Vehicle task execution method, device, processor, electronic device and vehicle

Abstract

This invention discloses a method, apparatus, processor, electronic device, and vehicle for task execution. The method includes: acquiring state information of a vehicle set, the task of the vehicle set, and road information of the vehicle set during the execution of the task. The state information indicates the idle state and / or location state of the vehicles in the vehicle set; the task is executed through vehicle transport operation instructions or transfer operation instructions; and the road information indicates the traffic congestion level of the current road segment and / or the environmental condition of the road. Based on the state information, the task, and the road information, a target path for the vehicle is determined; based on the target path, a task execution strategy for the target vehicle in the vehicle set is determined; and according to the task execution strategy, the target vehicle is controlled to execute the task along the target path. This invention solves the technical problem of low efficiency in vehicle task execution.

Description

Technical Field

[0001] This invention relates to the field of vehicle logistics technology, and more specifically, to a vehicle task execution method, apparatus, processor, electronic device, and vehicle. Background Technology

[0002] Currently, during vehicle task execution, target path planning is often static and fixed, resource matching lacks real-time dynamism, and anomaly responses rely on manual intervention. This makes it difficult for vehicle logistics scheduling systems to adapt to complex and ever-changing real-world transportation scenarios, hindering efficient, intelligent, and adaptive end-to-end collaborative optimization. Therefore, the technical problem of low efficiency in vehicle task execution persists.

[0003] There is currently no effective solution to the aforementioned technical problems. Summary of the Invention

[0004] The present invention provides a vehicle task execution method, apparatus, processor, electronic device and vehicle to at least solve the technical problem of low efficiency in vehicle task execution.

[0005] According to one aspect of the present invention, a vehicle task execution method is provided. The method includes: acquiring state information of a vehicle set, a task assigned to the vehicle set, and road information of the vehicle set during the execution of the task, wherein the state information indicates the idle state and / or location state of the vehicles in the vehicle set, the task is executed via a vehicle transport operation command or transfer operation command, and the road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located; determining a target path for the vehicle based on the state information, the task, and the road information, wherein the target path represents a path that satisfies both the state information and road conditions during the execution of the task; determining a task execution strategy for a target vehicle in the vehicle set based on the target path, wherein the task execution strategy represents the rules by which the target vehicle executes the task; and controlling the target vehicle to execute the task along the target path according to the task execution strategy.

[0006] Optionally, according to the task execution strategy, controlling the target vehicle to perform the work task along the target path includes: in response to detecting a deviation in the target path or a delay in the execution of the work task, using the scheduling module to adjust the task execution strategy; and controlling the target vehicle to perform the work task along the target path according to the adjusted task execution strategy.

[0007] Optionally, the target path of the vehicle is determined based on status information, task information, and road information, including: generating virtual mirror information of the task information based on status information, task information, and road information, wherein the virtual mirror information is used to represent a virtual mirror body synchronized with the task information, the virtual mirror body is constructed in a digital twin platform, and is used to represent an information database constructed from status information, task information, and road information; generating an initial path set based on the virtual mirror information; and determining the target path from the initial path set using a decision model, wherein the decision model is obtained by training a neural network model using status information samples, task information samples, road information samples, and initial path set samples, the status information samples are used to represent the idle state and / or location state of vehicle samples in the vehicle set samples, the task information samples are executed through the transportation operation instructions or transfer operation instructions of the vehicle samples, and the road information samples are used to represent the traffic congestion level of the current road segment where the vehicle sample is located, and / or the environmental state of the road where the current road segment is located.

[0008] Optionally, a decision model is used to determine the target path from the initial path set, including: using the decision model to evaluate the initial path set and obtain evaluation results, wherein the evaluation results are used to represent the priority scores of the paths in the initial path set in terms of time efficiency, energy consumption, task timeliness, and resource utilization; and determining the target path from the initial path set based on the evaluation results.

[0009] Optionally, the method further includes: adjusting the weights of the parameters in the decision model based on the evaluation results; and using the adjusted decision model to evaluate the initial path set to obtain the target path.

[0010] Optionally, based on the evaluation results, the target path is determined from the initial path set, including: sorting the priority scores to obtain a sorting result; and determining the paths in the initial path set whose priority scores are greater than the priority score threshold as the target paths.

[0011] Optionally, from the initial path set, paths with priority scores greater than the priority score threshold in the sorting results are identified as target paths, including: in response to a path with a priority score greater than the priority score threshold in the sorting results being a single path, identifying that single path as the target path; the method further includes: in response to multiple paths with priority scores greater than the priority score threshold in the sorting results being multiple paths, allocating the multiple paths to multiple vehicles whose job task attribute information is within the attribute threshold range according to the collaborative scheduling strategy.

[0012] Optionally, the target route of the vehicle is determined based on status information, work tasks, and road information, including: determining the target route at each target time interval based on status information, work tasks, and road information.

[0013] According to another aspect of the present invention, a vehicle task execution device is also provided. The device may include: an acquisition unit, configured to acquire state information of a vehicle set, a task of the vehicle set, and road information of the vehicle set during the execution of the task, wherein the state information indicates the idle state and / or location state of the vehicles in the vehicle set, the task is executed through vehicle transport operation instructions or transfer operation instructions, and the road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located; a first determination unit, configured to determine a target path for the vehicle based on the state information, the task, and the road information, wherein the target path indicates a path that satisfies both the state information and the road information during the execution of the task; a second determination unit, configured to determine a task execution strategy for a target vehicle in the vehicle set based on the target path, wherein the task execution strategy indicates the rules for the target vehicle to execute the task; and an execution unit, configured to control the target vehicle to execute the task along the target path according to the task execution strategy.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided. The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method of the embodiments of the present invention.

[0015] According to another aspect of the present invention, a processor is also provided. The processor is used to run a program, wherein the program executes the methods of the embodiments of the present invention during runtime.

[0016] According to another aspect of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention during runtime.

[0017] According to another aspect of the present invention, a computer program product is also provided. This computer program product includes a computer program that, when executed by a processor, implements the methods described in the embodiments of the present invention.

[0018] According to another aspect of the present invention, a vehicle is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present application when it runs.

[0019] In this embodiment, if it is necessary to control a target vehicle to perform a work task along a target path, the status information of the vehicle set, the work task of the vehicle set, and the road information of the vehicle set during the execution of the work task can be obtained. The status information indicates the idle status and / or location status of the vehicles in the vehicle set. The work task is executed through the vehicle's transport operation instructions or transfer operation instructions. The road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located. Based on the status information, the work task, and the road information, the target path of the vehicle can be determined. The target path represents the path that satisfies both the status information and the road conditions during the execution of the work task. Based on the target path, the task execution strategy of the target vehicle in the vehicle set can be determined. The task execution strategy represents the rules by which the target vehicle performs the work task. The target vehicle can be controlled to perform the work task along the target path according to the task execution strategy. In other words, this application embodiment no longer regards the vehicle as an independent tool for executing preset instructions, but continuously integrates three key data streams: status information, task information, and road information, to determine the target path of the vehicle. Based on the target path, a personalized task execution strategy is generated for each target vehicle, thereby solving the technical problem of low efficiency in vehicle task execution and achieving the technical effect of improving the efficiency of vehicle task execution. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and constitute a part of this application, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0021] Figure 1 This is a flowchart of a vehicle task execution method according to an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of a vehicle logistics intelligent scheduling system based on digital twin and dynamic path planning according to an embodiment of the present invention.

