A method and system for traffic control at intersections within a wharf

Abstract

This invention relates to the field of intelligent traffic control technology, specifically a method and system for traffic control at intersections within a wharf. The method includes: acquiring vehicles waiting to pass in the controlled lanes, vehicle information, and initial urgency weights; calculating the urgency of vehicles waiting to pass based on the vehicle information and initial urgency weights; sorting the vehicles according to their urgency to obtain a release queue for the controlled lanes, and grouping the release queues into release groups; calculating the Pareto optimal solution set for the release groups using a multi-objective optimization algorithm based on a pre-set inter-group release order and vehicle urgency; adjusting the initial urgency weights based on the Pareto optimal solution set and iteratively obtaining an optimal travel time allocation set and adjusted release groups; and controlling the intersection traffic based on the inter-group release order, the optimal travel time allocation set, and the adjusted release groups. This method solves the problem of insufficiently scientific and reasonable traffic control at intersections within wharves in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of intelligent traffic control technology, specifically a traffic control method and system for intersections within a wharf. Background Technology

[0002] As a core hub for cargo loading, unloading, and transshipment, the efficiency of internal road traffic at the port directly impacts overall operational progress and costs. With the continuous growth of port throughput and the promotion of automated operation modes, the density of various operating vehicles such as container trucks, tractor-trailers, and forklifts within the port has increased significantly. Especially at key nodes such as intersections, frequent vehicle convergence has led to increasingly prominent traffic conflicts.

[0003] Currently, traffic control at dock intersections mostly relies on fixed-time traffic lights or simple sequential release methods, lacking dynamic perception and differentiated scheduling capabilities based on vehicle task urgency. However, dockside operations involve diverse vehicle types with significantly different task attributes: container trucks within the yard must strictly adhere to loading and unloading time windows, while external container trucks may face yard closure deadlines, and port engineering vehicles undertake emergency support tasks. If intersection control cannot achieve refined scheduling by combining vehicle task time limits and waiting times, high-urgency vehicles may experience delays due to prolonged waiting, leading to a chain reaction of problems such as ship congestion and yard congestion.

[0004] Therefore, there is an urgent need for a more scientific and reasonable method for traffic control at intersections within the wharf, so as to improve the operational efficiency of the wharf. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for traffic control at intersections within wharves, solving the problem that existing technologies do not provide scientific and reasonable traffic control at intersections within wharves.

[0006] To achieve the above objectives, one aspect of the present invention provides a traffic control method for an intersection within a wharf. The method includes: acquiring vehicles waiting to pass, vehicle information, and initial urgency weights for multiple controlled lanes at a target intersection within the wharf; calculating the vehicle urgency of the vehicles waiting to pass based on the vehicle information and the initial urgency weights; sorting the vehicles waiting to pass according to the vehicle urgency to obtain a release queue for the controlled lanes, and grouping the release queues into release groups; calculating the Pareto optimal solution set for the release groups using a multi-objective optimization algorithm based on a pre-set inter-group release order and the vehicle urgency; adjusting the initial urgency weights based on the Pareto optimal solution set and iteratively obtaining an optimal passage time allocation set and an adjusted release group; and controlling the intersection traffic based on the inter-group release order, the optimal passage time allocation set, and the adjusted release group.

[0007] This invention lays the data foundation for precise traffic control by acquiring information on vehicles waiting to pass, vehicle information, and initial urgency weights at target intersections within a wharf. It calculates vehicle urgency by combining vehicle information and initial weights, accurately quantifying vehicle priority. By sorting vehicles by urgency to form release queues and grouping them into release groups, it achieves orderly planning of vehicle passage. By combining the release order and urgency between groups, a multi-objective optimization algorithm is used to find the Pareto optimal solution set for the release groups, achieving scientific optimization of the release strategy. By adjusting the initial weights based on the optimal solution set and iterating, the optimal passage time allocation and adjusted release groups are obtained, achieving dynamic adaptation of control parameters. Finally, by combining the release order between groups, the optimal passage time allocation, and the adjusted release groups, traffic control is precisely matched to the actual traffic flow demand at intersections. This improves the rationality and scientific nature of traffic control at intersections within the wharf, effectively avoiding problems such as vehicle task overtime and waiting backlogs, and ensuring the orderly and efficient operation of the entire wharf.

[0008] Optionally, the information on vehicles waiting to pass includes a waiting time limit, the length of time already waited, and a task time limit. The step of calculating the vehicle passage urgency based on the vehicle passage information and the initial urgency weight includes: determining a first urgency for each vehicle based on the waiting time limit and the length of time already waited; determining a second urgency for each vehicle based on the task time limit and the length of time already waited; and using the initial urgency weight to weightedly fuse the first urgency and the second urgency to obtain the vehicle passage urgency for each vehicle.

[0009] This invention enriches the core data dimensions for urgency calculation by refining the information of vehicles waiting to pass into waiting time limits, already waited time, and task time limits. It determines the first urgency based on the waiting time limit and already waited time, accurately quantifying the passage demand in the vehicle waiting dimension. It determines the second urgency by combining the task time limit and already waited time, effectively measuring the urgency of the vehicle's task dimension. Finally, it obtains the vehicle passage urgency by weighting and fusing the two urgency dimensions using initial urgency weights. This achieves a comprehensive consideration and scientific quantification of multi-dimensional passage demand, improving the accuracy and rationality of vehicle passage urgency calculation.

[0010] Optionally, determining the first urgency of each of the vehicles waiting to pass based on the waiting time limit and the waiting time includes: associating the waiting time limit and the waiting time to obtain the single-vehicle waiting time limit and the single-vehicle waiting time for each of the vehicles waiting to pass; and calculating the ratio of the single-vehicle waiting time to the single-vehicle waiting time limit to obtain the first urgency of each of the vehicles waiting to pass.

[0011] This invention obtains unique waiting parameters for each vehicle by linking the waiting time limit and the already waited duration one by one, achieving accurate matching of vehicle waiting dimension data and avoiding parameter confusion. By calculating the ratio of the already waited duration to the waiting time limit, the first urgency is obtained, which intuitively quantifies the proportion of the waiting time of a single vehicle relative to its own time limit, making the determination of the urgency of the waiting dimension more objective and targeted, and improving the accuracy of the first urgency calculation.

[0012] Optionally, the step of calculating the Pareto optimal solution set of the release group based on the pre-set inter-group release order and the vehicle passage urgency using a multi-objective optimization algorithm includes: setting the number of passage rounds and the duration of each round; setting an initial population based on the number of passage rounds, the duration of each round, and the vehicle passage urgency; and iteratively obtaining the Pareto optimal solution set of the release group using a multi-objective optimization algorithm based on the pre-set inter-group release order and the initial population.

[0013] This invention establishes clear time constraints for optimizing release groups by setting the number of passage rounds and the duration of each round, thus ensuring the rationality of the control cycle. By combining the number of passage rounds, the duration of each round, and the urgency of vehicle passage to set an initial population, the algorithm iteration has realistic initial samples. Based on the release order between groups and the initial population, a multi-objective optimization algorithm is used to iteratively find the Pareto optimal solution set for the release groups, achieving multi-objective scientific optimization of the release strategy and improving the scientificity and rationality of the release strategy optimization at dock intersections.

[0014] Optionally, setting the initial population based on the number of passage rounds, the duration of a single round, and the urgency of vehicle passage includes: calculating the total control time based on the number of passage rounds and the duration of a single round; estimating the theoretically passable vehicles for each controlled lane based on the total control time; calculating the passable urgency of each controlled lane based on the theoretically passable vehicles and the urgency of vehicle passage; selecting the maximum value after grouping the passable urgency into release groups to obtain the passable reference urgency of the release group; allocating a single-group total passage time to each release group according to the proportion of the passable reference urgency and the total control time; and determining the initial population based on the single-group total passage time and the number of passage rounds, with the duration of a single round as a constraint.

