A vehicle fleet task allocation method, device, equipment and vehicle fleet management system

Abstract

The application provides a fleet task allocation method, device and equipment and a fleet management system, and relates to the technical field of fleet management. After a fleet agent obtains transportation task information, the information is broadcast to each vehicle agent. The vehicle agent acquires vehicle state data of the vehicle in real time in response to the transportation task information, determines a bidding result for the transportation task based on the transportation task information and the vehicle state data; wherein the vehicle state data is data acquired by the fleet agent from the vehicle in real time, so as to ensure the reliability of the bidding result. The fleet agent receives the reported bidding result and determines a target vehicle for executing the transportation task from the fleet based on the bidding result. Based on real-time and local vehicle state data, the task allocation decision is completed, the quality of task allocation is ensured, the fleet agent does not need to analyze microscopic state data of all vehicles, only needs to analyze the bidding result, the calculation amount is greatly reduced, the efficiency of task allocation is improved, and the task allocation decision is prevented from lagging behind.

Description

Technical Field

[0001] This application relates to the field of fleet management technology, and in particular to a fleet task allocation method, device, equipment and fleet management system. Background Technology

[0002] With the rapid development of artificial intelligence, the Internet of Things, and vehicle-to-everything (V2X) technologies, the modern transportation industry is undergoing a profound digital transformation. In the transportation industry, transportation tasks are typically assigned to fleets, which then allocate these tasks to suitable vehicles. The quality and efficiency of task allocation directly impact the fleet's operating costs, efficiency, and safety.

[0003] Currently, when allocating transportation tasks, each vehicle in the fleet first collects its own status data via its onboard terminal and uploads it to the cloud. Then, the cloud analyzes the status data of all vehicles and selects a suitable vehicle to perform the transportation task. This "cloud-centralized" task allocation method has the following drawbacks: First, the computational load on the cloud increases dramatically with the number of vehicles, leading to reduced efficiency and delayed decision-making. Second, cloud-based decisions rely on status data reported by the onboard terminals several minutes or even longer in advance, resulting in a decline in the quality of task allocation (e.g., assigning a vehicle about to run out of power to perform a long-distance transportation task).

[0004] Therefore, there is a need to provide a fleet task allocation method to improve the quality and efficiency of task allocation. Summary of the Invention

[0005] In view of this, this application provides a fleet task allocation method, which is applied to vehicle agents in a fleet management system; wherein the fleet management system includes fleet agents and vehicle agents corresponding one-to-one with vehicles in the fleet; the method includes: Receive transportation task information broadcast by the fleet agent; wherein the transportation task information includes at least the destination and weight of the transportation task; In response to the transportation task information, obtain the vehicle's real-time status data; Based on the transportation task information and the vehicle status data, the bidding results for the transportation task are determined; wherein, the bidding results include a first result indicating that the transportation task is abandoned, and a second result indicating that the transportation task is applied for. The bidding results are uploaded to the fleet agent so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

[0006] Optionally, the method further includes: Determine whether communication with the fleet agent has been interrupted; If there is no interruption, proceed to the step of receiving the transportation task information broadcast by the fleet agent; If interrupted, the election result of this vehicle agent is determined based on the specified election criteria; if the election result indicates that it has been elected as the leader agent among the vehicle agents, then it performs the corresponding transportation task allocation operation in place of the fleet agent; wherein, the specified election criteria include indicators used to characterize available computing resources.

[0007] Optionally, determining the election result of the vehicle's intelligent agent based on specified election indicators includes: Based on the specified election criteria, a priority score for the vehicle agent is determined; wherein the priority score is used to characterize the priority of the vehicle agent among all vehicle agents. The priority score and the unique identifier of the vehicle's intelligent agent are combined to form an election message; The election message is broadcast to other vehicle agents; wherein, the other vehicle agents are vehicle agents in the fleet management system other than the vehicle agent itself. Determine whether an election response message is received from the other vehicle intelligent agent within a preset time period; wherein the election response message is used to indicate that the priority of the other vehicle intelligent agent is higher than the priority of the current vehicle intelligent agent; If so, then the election result indicates that the candidate was not elected as the leading agent. If not, then the election result indicates that the candidate has been elected as the leading agent.

[0008] Optionally, the vehicle in question is the lead vehicle in the target convoy; wherein, the target convoy is a convoy composed of the target vehicles; correspondingly, the method further includes: When performing the transportation task, formation decision data is acquired; wherein, the formation decision data includes at least road condition information of the road to be traveled, environmental perception information, and vehicle status data of the target vehicle; Based on the formation decision data, a target formation mode is determined from the preset formation modes; wherein, the preset formation modes include at least an energy-saving formation mode and a risk-avoidance formation mode; According to the target formation mode, the target state sequence of the target vehicles is planned so that the target vehicles drive according to the target state sequence; wherein, the target state sequence includes the target position and target speed corresponding to each planning time.

[0009] Optionally, the method further includes: Obtain the abnormal monitoring data of the vehicle; wherein the abnormal monitoring data includes at least the vehicle status data and the driver status data; When the abnormal monitoring data is determined to meet the preset abnormal conditions, an alarm pop-up is displayed on the vehicle dashboard of the vehicle; wherein, the alarm pop-up includes a trigger node for instructing the driver to request a rest. When it is determined that the trigger node has been triggered, a rest request is sent to the fleet agent so that the fleet agent can stop broadcasting new transportation task information to the vehicle.

[0010] This application also provides a fleet task allocation method, which is applied to a fleet agent in a fleet management system; wherein the fleet management system further includes vehicle agents corresponding one-to-one with the vehicles in the fleet; the method includes: Acquire transportation task information and broadcast the transportation task information to each of the vehicle intelligent agents; wherein, the transportation task information includes at least the destination and task weight of the transportation task; Receive bidding results reported by each of the vehicle intelligent agents; wherein, the bidding results include a first result indicating that the vehicle intelligent agent abandons the execution of the transportation task, and a second result indicating that the vehicle intelligent agent applies to execute the transportation task; Based on the bidding results, target vehicles for performing the transportation tasks are determined from the fleet.

[0011] Optionally, the second result also includes bid prices and estimated arrival times; based on each of the bid results, determining the target vehicle from the fleet to perform the transportation task includes: The vehicle that corresponds to the second result among all the vehicles is selected as a candidate vehicle; The allocation scores of the candidate vehicles are obtained by weighted summation of the allocation scores of the candidate vehicles using preset scoring weights; wherein the allocation scoring indicators include at least the bid price and the estimated arrival time; the preset scoring weights include a first preset weight corresponding to the bid price and a second preset weight corresponding to the estimated arrival time. The vehicle with the highest assigned score among the candidate vehicles is selected as the target vehicle.

[0012] Optionally, the method further includes: Each vehicle intelligent agent obtains the historical estimated arrival time and historical actual arrival time of each vehicle when performing historical transportation tasks. The historical transportation tasks in which the actual arrival time is no later than the estimated arrival time are considered as on-time delivery tasks. The proportion of the on-time delivery tasks to the total number of historical transportation tasks is taken as the on-time delivery ratio; The preset scoring weights are adjusted based on the on-time delivery ratio.

[0013] Optionally, the method further includes: Each vehicle intelligent agent acquires the historical estimated arrival time, historical actual arrival time, and task execution data of each vehicle when performing historical transportation tasks; wherein, the task execution data includes at least one of vehicle status data, driver status data, and road condition information; Based on the task execution data, the vehicles are clustered to obtain vehicle groups. Determine the partial on-time delivery ratio of the vehicle group; wherein the partial on-time delivery ratio is the proportion of historical transportation tasks delivered on time by the vehicle group to all historical transportation tasks delivered by the vehicle group. The vehicle group with the lowest local on-time delivery rate among all the vehicle groups is designated as the abnormal vehicle group. The task execution data of the abnormal vehicle group is analyzed to determine the reasons for the delayed delivery of transportation tasks, and transportation task allocation suggestions are generated based on the reasons.

