Task adaptive adjustment method and device of autonomous driving truck and computer equipment

Abstract

The application relates to a task self-adaptive adjustment method, device and computer equipment of an automatic driving truck. The method comprises the following steps: obtaining a port transportation order; wherein the port transportation order comprises a transportation box type; identifying the combined port transportation order according to the transportation box type, and combining the port transportation order; generating a corresponding to-be-executed task for the combined port transportation order, directly generating a to-be-executed task for the uncombined port transportation order, and obtaining a to-be-executed task list; in the process of executing each to-be-executed task in the to-be-executed task list, receiving an actual box type of a container corresponding to each to-be-executed task sensed in real time; in response to the fact that the transportation box type is inconsistent with the actual box type, dynamically correcting the to-be-executed task list, and adjusting the execution sequence of each to-be-executed task. The method can improve the task execution efficiency of the automatic driving truck for port transportation.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a method, apparatus, and computer device for adaptive task adjustment of an autonomous truck. Background Technology

[0002] With the development of autonomous driving technology, autonomous trucks (AT) have emerged and are widely used in the construction of smart ports.

[0003] Currently, automated trucks in ports typically operate on a fixed single-container mission mode, transporting only one container at a time. When faced with a large demand for transporting 20-foot containers, this mode results in a high empty load rate for vehicles and significant waste of transport capacity. Furthermore, the traditional approach heavily relies on complete and accurate pre-set order information provided by the Terminal Operating System (TOS), which can easily lead to mission failures, all of which limit overall transportation efficiency. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, and computer equipment that can improve the transportation efficiency of autonomous trucks for ports, addressing the aforementioned technical problems.

[0005] Firstly, a method for adaptive task adjustment of an autonomous truck is provided, the method comprising: Obtain port transportation orders; these orders include the type of container being transported. Port shipping orders can be merged based on container type identification; For merged port transportation orders, corresponding tasks to be executed are generated; for unmerged port transportation orders, tasks to be executed are generated directly, resulting in a list of tasks to be executed. During the execution of each task in the task list, the actual container type of each task is received in real time. In response to discrepancies between the transport container type and the actual container type, the list of tasks to be executed is dynamically revised, and the execution order of each task is adjusted.

[0006] In some embodiments, identifying mergeable port shipping orders based on container type includes: Determine whether the container type recorded in two consecutive port transport orders is the first container type; If so, then the two consecutive port transportation orders will be identified as mergeable port transportation orders.

[0007] In some embodiments, before each task in the task list is executed, the method further includes: The execution order of each task in the task list is optimized and adjusted.

[0008] In some embodiments, the execution order of each task in the task list to be executed is optimized and adjusted, including: If the list of tasks to be executed includes both tasks of type 1 and tasks of type 2, the execution order of the tasks of type 2 will be adjusted to take precedence over the tasks of type 1.

[0009] In some embodiments, the port transportation order also includes delivery time information, and optimizes the execution order of each task in the task list, including: Obtain the current location information of the container truck, and calculate the transportation distance and loading waiting time for each task to be executed based on the current location information of the container truck; The predicted delivery time is calculated based on the transportation distance and loading waiting time, and the execution order of tasks with predicted delivery times greater than the actual delivery time is advanced.

[0010] In some embodiments, the execution order of each task in the task list to be executed is optimized and adjusted, including: Obtain trailer structure information; Determine the operational phase and target loading location of each task in the task list; the operational phase includes either the loading phase or the unloading phase. Based on the trailer structure information, the execution order of each task to be performed is optimized and adjusted by applying preset loading and unloading sequence control rules, wherein the loading and unloading sequence control rules are adapted to the trailer structure information.

[0011] In some embodiments, before each task in the task list is executed, the method further includes: Acquire trailer rated load information, preset loading and unloading sequence control rules, and port operation space information; Feasibility verification was conducted on the list of tasks to be performed based on the trailer's rated load capacity information, the preset loading and unloading sequence control rules, and the port's operational space information. In response to the feasibility verification of the task list, the steps for executing each task in the task list are then initiated. In response to the failure of the feasibility verification of the pending task list, the execution order of each pending task in the pending task list will be split or optimized and adjusted for the merged port transportation orders.

[0012] In some embodiments, in response to discrepancies between the transport container type and the actual container type, the list of tasks to be executed is dynamically revised, and the execution order of each task to be executed is adjusted, including: Generate new tasks to be executed and a new list of tasks to be executed based on the actual box type; Calculate the transportation distance and loading waiting time for each new task to be executed, and determine the total transportation distance and loading waiting time for all new tasks to be executed. The execution order of each new task in the new task list is dynamically adjusted to minimize the total cost of the total transportation distance and the total loading waiting time.

[0013] In some embodiments, during the execution of each task in the task list, the method further includes: When the task to be performed is to transport a single first container type, the actual loading position of the first container type container in the trailer is received in real time. The additional offset value is determined based on the actual loading position and fixed reference parameters; The docking position of the autonomous truck is dynamically determined based on the additional offset value.

[0014] Secondly, a task adaptive adjustment device for an autonomous truck is provided, the device comprising: The order acquisition module is used to acquire port transportation orders; these orders include the type of transport container. The order merging module is used to identify mergeable port transportation orders based on the type of shipping container and then merge them. The task generation module is used to generate corresponding tasks to be executed for merged port transportation orders, and to directly generate tasks to be executed for unmerged port transportation orders, thus obtaining a list of tasks to be executed. The real-time sensing module is used to receive the actual container type of each task in the task list during the execution of each task in the real-time sensing. The sequence adjustment module is used to dynamically correct the list of tasks to be executed and adjust the execution order of each task in response to inconsistencies between the transport container type and the actual container type.

[0015] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of a method for adaptively adjusting the task of an autonomous truck are described when the processor executes the computer program.