[0023] Figure 3 This is a flowchart of a whole vehicle logistics intelligent scheduling method based on digital twin and dynamic path planning according to an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram of a vehicle task execution device according to an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] According to an embodiment of the present invention, an embodiment of a vehicle task execution method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] Figure 1 This is a flowchart of a vehicle task execution method according to an embodiment of the present invention, such as... Figure 1 As shown, the method may include the following steps.

[0029] Step S102: Obtain the status information of the vehicle set, the work tasks of the vehicle set, and the road information of the vehicle set during the execution of the work tasks.

[0030] In the technical solution provided by step S102 of the present invention, the status information can be used to indicate the idle status and / or location status of vehicles in the vehicle pool, the operation task can be executed by the vehicle's transport operation instruction or transfer operation instruction, and the road information can be used to indicate the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental status of the road where the current road segment is located.

[0031] In this embodiment, to accurately determine the target path of the vehicles, the status information of the vehicle set, the task of the vehicle set, and the road information of the vehicle set during the execution of the task can be obtained. The aforementioned task can be simply referred to as a task.

[0032] Optionally, the aforementioned status information may be the current operating status of each vehicle in the vehicle central database, including but not limited to: whether the vehicle is in an idle state, en route state, loading / unloading state, or fault state, as well as the precise location status, remaining battery power or fuel level status, current load ratio status, and driver online status.

[0033] Optionally, the aforementioned status information can be periodically uploaded to the cloud-based dispatch platform by sensors and a vehicle-mounted telematics box (T-Box) system deployed on the vehicle terminal, ensuring a complete understanding of the availability and capability boundaries of each vehicle. The location status in the aforementioned status information can be obtained through reporting by the vehicle-mounted Global Positioning System (GPS) / BeiDou module.

[0034] Optionally, the aforementioned tasks can be transportation or transfer instructions issued by the vehicle's logistics management system and required to be executed by the vehicle, such as "pick up 30 vehicles from factory A and deliver them to port B" or "transfer 5 complete vehicles from warehouse C to showroom D". Each task can include structured parameters such as origin, destination, cargo type, quantity, priority, and time window (e.g., loading must be completed before 14:00). These tasks can be synchronized to the vehicle's dispatching platform in real time for subsequent route planning.

[0035] Optionally, the aforementioned road information can cover the dynamic environmental characteristics of the road segment where the vehicle is currently located and the surrounding road network, including traffic congestion index, real-time traffic flow, traffic restriction rules, construction barriers, weather conditions (such as heavy rain and fog), and road surface slipperiness (identified by the fusion of weather station and vehicle camera).

[0036] Optionally, the aforementioned road information can be accessed in real time through vehicle-to-everything (V2X) platforms, urban traffic management platforms, and IoT edge devices to achieve a high-precision mirroring of the physical road environment.

[0037] In this embodiment, by accurately sensing the vehicle's status information, dispatchable vehicles can be proactively identified, avoiding the assignment of tasks to vehicles under maintenance or with insufficient power, thus reducing the task failure rate from the source. By combining real-time road information, high-risk or high-delay road sections can be excluded before the target path is generated, making the initial path set more feasible. Simultaneously, the task is a refined decision oriented towards specific business objectives (e.g., on-time delivery, energy minimization). The synergy of the vehicle set's status information, the vehicle set's task, and the road information during the execution of the task lays a reliable data foundation for the subsequent determination of the target path.

[0038] Step S104: Determine the target path of the vehicle based on the status information, task, and road information.

[0039] In the technical solution provided by step S104 of the present invention, the target path can be used to represent the path that satisfies the state information and the road information during the process of the vehicle performing the operation task.

[0040] In this embodiment, based on the acquired vehicle status information, task information, and road information, a dynamic decision-making environment can be comprehensively constructed to determine the target path that each vehicle should follow during the execution of the task. This target path is not a fixed route, but a dynamically generated, adaptive driving sequence based on the current global state, thereby ensuring that the vehicle's status information continuously meets preset state conditions throughout the entire task execution process, while the environmental information of the roads traversed meets road conditions (e.g., safety and efficiency constraints).

[0041] Optionally, the aforementioned status conditions may include compliance requirements for vehicle operation at each stage of the task. For example, before a loading / unloading task, the vehicle must be in an "idle and nearby reachable" state, and the remaining battery or fuel level must meet the threshold required for the round trip (e.g., a minimum threshold). During transport, the vehicle must remain within the permitted speed limit range to avoid triggering electronic fence alarms due to speeding. Upon arrival at the target loading / unloading point, the vehicle must be synchronized with the occupancy status of the loading / unloading bay to prevent prolonged waiting times due to idling or conflicts.

[0042] Optionally, the aforementioned road conditions may include the accessibility of real-time traffic flow. For example, avoiding road sections with congestion levels exceeding moderate, detouring through temporarily closed or accident-prone sections, avoiding areas with low visibility due to heavy fog or snow, and prioritizing exemptions from traffic restrictions or dedicated freight lanes to ensure that the target route is physically safe, legal, and efficient.

[0043] Optionally, in order to determine the target route of the vehicle, a dynamic path planning engine in the digital twin platform can be used to construct a virtual mirror of the entire logistics transportation process based on high-precision maps and real-time IoT data (such as status information, work tasks and road information).

[0044] Optionally, the aforementioned dynamic path planning engine can generate a set of initial paths that meet basic physical feasibility requirements based on the start and end points of the task, the time window, and the vehicle type, combined with the high-precision map topology. Subsequently, a decision model trained based on reinforcement learning can be invoked to score each candidate path in the initial path set in multiple dimensions. These scoring dimensions can cover task completion time, energy cost, vehicle waiting time, path length, traffic risk index, and waiting probability at loading / unloading points, among others.

[0045] Optionally, the weight parameters of the decision-making model, having been continuously optimized using historical execution data, can automatically identify key influencing factors in different scenarios. For example, during morning and evening rush hours, more attention is paid to travel time, while in rainy or snowy weather, more attention is paid to visibility and road conditions. After normalization of the scoring results, the path with the highest overall score, whose state information meets the state conditions, and whose road information meets the road conditions can be selected as the target path.

[0046] In this embodiment, the target path can dynamically respond to changes through the above steps. For example, when a road segment suddenly becomes congested, a new target path can be reassessed and output without manual intervention, thereby significantly improving emergency response speed. Furthermore, since the target path decision is deeply integrated with vehicle status information and road information, secondary scheduling and task failures caused by insufficient vehicle battery power or loading / unloading port conflicts can be effectively avoided, reducing vehicle empty load rates and improving task on-time performance. In addition, by incorporating road information into core decision variables, high-risk areas can be proactively avoided, reducing accident rates and insurance payout costs, while optimizing overall transportation energy consumption. This achieves a refined and adaptive target path generation mechanism that is constrained by status information, guided by task requirements, and uses road information as a variable.