[0015] This invention calculates the total control time by using the number of passes and the duration of each pass, setting a unified time boundary for the population. By estimating the theoretically passable vehicles in each lane based on the total control time, it controls the traffic flow dispersal scale of each lane. Combining the urgency of vehicle passage, it calculates the urgency of lane passage, quantifies the overall dispersal demand of the lane, and obtains a reference value by taking the maximum urgency of passage for each release group. This aligns with the opposite synchronous release control rules, and the total time for each group is allocated according to the proportion of the reference urgency, achieving a scientific allocation of time resources. The initial population is determined by using the duration of each pass as a constraint, ensuring that individual population members meet the actual control requirements, thus improving the scientific and rational nature of the initial population setting.

[0016] Optionally, the step of obtaining the Pareto optimal solution set of the release group by iteratively using a multi-objective optimization algorithm based on a pre-set inter-group release order and the initial population includes: calculating the minimum vehicle task time limit margin and the minimum single-vehicle waiting time limit margin for individuals in the initial population based on the pre-set inter-group release order; performing a non-dominated sorting of the initial population with the objective of maximizing the minimum vehicle task time limit margin and the minimum single-vehicle waiting time limit margin; and iterating using a multi-objective optimization algorithm based on the non-dominated sorting until a pre-set convergence condition is met to obtain the Pareto optimal solution set.

[0017] This invention calculates the minimum task and waiting time margin for each individual in the population by combining the release order between groups, accurately anchoring the core objective of algorithm optimization. It uses the maximum minimum of both margins as the objective to perform non-dominated sorting of the initial population, achieving hierarchical selection of the optimal control scheme. Based on the non-dominated sorting iterative algorithm until convergence, it obtains a Pareto optimal solution set, ensuring the optimality and effectiveness of the solution. This improves the targeting and scientific nature of multi-objective optimization algorithm iteration. The selected optimal solution can take into account the passage requirements of both vehicle task and waiting dimensions, avoiding vehicle timeouts. This provides high-quality support for subsequent weight adjustments and time allocation, enhancing the optimization effect and practicality of the intersection release strategy.

[0018] Optionally, the step of adjusting the initial urgency weights based on the Pareto optimal solution set and iteratively obtaining the optimal passage time allocation set and the adjusted release group includes: obtaining the minimum value set of vehicle task time limit leeway and the minimum value set of single vehicle waiting time limit leeway based on the Pareto optimal solution set; extracting minimum value association pairs from the minimum value set of vehicle task time limit leeway and the minimum value set of single vehicle waiting time limit leeway, respectively; determining the initial optimal first minimum value and the initial optimal second minimum value based on the minimum value association pairs; and adjusting the initial urgency weights based on the initial optimal first minimum value and the initial optimal second minimum value and iteratively obtaining the optimal passage time allocation set and the adjusted release group.

[0019] This invention extracts the set of minimum dual-time-limit margins from the Pareto optimal solution set, accurately aggregates and optimizes core data, extracts minimum value association pairs and determines the initial optimal dual minimum values, locks the core reference basis for weight adjustment, adjusts and iterates the initial urgency weights based on the initial optimal dual minimum values, and obtains the optimal time allocation and adjusted release group, realizing the dynamic adaptation of control parameters and release group, and improving the accuracy of the optimal passage time allocation set and the adjusted release group.

[0020] Optionally, the step of adjusting the initial urgency weights based on the initial optimal first minimum value and the initial optimal second minimum value and iterating to obtain the optimal passage time allocation set and the adjusted passage group includes: when the initial optimal first minimum value is positive, calculating the absolute difference between the initial optimal first minimum value and the initial optimal second minimum value; adjusting the initial urgency weights with the goal of minimizing the absolute difference and iterating to obtain the optimal passage time allocation set and the adjusted passage group.

[0021] This invention accurately identifies the control status of vehicles with no timeout by determining that the initial optimal first minimum value is positive, clarifies the optimization direction of weight adjustment, and quantifies the balance of the two-dimensional time limit margin by calculating the absolute difference between this value and the initial optimal second minimum value. The initial urgency weight is adjusted iteratively with the goal of minimizing the absolute difference, resulting in the optimal time allocation and the adjusted release group. This achieves balanced optimization of two-dimensional control and further improves the accuracy and rationality of the optimal passage time allocation set and the adjusted release group.

[0022] Optionally, the step of adjusting the initial urgency weights based on the initial optimal first minimum value and the initial optimal second minimum value and iterating to obtain the optimal passage time allocation set and the adjusted passage group further includes: when the initial optimal first minimum value is negative, calculating the absolute value of the initial optimal first minimum value; adjusting the initial urgency weights with the goal of minimizing the absolute value and iterating to obtain the optimal passage time allocation set and the adjusted passage group.

[0023] This invention accurately identifies the timeout status of vehicle tasks by determining that the initial optimal minimum value is negative, anchors the core direction of weight adjustment, and quantifies the degree of timeout of vehicle task time limits by calculating the absolute value of this value, providing a quantitative basis for weight adjustment. The initial urgency weight is adjusted iteratively with the goal of minimizing the absolute value, resulting in the optimal time allocation and adjusted release group. This achieves targeted optimization of task timeout problems, improves the accuracy and targeting of urgency weight adjustment, effectively reduces the risk of vehicle task timeouts, makes the passage time allocation more in line with the core control needs of the task dimension, optimizes the operability of the intersection release strategy, and further ensures the timely progress of dock operations.

[0024] Another aspect of the present invention provides a traffic control system for an intersection within a wharf, comprising: a processor, an input device, an output device, and a memory, wherein the processor, the input device, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute a traffic control method for an intersection within a wharf as described in any of the preceding aspects of the present invention.

[0025] The present invention provides a traffic control system for intersections within a wharf, which is compact in structure, stable in performance, highly integrated and simple in composition. It can stably execute the traffic control method for intersections within a wharf provided in the preceding aspect of the present invention, further enhancing the overall applicability and practical application capability of the present invention. Attached Figure Description

[0026] Figure 1 This is a flowchart of a traffic control method for an intersection within a wharf, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of a traffic control system for a wharf intersection according to an embodiment of the present invention. Detailed Implementation

[0027] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0028] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0029] Please see Figure 1 In order to solve the problems in the prior art, in an alternative embodiment, such as Figure 1 The method for traffic control at intersections within a wharf, as shown, includes the following steps: Step S1: Obtain the information on vehicles waiting to pass and their initial urgency weights for multiple controlled lanes at the target intersection within the dock.

[0030] In this embodiment, all operating vehicles within the terminal, including in-yard trucks and external trucks, port operation tractors, and forklifts, will first receive standardized operating instructions issued by the Terminal Operating System (TOS) when undertaking tasks such as cargo loading and unloading, yard transfer, and port engineering. The TOS will automatically plan the optimal driving route for each vehicle based on the current yard operation progress, real-time traffic status of each intersection, cargo loading and unloading plan, and port operation process connection requirements. At the same time, it will synchronize the core parameters related to the task to the vehicle's on-board terminal and the control terminal of the target intersection, realizing two-way data linkage between the vehicle and the road, and avoiding vehicle driving path conflicts and information disconnection.