[0014] This application also provides a fleet management system, the system including a fleet agent and a vehicle agent corresponding one-to-one with the vehicles in the fleet; The vehicle agent is used to execute the first fleet task allocation method described above; The fleet agent is used to execute the second fleet task allocation method described above.

[0015] Optionally, the system also includes a vehicle dashboard corresponding to each vehicle agent and a fleet dashboard corresponding to each fleet agent. The vehicle dashboard is used to display the vehicle-level monitoring data of the vehicle; wherein, the vehicle-level monitoring data includes at least vehicle status data; The fleet dashboard is used to display fleet-level monitoring data; wherein, the fleet-level monitoring data includes at least fleet operation indicators determined based on the vehicle-level monitoring data.

[0016] This application also provides a fleet task allocation device, which is configured in a vehicle agent within a fleet management system; wherein the fleet management system includes fleet agents and vehicle agents corresponding one-to-one with the vehicles in the fleet; the device includes: The task information receiving module is used to receive transportation task information broadcast by the fleet intelligent agent; wherein, the transportation task information includes at least the destination and weight of the transportation task; The status data acquisition module is used to acquire real-time vehicle status data in response to the transportation task information. The bidding decision module is used to determine the bidding result for the transportation task based on the transportation task information and the vehicle status data; wherein the bidding result includes a first result indicating that the transportation task is abandoned, and a second result indicating that the transportation task is applied for. The bidding result uploading module is used to upload the bidding results to the fleet agent, so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

[0017] This application also provides a fleet task allocation device, the device being configured in a fleet management system's fleet agent; wherein, the fleet management system further includes vehicle agents corresponding one-to-one with the vehicles in the fleet; the device includes: The task information broadcasting module is used to acquire transportation task information and broadcast the transportation task information to each of the vehicle intelligent agents; wherein, the transportation task information includes at least the destination and task weight of the transportation task; The bidding result receiving module is used to receive the bidding results reported by each of the vehicle intelligent agents; wherein, the bidding result includes a first result indicating that the vehicle intelligent agent abandons the execution of the transportation task, and a second result indicating that the vehicle intelligent agent applies to execute the transportation task; The task allocation module is used to determine the target vehicle to perform the transportation task from the fleet based on the bidding results.

[0018] This application also provides an electronic device, the electronic device comprising: Memory, used to store computer programs; A processor is used to implement the steps of any of the above-described fleet task allocation methods when executing the computer program.

[0019] In summary, this application provides a fleet task allocation method, apparatus, equipment, and fleet management system. After obtaining transportation task information, the fleet agent broadcasts the information to each vehicle agent. In response to the transportation task information, each vehicle agent acquires its own vehicle status data in real time and determines the bidding results for the transportation task based on the transportation task information and the vehicle status data. The vehicle status data is obtained by the fleet agent from the vehicle's local data in real time to ensure the reliability of the bidding results. The fleet agent receives the bidding results reported by each vehicle agent and determines the target vehicle from the fleet to perform the transportation task based on the bidding results. By using real-time and local vehicle status data, task allocation decisions are made, ensuring the quality of task allocation. Furthermore, the fleet agent does not need to analyze the micro-state data of all vehicles, but only the bidding results, greatly reducing the computational load on the fleet agent and thus improving the efficiency of task allocation and avoiding delays in task allocation decisions. Attached Figure Description

[0020] Figure 1 A schematic diagram of a fleet management system provided in this application; Figure 2 A first flowchart illustrating a fleet task allocation method provided in this application; Figure 3 A second flowchart illustrating a fleet task allocation method provided in this application; Figure 4 A first structural schematic diagram of a fleet task allocation device provided in this application; Figure 5 A second structural schematic diagram of a fleet task allocation device provided in this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. Detailed Implementation

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0022] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0023] Please refer to Figure 1 , Figure 1 This is a schematic diagram of a fleet management system provided in this application.

[0024] The fleet management system includes fleet agents and vehicle agents corresponding one-to-one with the vehicles in the fleet. Specifically, this application pre-instantiates a vehicle agent for each vehicle in the fleet and provides computing resources for the vehicle agents using onboard edge devices or computing containers; it also instantiates a fleet agent for the entire fleet and provides computing resources for the fleet agents using cloud servers.

[0025] The fleet agent establishes communication connections with each vehicle agent. Furthermore, communication connections are also established between the vehicle agents in the fleet management system of this application. Each vehicle agent obtains a unique digital identity certificate from the root certificate authority (Root CA) beforehand; this certificate contains the vehicle agent's public key and identity information, and is digitally signed by the root CA so that each vehicle agent can verify the authenticity of the certificate and establish a communication connection.

[0026] Please refer to Figure 2 , Figure 2 This application provides a first flowchart illustrating a fleet task allocation method, which is applied to vehicle agents in the aforementioned fleet management system; the method includes: S11. Receive transportation task information broadcast by the fleet agent; wherein the transportation task information includes at least the destination and weight of the transportation task.

[0027] The aforementioned transportation task information can be sent by the customer to the fleet agent through the order agent. After receiving the transportation task information, the fleet agent broadcasts the transportation task information to each vehicle agent.

[0028] S12. In response to transportation task information, obtain real-time vehicle status data of this vehicle.

[0029] After receiving the transportation task information, the vehicle intelligent agent immediately obtains the vehicle status data. The vehicle status data is data obtained by the vehicle intelligent agent from the vehicle's local area in real time. Therefore, the vehicle status data has both real-time and local characteristics, which ensures that the vehicle intelligent agent can accurately assess whether the vehicle can complete the transportation task based on the real-time and local vehicle status data.

[0030] It is understandable that there is a one-to-one correspondence between a vehicle intelligence agent and a vehicle, and the aforementioned "this vehicle" refers to the vehicle corresponding to the vehicle intelligence agent; correspondingly, the vehicle status data of the vehicle obtained by the vehicle intelligence agent is the status data of the single vehicle corresponding to that vehicle intelligence agent.

[0031] The data sources for the aforementioned vehicle status data include, but are not limited to, the vehicle's CAN bus (Controller Area Network), GPS (Global Positioning System), ADAS (Advanced Driver Assistance System), and driver monitoring systems. Vehicle status data includes, but is not limited to, the vehicle's location, speed, acceleration, engine speed, fuel or electricity consumption, remaining fuel or battery charge, tire pressure, and coolant temperature.

[0032] S13. Based on transportation task information and vehicle status data, determine the bidding results for the transportation task; wherein, the bidding results include a first result indicating abandonment of the transportation task and a second result indicating application for the execution of the transportation task.

[0033] In this application, after the vehicle agent obtains its own vehicle status data, it does not directly send the vehicle status data to the fleet agent. Instead, it first completes a bidding decision based on the transportation task information and the vehicle status data, that is, determines the bidding result for the transportation task. On the one hand, the vehicle status data is data obtained by the vehicle agent from the vehicle's local storage in real time. By ensuring the real-time nature and locality of the vehicle status data, the reliability of the bidding result generated by the vehicle agent is guaranteed, thereby ensuring the reliability of the fleet agent's allocation of transportation tasks based on the bidding result. On the other hand, by having the vehicle agent determine the bidding result itself, it undertakes some of the related operations for transportation task allocation, reducing the computational load of the fleet agent and thus improving the efficiency of transportation task allocation.

[0034] For example, the vehicle status data acquired by the vehicle agent includes the vehicle's remaining battery power. In response to transportation task information, the vehicle agent obtains the vehicle's current remaining battery power; this remaining battery power is real-time and accurate. Therefore, when faced with a long-distance transportation task, the vehicle agent can accurately determine that the vehicle's remaining battery power is insufficient to support the completion of the long-distance transportation task, and thus chooses to abandon the long-distance transportation task, avoiding the situation in related technologies where inaccurate remaining battery power data obtained from the cloud leads to the assignment of vehicles with low battery power to perform long-distance transportation tasks.