[0016] The aforementioned autonomous driving truck task adaptive adjustment method, device, and computer equipment, targeting mixed container transportation scenarios oriented towards ports, generate composite tasks by identifying and merging consecutive first-type container orders based on the container type, breaking through the limitations of the fixed single-container mode and effectively reducing the vehicle's empty load rate. In addition, by introducing a real-time perception mechanism during execution, the perceived actual container type is compared with the transport container type recorded in the order, and the list of tasks to be executed is dynamically adjusted accordingly. For example, the order merging strategy is corrected, and the execution order of the tasks to be executed is dynamically adjusted, thereby reducing the vehicle's empty load rate and improving task execution efficiency during real-time dynamic transportation task execution. Attached Figure Description

[0017] Figure 1 This is an application environment diagram of the task adaptive adjustment method for autonomous trucks in some embodiments; Figure 2 This is a flowchart illustrating the task adaptive adjustment method for autonomous trucks in some embodiments; Figure 3 This is a flowchart illustrating the task adaptive adjustment method for autonomous trucks in some application examples. Figure 4 This is a structural block diagram of the task adaptive adjustment device for an autonomous truck in some embodiments; Figure 5 This is a diagram showing the internal structure of a computer device in some embodiments. Detailed Implementation

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

[0019] The task adaptive adjustment method for autonomous trucks provided in this application can be applied to computer equipment, and such as... Figure 1 The application environment shown is illustrated. Specifically, this method can be executed by a processor in a server deployed in a smart port dispatch and control center, a cloud computing platform, or an onboard computing unit of an autonomous truck.

[0020] In some embodiments, the method is illustrated using a server 102 in a smart port dispatch and control center. The server 102 acquires port transportation orders, which include container types. It identifies mergeable port transportation orders based on container types and merges them. For merged port transportation orders, corresponding tasks are generated to be executed; for unmerged port transportation orders, tasks are directly generated to be executed, resulting in a list of tasks to be executed. During the execution of each task in the task list, a real-time perception system installed on the autonomous driving truck 104 continuously senses the actual container type corresponding to each task. In response to discrepancies between the transport container type and the actual container type, the task list is dynamically corrected, and the execution order of each task is adjusted. The server 102 can communicate with the autonomous driving truck 104 via a network.

[0021] According to Figure 1 In practical applications, the autonomous truck 104 can be equipped with a real-time perception system. This system may include, but is not limited to, vision cameras and lidar mounted on the front of the truck, used to actively scan and identify the target container when approaching or aligning with it for a task. Alternatively, cameras and other vision sensors mounted on the rear of the truck can be used to dynamically perceive the relative position of the target container loaded on the trailer in real time. Of course, the number, method, and location of the real-time perception system can be adaptively adjusted according to different vehicle models and task requirements.

[0022] In some embodiments, such as Figure 2 As shown, a method for adaptive task adjustment of autonomous trucks is provided, which is applied to... Figure 1 Taking the server in the example, the following steps are included: Step S202: Obtain port transportation orders; wherein, the port transportation orders include the type of transport container.

[0023] In this context, a port transport order refers to an operational instruction unit generated by the Terminal Operating System (TOS) or upstream business system, requesting the transport of a specific container from one location within the port to another. The container type refers to the standard size specifications of the container requested for transport in the order. In general port scenarios, this can primarily be either a 20-foot container (often referred to as a "small container" or "single small container," corresponding to the first container type in this application) or a 40-foot container (often referred to as a "large container," corresponding to the second container type in this application).

[0024] For example, in practical application scenarios, in addition to recording the type of container to be transported, port transportation orders may also include, but are not limited to, key information necessary for generating a complete transportation task, such as task type, task initiation location (e.g., the starting location of the container to be transported), task target location (e.g., the delivery location of the container to be transported), order priority, and expected delivery time information.

[0025] Specifically, the server receives port transportation orders in real time or periodically from the Terminal Operating System (TOS) or related production management system through a pre-defined data interface. These port transportation orders constitute the initial port transportation order set, denoted as: O = {o1, o2, ..., o} n}

[0026] For each port transport order received i The server can parse and structure the data to extract the key information necessary for adaptive task generation, including at least the container type information Ci.size∈{20ft, 40ft}, and may also include, but is not limited to, the container location Ci.locC, the order priority Ci.priority, and the expected delivery time information Ci.time.

[0027] For example, upon receiving a set of port transportation orders, the server can preprocess these orders. First, the server can initially sort the orders according to their arrival time or a preset order priority. Next, the server can scan the sorted order sequence to identify orders with incomplete or questionable information such as container type or location, initialize a placeholder data entry, and mark the order as "pending confirmation." These "pending confirmation" orders are not simply discarded or have their process blocked; instead, they are allowed to proceed to subsequent stages, waiting for the autonomous truck to perform real-time identification and information completion upon arrival at the site via its onboard real-time perception system.

[0028] Step S204: Identify port transport orders that can be merged based on the type of transport container, and merge them.

[0029] Specifically, during the preprocessing of port transportation orders, the server iterates through the initially sorted port transportation orders and identifies the container type information carried in the port transportation orders. Based on the container type information, it can determine whether two consecutive port transportation orders can be merged. For orders that can be merged, they are marked as mergeable port transportation orders. During the order merging process, orders carrying mergeable port transportation order identifiers are identified and merged. Thus, two consecutive mergeable port transportation orders are merged into one order.

[0030] Step S206: Generate corresponding tasks to be executed for merged port transportation orders, and directly generate tasks to be executed for unmerged port transportation orders, thus obtaining a list of tasks to be executed.