[0031] When the driving path planned by TOS for a vehicle covers a target intersection that needs to be controlled, the control system of the target intersection will automatically identify the vehicle through the vehicle-road cooperative module and include it in the candidate passage queue of the target intersection, forming a vehicle waiting to pass. At the same time, the start time of the vehicle entering the candidate passage queue is accurately recorded in the database of the control system. This start time is used as the reference time for subsequent calculation of the vehicle's waiting time. The waiting time will be updated in real time according to the vehicle's status in the candidate queue. If the vehicle undergoes temporary detours, task adjustments, or other situations, the control system will automatically remove it from the candidate queue and terminate the time recording, thus avoiding data deviation in the time dimension from the root. While including vehicles in the candidate passage queue, the intersection control system automatically assigns structured vehicle information based on the vehicle's task type, vehicle type, and cargo priority. This information is the core data for subsequently calculating the urgency of vehicle passage. In addition to the vehicle ID used to uniquely identify the vehicle, it also includes the waiting time limit (i.e., the maximum waiting time at the target intersection from the time the vehicle enters the intersection candidate), the waiting time already elapsed, and the task time limit. The vehicle ID is compiled using the unified coding rules of the port to ensure that the identity of each vehicle waiting to pass is unique and to prevent confusion of vehicle information during the intersection control process. The task time limit is directly linked to the specific work tasks undertaken by the vehicle. If it is a cargo truck responsible for loading and unloading goods into and out of the port, the task time limit is the intersection passage deadline calculated backward from the cargo loading and unloading time and the yard handover time (understandably, allowance for error time according to specific needs). If it is a vehicle for transshipment operations within the port, the task time limit is the work process connection time planned by TOS. If it is an emergency engineering vehicle within the port, the task time limit is the start deadline of emergency operations. The length of the task time limit directly determines the time urgency of the vehicle's operation.

[0032] The setting of the maximum waiting time limit is flexibly configured in conjunction with the overall dispatch strategy of port vehicles, vehicle type and task type. The maximum waiting time limit for special operation vehicles, container trucks transporting urgent goods and emergency operation vehicles in the port will be appropriately shortened according to dispatch needs. The maximum waiting time limit for ordinary transfer operation vehicles and container trucks returning empty can be reasonably extended. The setting of this parameter also needs to take into account the traffic flow linkage and balance between the target intersection and surrounding intersections. Moreover, the maximum waiting time limit and the task time limit are independent control parameters. The basis and considerations for their setting are completely different. Therefore, their values ​​are generally different in the actual control process. After completing the screening and inclusion of vehicles waiting to pass and the comprehensive collection and recording of their information, the intersection control system will further retrieve the pre-configured initial urgency weights. These initial urgency weights are basic control parameters pre-set in the control system by the dock control personnel based on the historical traffic data of the target intersection, the patterns of peak and off-peak hours of dock operations, and vehicle operation priority rules. They include a weight coefficient corresponding to the first urgency and a weight coefficient corresponding to the second urgency, and the sum of the two weight coefficients is 1 (initially, they can also be set to two 0.5s). The initial urgency weights will be synchronously stored in the parameter library of the intersection control system to provide a benchmark coefficient for the subsequent weighted fusion calculation of vehicle passage urgency. At the same time, these initial weights allow control personnel to manually fine-tune them according to the real-time operation of the dock, taking into account both the preset nature of the parameters and the flexibility of actual control.

[0033] Step S2: Calculate the vehicle passage urgency of the vehicle to be passed based on the vehicle information to be passed and the initial urgency weight.

[0034] The calculation of the vehicle passage urgency based on the vehicle information and the initial urgency weight specifically includes the following sub-steps: Step S201: Determine the first urgency of each of the vehicles waiting to pass based on the waiting time limit and the waiting time.

[0035] The determination of the first urgency level of each vehicle waiting to pass, based on the waiting time limit and the already waited duration, specifically includes the following sub-steps: Step S20101: Associate the waiting time limit and the waiting time to obtain the single-vehicle waiting time limit and the single-vehicle waiting time for each of the vehicles waiting to pass.

[0036] In this embodiment, for each vehicle waiting to pass through the target intersection, its unique waiting time limit and the waiting time are precisely bound one-to-one, forming a unique single-vehicle waiting time limit and single-vehicle waiting time for each vehicle. The two parameters of each vehicle are independently collected and do not mix with each other, realizing vehicle-level independent matching of the core parameters of the waiting dimension.

[0037] Step S20102: Calculate the ratio of the waiting time of the single vehicle to the waiting time limit of the single vehicle to obtain the first urgency of each of the vehicles waiting to pass.

[0038] In this embodiment, an independent quantitative calculation is performed on each vehicle waiting to pass through the target intersection. The calculation is performed by dividing the vehicle's individual waiting time as the numerator and the vehicle's waiting time limit as the denominator. The ratio obtained is the first urgency of the vehicle's waiting dimension. This calculation method directly quantifies the proportion of a vehicle's waiting time relative to its own waiting time limit. The size of the ratio is positively correlated with the vehicle's waiting urgency. The larger the ratio, the more urgent the vehicle's waiting status is. A larger ratio indicates that the vehicle's waiting time is more urgent.

[0039] Step S202: Determine the second urgency of each of the vehicles waiting to pass based on the task time limit and the waiting time.

[0040] In this embodiment, the remaining execution time limit for each vehicle waiting to pass is first calculated. ,in For the vehicle's mission time limit, The waiting time for the vehicle is calculated; then, the maximum remaining execution time is extracted from all vehicles currently waiting to pass. As a benchmark reference value; second urgency The calculation formula is: This quantification method uses the vehicle with the least urgent task (longest remaining execution time) as a reference, and intuitively reflects the relative urgency of each vehicle in terms of task dimension through a ratio: when hour, This indicates that the vehicle is the least urgent vehicle in the current task dimension; when hour, This indicates that the vehicle's task has just reached its deadline; when hour, Furthermore, the longer the timeout, the larger the value, with no upper limit, accurately depicting the severity of vehicle task timeouts. Because... According to the real-time traffic flow changes at the target intersection, The dynamic updating of relative urgency accurately reflects the priority of vehicles in the current competitive traffic environment. This relative quantification method aligns perfectly with the needs of dynamic intersection management, allowing subsequent multi-objective optimization to focus on the vehicle combinations that most require priority passage. It should be noted that if all vehicles have zero remaining execution time (i.e., ... (If the denominator is 0 and division is not possible at this point), then the second urgency level of all vehicles is uniformly set to 1, indicating that the urgency level of all vehicles is the same at this time.

[0041] In summary, the second urgency level adopts a relative quantification method based on the global maximum remaining time limit. Its core advantage is that it does not require any preset empirical parameters and relies entirely on real-time traffic flow data to dynamically generate the urgency level scale. This avoids the problem of insufficient adaptability caused by fixed thresholds and ensures the comparability of urgency levels between different control cycles.

[0042] It should be noted that the core of this method is to quantify the relative urgency of vehicle tasks, rather than to accurately calculate the absolute remaining time for vehicles to complete tasks. The remaining execution time limit is only used as a relative basis for ranking urgency, not as a basis for actual operation time planning. Therefore, the urgency calculation is a relative quantification of all vehicles. Furthermore, since intersection traffic control uses short-term cyclic calculations, all vehicles use the same set of calculation rules, which has a strong tolerance for uncontrollable errors and factors.

[0043] Step S203: The first urgency and the second urgency are weighted and fused using the initial urgency weight to obtain the vehicle passage urgency of each of the vehicles to be passed.

[0044] In this embodiment, an independent weighted fusion calculation is performed for each vehicle waiting to pass through the target intersection. The initial urgency weight includes a first weight coefficient corresponding to the first urgency and a second weight coefficient corresponding to the second urgency, and the sum of the two weight coefficients is 1. During weighted fusion, the first urgency of the vehicle is first multiplied by the first weight coefficient to obtain the first weighted value, and then the second urgency of the vehicle is multiplied by the second weight coefficient to obtain the second weighted value. Finally, the two weighted values ​​are summed, and the sum is the vehicle passage urgency of the vehicle waiting to pass. This achieves a comprehensive consideration of the urgency of the waiting dimension and the task dimension, allowing the vehicle passage urgency to fully reflect the actual control priority of the vehicle.

[0045] Step S3: Sort the vehicles waiting to pass according to the urgency of vehicle passage to obtain the release queue of the controlled lane, and group the release queue to obtain the release group.