[0047] Step S106: Based on the target path, determine the task execution strategy for the target vehicles in the vehicle cluster.

[0048] In the technical solution provided by step S106 of the present invention, the task execution strategy can be used to represent the rules for the target vehicle to perform the operation task.

[0049] In this embodiment, based on the determined target path, a specific task execution strategy can be generated for the target vehicle throughout the entire process of performing the task. This task execution strategy is not simply based on the route driving instructions, but rather a composite execution rule system embedded with task rhythm, resource coordination, anomaly response, and operational procedures, used to guide the target vehicle to complete the complete operational process from departure, en route, loading and unloading operations to the completion of the task.

[0050] For example, if the target route indicates that the target vehicle will arrive at the loading / unloading area of ​​Port B at 14:15, and the scheduled idle time for that loading / unloading area is 14:00–14:30, a task execution strategy can be generated to "slow down and enter the standby area 5 minutes in advance, maintain idling speed, and turn off the vehicle's air conditioning to save power." If the target route includes multiple loading / unloading nodes, the task execution strategy can insert "unload first, then load" or "segmented loading / unloading" instructions, along with suggestions for the target vehicle's parking location (e.g., unloading positions near charging stations). If the target vehicle is detected to be an electric transport vehicle with less than 30% battery remaining, the task execution strategy can automatically trigger "prioritize loading / unloading points equipped with charging stations" and add "optional charging stations along the way" as route extension nodes, forming a three-in-one collaborative execution solution of "route + charging + task." Simultaneously, abnormal response rules can be embedded in the task execution strategy. For example, if the target vehicle deviates from the target route by more than 200 meters, a "route recalibration + driver voice reminder" mechanism can be automatically activated. If the loading / unloading port is temporarily occupied, the task execution strategy can trigger a linkage mechanism of "automatic reporting of waiting timeout + switching to a nearby alternative port", which can complete the rescheduling of the operation task without manual intervention.

[0051] Optionally, it provides a set of dynamic operation instructions that can be converted from static target paths into executable, monitorable, and adaptive ones, which solves the pain point of relying on drivers or dispatchers to make judgments based on experience in traditional logistics scheduling, and makes the vehicle execution process standardized, intelligent, and reproducible.

[0052] In the embodiments of this application, the task execution strategy serves as a key bridge connecting the target path and the operation task, realizing the automated, collaborative, and intelligent operation of the entire vehicle logistics chain.

[0053] Step S108: According to the task execution strategy, control the target vehicle to perform the operation task along the target path.

[0054] In the technical solution provided by step S108 of the present invention, according to the task execution strategy, precise, structured execution instructions with real-time response capability can be sent to the vehicle terminal of the target vehicle to realize automated and coordinated control of the entire process of vehicle driving, parking, loading and unloading, communication and abnormal response.

[0055] In this embodiment, the target vehicle is controlled to perform the operation task along the target path according to the task execution strategy. It is not a one-way navigation distribution, but a closed-loop execution mechanism built on the task execution strategy and the real-time feedback of the target vehicle, which ensures that each operation is accurately implemented within the preset business logic and physical constraints.

[0056] Optionally, key instructions in the task execution strategy can be pushed to the target vehicle one by one through an in-vehicle communication module, such as fourth-generation (4G) or fifth-generation (5G) mobile communication technology, or vehicle-to-everything (V2X) technology.

[0057] Optionally, the aforementioned key instructions may include: activating the target vehicle's automated driving assistance system to drive along the target path; automatically decelerating to the loading / unloading area preparation state at a specific mileage point; triggering a "wait / enter" prompt based on the loading / unloading port status; prompting for the nearest charging point and automatically planning a detour when the battery level is below a threshold; and triggering automatic correction or manual takeover requests when the deviation from the path exceeds the safety tolerance. Simultaneously, a pre-arrival notification of the target vehicle is sent to the vehicle-mounted terminal at the loading / unloading station, coordinating hoisting equipment, personnel scheduling, and work order preparation to achieve two-way vehicle-station collaboration.

[0058] Optionally, throughout the entire execution process, the on-board terminal can continuously transmit real-time data such as vehicle location, speed, door lock status, battery level changes, and driver operation behavior. This data is then compared in real-time with the expected execution strategy via a digital twin platform. If any deviation is detected (e.g., loading / unloading delays, route deviations, or abnormal parking), an adaptive correction mechanism in the task execution strategy is immediately triggered. This might involve automatically replanning the remaining route, dispatching nearby vehicles to assist, or notifying the dispatcher to intervene, ensuring that the task is not interrupted due to localized anomalies.

[0059] In this embodiment, by controlling the target vehicle to perform tasks along the target path according to the task execution strategy, the standardization of task execution is significantly improved, the driver error rate is reduced, the loading and unloading connection time is shortened, and the overall task completion cycle is stabilized. Furthermore, the automatic response capability to anomalies reduces unplanned delays and significantly enhances transportation reliability. Precise control of vehicle start-stop, speed range, and charging nodes reduces energy consumption, extends battery life, and lowers operating costs, laying a solid foundation for building an unmanned, self-evolving, and highly resilient intelligent logistics system.

[0060] In the embodiments of the present invention, steps S102 to S108 above, if it is necessary to control the target vehicle to perform a work task along the target path, can obtain the status information of the vehicle set, the work task of the vehicle set, and the road information of the vehicle set during the execution of the work task. The status information indicates the idle status and / or location status of the vehicles in the vehicle set. The work task is executed through the vehicle's transport operation instructions or transfer operation instructions. The road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located. Based on the status information, the work task, and the road information, the target path of the vehicle can be determined. The target path indicates the path that satisfies both the status information and the road information during the execution of the work task. Based on the target path, the task execution strategy of the target vehicle in the vehicle set can be determined. The task execution strategy indicates the rules for the target vehicle to perform the work task. The target vehicle can be controlled to perform the work task along the target path according to the task execution strategy. In other words, this application embodiment no longer regards the vehicle as an independent tool for executing preset instructions, but continuously integrates three key data streams: status information, task information, and road information, to determine the target path of the vehicle. Based on the target path, a personalized task execution strategy is generated for each target vehicle, thereby solving the technical problem of low efficiency in vehicle task execution and achieving the technical effect of improving the efficiency of vehicle task execution.

[0061] The method described in this embodiment will be further described below.

[0062] As an optional embodiment, step S108, controlling the target vehicle to perform the work task along the target path according to the task execution strategy, includes: in response to detecting a deviation in the target path or a delay in the execution of the work task, adjusting the task execution strategy using the scheduling module; and controlling the target vehicle to perform the work task along the target path according to the adjusted task execution strategy.

[0063] In this embodiment, during the process of controlling the target vehicle to perform the operation task along the target path according to the task execution strategy, the system can continuously monitor whether the target vehicle deviates from the target path or whether the operation task is delayed (e.g., loading and unloading timeout, arrival time later than the window period, disconnection from upstream and downstream nodes, etc.) through real-time comparison of vehicle-mounted sensors, GPS positioning, loading and unloading port status feedback and cloud digital twin.