[0031] Among them, the tasks to be executed refer to the specific transportation operations to be performed by the automated guided vehicle (AGV) trucks. The list of tasks to be executed is a sequence of multiple tasks arranged in a certain order (such as a preliminary generation order or a sorting based on preset rules), denoted as T = {t1, t2, ..., t...}. m}

[0032] Specifically, based on the merging results of step S204, the server generates a list of pending tasks, including multiple tasks arranged in sequence. For merged port transportation orders, a unified pending task is generated, containing the transportation needs of two containers. For unmerged individual port transportation orders, separate pending tasks for each container are generated. This list serves as the basis for subsequent optimization adjustments and dynamic corrections.

[0033] Step S208: During the execution of each task in the task list, receive the actual container type of each task in real time.

[0034] Specifically, when an autonomous truck proceeds to the yard to perform loading and unloading operations according to a list of tasks, before executing each task, the truck can actively scan and identify the target container as it approaches or aligns with it using its multimodal real-time perception system. This system can utilize features such as, but is not limited to, vision cameras and LiDAR mounted on the front of the truck. The real-time perception system, by running specific image recognition or point cloud algorithms, can analyze the actual container size (e.g., denoted as Ĉᵢ.size) in real time and transmit this information to a server via a network. This real-time perception behavior of the autonomous truck can be integrated throughout the execution of each task, aiming to establish a field information verification channel independent of the upstream system (TOS). This allows the server to promptly and dynamically detect discrepancies between the port transportation order records and the actual situation on-site.

[0035] Step S210: In response to the discrepancy between the transport container type and the actual container type, dynamically correct the list of tasks to be executed and adjust the execution order of each task to be executed.

[0036] Specifically, the server can compare the actual container type (Ĉᵢ.size) perceived in real time on-site by the autonomous driving truck in step S208 with the transport container type (Cᵢ.size) recorded in the port transport order.

[0037] For example, the discrepancy between the shipping container type and the actual container type can include at least two situations. The first situation is that the size data corresponding to the shipping container type and the actual container type are inconsistent. The second situation is that the port transportation order corresponding to the task to be executed is marked as "pending confirmation", and the value corresponding to its shipping container type is a "placeholder".

[0038] When the server detects an inconsistency (e.g., a 20-foot box is planned but a 40-foot box is actually delivered, or vice versa), it immediately triggers a dynamic correction mechanism. First, it corrects the list of tasks to be executed; for example, it splits a planned combined order into two single-box tasks, or it reassesses the feasibility of merging tasks. Furthermore, based on the updated correct box type, it can replan the optimal transportation route (P(t)) with the goal of minimizing the total transportation distance (Dᵢ) and waiting time (Wᵢ), thereby dynamically planning the optimal execution order of the tasks to be executed.

[0039] The aforementioned adaptive task adjustment method for autonomous trucks, targeting mixed container transportation scenarios oriented towards ports, generates composite tasks by identifying and merging consecutive first-type container orders based on the container type, breaking through the limitations of the fixed single-container mode and effectively reducing the vehicle's empty load rate. Furthermore, by introducing a real-time perception mechanism during execution, the perceived actual container type is compared with the container type recorded in the order, and the list of tasks to be executed is dynamically adjusted accordingly. For example, the order merging strategy is corrected, and the execution order of the tasks to be executed is dynamically adjusted. Thus, in the real-time dynamic transportation task execution process, the vehicle's empty load rate is reduced while the task execution efficiency is improved.

[0040] In some embodiments, identifying mergeable port transport orders based on container type includes: determining whether the container types recorded in two consecutive port transport orders are both the first container type; if so, determining the two consecutive port transport orders as mergeable port transport orders.

[0041] In this embodiment, the server can iterate through the pre-processed port transportation orders and check any two consecutive orders. It determines whether the "container type" information recorded in both orders is "first container type". In the general port scenario of this application, the first container type typically refers to (e.g., 20-foot) standard containers (i.e., "small containers"). If the determination result is "yes", meaning both orders request the transportation of 20-foot small containers, the server can jointly identify these two orders as "mergeable port transportation orders". Subsequently, the server can merge these two marked orders, for example, generating a "trailer task" that carries the transportation of two small containers.

[0042] In this embodiment, by setting two specific judgment conditions, namely "continuous before and after" and "all container types are the first type", the merging basis is clarified. This merging strategy can not only avoid overloading, but also does not disrupt the basic sequence of task execution, reducing the number of empty round trips caused by autonomous trucks performing single container tasks, thereby making full use of the vehicle's potential loading capacity and helping to increase the container volume of a single transport.

[0043] In some embodiments, before each task in the task list is executed, the method further includes: optimizing the execution order of each task in the task list.

[0044] In this embodiment, after generating the initial list of tasks to be executed, but before task execution begins, the server can perform an optimization adjustment. The server can comprehensively consider various preset constraints and optimization objectives to adjust the execution order of all tasks in the list. This embodiment, by adding an optimization adjustment step before actual task execution, allows the server to plan based on a more comprehensive set of tasks, rather than solely relying on basic metrics (e.g., arrival time order) used in the initial ordering of orders. This more efficiently avoids potential job conflicts, thereby further improving task execution efficiency.

[0045] In some embodiments, optimizing the execution order of each task in the task list includes: when the task list includes both a first box-shaped task and a second box-shaped task, adjusting the execution order of the second box-shaped task to take precedence over the first box-shaped task.

[0046] In this embodiment, as a specific implementation of the above-mentioned optimization and adjustment, the server can detect the composition of the task list to be executed. When there are both tasks to be executed for transporting the first type of box (e.g., a 20-foot small box) and tasks to be executed for transporting the second type of box (e.g., a 40-foot large box) in the list, the server can apply the following priority rule: adjust the execution order of the tasks to be executed for transporting the second type of box (large box) to take precedence over the tasks to be executed for transporting the first type of box (small box).