[0046] In this embodiment, for vehicles waiting to pass through each controlled lane at the target intersection (crossroads) of the dock, the urgency of vehicle passage is used as the core sorting criterion. The vehicles waiting to pass through each controlled lane are sorted independently in descending order of the numerical value. After sorting, a dedicated release queue is formed for each controlled lane. This release queue directly defines the passage priority of vehicles waiting to pass in the corresponding lane. The higher the vehicle passage urgency value, the higher the ranking in the release queue, and the more qualified the vehicle is to pass first. Simultaneously, based on the corresponding driving direction of each vehicle in the planned driving path of the terminal operating system, and combined with the driving function division of the controlled lanes, the release queues of each controlled lane are uniformly grouped. Vehicles traveling in the direction of going straight up or down are grouped into the straight-up / down / straight-down release group; vehicles traveling in the direction of going straight left or right are grouped into the straight-left / right release group; vehicles traveling in the direction of turning left up or down are grouped into the left-up / down / left-turn release group; and vehicles traveling in the direction of turning left left or right are grouped into the left-right / left-turn release group. This forms four control forms: two straight-up release groups and two left-turn release groups (this method illustrates defining four directions: up, down, left, and right, based on a certain direction). The division of release groups is highly compatible with the driving direction restrictions of the controlled lanes at the intersection, ensuring that the grouping results fully comply with the actual traffic control rules of the terminal intersection. It should be noted that right-turning vehicles at the target intersection do not conflict with traffic flow in other directions (i.e., free flow not controlled by traffic lights). Therefore, right-turning vehicles can continue to pass under the premise of ensuring safety and do not need to be included in the subsequent group release control, and are not within the control scope of this invention.

[0047] Setting independent passage rules for the aforementioned divided release groups is based on the core safety requirements and order control needs of traffic management at the dock intersection. Because the vehicle trajectories of the straight-through release group and the left-turn release group have obvious overlap in the intersection passage area, if different release groups are released at the same time, it is very easy to cause safety accidents such as driving conflicts and scrapes among dock operation vehicles. This will not only cause vehicle traffic to stop and reduce the overall traffic efficiency of the intersection, but also disrupt the overall vehicle scheduling rhythm formulated by the dock based on the operation plan, affecting the orderly progress of subsequent operations such as cargo loading and unloading and yard transfer.

[0048] Step S4: Based on the pre-set inter-group release order and the urgency of vehicle passage, calculate the Pareto optimal solution set of the release group using a multi-objective optimization algorithm.

[0049] The calculation of the Pareto optimal solution set for the release group based on a pre-set inter-group release order and the vehicle passage urgency using a multi-objective optimization algorithm includes: Step S401: Set the number of passes and the duration of each pass.

[0050] In this embodiment, the process of setting the number of passes and the duration of each pass as core time constraints for the multi-objective optimization control of the target intersection release group is described. Considering the real-time traffic flow at the target intersection, the efficiency of vehicle operations at the dock, and the design objective of reducing the computational load of subsequent multi-objective optimization algorithms, the number of passes and the duration of each pass are uniformly set for the two straight-through release groups and the two left-turn release groups. The right-turn release group does not require separate control parameters. The preferred number of passes is 5 passes. This value range avoids the problem of excessive algorithm computation and reduced iteration efficiency caused by too many passes, while ensuring the rationality of the control cycle. It accurately focuses on the vehicles with the highest urgency and the most urgent need for priority release in each controlled lane, aligning with the core control requirements of the intersection. The single-round duration is flexibly configured based on the average speed of vehicles operating at the dock, the actual traffic capacity of the intersection, and the number of vehicles released at one time, preferably 4-8 minutes. It is the total time for two straight-through release groups and two left-turn release groups to complete a complete and independent round of release according to the pre-set release order between groups. In the process of setting, the sufficiency of single-round release and the overall control rhythm of the intersection are taken into account, so as to avoid the inability of priority vehicles to pass effectively due to short duration and the traffic congestion at the intersection due to excessive duration.

[0051] Step S402: Set an initial population based on the number of passage rounds, the duration of a single round, and the urgency of vehicle passage.

[0052] The process of setting the initial population based on the number of passage rounds, the duration of a single round, and the urgency of vehicle passage specifically includes the following sub-steps: Step S40201: Calculate the total control time based on the number of passage rounds and the duration of a single round.

[0053] In this embodiment, the total control duration is the overall time boundary defined by the initial population construction. The number of passage rounds and the duration of each round are multiplied by the number of rounds that are uniformly set for the straight-through and left-turn passage groups in advance. The result is the total control duration of the target intersection at the dock. This duration is the core time benchmark for the subsequent estimation of the theoretically passable vehicles in each controlled lane, the calculation of the urgency of passage, and the allocation of the total passage duration for each group. It limits a unified time range for all subsequent quantitative calculations, ensuring that the calculation results of each step are carried out around the established control cycle and conform to the overall traffic control rhythm of the intersection.

[0054] Step S40202: Estimate the theoretical number of vehicles that can pass through each of the controlled lanes based on the total control duration.

[0055] In this embodiment, the actual traffic conditions at the dock intersection are first considered, and the time required for different types of dock operation vehicles, such as container trucks and tractor-trailers, to pass through the intersection is estimated. The differences in travel time caused by the differences in vehicle size between long and short vehicles are fully taken into account, and the estimated time for each type of vehicle to pass through the intersection is determined and a reference standard is formed.

[0056] Based on this standard, for each controlled lane at the target intersection, considering the defined total control time and the assumption that vehicles in the release queue of each controlled lane are already sorted and that the type of each vehicle is known, the estimated travel time of each vehicle is accumulated sequentially, starting from the highest-urgency vehicle at the head of the release queue, until the accumulated time is close to but does not exceed the total control time. The number of vehicles accumulated at this point is the theoretical number of vehicles that can pass through that controlled lane within the total control time. Each controlled lane completes this estimation process independently. Because in actual control, data is reacquired and a new round of calculation is started after 1-2 rounds of release, there may be a slight deviation in the estimated number of theoretically passable vehicles (estimated by a few more or less). This has no substantial impact on the subsequent iterative solution of the multi-objective optimization algorithm or the final control effect; it is only necessary to use this estimation result to control the approximate scale of traffic flow that can be diverted by each lane.

[0057] Step S40203: Based on the vehicle traffic urgency, calculate the traffic urgency of each of the controlled lanes using the theoretically passable vehicles.

[0058] In this embodiment, for each controlled lane at the target intersection, based on the theoretically passable number of vehicles in that lane and the release queue within that lane that has been sorted by vehicle urgency from highest to lowest, starting with the highest urgency vehicle at the head of the queue, vehicles matching the theoretically passable number are selected sequentially. The urgency of these selected vehicles is summed, and the sum is the passable urgency of that controlled lane. Each controlled lane performs this calculation independently without affecting others. The calculated passable urgency directly reflects the overall urgency of the high-urgency traffic flow that can be alleviated within the total control time of the corresponding controlled lane.

[0059] Step S40204: After grouping the passability urgency according to the release group, select the maximum value to obtain the passability reference urgency of the release group.

[0060] In this embodiment, the calculated urgency of each controlled lane is precisely grouped and categorized according to its corresponding straight-through release group and left-turn release group, so that the urgency of each controlled lane corresponds to its respective release group. Then, the urgency values ​​of all controlled lanes in each release group are compared, and the maximum value is selected as the reference urgency of the release group. This method of selection conforms to the control rules of simultaneous release in opposite directions. Because when simultaneous release in opposite directions is carried out, the controlled lane with low urgency will pass simultaneously with the controlled lane with high urgency, without the need to allocate additional time resources, taking the maximum value in the group can accurately reflect the core urgency relief needs of the release group.

[0061] Step S40205: Allocate a single group's total passage time to each of the release groups according to the ratio between the passable reference urgency and the total control time.