[0064] Optionally, if a deviation from the target path is detected, such as more than 200 meters off the route, or if the delay in the execution of the task exceeds a threshold, such as waiting time exceeding 5 minutes or failure to arrive at the loading / unloading point on time, the online adaptive adjustment mechanism of the scheduling module can be automatically triggered to adjust the task execution strategy.

[0065] Optionally, the aforementioned scheduling module, based on real-time environmental data and a historical event knowledge base integrated within the digital twin platform, can quickly assess the root cause of deviations or delays. For example, is it due to sudden traffic congestion, temporary occupancy of loading / unloading points, vehicle mechanical failure, or loss of navigation signals? Then, it can invoke built-in scheduling modules, such as an emergency scheduling artificial intelligence (AI) model, to dynamically generate new task execution strategies by combining current vehicle status information, available surrounding resources, remaining task time windows, and overall logistics network load.

[0066] For example, if an accident ahead causes congestion on the original target route, an adjusted task execution strategy can be generated, including detour routes, charging and refueling suggestions, and priorities for new loading and unloading points. If the queue at the loading and unloading point is too long, instructions such as "pause and wait, prioritize nearby available vehicles for loading and then return" or "switch to alternative loading and unloading areas" can be issued, and the driver and the station can be notified simultaneously to coordinate and prepare. The adjusted task execution strategy can be sent to the target vehicle in real time through a low-latency communication channel, and the on-board terminal will then execute the new instructions, including updating the navigation target route, adjusting the driving assistance mode, resetting the expected arrival time, and even actively triggering voice prompts to guide the driver to cooperate with the new process.

[0067] In this embodiment, when a deviation from the target path or a delay in task execution is detected, the scheduling module adjusts the task execution strategy. Then, according to the adjusted strategy, the target vehicle is controlled to perform the task along the target path. This enhances the flexibility of task execution, reduces the response time from the discovery of anomalies to the adjustment of the task execution strategy, and minimizes chain reactions and complaints caused by delays. Furthermore, it optimizes vehicle and resource utilization, avoiding ineffective detours or prolonged waiting due to rigid target paths, improving overall transportation efficiency, and reducing driver fatigue and operational stress. Moreover, by continuously recording the input conditions and output results of each task execution strategy adjustment, the vehicle logistics scheduling system upgrades from a passive to an active response, maintaining high availability, high stability, and high intelligence in complex and ever-changing operating environments, providing crucial execution guarantees for building an unmanned, adaptive intelligent logistics system.

[0068] As an optional embodiment, step S104, determining the target path of the vehicle based on state information, task, and road information, includes: generating virtual mirror information of the task based on state information, task, and road information, wherein the virtual mirror information is used to represent a virtual mirror body synchronized with the task, the virtual mirror body is constructed in a digital twin platform, and is used to represent an information database constructed from state information, task, and road information; generating an initial path set based on the virtual mirror information; and determining the target path from the initial path set using a decision model, wherein the decision model is obtained by training a neural network model using state information samples, task samples, road information samples, and initial path set samples, the state information samples are used to represent the idle state and / or location state of vehicle samples in the vehicle set samples, the task samples are executed through the transportation operation instructions or transfer operation instructions of the vehicle samples, and the road information samples are used to represent the traffic congestion level of the current road segment where the vehicle sample is located, and / or the environmental state of the road where the current road segment is located.

[0069] In this embodiment, the process of determining the target route of the vehicle no longer relies solely on a simple combination of static maps and rule engines. Instead, it constructs a virtual mirror information that is completely synchronized with the real logistics scenario, thereby achieving high-fidelity simulation and intelligent extrapolation of the entire lifecycle of the operation task.

[0070] Optionally, real-time vehicle status information (e.g., vehicle location, battery level, load, online status), work tasks (e.g., start and end points, cargo type, time window, priority), and road information (e.g., real-time congestion index, weather conditions, construction sections, traffic restriction rules) can be integrated and imported into a digital twin platform to construct a virtual mirror that is synchronized with the physical world in real time and evolves dynamically.

[0071] Optionally, the aforementioned virtual mirror is not a simple visualization model, but a structured, computable information repository with spatiotemporal correlation capabilities. It can not only record the current state of each element but also simulate the interaction relationships between them. For example, how will the energy consumption of an electric truck change when traveling on a steep slope during heavy rain? If the loading dock ahead is temporarily closed due to a scheduling conflict, will it cause subsequent vehicles to queue? This virtual environment, based on causal modeling, can pre-simulate possible path combinations and execution results.

[0072] Optionally, based on the aforementioned virtual mirror, an initial path set containing hundreds of candidate paths can be automatically generated. The paths in this initial path set not only satisfy basic geographical accessibility but also incorporate business constraints. These constraints include avoiding restricted hours, prioritizing freight lanes, and ensuring sufficient vehicle battery power for round trips. Subsequently, a deeply trained decision-making model (e.g., a reinforcement learning model) can be invoked. This model is an intelligent evaluation and optimization system based on a neural network model, and its training data comes from massive amounts of historical real-world operational samples. These samples include vehicle status samples (e.g., vehicle performance at different battery levels), task samples (e.g., differences in execution efficiency between tasks of different priorities), road information samples (e.g., the impact curve of smog on traffic speed), and the actual execution results of each historical path (e.g., time consumption, energy consumption, delay rate, and customer ratings).

[0073] Optionally, by fusing A Algorithms and reinforcement learning models can determine which path, under the current conditions, achieves the comprehensive goal of "low cost, short time, low risk, and balanced resources." They can not only identify explicit factors (such as distance and congestion) but also capture implicit patterns. For example, during morning rush hour, taking a 1.2-kilometer detour to avoid main roads, although increasing the distance, actually saves 7 minutes compared to a direct route because it reduces waiting time at three red lights.

[0074] In this embodiment, virtual mirror information of the task is generated based on status information, task details, and road information. Then, an initial path set is generated based on this virtual mirror information. A decision model is used to determine the target path from this initial path set, significantly improving the scientific rigor and adaptability of the target path. Because the decision model continuously absorbs new samples for online learning, the accuracy of the target path recommendation is improved. Through the pre-simulation capability of the virtual mirror, potential conflicts (e.g., multiple vehicles simultaneously occupying the same loading / unloading point) can be identified before actual execution, allowing for advance resource coordination and reducing the trigger rate of abnormal events.

[0075] As an optional implementation method, a decision model is used to determine the target path from the initial path set, including: using the decision model to evaluate the initial path set and obtain evaluation results, wherein the evaluation results are used to represent the priority scores of the paths in the initial path set in terms of time efficiency, energy consumption, task timeliness, and resource utilization; and determining the target path from the initial path set based on the evaluation results.

[0076] In this embodiment, in the process of determining the target path from the initial path set using the decision model, it is not simply a single-dimensional choice based on the shortest distance or the shortest time. Instead, it uses an intelligent evaluation mechanism that integrates a multi-objective evaluation system, namely the decision model, to give a comprehensive and dynamic priority score to each candidate path in the initial path set, thereby determining the target path from the initial path set.

[0077] Optionally, the above decision-making model is trained based on a neural network model, which can simultaneously quantify and weigh the comprehensive performance of the path in dimensions such as time efficiency, energy consumption, task timeliness, and resource utilization, and output a comprehensive score that reflects the overall value of the target path.