[0035] The process of determining the bidding results for vehicle intelligent agents is explained below.

[0036] First, the vehicle's intelligent agent determines whether it possesses the basic capability to perform the transportation task based on transportation mission information and vehicle status data. For example, based on the vehicle's current remaining battery / fuel level, it determines whether the vehicle's range is sufficient to reach the destination; if not, the bid result is directly designated as the first result, meaning the transportation mission is abandoned. It then determines whether the weight of the transportation mission exceeds the vehicle's maximum rated load; if so, the bid result is designated as the first result. Finally, based on the distance between the vehicle's current location and the destination, as well as real-time traffic information, it determines the estimated arrival time, i.e., the time to reach the destination; if this time exceeds the time stipulated in the transportation mission, the bid result is again designated as the first result.

[0037] If it is determined that the vehicle possesses the basic capability to perform the transportation task, the bidding result is designated as the second result. Based on this, the expected costs required to perform the transportation task are calculated; these expected costs may include energy costs, time costs, vehicle depreciation costs, and opportunity costs, etc., which are not specifically limited in this application. Then, based on the expected costs and target profit margin, the bid price is determined. Finally, the bid price and estimated arrival time, along with the second result, are sent to the fleet agent so that the fleet agent can perform subsequent transportation task allocation operations.

[0038] In addition, the vehicle intelligent agent can also record the execution deviation of the transportation task (that is, the deviation between the actual arrival time and the expected arrival time), and adjust the above bidding logic based on the execution deviation to improve the accuracy of bidding.

[0039] S14. Upload the bidding results to the fleet agent so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

[0040] After determining its own bidding results, the vehicle agent uploads them to the fleet agent. Upon receiving the bidding results from each vehicle, the fleet agent evaluates them to determine the target vehicle for the transportation task. This "bid-evaluation" mechanism significantly reduces the computational load on the fleet agent, thereby improving the efficiency of transportation task allocation. The process by which the fleet agent determines the target vehicle based on the bidding results will be described in subsequent embodiments and will not be elaborated here.

[0041] In summary, in this application, the fleet agent makes task allocation decisions based on real-time and local vehicle status data, ensuring the quality of task allocation; and the fleet agent does not need to analyze the micro-vehicle status data of all vehicles, but only needs to analyze the bidding results, which greatly reduces the computational load of the fleet agent, thereby improving the efficiency of task allocation and avoiding lag in task allocation decisions.

[0042] Based on the above embodiments: As an optional embodiment, the method further includes: Determine if communication with the fleet agent has been interrupted; If there is no interruption, proceed to the step of receiving transportation task information broadcast by the fleet agent; If interrupted, the election result of this vehicle agent is determined based on the specified election criteria; if the election result indicates that it has been elected as the leader agent among the vehicle agents, it will perform the corresponding transportation task allocation operation in place of the fleet agent; wherein, the specified election criteria include indicators used to characterize available computing resources.

[0043] In this embodiment, when communication between the vehicle intelligence agent and the fleet intelligence agent is interrupted, the decision-making power for the allocation of transportation tasks is decentralized. That is, each vehicle intelligence agent negotiates to elect a leader intelligence agent from among the vehicle intelligence agents. The leader intelligence agent then replaces the fleet intelligence agent to complete the corresponding transportation task allocation operation, thus avoiding the paralysis of the fleet due to a single point of failure.

[0044] Specifically, the vehicle agent and the fleet agent maintain heartbeat communication; if the vehicle agent detects that the heartbeat response of the fleet agent has timed out, it determines that the communication between itself and the fleet agent has been interrupted; if it has not timed out, it determines that the communication between itself and the fleet agent has not been interrupted.

[0045] If communication is uninterrupted, the vehicle agent will normally perform the steps described in the previous embodiment: receiving transportation task information, obtaining vehicle status data in response to the transportation task information, and determining the bidding result based on the transportation task information and vehicle status data. This embodiment will not elaborate on these steps here.

[0046] In the event of a communication interruption, the vehicle agent determines the election result based on its own designated election criteria. These criteria include at least indicators representing the vehicle agent's available computing resources to ensure that the elected leader agent has sufficient computing resources to support transportation task allocation. The election result includes both the outcome of being elected as the leader agent and the outcome of not being elected.

[0047] If the election result indicates that the leader agent has been elected, it will perform the following transportation task allocation operations on behalf of the fleet agents: after receiving the transportation task information, it will broadcast the transportation task information to each vehicle agent, receive the bidding results uploaded by each vehicle agent, and determine the target vehicle based on the bidding results. The logic of the leader agent determining the target vehicle based on the bidding results is the same as that of the fleet agent, which will be described in detail in subsequent embodiments and will not be repeated here.

[0048] If the election results indicate that it has not been elected as the leader agent, it will normally perform the operations of determining the bidding results and uploading the bidding results to the leader agent to assist the leader agent in completing the transportation task allocation.

[0049] The process of determining the election results of the vehicle's intelligent agent based on specified election criteria is explained below.

[0050] As an optional embodiment, the election result of the vehicle's intelligent agent is determined based on a specified election criterion, including: Based on the specified election criteria, the priority score of this vehicle's intelligent agent is determined; wherein, the priority score is used to characterize the priority of this vehicle's intelligent agent among all vehicle intelligent agents. The election message is composed of the priority score and the unique identifier of the vehicle's intelligent agent; The election message is broadcast to other vehicle agents; these other vehicle agents are vehicle agents in the fleet management system other than this vehicle agent. Determine whether an election response message is received from other vehicle intelligent agents within a preset time period; wherein, the election response message is used to indicate that the priority of other vehicle intelligent agents is higher than the priority of this vehicle intelligent agent; If so, then the election result indicates that the candidate was not elected as the leader agent; If not, then the election result is determined to represent the elected leader agent.

[0051] First, based on specified election criteria, the priority score of this vehicle's intelligent agent is determined, that is, its priority among all vehicle intelligent agents. Specifically, the specified election criteria include at least indicators representing available computing resources, such as the number of available CPU cores and the amount of available memory. The priority score is determined by weighted summation of the specified election criteria. The more computing resources available to a vehicle intelligent agent, the better it can meet the computing power required for transportation task allocation, and therefore its priority score is higher, and its priority among all vehicle intelligent agents is also higher. Of course, the specified election criteria mentioned above can also include the vehicle's location and remaining battery power, etc., but this embodiment does not specifically limit this.

[0052] Next, the priority score and the unique identifier of this vehicle's intelligent agent are combined to form an election message, which is then broadcast to other vehicle intelligent agents. Taking this vehicle's intelligent agent as VA-X and other vehicle intelligent agents as VA-Y as an example, VA-X sends the election message to VA-Y; if the priority score of VA-Y is higher than that of VA-X, then VA-Y will return an election response message to VA-X to suppress VA-X's election attempt; if the priority score of VA-Y is not higher than that of VA-X, then VA-Y will not reply with an election response message, or will reply with a message acknowledging VA-X's election eligibility.

[0053] If this vehicle agent receives an election response message from any other vehicle agent within a preset time period, it indicates that this vehicle agent is not the highest priority vehicle agent, and therefore the election result indicates that this vehicle agent has not been elected as the leader agent.

[0054] If, within a preset time period, this vehicle agent does not receive an election response message from any other vehicle agent, it indicates that this vehicle agent has the highest priority, and therefore the election result signifies that this vehicle agent has been elected as the leader agent. At this point, this vehicle agent can also broadcast a message to other vehicle agents declaring itself as the leader agent, so that other vehicle agents can send their bidding results to this vehicle agent.

[0055] Furthermore, to avoid a message storm caused by multiple vehicle agents simultaneously initiating elections, vehicle agents can initiate elections only after detecting a period of communication interruption between themselves and the fleet agents. Additionally, a heartbeat communication mechanism must be established between the leader vehicle agent and all other vehicle agents; if the leader vehicle agent experiences a communication failure, the other vehicle agents can then re-elect a leader.