[0062] In this embodiment, the reference urgency of passage for all straight-through and left-turn release groups is first summed to obtain the total value of the reference urgency. For each release group, its own reference urgency is compared with this total value to obtain the proportion of reference urgency for passage for each release group. This proportion directly reflects the weight of the core urgency relief needs of each release group in the overall demand. Then, using this proportion as the sole allocation basis, the predetermined total control time is multiplied by the proportion of each release group. The result is the total passage time for each release group, realizing the resource-biased control principle that the higher the reference urgency of passage, the longer the total passage time for each group. Moreover, the sum of the total passage times for each release group is completely consistent with the total control time, forming a closed loop of unbiased time allocation.

[0063] Step S40206: Using the single-round duration as a constraint, determine the initial population based on the total passage time of a single group and the passage round number.

[0064] In this embodiment, the four release groups (two straight-through release groups and two left-turn release groups) are first classified according to the pre-set passage rounds, with each group corresponding to a specific round dimension. Taking a passage round of 5 as an example, all release groups are divided into the first, second, third, fourth, and fifth rounds, so that each passage round contains an independent duration allocation dimension for the four release groups. Each release group has its own duration allocation item in each round, thus establishing a basic framework for multi-round refined duration allocation. Subsequently, for each release group in each round, a clear passage time allocation is defined to obtain a passage time interval (the passage time interval is used for the range during individual iterations). The lower limit of this interval is a preset 0 or the minimum passage time (the minimum passage time refers to the minimum time that each release group needs to be allocated if the release time is not 0; in this embodiment, it is preferably 15 seconds. In port vehicle passage, too short a time is not conducive to vehicle passage preparation and instead increases ineffective passage time). The upper limit is the preset single-round time (one group is allocated a single-round time, and the other three groups are allocated 0 time). That is, the passage time interval is expressed as: ,in, The duration of a single round of passage for the release group. The duration of a single round, This is the minimum passage time. Simultaneously, the four passage groups adhere to two core time constraints: Constraint 1 is a single-round summation constraint, meaning that in each round, the sum of the passage times allocated to the four passage groups must be exactly the same as the preset single-round duration. Constraint 2 is a group-wide total duration summation constraint, meaning that the sum of the passage times allocated to each passage group across all passage rounds must be strictly equal to its allocated total single-group passage time.

[0065] Next, under the premise of satisfying the constraints of allocation interval, single-round duration summation, and single-group total passage duration, a numerical allocation method under multiple constraints is used to generate the initial population individuals. The core of this method lies in solving a duration allocation matrix with boundary constraints and equality constraints. The rows and columns of the duration allocation matrix correspond to the passage rounds and release groups, respectively, and the matrix elements are the allocated passage durations of the corresponding release groups in that round. In specific implementation, a mixed integer programming method or heuristic search strategy can be used to first generate an initial solution that satisfies the single-group total passage duration constraint, and then iteratively correct it based on deviation to simultaneously satisfy the single-round duration summation constraint, finally obtaining a series of duration allocation schemes that meet all constraints. After combining each scheme with the pre-set inter-group release order, each combination scheme is an initial population individual, whose dimension corresponds to the three-dimensional mapping relationship between the round, release group, and passage duration.

[0066] Finally, by differentiating the passage time for different rounds and different release groups within the compliant time allocation range, multiple complete control schemes with different time allocation strategies are generated, all of which meet the allocation range and the two core constraints. All schemes together constitute the initial population of the multi-objective optimization algorithm, ensuring that all individuals in the population are generated within the preset range and have physical significance for direct implementation.

[0067] Step S403: Based on the pre-set inter-group release order and the initial population, a multi-objective optimization algorithm is used to iterate and obtain the Pareto optimal solution set of the release group.

[0068] The process of obtaining the Pareto optimal solution set for the release group by iteratively using a multi-objective optimization algorithm based on a pre-set inter-group release order and the initial population specifically includes the following sub-steps: Step S40301: Calculate the minimum vehicle task time limit margin and the minimum single vehicle waiting time limit margin for individuals in the initial population during passage, based on the pre-set inter-group release order.

[0069] In this embodiment, firstly, based on the pre-set inter-group release order, combined with the passage time scheme allocated by the individual to each release group in each round, the queuing order of vehicles waiting to pass in each controlled lane, and the estimated time for each vehicle (the estimated time is the aforementioned reference standard), the specific time point at which each vehicle waiting to pass in each controlled lane actually passes through the target intersection is calculated precisely. The time calculation dimension is consistent with the baseline time of the vehicle's waiting time, task time limit, and waiting time limit to ensure the effectiveness of the time difference calculation. Subsequently, for each vehicle with a calculated passage time point, two time margins are calculated: the task time limit margin for the vehicle is obtained by subtracting the actual time point at the intersection from the vehicle's own task time limit, and the waiting time limit margin for the vehicle is obtained by subtracting the actual time point at the intersection from the vehicle's own waiting time limit. After calculating the two margins for all vehicles waiting to pass for the individual, the minimum value is extracted from the passing task time limit margin of all vehicles as the minimum passing vehicle task time limit margin for the individual. The minimum value is also extracted from the passing waiting time limit margin of all vehicles as the minimum single vehicle waiting time limit margin for the individual. These two minimum values ​​together constitute the core objective value of the multi-objective optimization of the initial population individual.

[0070] Step S40302: Perform non-dominated sorting on the initial population with the goal of maximizing the minimum vehicle task time limit margin during passage and the maximum single vehicle waiting time limit margin during passage.

[0071] In this embodiment, the core optimization objective of non-dominated sorting is to maximize both the minimum vehicle task time limit margin and the minimum single vehicle waiting time limit margin during passage. This objective aligns with the core requirements of traffic management at the dock intersection, namely, to minimize the occurrence of task time limit and waiting time limit overdue for vehicles waiting to pass, thus ensuring the passage efficiency of vehicles with high urgency. First, for all individuals in the initial population, Pareto dominance relationships between individuals are determined based on the two core objective values ​​corresponding to each individual. If both objective values ​​of individual A in the population are greater than or equal to the two objective values ​​of individual B, and at least one objective value is strictly greater than that of individual B, then individual A is determined to dominate individual B, and individual B does not qualify for the current optimal level. If there is no mutual dominance relationship between two individuals, then they are determined to have a Pareto non-dominance relationship and meet the conditions for being classified into the same non-dominance level. Subsequently, following this judgment rule, all individuals in the initial population are screened layer by layer. All individuals not dominated by any other individual are assigned to the first non-dominated level, which is the optimal level within the population, representing the control scheme that best fits the dual-objective maximization. Individuals in the first non-dominated level are then removed, and the dominance relationship judgment and screening steps are repeated for the remaining population individuals to sequentially divide the population into second, third, and subsequent non-dominated levels, completing the hierarchical division of the entire initial population. To avoid population convergence towards a single objective and to ensure population diversity during algorithm iteration, crowding is further calculated for individuals within each non-dominated level. By calculating the distance between an individual and its neighboring individuals in both objective dimensions, the crowding value of the individual is obtained. A higher crowding value indicates a sparser distribution of the individual in the objective space, resulting in higher retention value. Individuals within the same level are then sorted from largest to smallest crowding value. Finally, by dividing the population into non-dominated levels and sorting the crowding within each level, a complete non-dominated sort of the initial population is achieved. The sorting results ensure that the individuals in the population converge towards the direction of maximizing both objectives, while also taking into account the diversity of the population. This provides a clear basis for individual selection for subsequent iterations of the multi-objective optimization algorithm, ensuring that the iteration process can continuously select high-quality time allocation control schemes that meet the actual control needs of the dock intersection.

[0072] Step S40303: Based on the non-dominated sorting, a multi-objective optimization algorithm is used to iterate until the pre-set convergence condition is met, and a Pareto optimal solution set is obtained.