[0078] Optionally, each candidate path in the initial path set is mapped to a set of structured feature vectors, including estimated travel time, cumulative elevation gain (affecting energy consumption), number of congested road segments and estimated delay time, whether it passes through reserved empty loading / unloading points, whether it matches the charging needs of electric vehicles, and whether it closely aligns with upstream / downstream task time windows. Subsequently, a decision model is used to perform nonlinear fusion calculations on these structured feature vectors, outputting four sub-scores. The time efficiency score reflects whether the target path can be completed within the task time window with a buffer; the energy consumption score comprehensively considers road gradient, vehicle speed fluctuations, air conditioning load, and vehicle power type (e.g., electric / fuel); the task punctuality score assesses whether the target path can ensure timely arrival at critical nodes, avoiding cascading delays caused by delays; and the resource utilization score measures whether the target path helps to revitalize idle resources, such as whether it can drive empty vehicles back for loading, reduce waiting conflicts at loading / unloading points, and promote the priority execution of short-distance, high-frequency tasks by electric vehicles. Finally, the four types of scores are weighted according to priority and combined into a total score, which serves as the priority score for the target path.

[0079] For example, in a cross-provincial vehicle transportation dispatch, two routes were generated: Route A is a straight highway route, estimated to take 3 hours, but it passes through two morning rush hour congestion areas, with an estimated delay of 22 minutes, and another vehicle is already operating at the final loading / unloading point, with an estimated wait of 15 minutes. Route B is a secondary trunk line that detours 15 kilometers, estimated to take 3 hours and 20 minutes, but it is unobstructed throughout, has no congestion, the final loading / unloading point is currently available and equipped with charging stations, and the vehicle is an electric vehicle with 38% battery remaining. In this case, although the decision-making model perceives Route B as slightly inferior in "time efficiency," it saves 12% of battery power in "energy consumption" due to reduced rapid acceleration / braking, ensures on-time arrival in "task punctuality" because there is no waiting, and achieves precise matching of electric vehicles and charging resources in "resource utilization." After comprehensive weighted calculation, Route B has a higher total score than Route A and is ultimately selected as the target route.

[0080] In this embodiment, an intelligent decision-making hub for multi-objective collaborative optimization is constructed, avoiding the local optimization trap caused by the optimization of a single indicator in traditional scheduling systems. For example, on-time performance may be sacrificed to save distance, or energy waste may be amplified in pursuit of speed. A decision model is used to evaluate the initial path set, and then, based on the evaluation results, the target path is determined from the initial path set, making the selection of the target path more effective.

[0081] As an optional embodiment, the method further includes: adjusting the weights of the parameters in the decision model based on the evaluation results; and using the adjusted decision model to evaluate the initial path set to obtain the target path.

[0082] In this embodiment, after the initial path set is evaluated and the target path is selected based on the decision model, a closed-loop self-optimization mechanism can be further introduced. That is, the parameter weights of each evaluation dimension within the decision model are dynamically adjusted according to the actual feedback data after the task is executed, thereby achieving continuous evolution and performance improvement of the decision model.

[0083] Optionally, after a vehicle completes its task, real-time execution data can be automatically collected, including: the deviation between actual and estimated path time, the difference between actual and predicted energy consumption, whether the vehicle arrived at the loading / unloading node on time, whether the path selection caused other vehicles to wait or resource conflicts, driver operation feedback, and customer satisfaction ratings. This feedback information can be used as "supervisory signals" input into the training module of the decision-making model. Through backpropagation or reinforcement learning mechanisms, the model can automatically analyze which weight settings led to evaluation biases. For example, if the "shortest distance" path is repeatedly selected, but the electric vehicle frequently breaks down midway due to avoiding charging stations, the weight of the "path length" dimension can be automatically reduced, while the weights of "energy matching degree" and "charging point accessibility" can be increased. Furthermore, if a preference for fast-response paths for high-priority tasks leads to a long-term backlog of low-priority tasks, a "task fairness" factor can be automatically introduced to appropriately increase the weight of the timeliness score for low-priority tasks, achieving a global balance in resource allocation.

[0084] Optionally, this dynamic adjustment of parameter weights is not done offline retraining, but rather in real time through incremental learning on edge computing nodes or cloud services. This ensures that the decision-making model remains synchronized with the current operational status of the logistics network, vehicle composition (e.g., changes in the ratio of fuel vehicles to electric vehicles), and seasonal traffic patterns (e.g., holiday congestion patterns). Each task execution is a fine-tuning of the decision-making model, resulting in a more accurate target path.

[0085] In this embodiment, based on the evaluation results, the weights of the parameters in the decision-making model are adjusted. Then, the adjusted decision-making model is used to evaluate the initial path set to obtain the target path. This breaks the rigid pattern of traditional scheduling algorithms that are trained once and remain unchanged for a long time, and constructs a self-evolving intelligent agent capable of sensing environmental evolution, understanding changes in business requirements, and proactively optimizing decision logic. Therefore, it no longer relies on periodic updates based on human expert experience, but is driven by real data, improving the accuracy of the target path.

[0086] As an optional implementation method, the target path is determined from the initial path set based on the evaluation results, including: sorting the priority scores to obtain a sorting result; and determining the paths in the initial path set whose priority scores are greater than the priority score threshold as the target paths.

[0087] In this embodiment, after the decision model performs a multi-dimensional evaluation of each path in the initial path set and obtains a comprehensive priority score in terms of time efficiency, energy consumption, task timeliness and resource utilization, it does not select only the single highest-scoring path as the target path. Instead, it adopts a hierarchical screening mechanism to determine the paths with priority scores greater than the priority score threshold in the ranking results as the target paths.

[0088] Optionally, the priority scores of candidate paths in the initial path set are sorted from high to low to form a clear priority sequence. Then, a dynamic priority score threshold is set. This threshold is not fixed but adaptively adjusted based on the urgency of the current scheduling scenario, resource scarcity, and overall system load. For example, during peak periods or times of frequent abnormal events, the priority score threshold can be appropriately lowered to expand the range of selectable paths and improve scheduling flexibility. When resources are abundant and tasks are stable, the priority score threshold can be increased to ensure the selection of a suitable target path.

[0089] Optionally, paths with priority scores greater than the priority score threshold are selected from the sorting results to form the final set of target paths, and then further optimized paths are selected from these to obtain the target paths.

[0090] For example, in a cross-provincial transportation dispatch, ten candidate routes are evaluated. Five of these routes have priority scores greater than the priority score threshold. The highest-scoring route carries a slight risk due to its passage through a section of highway about to be constructed. The second-highest-scoring route, although slightly longer, offers overall stability, available loading and unloading points, and optimal battery matching. Both routes can be included in the target route set, while the second-best route is pre-loaded into the vehicle terminal as an emergency backup plan. When the vehicle encounters an actual road closure due to construction ahead, there is no need to recalculate; it can immediately switch to the backup route, achieving a zero-delay response.