[0056] Furthermore, for large convoys, due to terrain limitations and other reasons, the entire convoy may be divided into multiple convoy groups. While communication between convoy groups is impossible, communication between the vehicle agents within each convoy group is possible. In this case, the vehicle agents within each convoy group can elect a leader agent following the steps described above. When communication between the convoy groups is restored, the vehicle agent with the highest priority is elected as the final leader agent, taking unified control of the transportation task allocation operation.

[0057] Finally, when the fleet agent re-establishes communication with each vehicle agent, it broadcasts a regression message to all vehicle agents. Upon receiving the regression message, the leader agent relinquishes the decision-making authority for transportation task allocation back to the fleet agent and synchronizes all information from the decision-making period (including but not limited to the allocation and execution status of transportation tasks) with the fleet agent, enabling the fleet agent to take over the task allocation work.

[0058] In summary, this embodiment can delegate the task allocation decision-making authority to the vehicle intelligence agent when communication between the fleet intelligence agent and the vehicle intelligence agent is interrupted; when communication between the fleet intelligence agent and the vehicle intelligence agent is restored, the decision-making authority is returned to the fleet intelligence agent, ensuring the continuity of transportation task allocation and execution, and avoiding fleet paralysis due to single point of failure.

[0059] As an optional embodiment, this vehicle is the lead vehicle in the target convoy; wherein, the target convoy is a convoy composed of various target vehicles; correspondingly, the method further includes: When performing transportation tasks, acquire formation decision data; the formation decision data includes at least road condition information of the road to be traveled, environmental perception information, and vehicle status data of the target vehicles; Based on formation decision data, a target formation mode is determined from preset formation modes; wherein, the preset formation modes include at least an energy-saving formation mode and a risk-avoidance formation mode; According to the target formation mode, the target state sequence of the target vehicles is planned so that the target vehicles can drive according to the target state sequence; wherein, the target state sequence includes the target position and target speed at each planning time.

[0060] When multiple target vehicles jointly perform the same transportation task, they form a target convoy. To reduce the overall energy consumption of the convoy and improve its driving safety, this embodiment provides a convoy driving mechanism, in which the lead vehicle in the target convoy is responsible for coordinating the driving speed and position of each following vehicle. The aforementioned convoy driving mechanism can be manually triggered by the convoy manager, or it can be automatically triggered by the vehicle's intelligent agent when it detects insufficient vehicle battery power or complex road conditions; this embodiment does not limit this.

[0061] During the transportation mission, the lead vehicle acquires platooning decision data. This data includes at least road condition information, environmental perception information, and vehicle status data of the target vehicles. Road condition information includes, but is not limited to, the congestion index, construction zones, speed limits, and gradient information of the road to be traversed. Environmental perception information includes, but is not limited to, obstacle information, pedestrian information, vehicle information, traffic signs, and weather conditions collected by onboard environmental perception sensors. Vehicle status data includes, but is not limited to, vehicle position, speed, acceleration, remaining battery power, and load.

[0062] Subsequently, the lead vehicle determines the target formation mode based on the formation decision data. For example, if it detects that there are vehicles with insufficient remaining battery power in the target convoy, or if it detects that there is congestion or construction on the road ahead, it determines the target formation mode to be an energy-saving formation mode; when it detects that there is a sudden obstacle on the road ahead, it determines the target formation mode to be a hazard avoidance formation mode.

[0063] Ultimately, the lead vehicle's vehicle agent plans the target state sequence for each vehicle in the target convoy according to the target formation mode. As an optional implementation, when the target formation mode is an energy-saving formation mode, the target state sequence for each target vehicle is planned based on formation decision data, with the goal of minimizing the total energy consumption of the target convoy; when the target formation mode is a risk-avoidance formation mode, the target state sequence for each target vehicle is planned based on formation decision data, with the goal of maximizing the safety margin of the target convoy.

[0064] In summary, this embodiment utilizes the communicability between vehicle agents, allowing the lead vehicle's agent to autonomously complete platooning planning, reducing reliance on platoon agents and ensuring efficient collaborative driving of all target vehicles under various conditions, thereby reducing overall platoon energy consumption and improving driving safety.

[0065] As an optional embodiment, the method further includes: Obtain abnormal monitoring data for this vehicle; the abnormal monitoring data shall include at least vehicle status data and driver status data; When abnormal monitoring data is determined to meet the corresponding preset abnormal conditions, an alarm pop-up is displayed on the vehicle dashboard; the alarm pop-up includes a trigger node to instruct the driver to request a rest. When a trigger node is determined to be triggered, a rest request is sent to the fleet agent to cause the fleet agent to suspend broadcasting new transportation task information to the vehicle.

[0066] In this embodiment, the fleet management system also includes a vehicle dashboard corresponding to each vehicle. In addition to displaying the vehicle's status data, driver status data, task execution data, and environmental perception data, the vehicle dashboard can also interact with the driver, providing the driver with an entry point to intervene in the transportation task allocation operation, thus realizing true human-machine collaboration.

[0067] Specifically, the vehicle's intelligent agent continuously acquires anomaly monitoring data from various data sources; among which, anomaly monitoring data includes at least vehicle status data and driver status data; vehicle status data will not be elaborated here; driver status data includes, but is not limited to, fatigue index, distraction duration, and driving behavior collected through the driver monitoring system.

[0068] After acquiring the anomaly monitoring data, it is determined whether the anomaly monitoring data meets the corresponding preset anomaly conditions. For example, if the anomaly monitoring data is a fatigue index, the corresponding preset anomaly condition is that the fatigue index is higher than a preset index; if the anomaly monitoring data is distraction duration, the corresponding preset anomaly condition is that the distraction duration exceeds a preset duration; if the anomaly monitoring data is the number of dangerous driving behaviors (such as sudden acceleration, sudden braking, and sharp turns), the preset anomaly condition is that the number of dangerous driving behaviors per unit mileage is higher than a preset number. The preset anomaly conditions can be set according to actual needs, and this embodiment does not impose any special limitations.

[0069] If the abnormal monitoring data meets the corresponding preset abnormal conditions, the vehicle dashboard's alarm mechanism is triggered. That is, an alarm pop-up is displayed on the vehicle dashboard; this pop-up includes a trigger node to instruct the driver to request a rest. This trigger node is equivalent to an interactive control provided by the vehicle dashboard, which the driver can trigger. When the vehicle agent determines that the node has been triggered, it sends a rest request (including the vehicle's unique identifier and the reason for the rest) to the fleet agent, causing the fleet agent to suspend broadcasting new transport task information to the vehicle. Furthermore, if the vehicle is currently performing a transport task, the fleet agent can transfer the task to other vehicles to ensure the safety of vehicles and personnel.

[0070] In summary, in this embodiment, the driver is no longer a passive, dumb terminal performing transportation tasks. Instead, the driver is empowered through trigger nodes on the vehicle dashboard, enabling the driver to intervene in the allocation of transportation tasks based on their own status. This achieves human-machine collaborative decision-making, which helps improve vehicle driving safety and ensures the safe completion of transportation tasks.

[0071] Please refer to Figure 3 , Figure 3 This application provides a second flowchart illustrating a fleet task allocation method, which is applied to a fleet agent in a fleet management system. The method includes: S21. Obtain transportation task information and broadcast the transportation task information to each vehicle intelligent agent; wherein, the transportation task information includes at least the destination and weight of the transportation task.

[0072] S22. Receive the bidding results reported by each vehicle intelligent agent; wherein, the bidding results include a first result indicating that the vehicle intelligent agent abandons the transportation task, and a second result indicating that the vehicle intelligent agent applies to perform the transportation task.

[0073] S23. Based on the results of each bid, determine the target vehicles from the fleet to perform the transportation tasks.