[0073] In this embodiment, the initial population after completing the non-dominated hierarchy division and congestion ranking is used as the initial sample for iteration. During the iteration process, the core operation rules of selection, crossover, and mutation of the multi-objective optimization algorithm are strictly followed. In the selection of individuals in the population, individuals with better non-dominated hierarchies and larger congestion values ​​are given priority to enter the next generation of the population. This ensures that the population as a whole converges towards maximizing the dual objectives of minimizing the vehicle task time limit margin and minimizing the single vehicle waiting time limit margin during passage, while maintaining the diversity of the population and avoiding the algorithm getting trapped in local optima. For each newly generated population in each iteration, the complete operations of steps S40301 and S40302 are repeated, that is, the two core objective values ​​of each individual in the population are calculated in sequence, the Pareto dominance relationship between individuals is determined and the non-dominated hierarchy is divided, and the congestion degree within the same level is calculated and ranked. This ensures that the population of each iteration is a high-quality candidate solution that meets the control requirements of the dock intersection. In this embodiment, the pre-set convergence condition is a combination of multi-dimensional judgment criteria, including three judgment criteria: the number of iterations reaches a preset threshold, the two core objective values ​​of the first non-dominated level of the population show no significant fluctuations for multiple generations, and the Pareto front tends to stabilize. During the iteration process, the iteration state of the current population is judged in real time. When any preset convergence condition is met, the algorithm iteration is terminated immediately. Finally, all individuals in the final population at the first non-dominated level at the time of iteration termination are selected. The control schemes corresponding to the individuals in this level have no Pareto dominance relationship with each other and are all optimal release time allocation schemes under the dual objective maximization. These individuals are integrated to form the Pareto optimal solution set of the release group. Each solution in this set corresponds to a set of implementable port and intersection release group time allocation schemes.

[0074] The introduction of the Pareto optimal solution set enables this method to scientifically balance the two core management objectives of ensuring task time limits and ensuring fairness in waiting time limits. Each solution in the set represents an equilibrium state that cannot improve one objective without compromising the other, providing a rich pool of candidate solutions for subsequent weight adjustments, rather than a single fixed solution, thus significantly improving the flexibility and robustness of decision-making.

[0075] Step S5: Based on the Pareto optimal solution set, adjust the initial urgency weights and iterate to obtain the optimal passage time allocation set and the adjusted passage group.

[0076] Specifically, adjusting the initial urgency weights based on the Pareto optimal solution set and iteratively obtaining the optimal passage time allocation set and the adjusted passage group includes the following sub-steps: Step S501: Obtain the minimum set of vehicle task time limit margins and the minimum set of single vehicle waiting time limit margins during passage based on the Pareto optimal solution set.

[0077] In this embodiment, each optimal solution in the Pareto optimal solution set corresponds to a set of passage time allocation schemes adapted to the control of the target intersection at the dock. Furthermore, in the initial target value calculation, each optimal solution has a uniquely matched minimum vehicle task time limit margin and minimum single-vehicle waiting time limit margin during passage. Data is extracted from the Pareto optimal solution set according to the one-to-one correspondence principle. The minimum vehicle task time limit margin corresponding to each optimal solution in the set is extracted sequentially. All extracted values ​​are then aggregated and integrated to form a set of minimum vehicle task time limit margins during passage. Simultaneously, using the same one-to-one correspondence principle, the minimum single-vehicle waiting time limit margin corresponding to each optimal solution is extracted, completing another set of aggregated values ​​to obtain a set of minimum single-vehicle waiting time limit margins during passage. The number of elements in the two minimum sets is exactly the same as the number of optimal solutions in the Pareto optimal solution set, and the elements at the same position in the two sets correspond to two quantitative indicators of the same optimal solution in the Pareto optimal solution set, ensuring that the corresponding minimum value association pairs can be accurately extracted based on position matching in the future.

[0078] Step S502: Extract minimum value association pairs from the minimum value set of vehicle task time limit leeway during passage and the minimum value set of single vehicle waiting time limit leeway during passage, respectively.

[0079] In this embodiment, firstly, a global numerical filter is performed on the set of minimum vehicle task time limit margins during passage to extract the global minimum value in the set, and the unique position index of the minimum value in the set is recorded. Then, based on the position index, the value corresponding to the same position index is accurately extracted from the set of minimum single vehicle waiting time limit margins during passage. The extracted global minimum task time limit margin is matched and combined with the waiting time limit margin value at the same position to obtain a unique minimum value association pair. The two values ​​in the association pair correspond to two quantitative indicators of the same optimal solution in the Pareto optimal solution set, ensuring the uniqueness and effectiveness of data matching.

[0080] Step S503: Determine the initial optimal first minimum value and the initial optimal second minimum value based on the minimum value association pair.

[0081] In this embodiment, a first round of screening is conducted on all minimum value association pairs to select those with a positive minimum vehicle task time limit margin. This screening condition prioritizes ensuring that there is no risk of vehicle task time limit exceeding the limit. If such a positive minimum value association pair exists, the absolute value of the difference between the minimum vehicle task time limit margin and the minimum single vehicle waiting time limit margin is calculated for each of these association pairs. The association pair with the smallest absolute value is selected, and the minimum vehicle task time limit margin in this pair is determined as the initial optimal first minimum value. The corresponding minimum single vehicle waiting time limit margin is determined as the initial optimal second minimum value. The smallest absolute value of the difference represents a more balanced control status between the two margin dimensions, which is the optimal choice in this case.

[0082] If no association pair with a positive minimum vehicle task time limit leeway is found after the first round of screening, meaning all association pairs have a risk of task time limit exceeding the limit, then a global numerical comparison is performed on the minimum vehicle task time limit leeway among all minimum value association pairs. The association pair with the largest value is selected, and the minimum vehicle task time limit leeway ...

[0083] The initial optimal first minimum value and the initial optimal second minimum value satisfy the following formula: in, To establish the indexes of the initial optimal first minimum and the initial optimal second minimum in the minimum-value association pair. To obtain the index of the associative pair that minimizes the expression within the parentheses, To obtain the index of the associative pair that maximizes the expression within the parentheses, For the first The minimum value is associated with the minimum vehicle task time limit margin when passing through the middle. For the first The minimum value is associated with the minimum waiting time limit margin for a single vehicle when passing through the middle. For the set of minimum value association pairs, For each minimum value associated with a vehicle task time limit margin greater than zero during passage, This is a minimum value association pair where there is no minimum vehicle task time limit margin greater than zero during passage.

[0084] It should be noted that in this step, the global minimum value in the set of minimum vehicle task time limit margins and its associated waiting time limit margins are selected as the initial optimal first minimum value and the initial optimal second minimum value. This is based on the "barrel effect" design concept, which prioritizes ensuring the service level of the most urgent vehicles and avoids any vehicle affecting the overall terminal operation due to severe task timeouts. Although this strategy may sacrifice some overall traffic efficiency, in port operation scenarios, ensuring the seriousness of task time limits and preventing excessive delays by individual vehicles from causing a chain reaction are the primary goals of intersection control. This design ensures that weight adjustments always focus on vehicles that most need priority passage, reflecting the "safety net" control philosophy, and is highly consistent with the actual operational needs of the terminal.

[0085] Step S504: Adjust the initial urgency weights according to the initial optimal first minimum value and the initial optimal second minimum value, and iterate to obtain the optimal passage time allocation set and the adjusted passage group.

[0086] The process of adjusting the initial urgency weights based on the initial optimal first minimum value and the initial optimal second minimum value, and iteratively obtaining the optimal passage time allocation set and the adjusted passage group, specifically includes the following sub-steps: Step S50401: When the initial optimal first minimum value is positive, calculate the absolute difference between the initial optimal first minimum value and the initial optimal second minimum value.

