time power information reported by the target transport vehicle.

[0144] Deviation determination unit 200 is used to determine whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transport order;

[0145] The position prediction unit 300 is used to predict the target deviation position of the target transport vehicle based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame when the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame.

[0146] The site information acquisition unit 400 is used to retrieve all replenishment sites within a preset range of the target deviation location through the replenishment network geographic information system, and to acquire the real-time site status information of each replenishment site.

[0147] The probability determination unit 500 is used to determine the vehicle refueling probability value based on the real-time station status information and the vehicle's current battery level information.

[0148] The warning triggering unit 600 is used to determine that the probability value of the vehicle recharging is less than or equal to the probability threshold, and to determine that the vehicle has not recharged based on the real-time power change information of the vehicle, thereby triggering a warning that the vehicle has deviated from the transportation route.

[0149] like Figure 5 As shown, Figure 5 The above is a schematic diagram of the hardware structure of the online monitoring system for logistics transportation in some embodiments of this application. This application also provides an online monitoring system for logistics transportation, including a memory 1000 and a processor 2000. The memory 1000 is used to store computer-readable instructions, and the processor 2000 is used to call the computer-readable instructions to execute the online monitoring method for logistics transportation as described above.

[0150] The processor 2000 provides computing and control capabilities to control the online monitoring system for logistics transportation to perform corresponding tasks. For example, it controls the online monitoring system for logistics transportation to perform the online monitoring method for logistics transportation in any of the above method embodiments. The method includes: acquiring real-time location information and real-time battery information reported by the target transport vehicle; determining whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transportation order; predicting the target deviation position of the target transport vehicle based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame when the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame; retrieving all refueling stations within a preset range of the target deviation position through a refueling network geographic information system and acquiring the real-time station status information of each refueling station; determining the vehicle refueling probability value based on the real-time station status information and the vehicle's current battery information; determining that the vehicle refueling probability value is less than or equal to a probability threshold, and determining that the vehicle is not refueling or there is no expected refueling based on the vehicle's real-time battery change information and the vehicle's movement status, triggering an early warning of vehicle deviation from the transportation route.

[0151] The processor 2000 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0152] The memory 1000, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the online monitoring method for logistics transportation in the embodiments of this application. The processor 2000 can implement the online monitoring method for logistics transportation in any of the above method embodiments by running the non-transitory software programs, instructions, and modules stored in the memory 1000.

[0153] Specifically, memory 1000 may include volatile memory (VM), such as random access memory (RAM); memory 1000 may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), or other non-transitory solid-state storage devices; memory 1000 may also include combinations of the above types of memory.

[0154] In summary, the logistics transportation online monitoring system of this application adopts the technical solution of any of the above-mentioned logistics transportation online monitoring method embodiments. Therefore, it has at least the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated here.

[0155] This application also provides a computer-readable storage medium, such as a memory including program code, which can be executed by a processor to complete the online monitoring method for logistics transportation described above. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CDROM), magnetic tape, floppy disk, or optical data storage device, etc.

[0156] This application also provides a computer program product comprising one or more lines of program code stored in a computer-readable storage medium. The processor of the early warning system reads the program code from the computer-readable storage medium and executes the program code to complete the steps of the online logistics transportation monitoring method provided in the above embodiments.

[0157] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program or program code related to hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0158] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0159] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software and a general-purpose hardware platform, or of course, using hardware. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0160] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for online monitoring of logistics transportation, characterized in that, The method includes: Obtain the real-time location and battery level information reported by the target transport vehicle; The vehicle is determined to have deviated from the target reference trajectory frame based on its real-time location information, wherein the target reference trajectory frame is determined based on the logistics transportation order. When a vehicle deviates from the target reference trajectory frame, the target deviation position of the target transport vehicle is predicted based on the historical position sequence information before the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame; The system retrieves all recharge stations within a preset range of the target's deviation location using the recharge network geographic information system, and obtains the real-time station status information of each recharge station. The vehicle refueling probability value is determined based on the real-time site status information and the vehicle's current battery level information. If the probability value of vehicle recharging is determined to be less than or equal to the probability threshold, and the vehicle is determined not to be recharging or not expected to be recharging based on the real-time battery level change information and vehicle movement status, an early warning of vehicle deviation from the transportation route is triggered.

2. The logistics transport online monitoring method of claim 1, wherein, Before determining whether the vehicle has deviated from the target reference trajectory frame based on the real-time location information of the target transport vehicle, the method further includes: Obtain the transportation order information and current transportation constraint information of the target transport vehicle, wherein the transportation order information includes origin information, destination information and route node information; The initial reference trajectory frame is determined based on the starting point information, ending point information, and route node information. The target reference trajectory frame is obtained by dynamically adjusting the initial reference trajectory frame based on the current transportation constraint information.