[0074] For a detailed description of the fleet task allocation method executed by the fleet agent, please refer to the above-described embodiment of the fleet task allocation method executed by the vehicle agent, which will not be repeated here.

[0075] Based on the above embodiments: As an optional embodiment, the second result also includes bid prices and estimated arrival times; based on the bid results, target vehicles for the transportation mission are identified from the fleet, including: The vehicle that corresponds to the second result among all vehicles is selected as a candidate vehicle. Using preset scoring weights, the allocation scoring indicators of candidate vehicles are weighted and summed to obtain the allocation score of the candidate vehicles; wherein, the allocation scoring indicators include at least the bid price and the estimated arrival time; the preset scoring weights include a first preset weight corresponding to the bid price and a second preset weight corresponding to the estimated arrival time. The vehicle with the highest score among the candidate vehicles will be designated as the target vehicle.

[0076] After receiving the bidding results reported by each vehicle agent, the fleet agent needs to select the most suitable target vehicle from multiple candidate vehicles applying to perform the transportation task. In this embodiment, a scoring mechanism is provided to determine the target vehicle based on preset scoring weights, comprehensively considering the bid price and the estimated arrival time, in order to achieve globally optimal task allocation.

[0077] First, within a defined time window, the fleet collects the bidding results reported by all vehicle agents. Vehicles with a second-place bid (i.e., those applying to perform the transportation task) are designated as candidate vehicles. Vehicles with a first-place bid (i.e., those abandoning the transportation task), as well as vehicles that did not report their bidding results to the fleet agent, are removed from the candidate vehicle list for this transportation task.

[0078] For each candidate vehicle, the allocation score is obtained by weighting and summing the allocation score indicators of the candidate vehicle using preset scoring weights. As an optional embodiment, the bid price of the candidate vehicle is normalized based on the highest and lowest bid prices; the expected arrival time of the candidate vehicle is normalized based on the earliest and latest expected arrival times; and the normalized bid price and expected arrival time are weighted and summed using a first preset weight and a second preset weight to obtain the allocation score of the candidate vehicle. The lower the bid price and the earlier the expected arrival time of a candidate vehicle, the higher its corresponding allocation score, and the more likely it is to be selected as the target vehicle.

[0079] Of course, as an optional implementation, the vehicle's historical reputation score can also be used as an allocation metric. The historical reputation score is determined based on the vehicle's historical on-time delivery rate (i.e., the proportion of on-time completed tasks out of the total number of tasks) and customer evaluation scores. Introducing historical reputation scores allows the vehicle agent to provide more realistic and reliable bidding results and incentivizes drivers to maintain good service quality.

[0080] It can be seen that the first and second preset weights affect the degree to which transportation task allocation emphasizes bid price and estimated arrival time. When the first preset weight is larger, the fleet agent is more inclined to allocate transportation tasks to candidate vehicles with lower bid prices; when the second preset weight is larger, the fleet agent is more inclined to allocate transportation tasks to candidate vehicles with earlier estimated arrival times. Based on this, this application also adjusts the preset scoring weights according to the on-time delivery ratio of transportation tasks, and the process is described in detail below.

[0081] As an optional embodiment, the method further includes: Each vehicle's intelligent agent acquires the historical estimated arrival time and historical actual arrival time when each vehicle performed a historical transportation task. The historical transportation tasks whose actual arrival time is no later than the historical estimated arrival time are considered as on-time delivery tasks. The proportion of on-time delivery tasks to historical transportation tasks is used as the on-time delivery ratio. Adjust the preset scoring weights based on the on-time delivery ratio.

[0082] In this embodiment, the fleet agent acquires the historical estimated arrival time and historical actual arrival time of each vehicle for historical transportation tasks. For each historical transportation task, if the historical actual arrival time is not later than the historical estimated arrival time, the task is considered an on-time delivery task; otherwise, the task is considered a delayed delivery task. The proportion of on-time delivery tasks to historical transportation tasks is used as the on-time delivery ratio.

[0083] As an optional implementation, a target ratio range is preset. If the on-time delivery ratio is lower than the lower limit of the target ratio range, it indicates that the fleet as a whole has a tendency to delay delivery tasks. Therefore, subsequent transportation task allocation needs to place more emphasis on timeliness. Based on this, a second preset weight is increased and the first preset weight is decreased, making the fleet agent more inclined to allocate transportation tasks to vehicles with earlier expected arrival times. Of course, it is necessary to limit the second preset weight from falling below a preset weight threshold to avoid transportation task allocation being overly biased towards timeliness while ignoring costs.

[0084] Correspondingly, if the on-time delivery rate exceeds the upper limit of the target rate range, it indicates that the fleet's on-time delivery of transportation tasks meets the requirements, and costs can be appropriately optimized. Based on this, the first preset weight is increased and the second preset weight is decreased, making the fleet agent more inclined to allocate transportation tasks to vehicles with lower bids. Similarly, it is also necessary to limit the first preset weight below a preset weight threshold to avoid the allocation of transportation tasks being overly biased towards reducing costs while ignoring timeliness.

[0085] In summary, the fleet agent dynamically adjusts the preset scoring weights used in task allocation based on the fleet's on-time delivery ratio, thereby continuously optimizing the task allocation scheme and balancing timeliness and cost.

[0086] As an optional embodiment, the method further includes: Each vehicle's intelligent agent acquires the historical estimated arrival time, historical actual arrival time, and task execution data for each vehicle when performing historical transportation tasks; among which, the task execution data includes at least one of vehicle status data, driver status data, and road condition information; Based on the task execution data, each vehicle is clustered to obtain each vehicle group; Determine the partial on-time delivery ratio of the vehicle group; where the partial on-time delivery ratio is the proportion of historical transportation tasks delivered on time by the vehicle group out of all historical transportation tasks delivered by the vehicle group. The vehicle group with the lowest local on-time delivery rate among all vehicle groups is identified as an abnormal vehicle group. Analyze the task execution data of abnormal vehicle groups to determine the reasons for the delayed delivery of transportation tasks, and generate transportation task allocation suggestions based on the reasons.

[0087] In this embodiment, in addition to acquiring the historical estimated arrival time and historical actual arrival time of each vehicle when performing historical transportation tasks, the fleet agent also acquires the task execution data of the vehicles when performing historical transportation tasks. The task execution data includes at least one of vehicle status data, driver status data, and road condition information, the specific content of which can be referred to the foregoing embodiments and will not be repeated here.

[0088] It should be noted that the fleet agent does not need to obtain the historical estimated arrival time, historical actual arrival time, and task execution data for all historical transportation tasks performed by vehicles. Instead, it only needs to obtain the data corresponding to each historical transportation task performed by the vehicle within the abnormal time window. As an optional embodiment, the fleet agent inputs the fleet's historical on-time delivery ratio into the Long Short-Term Memory (LSTM) network, so that the LTM network can determine the expected on-time delivery ratio for the next few hours based on the historical on-time delivery ratio. If the actual on-time delivery ratio is consistently lower than the expected on-time delivery ratio, an anomaly warning is triggered, and the time window in which the actual on-time delivery ratio is consistently lower than the expected on-time delivery ratio is designated as the aforementioned abnormal time window.

[0089] After obtaining the task execution data of each vehicle, an unsupervised clustering algorithm is used to cluster the vehicles into vehicle groups. For each vehicle group, the local on-time delivery ratio is determined; the local on-time delivery ratio is the proportion of historical transport tasks delivered on time by the vehicle group out of all historical transport tasks delivered by the group. The vehicle group with the lowest local on-time delivery ratio is identified as an abnormal vehicle group. Correlation analysis is performed on the task execution data of abnormal vehicle groups to determine the reasons for delayed delivery, allowing for optimization of the transport task allocation logic. For example, if the fatigue index of abnormal vehicle groups is consistently high, the fatigue index is considered the reason for delayed delivery, and transport tasks are more likely to be assigned to drivers with lower fatigue indices.