[0091] In this embodiment, the above steps can compress the response time for route switching in abnormal situations. Secondly, due to the availability of multiple high-quality alternative routes, vehicle waiting time and task delay rates are significantly reduced, improving the overall stability of task completion. Drivers can receive multi-route selection suggestions on the in-vehicle terminal, enhancing trust in human-machine collaboration and reducing resistance caused by the "unchangeable single path." Furthermore, it provides rich sample data for subsequent route review and decision model optimization.

[0092] As an optional embodiment, from the initial path set, the path with a priority score greater than the priority score threshold in the ranking result is determined as the target path, including: in response to the path with a priority score greater than the priority score threshold in the ranking result being a path, determining one path as the target path; the method further includes: in response to the path with a priority score greater than the priority score threshold in the ranking result being multiple paths, according to the cooperative scheduling strategy, assigning the multiple paths to multiple vehicles whose attribute information of the task is within the attribute threshold range.

[0093] In this embodiment, after filtering out paths with priority scores greater than the priority score threshold from the initial path set, the paths that meet the conditions are not simply assigned to a single vehicle. Instead, a differentiated path allocation strategy is adopted based on the complexity of the actual scheduling scenario to achieve a fine-grained matching of resources and tasks.

[0094] Optionally, when only one path in the ranking results has a priority score greater than the priority score threshold, that path can be directly designated as the target path and assigned to the target vehicle corresponding to the current task. This ensures the certainty of the decision and the uniqueness of the execution, avoiding confusion or resource waste caused by multiple path selections. This single-path locking mechanism is suitable for scenarios with extremely high task priority, extremely short time windows, or highly scarce vehicle resources. For example, critical tasks such as emergency whole-vehicle shipments, transportation of high-value vehicle models, or concentrated nighttime outbound shipments.

[0095] Optionally, when multiple paths in the sorting results satisfy the priority score being greater than the priority score threshold, a collaborative scheduling strategy can be initiated. Based on the relationship between the attribute information of the task and the attribute threshold, these high-quality paths can be intelligently allocated to multiple suitable target vehicles, thereby achieving parallel optimization of resources and network-level collaboration.

[0096] Optionally, the aforementioned attribute information may include the type of goods for the task (e.g., high-end cars, engineering vehicles), transportation time window (e.g., delivery must be made before 8:00 AM), loading and unloading requirements (e.g., hydraulic tailgate and vibration damping device required), vehicle type (e.g., electric / fuel), remaining battery / fuel level, driver qualifications, and location.

[0097] Optionally, the above attribute threshold ranges are allowed to be matched within a reasonable range. For example, if the job task requires "electric vehicles must be used," then matching must be done within the electric vehicle pool. If the job task is "non-emergency general transportation," then electric vehicles with a battery level of 30% or higher are allowed to participate in the task.

[0098] For example, in a multi-vehicle shuttle dispatch within the park, four routes with priority scores all above the priority score threshold were identified, each leading to a different loading / unloading point. At this point, intelligent matching can be performed based on the status of the six vehicles currently on standby: three electric trucks, two gasoline trucks, and one low-battery electric vehicle. The shortest and smoothest route is assigned to the electric truck with sufficient battery power and the closest location, prioritizing high-efficiency tasks. Another route, which detours but avoids congestion and has lower energy consumption, is assigned to the gasoline truck for bulk transport. Simultaneously, a route requiring a charging station stop is assigned to the electric truck with only 28% battery remaining, achieving integrated "transportation + charging." The remaining vehicles, due to mismatched attributes (e.g., lack of a tailgate, lack of special vehicle qualifications), are not assigned, avoiding forced assignment and secondary dispatching. The entire process requires no manual intervention and can complete multi-dimensional matching in milliseconds.

[0099] In this embodiment, multi-dimensional matching of target paths and vehicles further reduces the empty-run rate of target vehicles and improves resource utilization. Secondly, parallel allocation of multiple paths enables multiple tasks to be executed simultaneously, improving task execution efficiency. By allocating high-energy-consuming paths to fuel vehicles, high-time-efficiency paths to nearby vehicles, and refueling paths to vehicles with low battery levels, deep synergy between energy structure, spatial layout, and task requirements is achieved.

[0100] As an optional embodiment, step S104, determining the target path of the vehicle based on status information, work task, and road information, includes: determining the target path at each target time interval based on status information, work task, and road information.

[0101] In this embodiment, in the process of determining the target path based on status information, task information, and road information, the target path is not fixed after being planned once when the task is started. Instead, a periodic dynamic replanning mechanism is used as the core. Every time the target time period is reached, the target path is recalculated and determined based on the real-time updated status information of the vehicle, the latest status of the task, and the dynamically changing road information of the entire area.

[0102] Optionally, each replanning process can re-collect data such as the vehicle's precise location, remaining battery / fuel level, load status, surrounding environment perceived by onboard sensors (e.g., sudden obstacles, emergency braking), changes in task priority (e.g., urgent customer orders or cancellations), whether loading / unloading points are available or occupied, rain / icing warnings from weather platforms, and temporary traffic restrictions or accident reports issued by traffic management departments. This data can be synchronized in real time to the digital twin platform, forming a virtual mirror image of the current moment. Then, a dynamic path planning engine can be invoked to re-simulate feasible paths in the virtual mirror at target time intervals (e.g., 5 minutes), combining this with a reinforcement learning model for multi-objective evaluation, and outputting the target path.

[0103] Optionally, the target route can remain exactly the same as the initial plan, or it can be significantly adjusted. For example, if the original route is blocked due to a sudden traffic accident ahead, a new detour route can be planned within 3 seconds. Or, if a loading / unloading point completes its work ahead of schedule, the route for the next work task can be proactively optimized to connect to the nearest point, reducing empty driving distance.

[0104] In this embodiment of the invention, if it is necessary to control a target vehicle to perform a work task along a target path, the status information of the vehicle set, the work task of the vehicle set, and the road information of the vehicle set during the execution of the work task can be obtained. The status information indicates the idle status and / or location status of the vehicles in the vehicle set. The work task is executed through the vehicle's transport operation instructions or transfer operation instructions. The road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located. Based on the status information, the work task, and the road information, the target path of the vehicle can be determined. The target path represents the path that satisfies both the status information and the road conditions during the execution of the work task. Based on the target path, the task execution strategy of the target vehicle in the vehicle set can be determined. The task execution strategy represents the rules for the target vehicle to perform the work task. The target vehicle can be controlled to perform the work task along the target path according to the task execution strategy. In other words, this application embodiment no longer regards the vehicle as an independent tool for executing preset instructions, but continuously integrates three key data streams: status information, task information, and road information, to determine the target path of the vehicle. Based on the target path, a personalized task execution strategy is generated for each target vehicle, thereby solving the technical problem of low efficiency in vehicle task execution and achieving the technical effect of improving the efficiency of vehicle task execution.

[0105] The technical solutions of the embodiments of the present invention will be illustrated below with reference to preferred embodiments.

[0106] Currently, the whole vehicle logistics scheduling has the following problems: rigid route planning, inefficient resource allocation, delayed emergency response, and insufficient multi-objective optimization.