[0047] This adjustment in this embodiment is based on the physical characteristics of the trailer. The 40-foot large container occupies the entire length of the trailer, and its loading and unloading operations have more stringent space requirements. Arranging smaller container tasks before it may cause operational interference or require additional movement and adjustments. Therefore, prioritizing the transportation sequence of large container tasks is to meet the spatial constraints of the trailer structure and ensure smooth and safe loading and unloading operations.

[0048] In some embodiments, the execution order of each task to be executed in the task list is optimized and adjusted, including: obtaining the current truck location information, calculating the transportation distance and loading waiting time of each task to be executed based on the current truck location information; calculating the predicted delivery time based on the transportation distance and loading waiting time, and advancing the execution order of tasks whose predicted delivery time is greater than the delivery time information.

[0049] In this embodiment, in addition to information about the container type, the port transportation order may also include delivery time information (Ci.time), which is the expected time window for the container to reach its destination. When optimizing the execution order, the server can further obtain vehicle-related information about the trucks, such as their current location. Based on the current truck location information, the server can further combine the container location (Ci.loc) in each pending task to dynamically calculate the transportation distance (Di) required to execute the task. Simultaneously, the server can combine the real-time operation queues and status of port equipment (such as quay cranes and yard cranes) to predict the possible loading wait time (Wi) at the loading point in the pending task. Subsequently, the server can combine the transportation distance (e.g., converted to time) and the loading wait time to calculate the predicted delivery time for completing the pending task. The execution order of pending tasks with predicted delivery times later than the delivery time specified in the order (i.e., potentially timed out) is then prioritized in the list.

[0050] This embodiment can proactively respond to the urgency and timeliness requirements of business operations, and improve the on-time delivery rate of orders while meeting the basic execution sequence, thereby improving the reliability of port transportation operations.

[0051] In some embodiments, optimizing the execution order of each task in the task list includes: obtaining trailer structure information; determining the operation stage and target loading position of the container corresponding to each task in the task list; wherein the operation stage includes a loading stage or an unloading stage; and optimizing the execution order of each task based on the trailer structure information by applying a preset loading and unloading sequence control rule, wherein the loading and unloading sequence control rule is adapted to the trailer structure information.

[0052] In this embodiment, the server can further obtain the trailer structure information (V.trailer) of the autonomous driving truck. The trailer structure information defines the physical configuration, load-bearing surface layout, and connection relationship with the preceding vehicle. For each item in the list of tasks to be executed, the server can determine its corresponding operation stage (whether it is to go to the yard for "loading" or to the quay crane / yard crane for "unloading"), and determine the target loading position of the container on the trailer in that task (for example, for a twin-trailer task, it can represent whether the two small containers are located at the "front" or "rear" of the trailer; for a single-container task, it can represent the position it occupies alone).

[0053] Subsequently, the server can optimize and adjust the execution order of various tasks in the task list based on the acquired trailer structure information and by applying preset loading and unloading sequence control rules adapted to the trailer's physical characteristics. The core logic of these loading and unloading sequence control rules is universal, ensuring the continuity and safety of loading and unloading operations. For tasks awaiting loading, the execution order can be prioritized according to the target loading position of the containers on the trailer, from front to back. That is, the task of loading the "front container" is executed first, followed by the task of loading the "rear container".

[0054] For tasks awaiting execution during the unloading phase, the execution order should be prioritized according to the target loading position of the containers on the trailer, from back to front. That is, the task of unloading the "rear container" should be executed first, followed by the task of unloading the "front container".

[0055] It should be noted that the above loading and unloading sequence control rules are only one example. The specific loading and unloading sequence control rules are not limited to this. They can be adapted to different port operating procedures and loading and unloading process standards, depending on different ports, trailer structures, and operational requirements.

[0056] In this embodiment, by explicitly using trailer structure information as optimization input and combining the operation stage with the target container loading location, a task execution sequence highly compatible with the trailer structure can be dynamically generated. This effectively adapts to the operational characteristics of port cranes and other loading and unloading equipment and the spatial layout of the trailer, thereby ensuring smooth and safe loading and unloading operations and improving task execution efficiency.

[0057] In some embodiments, before each task in the task list is executed, the method further includes: obtaining trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; performing feasibility verification on the task list based on the trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; in response to the task list passing the feasibility verification, proceeding to the step of executing each task in the task list; and in response to the task list failing the feasibility verification, splitting or optimizing the execution order of each task in the task list for merged port transportation orders.

[0058] In this embodiment, after generating the list of tasks to be executed, and before the tasks begin execution, the server can perform multi-dimensional feasibility verification to ensure that the generated plan is physically and operationally feasible, thereby avoiding inefficiency caused by unreliable plans. The steps of this multi-dimensional feasibility verification can specifically include: 1. Obtain verification parameters.

[0059] The server can obtain three core parameter information: Trailer rated load capacity (V.capacity): This is the maximum safe load that the trailer can withstand.

[0060] Preset loading and unloading sequence control rules: such as the "front loading and rear unloading" general operation logic adapted to the trailer structure, as mentioned above.

[0061] Port operation space information includes physical space constraints such as the width of the yard passageway, turning radius, operating radius of the target quay crane, and safety distance from adjacent equipment.

[0062] 2. Conduct feasibility verification.

[0063] Based on the above parameter information, the server can perform item-by-item or overall verification of the list of tasks to be executed, which may include, but is not limited to: Load verification: Calculate the total weight of all containers involved in the tasks to be performed in the list, and verify whether it exceeds the trailer's rated load capacity.

[0064] Sequence and Space Verification: Combining loading and unloading sequence control rules, the task execution process is simulated to verify whether the planned loading / unloading sequence will prevent the operation from being completed within the constraints of port operating space due to trailer structure or container placement.

[0065] 3. Verification result processing.