3. The logistics transport online monitoring method of claim 2, wherein, The current transportation constraint information includes current road congestion information and current traffic control information. The step of dynamically adjusting the initial reference trajectory frame based on the current transportation constraint information to obtain the target reference trajectory frame includes: Obtain real-time traffic information of the path covered by the initial reference trajectory frame, and identify target congested road sections whose traffic congestion level exceeds a preset threshold; Based on the congestion level and congestion length of the target congested road segment, the boundary width of the initial reference trajectory frame around the target congested road segment is dynamically expanded according to preset rules to obtain the candidate reference trajectory frame; Obtain current traffic control information and identify target road sections and areas where passage is prohibited or restricted due to traffic control. Based on the real-time road network, one or more detour routes are planned for the target road segment area, and the candidate reference trajectory boxes are adjusted according to the detour routes to obtain the target reference trajectory box.

4. The logistics transport online monitoring method of claim 1, wherein, The step of predicting the target deviation position of the target transport vehicle based on the historical position sequence information of the vehicle before it deviates from the target reference trajectory frame includes: Obtain the historical position sequence of the target transport vehicle within a preset time period before it deviates from the target reference trajectory frame; Based on the historical location sequence, a path prediction model is used to generate multiple candidate vehicle routes and estimate the probability of occurrence of each candidate route. Calculate the intersection point between each candidate path and the boundary line of the target reference trajectory box; The intersection points of each candidate path are weighted and fused based on their probability of occurrence, and the calculated weighted average position is determined as the target deviation position.

5. The logistics transport online monitoring method of claim 4, wherein, The step of generating multiple candidate vehicle routes and estimating the probability of occurrence of each candidate route based on the historical location sequence using a route prediction model includes: The historical location sequence is mapped to the digital road network to determine the historical travel segment sequence of the vehicle in the road network; Based on the historical travel route sequence, the vehicle's direction of travel and initial route are determined; Starting from the initial road segment of the vehicle in the road network and guided by the direction of travel, multiple paths are explored in the digital road network to generate multiple candidate paths that start from the initial road segment and conform to the direction of vehicle travel. For each candidate path, calculate its directional matching degree with the historical location sequence, and obtain the average road grade of the path; The probability of occurrence of the candidate path is estimated based on the matching degree and the average road grade; wherein, the candidate path with a higher matching degree with the historical location sequence and a higher average road grade is assigned a higher probability of occurrence.

6. The logistics transport online monitoring method of claim 1, wherein, The step of determining the vehicle recharging probability value based on the station status information and the vehicle's current battery level information includes: Based on the real-time site status information, candidate power replenishment sites that are idle or about to become idle are selected. Calculate the estimated energy consumption of the target transport vehicle as it travels from the target deviation position to each of the candidate refueling stations; The vehicle's current battery level is compared with the estimated energy consumption to obtain a first probability sub-value for each candidate charging station. Based on the path distance between the target deviation position and each of the candidate refueling stations, a second probability sub-value is obtained for each candidate refueling station; Based on the first probability sub-value and the second probability sub-value, the vehicle refueling probability value for each candidate refueling station is obtained; The maximum value among all candidate refueling station vehicle refueling probability values ​​is determined as the final vehicle refueling probability value.

7. The logistics transport online monitoring method of claim 1, wherein, After triggering the warning that the vehicle has deviated from the transport route, the method further includes: Generate an early warning report that includes the target deviation location, real-time vehicle battery level, a list of nearby charging stations, and a recommended return route; The warning report is pushed to the cloud monitoring platform and the vehicle terminal of the target transport vehicle; If the driver confirmation information is received from the vehicle terminal and the deviation is reasonable, the warning is lifted and the reason for the deviation is recorded; if no confirmation information is received or the deviation is unreasonable, the cloud monitoring platform initiates a manual intervention process.

8. The logistics transport online monitoring method of claim 1, wherein, The determination that the vehicle does not need recharging based on real-time battery level changes and vehicle movement includes: The real-time trajectory of the vehicle after deviation is obtained, and the trajectory feature vector is extracted. The trajectory feature vector includes the average speed change rate, the direction angle change frequency, and the distance change rate to the nearest refueling station. The trajectory feature vector is input into a pre-trained power replenishment intention recognition model, which is trained based on the correlation between vehicle motion patterns and final power replenishment behavior in historical deviation cases. Obtain the expected energy replenishment score output by the model. If the score is lower than a preset threshold, it is determined that the vehicle does not have an expected energy replenishment.

9. An online monitoring device for logistics transportation, characterized in that, include: The status acquisition unit is used to acquire the real-time location information and real-time power information reported by the target transport vehicle. The deviation judgment unit is used to determine whether the vehicle deviates from the target reference trajectory frame based on the real-time location information of the target transport vehicle, wherein the target reference trajectory frame is determined based on the logistics transport order; A position prediction unit is used to predict the target deviation position of the target transport vehicle based on historical position sequence information before the vehicle deviates from the target reference trajectory frame when the vehicle deviates from the target reference trajectory frame, wherein the target deviation position represents the initial position of the vehicle deviating from the target reference trajectory frame; The site information acquisition unit is used to retrieve all replenishment sites within a preset range of the target deviation location through the replenishment network geographic information system, and to acquire the real-time site status information of each replenishment site. The probability determination unit is used to determine the vehicle recharging probability value based on the real-time site status information and the vehicle's current battery level information. The warning triggering unit is used to determine that the probability value of the vehicle recharging is less than or equal to the probability threshold, and to determine that the vehicle has not recharged based on the real-time power change information of the vehicle, thereby triggering a warning that the vehicle has deviated from the transportation route.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images