[0107] Specifically, rigid route planning manifests in traditional dispatch systems' reliance on static road networks and fixed routes, making them unable to respond in real time to dynamic factors such as traffic congestion and weather changes. Inefficient resource allocation relies on manual experience in matching vehicles with loading / unloading resources, leading to high empty load rates and long vehicle waiting times. Delayed emergency response requires manual intervention for abnormal events (such as vehicle breakdowns and traffic control), with response delays exceeding 30 minutes. Insufficient multi-objective optimization makes it difficult to simultaneously consider transportation costs, time efficiency, energy consumption control, and load factor improvement. Therefore, the technical problem of low vehicle task execution efficiency persists.

[0108] To address the aforementioned issues, this invention provides a vehicle logistics intelligent scheduling system based on digital twins and dynamic path planning. Through a deep fusion mechanism of digital twins and real-time logistics data, a dynamic path planning algorithm based on reinforcement learning, an optimization model for multi-resource collaborative scheduling, and an AI autonomous decision-making and learning system for emergency events, the system achieves the technical effect of improving the efficiency of vehicle task execution.

[0109] The embodiments of the present invention will be further described below.

[0110] Figure 2 This is a schematic diagram of a vehicle logistics intelligent scheduling system based on digital twins and dynamic path planning according to an embodiment of the present invention, as shown below. Figure 2 As shown, the intelligent logistics dispatch system 200 includes: a digital twin platform 201, a dynamic route planning engine 202, a resource matching module 203, and an emergency dispatch AI module 204.

[0111] The Digital Twin Platform 201 constructs a virtual mirror of the entire logistics and transportation process based on high-precision maps and real-time IoT data.

[0112] Dynamic path planning engine 202, integrating A Algorithms and reinforcement learning models are used to calculate the optimal path in real time.

[0113] The resource matching module 203 uses a multi-objective optimization algorithm to achieve dynamic matching of vehicles, drivers, and loading / unloading platforms.

[0114] The emergency dispatch AI module 204 automatically generates emergency dispatch plans based on a historical event database and real-time traffic conditions.

[0115] Figure 3 This is a flowchart of a vehicle logistics intelligent scheduling method based on digital twins and dynamic path planning according to an embodiment of the present invention, as shown below. Figure 3 As shown, the method includes the following steps.

[0116] Step S301: Access data.

[0117] In this embodiment, order information, vehicle status, road condition data, and weather information can be received.

[0118] Step S302: Update the digital twin platform.

[0119] In this embodiment, the physical world state can be synchronized to a virtual platform (e.g., a digital twin platform) in real time.

[0120] Step S303: Perform route planning.

[0121] In this embodiment, the dynamic route planning engine can re-evaluate and output a recommended route every 5 minutes.

[0122] Step S304: Perform resource allocation.

[0123] In this embodiment, the intelligent logistics scheduling system can automatically allocate vehicles and loading / unloading resources and push tasks to the vehicle terminal.

[0124] Step S305: Perform anomaly monitoring.

[0125] In this embodiment, anomalies such as vehicle trajectory deviation and delay can be monitored in real time. If vehicle trajectory deviation or delay occurs, emergency dispatch can be triggered.

[0126] Step S306: Conduct feedback learning.

[0127] In this embodiment, the algorithm model (e.g., a reinforcement learning model) can be optimized based on the actual transportation results.

[0128] In this embodiment of the invention, the above steps can reduce the route planning response time to ≤3 seconds; reduce the vehicle empty load rate to ≥35%; increase the success rate of automatic scheduling for abnormal events to ≥90%; and improve the overall transportation efficiency to ≥25%.

[0129] The following example, using a specific scenario, will provide further explanation.

[0130] In cross-provincial full-truckload transportation scenarios, the logistics intelligent dispatch system simultaneously dispatches 200+ vehicles, covering 5 main routes; it avoids 3 accident-prone road sections in real time and automatically adjusts routes; the average vehicle waiting time is reduced from 45 minutes to 12 minutes.

[0131] In short-haul dispatching within the park, vehicles are dynamically allocated to available loading and unloading points based on the status of the loading and unloading towers; mixed dispatching of electric trucks and fuel vehicles is supported, with priority given to allocating electric vehicles to nearby tasks.

[0132] According to embodiments of the present invention, a vehicle task execution device is also provided. It should be noted that this vehicle task execution device can be used to execute the vehicle task execution method described in the above embodiments.

[0133] Figure 4 This is a schematic diagram of a vehicle task execution device according to an embodiment of the present invention, such as... Figure 4 As shown, the vehicle's task execution device 400 may include: an acquisition unit 402, a first determination unit 404, a second determination unit 406, and an execution unit 408.

[0134] The acquisition unit 402 is used to acquire the status information of the vehicle set, the work tasks of the vehicle set, and the road information of the vehicle set during the execution of the work tasks. The status information is used to indicate the idle status and / or location status of the vehicles in the vehicle set. The work tasks are executed through the vehicle's transportation operation instructions or transfer operation instructions. The road information is used to indicate the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental status of the road where the current road segment is located.

[0135] The first determining unit 404 is used to determine the target path of the vehicle based on the state information, the task, and the road information. The target path is used to represent the path that makes the state information meet the state conditions and the road information meet the road conditions during the process of the vehicle performing the task.

[0136] The second determining unit 406 is used to determine the task execution strategy of the target vehicle in the vehicle set based on the target path, wherein the task execution strategy is used to represent the rules for the target vehicle to perform the operation task.

[0137] The execution unit 408 is used to control the target vehicle to perform the operation task along the target path according to the task execution strategy.

[0138] Optionally, the execution unit 408 includes: a first adjustment subunit, used to adjust the task execution strategy using the scheduling module in response to detecting a deviation in the target path or a delay in the execution of the task; and a control subunit, used to control the target vehicle to execute the task along the target path according to the adjusted task execution strategy.

[0139] Optionally, the first determining unit 404 includes: a first generating subunit, used to generate virtual mirror information of the task based on state information, task, and road information, wherein the virtual mirror information is used to represent a virtual mirror body synchronized with the task, the virtual mirror body is constructed in a digital twin platform, and is used to represent an information database constructed from state information, task, and road information; a second generating subunit, used to generate an initial path set based on the virtual mirror information; and a first determining subunit, used to determine a target path from the initial path set using a decision model, wherein the decision model is obtained by training a neural network model using state information samples, task samples, road information samples, and initial path set samples, the state information samples are used to represent the idle state and / or location state of vehicle samples in the vehicle set samples, the task samples are executed by the transportation operation instructions or transfer operation instructions of the vehicle samples, and the road information samples are used to represent the traffic congestion level of the current road segment where the vehicle sample is located, and / or the environmental state of the road where the current road segment is located.

[0140] Optionally, the first determining subunit includes: an evaluation subunit, used to evaluate the initial path set using a decision model to obtain an evaluation result, wherein the evaluation result is used to represent the priority score of the paths in the initial path set in terms of time efficiency, energy consumption, task timeliness, and resource utilization; and a second determining subunit, used to determine the target path from the initial path set based on the evaluation result.

[0141] Optionally, the vehicle's task execution device 400 further includes: an adjustment unit for adjusting the weights of parameters in the decision model based on the evaluation results; and an evaluation unit for evaluating the initial path set using the adjusted decision model to obtain the target path.