[0066] The server can adopt different branch processing logic based on the verification result: Once the list of tasks to be executed passes the feasibility verification, the plan is confirmed to be safe and feasible. The server then initiates the tasks, allowing the autonomous truck to enter the subsequent actual task execution process.

[0067] If the list of tasks to be executed fails the feasibility verification in any dimension, dynamic adjustment measures will be triggered immediately. Adjustment measures include, but are not limited to: splitting merged port transportation orders (e.g., splitting an overweight "trailer task" into two single-container tasks), or re-optimizing the execution order of each task in the list of tasks to be executed (e.g., adjusting the loading order of large and small containers to adapt to space constraints), until a new list of tasks that can pass the feasibility verification is generated.

[0068] It is worth noting that, when optimizing and adjusting the execution order and verifying feasibility, the information used, such as the current location of the truck, trailer structure information, trailer rated load information, and port operating space information, can all be collected by the real-time perception system loaded on the autonomous truck during the task execution process and transmitted to the server so that the server can dynamically optimize and adjust the execution order of the tasks to be executed.

[0069] In some embodiments, in response to inconsistencies between the transport container type and the actual container type, the list of tasks to be executed is dynamically modified, and the execution order of each task to be executed is adjusted, including: generating new tasks to be executed and a new list of tasks to be executed based on the actual container type; calculating the transport distance and loading waiting time for each new task to be executed, and determining the total transport distance and total loading waiting time for all new tasks to be executed; and dynamically adjusting the execution order of each new task to be executed in the new list of tasks to be executed, so as to minimize the overall cost of the total transport distance and the total loading waiting time.

[0070] In this embodiment, when the server detects in real-time that the preset transport container type (Cᵢ.size) of one or more tasks is inconsistent with the actual container type (Ĉᵢ.size), the following steps can be performed to dynamically adjust the execution order of the tasks in the task list: The server can correct the affected list of pending tasks based on the updated and correct actual container size (Ĉᵢ.size). For example, this may include splitting the original merged tasks, changing the task type (e.g., from a twin-cargo task to a single-container task), or re-identifying and merging any mergeable port transport orders, thereby generating a new list of pending tasks. The generation, preprocessing, and optimization of the new list of pending tasks can refer to the corresponding methods for processing the initial list of pending tasks described in the embodiments above.

[0071] For each task to be executed in the new list, the server can calculate its estimated transport distance (Dᵢ) and predicted loading wait time (Wᵢ).

[0072] Then, the server uses an objective function to minimize the combined cost of the total transport distance (ΣDᵢ) and the total waiting time (ΣWᵢ) of all new tasks to be executed.

[0073] For example, the objective function can be: Here, α and β are weighting coefficients used to balance distance cost and time cost.

[0074] The server can dynamically adjust the execution order of each task in the new list of tasks to be executed based on optimization algorithms (such as heuristic search, genetic algorithms, etc.), continuously calculate the comprehensive cost value under different orders, and finally select the execution order that minimizes the cost value as the final adjustment scheme.

[0075] This embodiment aims to minimize the overall cost of transportation distance and loading waiting time. Based on dynamically correcting information deviations, it dynamically calculates the execution sequence with the best global efficiency based on the corrected data. This improves the adaptability and operational efficiency of port transportation in uncertain port environments where environmental factors such as truck location, port conditions, trailer structure, and container type change dynamically or where information is missing.

[0076] In some embodiments, during the execution of each task in the task list, the method further includes: when the task to be executed is a task of transporting a single first container type, receiving real-time sensing of the actual loading position of the first container type container in the trailer; determining an additional offset value based on the actual loading position and fixed reference parameters; and dynamically determining the docking position of the autonomous driving truck based on the additional offset value.

[0077] In some embodiments, when the task to be performed is to transport a single first-type box, to address the practical problem that the loading position of the box on the trailer is not fixed (it may be located at the front, rear, or shifted due to manual operation, bumps, etc.), this embodiment can also add a precise position adaptive mechanism during execution. Specific steps are as follows: Real-time position awareness: After the container is loaded or during transportation, the autonomous truck uses its onboard real-time perception system to perceive the actual loading position (Ĉᵢ.pos) of the first container type in the trailer. This actual loading position can indicate not only whether it is located in the "FrontSlot" or "RearSlot", but also the horizontal offset (Offset_x).

[0078] Furthermore, the server can perform calculations based on the actual loading position perceived by the autonomous truck and fixed reference parameters. These fixed reference parameters may include, but are not limited to, the tractor head length (L_head), the standard geometry of the trailer (V.trailer_geom), and a preset safety margin (SafetyMargin). The server can apply a preset calculation model to determine an additional offset value (Δs) to compensate for positional deviations. For example, if the actual loading position of the small box is at the rear of the trailer, Δs will be larger; if there is a lateral offset, Δs can also include a corresponding correction component.

[0079] The stopping position (p_target) of an automated guided vehicle (AGV) truck for loading and unloading operations can be determined by superimposing a calculated additional offset value (Δs) onto a reference stopping point (p_ref). This dynamically and accurately determines the actual stopping position of the AAV truck for this task, i.e., p_target = p_ref + Δs. The reference stopping point (p_ref) is usually determined based on the standard operating point of the target loading and unloading equipment (such as a quay crane).

[0080] In this embodiment, by sensing the actual loading position of the container in real time and dynamically calculating the matching parking position, the risk of misalignment of loading and unloading equipment (such as spreaders), repeated manual fine-tuning, or even operation interruption caused by this is eliminated. Furthermore, in the scenario of single small container (single first container type) transportation, the adaptive capability and efficiency of automated operation are improved.

[0081] In some embodiments, reference may be made to Figure 3 As shown, Figure 3 The diagram illustrates a flowchart of a task adaptive adjustment method for autonomous trucks in some application examples. Specifically, it may include the following steps: S301: Obtain the collection of port transportation orders.