[0087] In this embodiment, the initial optimal first minimum value is first evaluated for positive or negative value. A positive value indicates that under the control scheme corresponding to the current Pareto optimal solution, all vehicles waiting to pass through the target intersection have no timeout issues, and the core passage needs of vehicles in the task dimension have been effectively guaranteed. At this point, the optimization objective of weight adjustment will shift from avoiding task timeouts to achieving a two-dimensional balance between task timeout margin and waiting timeout margin. For the determination result that the initial optimal first minimum value is positive, the absolute value difference operation is directly performed using the initial optimal first minimum value and its uniquely matching initial optimal second minimum value as calculation parameters. The absolute value of the difference between the two values ​​is calculated to obtain the absolute difference between the initial optimal first minimum value and the initial optimal second minimum value. The magnitude of this absolute difference is negatively correlated with the balance of the two-dimensional control state. The smaller the value, the closer the vehicle's task timeout margin and waiting timeout margin are, and the more balanced the two-dimensional passage control state.

[0088] Step S50402: With the goal of minimizing the absolute difference, the initial urgency weight is adjusted and iterated to obtain the optimal passage time allocation set and the adjusted passage group.

[0089] In this embodiment, the core optimization objective is to minimize the absolute difference between the initial optimal first minimum value and the initial optimal second minimum value. Dynamic adjustments and multiple rounds of iterative calculations are performed on the initial urgency weights to ultimately obtain the optimal passage time allocation set and adjusted release groups that adapt to the actual control needs of the target wharf intersection. The initial urgency weights include a first weight coefficient corresponding to the first urgency and a second weight coefficient corresponding to the second urgency, and the sum of these two coefficients is 1. Within a reasonable range of values ​​for these weight coefficients, the first weight coefficients can be adjusted according to a predetermined step size (e.g., with a step size of 0.05, within the range [0,1]). The corresponding second weighting coefficient is The two coefficients are fine-tuned. During the fine-tuning process, the weights can be adjusted to skew towards the task time limit margin or waiting time limit margin based on the changing trend of the absolute difference. After each adjustment of the weight coefficients, the aforementioned complete technical process of calculating vehicle passage urgency, sorting and dividing the control lane release queue, setting the initial population, iteratively solving the Pareto optimal solution set using the multi-objective optimization algorithm (NSGA-II), determining the initial optimal first minimum and the initial optimal second minimum, and calculating the absolute difference between the two is re-executed to obtain the absolute difference corresponding to the current round of weight adjustment. The absolute difference of this round is compared with the absolute difference of the previous iteration. The iterative operation of fine-tuning the weight coefficients, recalculating the entire process, and comparing the absolute difference is continuously executed until the global minimum absolute difference is obtained. At this point, the task time limit margin and the waiting time limit margin of the vehicle passage reach a dual-dimensional optimal equilibrium state. The urgency weight in this state is the adjusted optimal urgency weight. Based on this optimal urgency weight, the optimal release time allocation scheme within the Pareto optimal solution set of the corresponding iteration round is extracted and integrated to form the optimal passage time allocation set. Simultaneously, the vehicle passage urgency of all vehicles waiting to pass is recalculated according to this optimal weight, and the vehicle composition and sorting of the original release group are optimized and adjusted to obtain an adjusted release group adapted to the optimal passage time allocation. The convergence condition includes either the number of iterations reaching a preset threshold or the absolute difference decreasing to a preset minimum value.

[0090] Step S50403: When the initial optimal first minimum value is negative, calculate the absolute value of the initial optimal first minimum value.

[0091] In this embodiment, the initial optimal first minimum value is first assessed for its positive or negative value. A negative value indicates that under the control scheme corresponding to the current Pareto optimal solution, there are situations where the actual passage time of vehicles waiting to pass through the target intersection exceeds their task time limit. The core passage needs of the vehicles in the task dimension are not effectively met. In this case, the optimization objective of the weight adjustment will focus on avoiding vehicle task time limit overruns and reducing the degree of overruns in the task dimension. For the determination result that the initial optimal first minimum value is negative, the absolute value of the initial optimal first minimum value is directly performed to obtain the absolute value of the initial optimal first minimum value. The magnitude of this absolute value is positively correlated with the degree of overrun of the vehicle task time limit. The smaller the value, the shorter the overrun of the vehicle task time limit and the milder the overrun, and vice versa.

[0092] Step S50404: With the goal of minimizing the absolute value, adjust the initial urgency weight and iterate to obtain the optimal passage time allocation set and the adjusted passage group.

[0093] In this embodiment, the core optimization objective is to minimize the absolute value of the initial optimal first minimum value. Dynamic adjustments and multiple rounds of iterative calculations are performed on the initial urgency weights to address the vehicle task timeout problem and obtain a suitable optimal passage time allocation set and adjusted release groups. The initial optimal first minimum value is negative, and its absolute value reflects the degree of vehicle task timeout. The core optimization is to reduce this timeout degree by adjusting the weights until a globally minimum absolute value is obtained. During adjustment, the weight coefficients are gradient-shifted towards the second urgency of the task dimension, increasing the proportion of the second weight coefficient to highlight the control priority of vehicle task timeouts. After each fine-tuning of the initial urgency weight coefficients, the aforementioned complete technical process of calculating vehicle passage urgency, sorting and dividing the control lane release queue, setting the initial population, iteratively solving the Pareto optimal solution set using a multi-objective optimization algorithm, determining the initial optimal first minimum value and the initial optimal second minimum value, and calculating the absolute value of the initial optimal first minimum value is re-executed to obtain the corresponding absolute value after this round of weight adjustment. The absolute value of this round is compared with the absolute value of the previous iteration. Iterative operations of fine-tuning the weight coefficients to the task dimension, recalculating the entire process, and comparing absolute values ​​are continuously performed until the globally minimum absolute value is obtained. At this point, the timeout level of the vehicle task limit is minimized, or even a timeout-free control effect is achieved. The urgency weight in this state is the adjusted optimal urgency weight. Based on this optimal urgency weight, the optimal release time allocation scheme within the Pareto optimal solution set of the corresponding iteration round is extracted and integrated to form the optimal passage time allocation set. Simultaneously, the vehicle passage urgency of all vehicles waiting to pass is recalculated based on this optimal weight. The vehicle composition and sorting of the original release group are then optimized and adjusted to obtain an adjusted release group adapted to the optimal passage time allocation.

[0094] Through nested iterations of outer-layer weight optimization and inner-layer multi-objective optimization, an adaptive closed-loop mechanism of "goal-oriented - solution generation - effect evaluation - parameter correction" is formed. The outer-layer weight adjustment is responsible for calibrating the relative importance of the two urgency dimensions, while the inner-layer optimization is responsible for generating the optimal release solution under the weights. The two work together and approach each other round by round, ultimately enabling the control strategy to simultaneously meet the dual requirements of avoiding task timeouts and ensuring fairness in waiting.

[0095] Step S6: Control the traffic at the intersection based on the inter-group release order, the optimal passage time allocation set, and the adjusted release group.

[0096] In this embodiment, following the pre-set inter-group release order, the optimal passage time allocation set obtained through iteration is retrieved to match the passage time parameters for each adjusted release group. According to the vehicle composition and sorting of the adjusted release group, each release group is sequentially controlled in a round-robin manner. The passage time period of each release group is adjusted according to the corresponding passage time, guiding vehicles in the group to pass through the intersection in an orderly manner according to the urgency of passage, thereby achieving refined and orderly traffic management at the target intersection of the wharf.

[0097] Then, by re-acquiring the vehicles waiting to pass in the newly added intersection queuing sequence, the information of the vehicles waiting to pass, and the initial urgency weight, the initial weight here adopts the adjusted initial urgency weight mentioned above, and thus a new iterative loop control is carried out.