[0090] For example, at 14:15, the fleet dashboard corresponding to the fleet agent displays the following alarm message: "The on-time delivery rate of the fleet in the East China region decreased by 20% from 13:00, deviating from the predicted on-time delivery rate. Please pay attention." At this time, the fleet agent pulls the task execution data of all vehicles performing transportation tasks in the East China region between 13:00 and 14:15. These vehicles are clustered to obtain vehicle group A, vehicle group B, and vehicle group C. Among them, the local on-time delivery rate of vehicle group A is 55%, and its common characteristics are: average vehicle speed less than 20km / h, low fatigue index, and normal power consumption; the local on-time delivery rate of vehicle group B is 95%, and its common characteristics are: average vehicle speed of 60km / h and all indicators are normal; the local on-time delivery rate of vehicle group C is 85%, and its common characteristics are: low remaining vehicle power.

[0091] It is evident that the on-time delivery rate for vehicle group A is significantly low, with a common characteristic being low-speed driving despite driver fatigue. Furthermore, based on task execution data, it was found that heavy rainfall occurred in the East China region during the task execution period for vehicle group A, causing widespread congestion on urban elevated roads and main thoroughfares, reducing the average vehicle speed to below 20 km / h. Therefore, the heavy rainfall was identified as the primary cause of the decline in the on-time delivery rate. Based on this, task allocation recommendations are generated: Given the continued heavy rainfall in the East China region for the next two hours, it is recommended to suspend the allocation of new long-distance orders to vehicles in this region; priority should be given to assigning transportation tasks to vehicles with sufficient battery power located in other non-congested areas.

[0092] Furthermore, the fleet agent can utilize Long Short-Term Memory (LSTM) networks for fault rate prediction, power consumption prediction, on-time delivery rate prediction, and vehicle cooperative behavior prediction. Fault rate prediction can be based on historical fault data and environmental information to predict the probability of future faults. Power consumption prediction can be based on vehicle battery level and task requirements to predict future power consumption. On-time delivery rate prediction can be based on historical on-time delivery rates and real-time road conditions to predict the on-time delivery rate of future transportation tasks. Vehicle system behavior prediction can be based on decision tree or random forest models to analyze the relationship between cooperative behavior and the environment and transportation tasks, predicting the likelihood of cooperative driving between vehicles.

[0093] Predictive metrics provide fleet agents with a more comprehensive basis for task allocation. For example, if a high vehicle failure rate is predicted for a certain area, tasks will be avoided in that area; if a task is predicted to consume a lot of power, vehicles with sufficient power will be prioritized for the transportation task; if a task is predicted to have an unsatisfactory on-time performance, vehicles with good on-time performance will be prioritized for the transportation task, or the driving routes of the vehicles will be replanned; if a high demand for vehicle coordination is predicted for a certain area, vehicle platooning or vehicle coordination strategies will be adjusted in advance.

[0094] like Figure 1 As shown, this application also provides a fleet management system, which includes a fleet agent and vehicle agents corresponding one-to-one with the vehicles in the fleet; The vehicle intelligent agent is used to execute the first fleet task allocation method described above; The fleet agent is used to execute the second fleet task allocation method described above.

[0095] For a detailed description of the fleet management system provided in this application, please refer to the above-described embodiment of the fleet task allocation method; further details will not be elaborated upon here.

[0096] Based on the above embodiments: As an optional embodiment, the system also includes a vehicle dashboard corresponding to a vehicle agent and a fleet dashboard corresponding to a fleet agent. The vehicle dashboard is used to display the vehicle-level monitoring data of this vehicle; the vehicle-level monitoring data includes at least vehicle status data. The fleet dashboard is used to display fleet-level monitoring data; the fleet-level monitoring data includes at least the fleet operation indicators determined based on vehicle-level monitoring data.

[0097] In this embodiment, the fleet management system is equipped with a two-layer dashboard, namely a vehicle dashboard corresponding one-to-one with the vehicle intelligent agent, and a fleet dashboard corresponding to the fleet intelligent agent.

[0098] The vehicle dashboard displays the vehicle's level monitoring data. This data includes at least vehicle location, speed, acceleration, engine speed, fuel / electricity consumption, remaining battery / fuel level, tire pressure, and coolant temperature, allowing managers to stay informed about the vehicle's status.

[0099] In addition, vehicle-level monitoring data can also include task execution data, environmental perception data, driver status data, equipment health data, and communication status data. Task execution data includes the identifier of the currently executing transportation task, its destination, estimated arrival time, and task progress. Environmental perception data includes congestion index, construction conditions, and speed limits on the road ahead. Driver status data includes driver fatigue levels determined by analysis of steering wheel grip strength, blink frequency, and voice tone. Equipment health data includes the vehicle's fault codes, battery health, brake pad wear, and tire wear. Communication status data includes communication quality, signal strength, and communication latency between the vehicle's intelligent agent and the fleet's intelligent agents.

[0100] Based on this, the vehicle intelligent agent can process the vehicle-level monitoring data and display it on the vehicle dashboard in the form of heatmaps or Gantt charts. This allows the driver to intuitively understand the vehicle's operating status, improving driving safety and efficiency. When abnormalities are detected in the vehicle-level monitoring data, the vehicle intelligent agent controls the vehicle dashboard to display a highlighted warning, enabling the driver to intervene in advance and avoid accidents caused by sudden malfunctions.

[0101] Fleet dashboards are used to display fleet-level monitoring data. Fleet-level monitoring data includes fleet operational metrics determined based on vehicle-level monitoring data. These metrics include, but are not limited to, overall on-time delivery rate, average task completion time, vehicle utilization, cost per kilometer, and cost per unit task. Fleet dashboards can display these operational metrics on a map in card format, allowing managers to intuitively understand the fleet's operational status.

[0102] The fleet signboard serves as an external service interface. For example... Figure 1As shown, for the order agent, the fleet dashboard displays the fleet's operational metrics to help it secure more transport tasks. For the charging agent, when the fleet dashboard displays "3 vehicles in region A have less than 15% battery," the fleet agent automatically sends a "reservation for charging" request to the charging agent.

[0103] Furthermore, managers can simulate vehicle dispatching through fleet dashboards to analyze the impact on fleet operating metrics, assisting in decision-making regarding transportation task allocation and reducing trial-and-error costs. For example, a manager can drag transportation task X to a designated area; the fleet agent automatically analyzes the impact of having vehicle B in the designated area perform transportation task X on overall fuel consumption and on-time delivery rate.

[0104] A two-tiered Kanban system, consisting of a fleet Kanban board and vehicle Kanban boards, can form a negative feedback system. The Kanban boards display the system output in real time, and the fleet and vehicle agents use the data displayed as input to adjust their decision-making parameters. For example, if the Kanban board displays the delay rate of transportation tasks, the vehicle agent can adjust its bidding for subsequent transportation tasks based on this delay rate to gain more rest time; the fleet agent can adjust its task allocation bias based on the delay rate, placing greater emphasis on the timeliness of vehicle task completion, thus enabling the fleet to continuously approach the goals of high on-time performance and low energy consumption.

[0105] Please refer to Figure 4 , Figure 4 This application provides a first structural schematic diagram of a fleet task allocation device, which is configured in a vehicle agent within a fleet management system; the device includes: The task information receiving module 11 is used to receive transportation task information broadcast by the fleet intelligent agent; wherein, the transportation task information includes at least the destination and weight of the transportation task. The status data acquisition module 12 is used to acquire real-time vehicle status data in response to transportation task information. The bidding decision module 13 is used to determine the bidding results for the transportation task based on the transportation task information and vehicle status data; wherein, the bidding results include a first result indicating abandoning the execution of the transportation task, and a second result indicating applying to execute the transportation task; The bidding result upload module 14 is used to upload the bidding results to the fleet agent so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

[0106] For a detailed description of the fleet task allocation device provided in this application, please refer to the embodiments of the above-described fleet task allocation method; this application will not repeat the details here.