[0082] S302: Preprocess multiple port transportation orders to obtain the port transportation orders after initial sorting.

[0083] S303: Port transport orders with missing information should be marked as "pending confirmation".

[0084] S304: Traverse the port transportation orders after the initial sorting, identify two consecutive orders where the "transport container type" information is both "first container type", determine the two consecutive orders as "mergeable port transportation orders", and merge them.

[0085] S305: Generate a list of tasks to be executed based on the order merging results, which includes multiple tasks to be executed in a certain execution order.

[0086] S306: Optimize and adjust the execution order of each task in the task list based on the transport container type, delivery time information, truck location information, trailer structure information, and operation stage.

[0087] S307: Perform feasibility verification on the optimized list of tasks to be executed.

[0088] Determine whether the verification passed. If yes, proceed to step S308; otherwise, proceed to step S309.

[0089] S308: Instructs autonomous trucks to execute tasks based on a list of pending tasks after feasibility verification has been passed.

[0090] S309: Split or optimize the execution order of each task in the list of tasks to be executed until the feasibility is verified.

[0091] S310: When the actual container type is received in real time through the on-board real-time perception system of the autonomous driving truck, the transport container type recorded in the corresponding order is compared with the actual container type.

[0092] Determine whether the two are consistent. If yes, proceed to step S311; otherwise, proceed to step S312.

[0093] S311: Continue executing the task, and when encountering a single first box type task, adaptively adjust the position of the parking point.

[0094] S312: Generate new tasks to be executed and a new list of tasks to be executed based on the actual box type.

[0095] S313: Calculate the total transportation distance and total loading waiting time of all new tasks to be executed, and dynamically adjust the execution order of each new task to be executed so that the total cost of the total transportation distance and total waiting time is minimized.

[0096] It should be understood that, although Figure 2 and Figure 3 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 and Figure 3At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0097] In some embodiments, such as Figure 4 As shown, an adaptive task adjustment device for an autonomous truck is provided, comprising: an order acquisition module 410, an order merging module 420, a task generation module 430, a real-time perception module 440, and a sequence adjustment module 450, wherein: The order acquisition module 410 is used to acquire port transportation orders; wherein, the port transportation orders include the type of transport container; The order merging module 420 is used to identify mergeable port transportation orders based on the type of transport container and to merge them. The task generation module 430 is used to generate corresponding tasks to be executed for merged port transportation orders and to directly generate tasks to be executed for unmerged port transportation orders, thus obtaining a list of tasks to be executed. The real-time sensing module 440 is used to receive the actual container type of each of the tasks to be executed in real-time during the execution of each of the tasks to be executed in the task list. The sequence adjustment module 450 is used to dynamically correct the list of tasks to be executed and adjust the execution order of each task to be executed in response to the inconsistency between the transport container type and the actual container type.

[0098] In some embodiments, the order merging module 420 is used to determine whether the container types recorded in two consecutive port transport orders are both the first container type; if so, the two consecutive port transport orders are determined as mergeable port transport orders.

[0099] In some embodiments, the sequence adjustment module 450 is further configured to optimize the execution order of each task in the task list.

[0100] In some embodiments, the order adjustment module 450 is further configured to adjust the execution order of the second box-type task to take precedence over the first box-type task when the task to be executed list includes both the first box-type task to be executed and the second box-type task to be executed.

[0101] In some embodiments, the sequence adjustment module 450 is further configured to obtain the current truck location information, calculate the transportation distance and loading waiting time of each task to be executed based on the current truck location information, calculate the predicted delivery time based on the transportation distance and loading waiting time, and advance the execution order of tasks to be executed whose predicted delivery time is greater than the delivery time information.

[0102] In some embodiments, the sequence adjustment module 450 is further configured to acquire trailer structure information; determine the operation stage corresponding to each task to be executed in the task list and the target loading position of the container; wherein the operation stage includes a loading stage or an unloading stage; and optimize and adjust the execution order of each task to be executed by applying a preset loading and unloading sequence control rule according to the trailer structure information, wherein the loading and unloading sequence control rule is adapted to the trailer structure information.

[0103] In some embodiments, the sequence adjustment module 450 is further configured to acquire trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; perform feasibility verification on the list of tasks to be executed based on the trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; in response to the list of tasks to be executed passing the feasibility verification, proceed to the step of executing each task in the list of tasks to be executed; in response to the list of tasks to be executed failing the feasibility verification, split or optimize the execution order of each task in the list of tasks to be executed for the merged port transportation order.

[0104] In some embodiments, the sequence adjustment module 450 is further configured to generate new tasks to be executed and a new list of tasks to be executed based on the actual container type; calculate the transportation distance and loading waiting time of each new task to be executed, and determine the total transportation distance and loading waiting time of all new tasks to be executed; dynamically adjust the execution order of each new task to be executed in the new list of tasks to be executed, so as to minimize the overall cost of the total transportation distance and the total loading waiting time.

[0105] The real-time perception module 440 is also used to receive the actual loading position of the container of the first type of container in the trailer in real time when the task to be performed is to transport a single first type of container; determine the additional offset value according to the actual loading position and fixed reference parameters; and dynamically determine the parking position of the autonomous driving truck according to the additional offset value.

[0106] Specific limitations regarding the task adaptive adjustment device for autonomous trucks can be found in the above-mentioned limitations on the task adaptive adjustment method for autonomous trucks, and will not be repeated here. Each module in the aforementioned task adaptive adjustment device for autonomous trucks can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0107] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a task adaptive adjustment method for autonomous driving trucks.