[0098] It should be noted that although this method involves two-layer iteration (outer layer weight step adjustment, inner layer multi-objective optimization), in practical engineering applications, the real-time requirements can be ensured through the following mechanisms: First, a rolling optimization strategy is adopted, optimizing only one future control cycle (e.g., 5 passage rounds) each time, and the initial weights of the current cycle directly inherit the optimization results of the previous cycle, so that the outer layer weight adjustment can usually converge quickly within 3-5 steps; Second, the iterative process of the multi-objective optimization algorithm can make full use of historical Pareto front solutions as the initial population, which greatly reduces the computation time.

[0099] This invention addresses the unique scenario of intersections within a wharf. Its core design logic revolves around "prioritizing task urgency to prevent chain-reaction risks," fundamentally differing from the paradigm of general urban transportation that prioritizes overall traffic efficiency. In urgency calculation, the task dimension (secondary urgency) has no upper limit and dominates weighted fusion, aiming to quantify timeout levels and prioritize the clearance of delayed vehicles. When all vehicles exceed their task timeouts, a unified value is assigned, with the waiting dimension determining the order for efficient clearance. Optimization focuses on the most urgent individual vehicles (the "weakest link" effect), as the delay of a single critical vehicle can trigger a global paralysis; sacrificing some overall efficiency for the stability of the operational chain is a necessary trade-off. The release order between groups strictly adheres to traffic safety rules and cannot be dynamically adjusted. The time allocation matrix only defines the time allotted to each release group; actual releases are executed sequentially according to a preset order, ensuring no safety conflicts. Other designs, such as the theoretically passable vehicle number estimation providing only a general range for the initial population and the unified sorting of new and old vehicles, are all based on actual wharf operation experience and have been validated.

[0100] like Figure 2 As shown, in another aspect, the present invention also provides a traffic control system for an intersection within a wharf, comprising: a processor, an input device, an output device, and a memory, wherein the processor, the input device, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute relevant steps of a relevant embodiment of a traffic control method for an intersection within a wharf according to the present invention.

[0101] This invention provides a traffic control system for intersections within a wharf. The functional components can be integrated into a single processing unit, or each component can exist independently, or two or more components can be integrated into one unit. The integrated components can be implemented in hardware or software.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A traffic control method for intersections within a wharf, characterized in that, The method includes: Obtain information on vehicles waiting to pass and their initial urgency weights in multiple controlled lanes at the target intersection within the wharf. The vehicle passage urgency of the vehicles to be passed is calculated based on the information of the vehicles to be passed and the initial urgency weight; The vehicles waiting to pass are sorted according to their urgency to obtain the release queue of the controlled lane, and the release queue is then grouped to obtain the release group. Based on the pre-set inter-group release order and the vehicle passage urgency, a multi-objective optimization algorithm is used to calculate the Pareto optimal solution set of the release group; Based on the Pareto optimal solution set, the initial urgency weights are adjusted and iterated to obtain the optimal passage time allocation set and the adjusted passage group, including: The minimum set of vehicle task time limit margin and the minimum set of single vehicle waiting time limit margin during passage are obtained based on the Pareto optimal solution set. Extract minimum value association pairs from the minimum set of vehicle task time limit leeway during passage and the minimum set of single vehicle waiting time limit leeway during passage, respectively; Based on the minimum value association pair, determine the initial optimal first minimum value and the initial optimal second minimum value; The initial urgency weights are adjusted and iterated based on the initial optimal first minimum value and the initial optimal second minimum value to obtain the optimal passage time allocation set and the adjusted passage group; Traffic at the intersection is controlled based on the inter-group release order, the optimal passage time allocation set, and the adjusted release group.

2. The traffic control method for intersections within a wharf according to claim 1, characterized in that, The information on vehicles waiting to pass includes waiting time limit, accumulated waiting time, and task time limit. The calculation of the vehicle passage urgency based on the information on vehicles waiting to pass and the initial urgency weight includes: The first urgency level of each of the vehicles waiting to pass is determined based on the waiting time limit and the already waited time. The second urgency level of each of the vehicles waiting to pass is determined based on the task time limit and the waiting time. The first urgency and the second urgency are weighted and fused using the initial urgency weight to obtain the vehicle urgency of each vehicle to be passed.

3. The traffic control method for intersections within a wharf according to claim 2, characterized in that, Determining the first urgency level of each vehicle waiting to pass based on the waiting time limit and the already waited duration includes: The waiting time limit and the waiting time duration are correlated to obtain the single-vehicle waiting time limit and single-vehicle waiting time duration for each of the vehicles waiting to pass. The first urgency level of each vehicle waiting to pass is obtained by calculating the ratio of the waiting time of the vehicle to the waiting limit of the vehicle.

4. The traffic control method for intersections within a wharf according to claim 1, characterized in that, The calculation of the Pareto optimal solution set for the release group based on the pre-set inter-group release order and the vehicle passage urgency using a multi-objective optimization algorithm includes: Set the number of rounds and the duration of each round; An initial population is set based on the number of passage rounds, the duration of a single round, and the urgency of vehicle passage; Based on the pre-set inter-group release order and the initial population, a multi-objective optimization algorithm is used to iteratively obtain the Pareto optimal solution set for the release group.

5. The traffic control method for intersections within a wharf according to claim 4, characterized in that, The initial population setting based on the number of passage rounds, the duration of a single round, and the urgency of vehicle passage includes: The total control time is calculated based on the number of passage rounds and the duration of each round. Based on the total control duration, the theoretical number of vehicles that can pass through each of the controlled lanes is estimated. Based on the vehicle traffic urgency, the passability urgency of each of the controlled lanes is calculated using the theoretically passable vehicles. The maximum value is selected after grouping the passability urgency according to the release group to obtain the passability reference urgency of the release group; The total passage time for each of the release groups is allocated according to the ratio of the passable reference urgency and the total control time. Using the single-round duration as a constraint, the initial population is determined based on the total passage time of a single group and the passage round number.

6. The traffic control method for intersections within a wharf according to claim 4, characterized in that, The Pareto optimal solution set for the release group obtained by iteratively using a multi-objective optimization algorithm based on a pre-set inter-group release order and the initial population includes: Based on the pre-set inter-group release order, calculate the minimum vehicle task time limit margin and the minimum single vehicle waiting time limit margin for individuals in the initial population during passage. The initial population is sorted non-dominated with the goal of maximizing the minimum vehicle task time limit margin and the minimum single vehicle waiting time limit margin during passage. Based on the non-dominated sorting, a multi-objective optimization algorithm is used to iterate until the pre-set convergence condition is met, and a Pareto optimal solution set is obtained.

7. The traffic control method for intersections within a wharf according to claim 1, characterized in that, The step of adjusting the initial urgency weight based on the initial optimal first minimum value and the initial optimal second minimum value, and iteratively obtaining the optimal passage time allocation set and the adjusted passage group, includes: When the initial optimal first minimum value is determined to be positive, the absolute difference between the initial optimal first minimum value and the initial optimal second minimum value is calculated; With the goal of minimizing the absolute difference, the initial urgency weights are adjusted and iterated to obtain the optimal passage time allocation set and the adjusted passage group.

8. The traffic control method for intersections within a wharf according to claim 7, characterized in that, The step of adjusting the initial urgency weight based on the initial optimal first minimum value and the initial optimal second minimum value, and iteratively obtaining the optimal passage time allocation set and the adjusted passage group, further includes: When the initial optimal first minimum value is determined to be negative, the absolute value of the initial optimal first minimum value is calculated; With the goal of minimizing the absolute value, the initial urgency weights are adjusted and iterated to obtain the optimal passage time allocation set and the adjusted passage group.

9. A traffic control system for intersections within a wharf, characterized in that, include: The system includes a processor, an input device, an output device, and a memory, all interconnected, wherein the memory stores a computer program comprising program instructions, and the processor is configured to invoke the program instructions to execute a traffic control method for an intersection within a dock as described in any one of claims 1 to 8.

Images