[0107] Based on the above embodiments: As an optional embodiment, the device further includes: The communication detection module is used to determine whether the communication with the fleet intelligent agent is interrupted; if it is not interrupted, the task information receiving module 11 is triggered; if it is interrupted, the decision degradation module is triggered. The decision demotion module is used to determine the election results of the vehicle's intelligent agent based on specified election indicators. If the election result indicates that the vehicle has been elected as the leader intelligent agent among the vehicle's intelligent agents, it will perform the corresponding transportation task allocation operation on behalf of the fleet intelligent agents. The specified election indicators include indicators used to characterize available computing resources.

[0108] As an optional implementation, the decision degradation module includes: The priority determination module is used to determine the priority score of the vehicle's intelligent agent based on specified election indicators; wherein, the priority score is used to characterize the priority of the vehicle's intelligent agent among all vehicle intelligent agents. The election message construction module is used to combine the priority score and the unique identifier of the vehicle's intelligent agent to form an election message; The election message broadcasting module is used to broadcast election messages to other vehicle intelligent agents; among them, other vehicle intelligent agents are vehicle intelligent agents in the fleet management system other than this vehicle intelligent agent; The election result determination module is used to determine whether an election response message is received from other vehicle intelligent agents within a preset time period; wherein, the election response message is used to indicate that the priority of other vehicle intelligent agents is higher than the priority of this vehicle intelligent agent; if yes, the election result indicates that it was not elected as the leader intelligent agent; if no, the election result indicates that it was elected as the leader intelligent agent. The demotion execution module is used to perform the corresponding transportation task allocation operation on behalf of the fleet agent when the election result indicates that the agent has been elected as the leader agent among the vehicle agents.

[0109] As an optional embodiment, this vehicle is the lead vehicle in the target convoy; wherein, the target convoy is a convoy composed of various target vehicles; correspondingly, the device further includes: The formation data acquisition module is used to acquire formation decision data when performing transportation tasks; the formation decision data includes at least road condition information of the road to be traveled, environmental perception information, and vehicle status data of the target vehicles. The formation mode determination module is used to determine the target formation mode from the preset formation modes based on formation decision data; wherein the preset formation modes include at least an energy-saving formation mode and a risk-avoidance formation mode; The formation planning module is used to plan the target state sequence of target vehicles according to the target formation mode, so that the target vehicles can drive according to the target state sequence; wherein, the target state sequence includes the target position and target speed at each planning time.

[0110] As an optional embodiment, the device further includes: The anomaly monitoring module is used to acquire anomaly monitoring data for this vehicle; the anomaly monitoring data includes at least vehicle status data and driver status data. The alarm module is used to display an alarm pop-up on the vehicle dashboard when abnormal monitoring data is determined to meet preset abnormal conditions; the alarm pop-up includes a trigger node to instruct the driver to request a rest. The rest request module is used to send a rest request to the fleet agent when a trigger node is determined to be triggered, so that the fleet agent can stop broadcasting new transportation task information to its own vehicle.

[0111] Please refer to Figure 5 , Figure 5 This application provides a second structural schematic diagram of a fleet task allocation device, which is configured in a fleet agent within a fleet management system; the device includes: The task information broadcasting module 21 is used to acquire transportation task information and broadcast the transportation task information to each vehicle intelligent agent; wherein, the transportation task information includes at least the destination and task weight of the transportation task; The bidding result receiving module 22 is used to receive the bidding results reported by each vehicle intelligent agent; wherein, the bidding result includes a first result indicating that the vehicle intelligent agent abandons the execution of the transportation task, and a second result indicating that the vehicle intelligent agent applies to execute the transportation task; The task allocation module 23 is used to determine the target vehicles to perform transportation tasks from the fleet based on the results of each bid.

[0112] For a detailed description of the fleet task allocation device provided in this application, please refer to the embodiments of the above-described fleet task allocation method; this application will not repeat the details here.

[0113] Based on the above embodiments: As an optional embodiment, the second result also includes the bid price and estimated arrival time; the task allocation module 23 includes: The candidate vehicle determination module is used to identify the vehicles that correspond to the second result among all vehicles as candidate vehicles. The allocation score determination module is used to perform a weighted summation of the allocation score indicators of candidate vehicles using preset score weights to obtain the allocation score of the candidate vehicles; wherein, the allocation score indicators include at least the bid price and the estimated arrival time; the preset score weights include a first preset weight corresponding to the bid price and a second preset weight corresponding to the estimated arrival time. The target vehicle determination module is used to assign the vehicle with the highest score among the candidate vehicles as the target vehicle.

[0114] As an optional embodiment, the device further includes: The first acquisition module is used to acquire the historical estimated arrival time and historical actual arrival time of each vehicle when it performed historical transportation tasks through each vehicle intelligent agent. The on-time task determination module is used to identify tasks in the history of transportation tasks whose actual arrival time is no later than the historical estimated arrival time as on-time delivery tasks. The on-time delivery ratio determination module is used to determine the proportion of on-time delivery tasks to historical transportation tasks as the on-time delivery ratio. The weighting adjustment module is used to adjust the preset scoring weights based on the on-time delivery ratio.

[0115] As an optional embodiment, the device further includes: The second acquisition module is used to acquire, through each vehicle intelligent agent, the historical estimated arrival time, historical actual arrival time, and task execution data of each vehicle when performing historical transportation tasks; wherein, the task execution data includes at least one of vehicle status data, driver status data, and road condition information; The clustering module is used to cluster vehicles based on task execution data to obtain vehicle groups; The local proportion determination module is used to determine the local on-time delivery proportion of the vehicle group; where the local on-time delivery proportion is the proportion of historical transportation tasks delivered on time by the vehicle group to all historical transportation tasks delivered by the vehicle group. The abnormal vehicle group identification module is used to identify the vehicle group with the lowest local on-time delivery rate among all vehicle groups as the abnormal vehicle group. The anomaly analysis module is used to analyze the task execution data of abnormal vehicle groups, determine the reasons for the delayed delivery of transportation tasks by abnormal vehicle groups, and generate transportation task allocation suggestions based on the reasons.

[0116] Please refer to Figure 6 , Figure 6 This application provides a schematic diagram of the structure of an electronic device, which includes: Memory 61 is used to store computer programs; Processor 62 is used to implement the steps of any of the above-described vehicle task allocation methods when executing a computer program.

[0117] The aforementioned memory 61 includes all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROMs and DVD-ROMs. The processor and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0118] In some embodiments, the electronic device may further include a display screen, input / output interfaces, communication interfaces, a power supply, and a communication bus, etc. Those skilled in the art will understand that... Figure 6 The structures shown do not constitute a limitation on the electronic device and may include more components than those shown.

[0119] For a detailed description of the electronic equipment provided in this application, please refer to the embodiments of the above-described fleet task allocation method; this application will not repeat the details here.

[0120] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily intended to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation thereof.

[0121] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0122] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for assigning tasks to a fleet of vehicles, characterized in that, The method is applied to vehicle agents in a fleet management system; wherein the fleet management system includes fleet agents and vehicle agents corresponding one-to-one with vehicles in the fleet; the method includes: Receive transportation task information broadcast by the fleet agent; wherein the transportation task information includes at least the destination and weight of the transportation task; In response to the transportation task information, obtain the vehicle's real-time status data; Based on the transportation task information and the vehicle status data, the bidding results for the transportation task are determined; wherein, the bidding results include a first result indicating that the transportation task is abandoned, and a second result indicating that the transportation task is applied for. The bidding results are uploaded to the fleet agent so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

2. The fleet task allocation method as described in claim 1, characterized in that, The method further includes: Determine whether communication with the fleet agent has been interrupted; If there is no interruption, proceed to the step of receiving the transportation task information broadcast by the fleet agent; If interrupted, the election result of this vehicle agent is determined based on the specified election criteria; if the election result indicates that it has been elected as the leader agent among the vehicle agents, then it performs the corresponding transportation task allocation operation in place of the fleet agent; wherein, the specified election criteria include indicators used to characterize available computing resources.