[0108] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0109] In some embodiments, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: obtaining port transportation orders; wherein the port transportation orders include container types; identifying mergeable port transportation orders based on the container types and merging them; generating corresponding tasks to be executed for merged port transportation orders and directly generating tasks to be executed for unmerged port transportation orders, thereby obtaining a list of tasks to be executed; during the execution of each task to be executed in the list of tasks to be executed, receiving the actual container type of each task to be executed in real time; and dynamically correcting the list of tasks to be executed and adjusting the execution order of each task to be executed in response to inconsistencies between the container types and the actual container types.

[0110] In some embodiments, when the processor executes the computer program, it further implements the following steps: determining whether the container types recorded in two consecutive port transport orders are both the first container type; if so, determining the two consecutive port transport orders as mergeable port transport orders.

[0111] In some embodiments, when the processor executes a computer program, it further performs the following steps: optimizing the execution order of each task in the task list to be executed.

[0112] In some embodiments, when the processor executes a computer program, it further implements the following steps: when the list of tasks to be executed includes both a first box-type task to be executed and a second box-type task to be executed, the execution order of the second box-type task to be executed is adjusted to take precedence over the first box-type task to be executed.

[0113] In some embodiments, when the processor executes the computer program, it further implements the following steps: obtaining the current truck location information, calculating the transportation distance and loading waiting time of each task to be executed based on the current truck location information; calculating the predicted delivery time based on the transportation distance and loading waiting time, and advancing the execution order of tasks to be executed whose predicted delivery time is greater than the delivery time information.

[0114] In some embodiments, when the processor executes the computer program, it further implements the following steps: obtaining trailer structure information; determining the operation stage corresponding to each task to be executed in the task list and the target loading position of the container; wherein the operation stage includes a loading stage or an unloading stage; and optimizing and adjusting the execution order of each task to be executed by applying a preset loading and unloading sequence control rule according to the trailer structure information, wherein the loading and unloading sequence control rule is adapted to the trailer structure information.

[0115] In some embodiments, when the processor executes the computer program, it further implements the following steps: acquiring trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; performing feasibility verification on the list of tasks to be executed based on the trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; in response to the list of tasks to be executed passing the feasibility verification, proceeding to the step of executing each task in the list of tasks to be executed; and in response to the list of tasks to be executed failing the feasibility verification, splitting or optimizing the execution order of each task in the list of tasks to be executed for the merged port transportation order.

[0116] In some embodiments, when the processor executes the computer program, it further performs the following steps: generating new tasks to be executed and a new list of tasks to be executed based on the actual box type; calculating the transportation distance and loading waiting time of each new task to be executed, and determining the total transportation distance and total loading waiting time of all new tasks to be executed; dynamically adjusting the execution order of each new task to be executed in the new list of tasks to be executed, so as to minimize the overall cost of the total transportation distance and the total loading waiting time.

[0117] In some embodiments, when the processor executes the computer program, it further implements the following steps: when the task to be performed is to transport a single first container type, receiving real-time sensing of the actual loading position of the first container type container in the trailer; determining an additional offset value based on the actual loading position and fixed reference parameters; and dynamically determining the docking position of the autonomous driving truck based on the additional offset value.

[0118] In some embodiments, a computer-readable storage medium is provided having a computer program stored thereon. When executed by a processor, the computer program performs the following steps: obtaining port transportation orders; wherein the port transportation orders include container types; identifying mergeable port transportation orders based on the container types and merging them; generating corresponding tasks to be executed for merged port transportation orders and directly generating tasks to be executed for unmerged port transportation orders, thereby obtaining a list of tasks to be executed; during the execution of each task to be executed in the list of tasks to be executed, receiving the actual container type of each task to be executed in real time; and dynamically correcting the list of tasks to be executed and adjusting the execution order of each task to be executed in response to inconsistencies between the container types and the actual container types.

[0119] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: determining whether the container types recorded in two consecutive port transport orders are both the first container type; if so, determining the two consecutive port transport orders as mergeable port transport orders.

[0120] In some embodiments, when a computer program is executed by a processor, it further performs the following steps: optimizing the execution order of each task in the task list to be executed.

[0121] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: when the list of tasks to be executed includes both a first box-type task to be executed and a second box-type task to be executed, the execution order of the second box-type task to be executed is adjusted to take precedence over the first box-type task to be executed.

[0122] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: obtaining the current truck location information, calculating the transportation distance and loading waiting time of each task to be executed based on the current truck location information; calculating the predicted delivery time based on the transportation distance and loading waiting time, and advancing the execution order of tasks to be executed whose predicted delivery time is greater than the delivery time information.

[0123] In some embodiments, when the computer program is executed by the processor, it further performs the following steps: obtaining trailer structure information; determining the operation stage corresponding to each task to be executed in the task list and the target loading position of the container; wherein the operation stage includes a loading stage or an unloading stage; and optimizing and adjusting the execution order of each task to be executed by applying a preset loading and unloading sequence control rule according to the trailer structure information, wherein the loading and unloading sequence control rule is adapted to the trailer structure information.

[0124] In some embodiments, when the computer program is executed by the processor, it further implements the following steps: acquiring trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; performing feasibility verification on the list of tasks to be executed based on the trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; in response to the list of tasks to be executed passing the feasibility verification, proceeding to the step of executing each task in the list of tasks to be executed; in response to the list of tasks to be executed failing the feasibility verification, splitting or optimizing the execution order of each task in the list of tasks to be executed for the merged port transportation order.

[0125] In some embodiments, when the computer program is executed by the processor, it further performs the following steps: generating new tasks to be executed and a new list of tasks to be executed based on the actual box type; calculating the transportation distance and loading waiting time of each new task to be executed, and determining the total transportation distance and total loading waiting time of all new tasks to be executed; dynamically adjusting the execution order of each new task to be executed in the new list of tasks to be executed, so as to minimize the overall cost of the total transportation distance and the total loading waiting time.