[0142] Optionally, the second determining subunit includes: a sorting subunit, used to sort the priority scores to obtain a sorting result; and a third determining subunit, used to determine the paths with priority scores greater than the priority score threshold in the sorting result from the initial path set as target paths.

[0143] Optionally, the third determining subunit includes: a fourth determining subunit, used to determine a path as the target path in response to the path in the sorting result having a priority score greater than the priority score threshold as a path; the task execution device 400 of the vehicle further includes: an allocation unit, used to allocate multiple paths to multiple vehicles whose attribute information of the task is within the attribute threshold range in response to the path in the sorting result having a priority score greater than the priority score threshold as multiple paths, according to the collaborative scheduling strategy.

[0144] Optionally, the first determining unit 404 includes a fifth determining subunit, used to determine the target path based on status information, task information, and road information at each target time interval.

[0145] In this embodiment of the invention, the acquisition unit 402 acquires the status information of the vehicle set, the work task of the vehicle set, and the road information of the vehicle set during the execution of the work task. The status information indicates the idle status and / or location status of the vehicles in the vehicle set. The work task is executed through the vehicle's transport operation instructions or transfer operation instructions. The road information indicates the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental state of the road where the current road segment is located. The first determining unit 404 determines the target path of the vehicle based on the status information, the work task, and the road information. The target path represents the path that satisfies both the status information and the road information during the execution of the work task. The second determining unit 406 determines the task execution strategy of the target vehicle in the vehicle set based on the target path. The task execution strategy represents the rules by which the target vehicle executes the work task. The execution unit 408 controls the target vehicle to execute the work task along the target path according to the task execution strategy, thus solving the technical problem of low efficiency in vehicle task execution and achieving the technical effect of improving the efficiency of vehicle task execution.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0147] According to embodiments of the present invention, a computer-readable storage medium is also provided, the storage medium including a stored program, wherein the program executes the methods described above.

[0148] According to an embodiment of the present invention, a processor is also provided for running a program, wherein the program executes the methods described above during runtime.

[0149] According to another aspect of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present invention during runtime.

[0150] Embodiments of this application also provide a computer program product. Optionally, in this embodiment, the computer program product may include a computer program that, when executed by a processor, implements the methods described in the embodiments of this application.

[0151] According to another aspect of the present invention, a vehicle is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods of various embodiments of the present application when it runs.

[0152] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0154] The units described as separate components may or may not be physically separate. 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0157] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for executing a vehicle's tasks, characterized in that, include: The system acquires the status information of the vehicle set, the work tasks of the vehicle set, and the road information of the vehicle set during the execution of the work tasks. The status information is used to indicate the idle status and / or location status of the vehicles in the vehicle set. The work tasks are executed through the transportation operation instructions or transfer operation instructions of the vehicles. The road information is used to indicate the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental status of the road where the current road segment is located. Based on the status information, the task, and the road information, the target path of the vehicle is determined, wherein the target path represents the path in which the status information satisfies the status conditions and the road information satisfies the road conditions during the execution of the task by the vehicle. Based on the target path, a task execution strategy for the target vehicles in the vehicle cluster is determined, wherein the task execution strategy is used to represent the rules by which the target vehicles perform the operation task; According to the task execution strategy, control the target vehicle to perform the operation task along the target path.

2. The method according to claim 1, characterized in that, According to the task execution strategy, controlling the target vehicle to perform the operation task along the target path includes: In response to the detection of a deviation in the target path or a delay in the execution of the job task, the scheduling module is used to adjust the task execution strategy; According to the adjusted task execution strategy, the target vehicle is controlled to perform the operation task along the target path.

3. The method according to claim 1, characterized in that, Based on the status information, the task, and the road information, the target path of the vehicle is determined, including: Based on the status information, the task, and the road information, a virtual mirror information of the task is generated. The virtual mirror information is used to represent a virtual mirror body synchronized with the task. The virtual mirror body is constructed in a digital twin platform and is used to represent an information database constructed from the status information, the task, and the road information. Based on the virtual image information, an initial path set is generated; The target path is determined from the initial path set using a decision model, wherein the decision model is obtained by training a neural network model using state information samples, task samples, road information samples, and initial path set samples. The state information samples are used to represent the idle state and / or location state of vehicle samples in the vehicle set samples. The task samples are executed by the transportation operation instructions or transfer operation instructions of the vehicle samples. The road information samples are used to represent the traffic congestion level of the current road segment where the vehicle sample is located, and / or the environmental state of the road where the current road segment is located.

4. The method according to claim 3, characterized in that, Using a decision model, the target path is determined from the initial path set, including: The decision model is used to evaluate the initial path set to obtain evaluation results, wherein the evaluation results are used to represent the priority scores of the paths in the initial path set in terms of time efficiency, energy consumption, task timeliness, and resource utilization. Based on the evaluation results, the target path is determined from the initial path set.

5. The method according to claim 4, characterized in that, The method further includes: Based on the evaluation results, the weights of the parameters in the decision model are adjusted; The initial path set is evaluated using the adjusted decision model to obtain the target path.

6. The method according to claim 4, characterized in that, Based on the evaluation results, the target path is determined from the initial path set, including: The priority scores are sorted to obtain the sorting results; From the initial path set, the paths whose priority scores are greater than the priority score threshold in the sorting results are determined as the target paths.

7. The method according to claim 6, characterized in that, From the initial path set, paths in the sorting results whose priority scores are greater than the priority score threshold are identified as the target paths, including: In response to the ranking result, a path whose priority score is greater than the priority score threshold is identified as a path, and the path is determined as the target path. The method further includes: In response to the fact that there are multiple paths in the sorting result whose priority scores are greater than the priority score threshold, the multiple paths are respectively assigned to multiple vehicles whose attribute information of the task is within the attribute threshold range, according to the collaborative scheduling strategy.

8. The method according to any one of claims 1 to 7, characterized in that, Based on the status information, the task, and the road information, the target path of the vehicle is determined, including: At each target time interval, the target path is determined based on the status information, the task, and the road information.

9. A vehicle task execution device, characterized in that, include: The acquisition unit is used to acquire the status information of the vehicle set, the operation task of the vehicle set, and the road information of the vehicle set during the execution of the operation task. The status information is used to indicate the idle status and / or location status of the vehicles in the vehicle set. The operation task is executed by the transportation operation command or transfer operation command of the vehicle. The road information is used to indicate the traffic congestion level of the current road segment where the vehicle is located, and / or the environmental status of the road where the current road segment is located. The first determining unit is configured to determine the target path of the vehicle based on the state information, the task, and the road information, wherein the target path represents the path in which the state information satisfies the state conditions and the road information satisfies the road conditions during the process of the vehicle performing the task. The second determining unit is used to determine the task execution strategy of the target vehicles in the vehicle cluster based on the target path, wherein the task execution strategy is used to represent the rules for the target vehicles to perform the operation task; An execution unit is used to control the target vehicle to perform the operation task along the target path in accordance with the task execution strategy.

10. A processor, characterized in that, The processor is used to run a program, wherein the program is executed by the processor to perform the method according to any one of claims 1 to 8.

11. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

12. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

Images