3. The fleet task allocation method as described in claim 2, characterized in that, The process of determining the election results of the vehicle's intelligent agent based on specified election indicators includes: Based on the specified election criteria, a priority score for the vehicle agent is determined; wherein the priority score is used to characterize the priority of the vehicle agent among all vehicle agents. The priority score and the unique identifier of the vehicle's intelligent agent are combined to form an election message; The election message is broadcast to other vehicle agents; wherein, the other vehicle agents are vehicle agents in the fleet management system other than the vehicle agent itself. Determine whether an election response message is received from the other vehicle intelligent agent within a preset time period; wherein the election response message is used to indicate that the priority of the other vehicle intelligent agent is higher than the priority of the current vehicle intelligent agent; If so, then the election result indicates that the candidate was not elected as the leading agent. If not, then the election result indicates that the candidate has been elected as the leading agent.

4. The fleet task allocation method as described in claim 1, characterized in that, The vehicle in question is the lead vehicle in the target convoy; wherein the target convoy is a convoy composed of the target vehicles; correspondingly, the method further includes: When performing the transportation task, formation decision data is acquired; wherein, the formation decision data includes at least road condition information of the road to be traveled, environmental perception information, and vehicle status data of the target vehicle; Based on the formation decision data, a target formation mode is determined from the preset formation modes; wherein, the preset formation modes include at least an energy-saving formation mode and a risk-avoidance formation mode; According to the target formation mode, the target state sequence of the target vehicles is planned so that the target vehicles drive according to the target state sequence; wherein, the target state sequence includes the target position and target speed corresponding to each planning time.

5. The fleet task allocation method as described in claim 1, characterized in that, The method further includes: Obtain the abnormal monitoring data of the vehicle; wherein the abnormal monitoring data includes at least the vehicle status data and the driver status data; When the abnormal monitoring data is determined to meet the preset abnormal conditions, an alarm pop-up is displayed on the vehicle dashboard of the vehicle; wherein, the alarm pop-up includes a trigger node for instructing the driver to request a rest. When it is determined that the trigger node has been triggered, a rest request is sent to the fleet agent so that the fleet agent can stop broadcasting new transportation task information to the vehicle.

6. A method for assigning tasks to a fleet of vehicles, characterized in that, The method is applied to fleet agents in a fleet management system; wherein, the fleet management system further includes vehicle agents corresponding one-to-one with the vehicles in the fleet; the method includes: Acquire transportation task information and broadcast the transportation task information to each of the vehicle intelligent agents; wherein, the transportation task information includes at least the destination and task weight of the transportation task; Receive bidding results reported by each of the vehicle intelligent agents; wherein, the bidding results include a first result indicating that the vehicle intelligent agent abandons the execution of the transportation task, and a second result indicating that the vehicle intelligent agent applies to execute the transportation task; Based on the bidding results, target vehicles for performing the transportation tasks are determined from the fleet.

7. The fleet task allocation method as described in claim 6, characterized in that, The second result also includes bid prices and estimated arrival times; based on each of the bid results, the target vehicle to perform the transportation task is determined from the fleet, including: The vehicle that corresponds to the second result among all the vehicles is selected as a candidate vehicle; The allocation scores of the candidate vehicles are obtained by weighted summation of the allocation scores of the candidate vehicles using preset scoring weights; wherein the allocation scoring indicators include at least the bid price and the estimated arrival time; the preset scoring weights include a first preset weight corresponding to the bid price and a second preset weight corresponding to the estimated arrival time. The vehicle with the highest assigned score among the candidate vehicles is selected as the target vehicle.

8. The fleet task allocation method as described in claim 7, characterized in that, The method further includes: Each vehicle intelligent agent obtains the historical estimated arrival time and historical actual arrival time of each vehicle when performing historical transportation tasks. The historical transportation tasks in which the actual arrival time is no later than the estimated arrival time are considered as on-time delivery tasks. The proportion of the on-time delivery tasks to the total number of historical transportation tasks is taken as the on-time delivery ratio; The preset scoring weights are adjusted based on the on-time delivery ratio.

9. The fleet task allocation method as described in claim 6, characterized in that, The method further includes: Each vehicle intelligent agent acquires the historical estimated arrival time, historical actual arrival time, and task execution data of each vehicle when performing historical transportation tasks; wherein, the task execution data includes at least one of vehicle status data, driver status data, and road condition information; Based on the task execution data, the vehicles are clustered to obtain vehicle groups. Determine the partial on-time delivery ratio of the vehicle group; wherein the partial on-time delivery ratio is the proportion of historical transportation tasks delivered on time by the vehicle group to all historical transportation tasks delivered by the vehicle group. The vehicle group with the lowest local on-time delivery rate among all the vehicle groups is designated as the abnormal vehicle group. The task execution data of the abnormal vehicle group is analyzed to determine the reasons for the delayed delivery of transportation tasks, and transportation task allocation suggestions are generated based on the reasons.

10. A fleet management system, characterized in that, The system includes a fleet agent and vehicle agents that correspond one-to-one with the vehicles in the fleet. The vehicle agent is used to execute the fleet task allocation method according to any one of claims 1 to 5; The fleet agent is used to execute the fleet task allocation method according to any one of claims 6 to 9.

11. The fleet management system as described in claim 10, characterized in that, The system also includes vehicle dashboards corresponding one-to-one with the vehicle intelligent agents, and fleet dashboards corresponding to the fleet intelligent agents; The vehicle dashboard is used to display the vehicle-level monitoring data of the vehicle; wherein, the vehicle-level monitoring data includes at least vehicle status data; The fleet dashboard is used to display fleet-level monitoring data; wherein, the fleet-level monitoring data includes at least fleet operation indicators determined based on the vehicle-level monitoring data.

12. A fleet task allocation device, characterized in that, The device is configured in a vehicle agent within a fleet management system; wherein the fleet management system includes fleet agents and vehicle agents corresponding one-to-one with the vehicles in the fleet; the device includes: The task information receiving module is used to receive transportation task information broadcast by the fleet intelligent agent; wherein, the transportation task information includes at least the destination and weight of the transportation task; The status data acquisition module is used to acquire real-time vehicle status data in response to the transportation task information. The bidding decision module is used to determine the bidding result for the transportation task based on the transportation task information and the vehicle status data; wherein the bidding result includes a first result indicating that the transportation task is abandoned, and a second result indicating that the transportation task is applied for. The bidding result uploading module is used to upload the bidding results to the fleet agent, so that the fleet agent can determine the target vehicle to perform the transportation task from the fleet based on the bidding results reported by each vehicle agent.

13. A fleet task allocation device, characterized in that, The device is configured in a fleet management system's fleet agent; wherein, the fleet management system also includes vehicle agents corresponding one-to-one with the vehicles in the fleet; the device includes: The task information broadcasting module is used to acquire transportation task information and broadcast the transportation task information to each of the vehicle intelligent agents; wherein, the transportation task information includes at least the destination and task weight of the transportation task; The bidding result receiving module is used to receive the bidding results reported by each of the vehicle intelligent agents; wherein, the bidding result includes a first result indicating that the vehicle intelligent agent abandons the execution of the transportation task, and a second result indicating that the vehicle intelligent agent applies to execute the transportation task; The task allocation module is used to determine the target vehicle to perform the transportation task from the fleet based on the bidding results.

14. An electronic device, characterized in that, The electronic device includes: Memory, used to store computer programs; A processor, configured to implement the steps of the fleet task allocation method as described in any one of claims 1 to 5 or 6 to 9 when executing the computer program.

Images