[0126] In some embodiments, when the computer program is executed by the processor, it further performs the following steps: when the task to be performed is to transport a single first container type, receiving real-time sensing of the actual loading position of the first container type container in the trailer; determining an additional offset value based on the actual loading position and fixed reference parameters; and dynamically determining the docking position of the automated driving truck based on the additional offset value.

[0127] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0128] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0129] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the characters in this article generally indicate that the preceding and following related objects have an "or" relationship.

[0130] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0131] It should be noted that in the embodiments of this application, if user information or user data are involved, the user's authorization and consent are required before acquisition and processing. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

Claims

1. A method for adaptively adjusting the task of an autonomous truck, the method comprising: Obtain port transportation orders; wherein, the port transportation orders include the type of transport container; Port transport orders can be merged based on the type of transport container identified; For merged port transportation orders, corresponding tasks to be executed are generated; for unmerged port transportation orders, tasks to be executed are generated directly, resulting in a list of tasks to be executed. Acquire trailer rated load information, preset loading and unloading sequence control rules, and port operation space information; The feasibility of the list of tasks to be executed is verified based on the trailer's rated load capacity information, the preset loading and unloading sequence control rules, and the port operation space information. In response to the failure of the feasibility verification of the list of tasks to be executed, the execution order of each task in the list of tasks to be executed in the merged port transportation order is split or optimized. During the execution of each of the tasks in the list of tasks to be executed, the actual container type corresponding to each task to be executed is perceived in real time by the perception system mounted on the autonomous truck. In response to the discrepancy between the transport container type and the actual container type, the list of tasks to be executed is dynamically corrected, and the execution order of each task to be executed is adjusted. When the task to be performed is to transport a single first-type container, the actual loading position of the first-type container in the trailer is received in real time. The additional offset value corresponding to the container is determined based on the actual loading position and fixed reference parameters. The parking position of the autonomous truck is dynamically determined based on the additional offset value.

2. The method according to claim 1, characterized in that, The step of identifying mergeable port transport orders based on the type of transport container includes: Determine whether the container type recorded in two consecutive port transport orders is the first container type; If so, the two consecutive port transport orders are identified as the mergeable port transport orders.

3. The method according to claim 1, characterized in that, Before each of the tasks in the list of tasks to be executed is executed, the method further includes: The execution order of each task in the task list is optimized and adjusted.

4. The method according to claim 3, characterized in that, The optimization and adjustment of the execution order of each task in the task list includes: If the list of tasks to be executed includes both tasks of the first box type and tasks of the second box type, the execution order of the tasks of the second box type will be adjusted to take precedence over the tasks of the first box type; and / or The port transportation order also includes delivery time information. The optimization and adjustment of the execution order of each task in the task-to-be-executed list includes: Obtain the current truck location information, and calculate the transportation distance and loading waiting time for each of the tasks to be executed based on the current truck location information; The predicted delivery time is calculated based on the transportation distance and the loading waiting time, and the execution order of tasks whose predicted delivery time is greater than the actual delivery time is advanced.

5. The method according to claim 3, characterized in that, The optimization and adjustment of the execution order of each task in the task list includes: Obtain trailer structure information; Determine the operation stage and target loading location of the container for each of the tasks to be executed in the list of tasks to be executed; wherein, the operation stage includes a loading stage or an unloading stage; Based on the trailer structure information, a preset loading and unloading sequence control rule is applied to optimize and adjust the execution order of each task to be executed, wherein the loading and unloading sequence control rule is adapted to the trailer structure information.

6. The method according to claim 1, characterized in that, After verifying the feasibility of the list of tasks to be executed based on the trailer's rated load capacity information, the preset loading and unloading sequence control rules, and the port operating space information, the method further includes: In response to the feasibility verification of the list of tasks to be executed, the step of executing each of the tasks in the list of tasks to be executed is initiated.

7. The method according to any one of claims 1 to 6, characterized in that, The response to the discrepancy between the transport container type and the actual container type, dynamically correcting the list of tasks to be executed and adjusting the execution order of each task to be executed, includes: Generate new tasks to be executed and a new list of tasks to be executed based on the actual box type; Calculate the transportation distance and loading waiting time for each of the new tasks to be executed, and determine the total transportation distance and loading waiting time for all the new tasks to be executed. The execution order of each new task in the new task list is dynamically adjusted to minimize the combined cost of the total transportation distance and the total loading waiting time.

8. A task adaptive adjustment device for an autonomous driving truck, characterized in that, The device includes: The order acquisition module is used to acquire port transportation orders; wherein, the port transportation orders include the type of transport container; The order merging module is used to identify mergeable port transportation orders based on the type of transport container and to merge them. The task generation module is used to generate corresponding tasks to be executed for merged port transportation orders and directly generate tasks to be executed for unmerged port transportation orders, thus obtaining a list of tasks to be executed; it acquires trailer rated load information, preset loading and unloading sequence control rules, and port operating space information; it performs feasibility verification on the list of tasks to be executed based on the trailer rated load information, the preset loading and unloading sequence control rules, and the port operating space information; in response to the list of tasks to be executed failing the feasibility verification, it splits or optimizes the execution order of each task in the list of tasks to be executed for the merged port transportation orders; The real-time perception module is used to, during the execution of each of the tasks in the task list, perceive the actual container type corresponding to each task in real time through the perception system mounted on the autonomous truck; when the task is to transport a single first container type, it receives the real-time perception of the actual loading position of the first container type container in the trailer; determines the additional offset value corresponding to the container based on the actual loading position and fixed reference parameters; and dynamically determines the stopping position of the autonomous truck based on the additional offset value. The sequence adjustment module is used to dynamically correct the list of tasks to be executed and adjust the execution order of each task to be executed in response to the inconsistency between the transport container type and the actual container type.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

Images