Charging pile dynamic distribution method under multi-forklift collaborative operation scene

Abstract

The invention discloses a charging pile dynamic distribution method in a multi-forklift collaborative operation scene, and relates to the technical field of intelligent logistics dispatch, and the method comprises the steps: S1, based on a current traffic flow prediction result, generating the predicted arrival time of a forklift at each charging pile, and S2, according to the predicted arrival time, determining the charging pile according to the predicted arrival time. The method comprises the steps of S1, calculating the actual waiting duration and the corresponding priority weight of each forklift, S3, dynamically regulating and controlling the charging queuing sequence of the forklifts based on the priority weights, and S4, sending the regulated queuing sequence to a target forklift to guide the target forklift to complete charging pile distribution. According to the charging pile dynamic allocation method in the multi-forklift collaborative operation scene, the charging queuing sequence is regulated and controlled according to the current traffic flow prediction result, so that the problems of congestion and overlong waiting time are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics scheduling technology, and in particular to a method for dynamically allocating charging piles in a scenario where multiple forklifts work together. Background Art

[0002] In modern intelligent warehousing and logistics systems, electric forklifts are widely used in multi-forklift collaborative operations due to their high efficiency and environmental friendliness. With the increasing number of forklifts and the increasing workload, the frequent need for charging has made charging resource scheduling a key factor restricting system efficiency.

[0003] Currently, commonly used forklift charging scheduling strategies mostly adopt fixed allocation or first-come, first-served charging methods, which can meet basic needs when the system is small. However, in large-scale scenarios, when multiple forklifts request charging at the same time and the number of charging piles is limited, it is very easy to cause congestion in the charging area and excessive waiting time for charging, which in turn affects the execution of forklift tasks and the overall logistics scheduling efficiency. In addition, existing charging queuing strategies generally lack the ability to comprehensively analyze information such as the remaining power of the forklift, the urgency of the task, and the current traffic flow, and do not introduce predictive models to predict future traffic conditions or charging demand trends, making it difficult to adjust the queue order or optimize the allocation of charging resources in a timely manner. This static, passive charging scheduling mechanism shows obvious limitations in highly dynamic operating environments. Summary of the Invention

[0004] The purpose of the present invention is to provide a dynamic allocation method for charging piles in a multi-forklift collaborative operation scenario, and to adjust the charging queue sequence according to the current traffic flow prediction results to solve the problems of congestion and long waiting time.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario, the method comprising:

[0006] S1. Generate the estimated arrival time of the forklift at each charging station based on the current traffic flow prediction results. This includes using the mapping system to obtain the path distance from the forklift's current location to each charging station, giving the current actual average speed of each path based on traffic sensor data or historical traffic flow, calculating the estimated arrival time of the forklift at each charging station, and storing all arrival times of each forklift in a matrix for subsequent analysis.

[0007] S2. Calculate the actual waiting time and corresponding priority weight of each forklift based on the estimated arrival time. This includes counting the number of forklifts currently or expected to arrive at each charging station, setting the maximum number of charges that the station can handle simultaneously as a constant, and calculating a priority score for each forklift.

[0008] S3. Dynamically adjust the charging queue order of forklifts based on priority weights, including collecting the priority score of each forklift, calculating the sum of the priority scores of all forklifts, calculating the probability of each forklift being dispatched and being noticed, and sorting the forklifts from high to low according to the probability of dispatching and being noticed, which will be used as the queue order for the next round of charging;

[0009] S4. Send the adjusted queue order to the target forklift to guide it to complete the charging pile allocation, including obtaining the current position coordinates of all forklifts, obtaining the position coordinates of all target charging piles, calculating the straight-line distance between each pair of forklifts and piles, performing nearest neighbor matching on the current queue order, determining the final allocation relationship, and sending the allocation result to the corresponding forklift through the network interface.

[0010] Preferably, the specific method for calculating the estimated arrival time of the forklift to each charging pile in S1 includes using a map system or a path planning engine to obtain the shortest travel path from the current position of the forklift to the target charging pile, calculating the length of each section in the path and summing them to obtain the total path distance, calling a traffic flow sensor or a historical traffic database to obtain the real-time traffic speed of each section of the road on the path, and dividing the path distance from the current position of the forklift to the charging pile by the actual traffic speed of the current section to obtain the estimated arrival time of the forklift to each charging pile.

[0011] Preferably, the specific method for calculating the priority score in S2 includes, for each charging pile, counting the number of forklifts currently or expected to arrive at the charging pile in queue, setting a standard value for each charging pile, indicating the number of forklifts that the charging pile can simultaneously accept and process at the same time, and converting the current queue number of the charging pile where the forklift is located into the ratio of the maximum number of forklifts that the charging pile can serve, and then using this ratio to derive a priority score, recording this score for each forklift, and sorting them according to the score. Forklifts with higher scores will be given priority in subsequent scheduling to enter the front of the queue.

[0012] Preferably, the specific determination method of setting a standard value for each charging pile in S2 includes marking the maximum number of interfaces for connecting forklifts at the same time on each charging pile device when it leaves the factory as the standard value set for the charging pile.

[0013] Preferably, the specific method for calculating the probability of each forklift being paid attention to in scheduling in S3 includes collecting the priority scores calculated for each forklift in S2, summing up the priority scores of all the above forklifts, and finding their sum. The priority score of each forklift is divided by the total score just obtained to obtain a numerical value in the form of a percentage. This numerical value represents the probability of the forklift being selected in the current round of scheduling. According to the probability of attention of each forklift, they are sorted from high to low. The sorting result will be used as the order of the charging queue in the next round. The forklifts with higher probabilities are ranked first and get priority in charging safety.

[0014] Preferably, the specific calculation method for calculating the straight-line distance between each pair of forklifts and charging piles in S4 includes reading the two-dimensional coordinate position of each forklift in the current storage site, reading the fixed coordinate values of all charging piles on the site map, and for each pair of forklifts and target charging piles, subtracting the values of the forklift position and the pile position in two directions respectively to obtain the difference in the horizontal and vertical directions, squaring the differences in the two directions just obtained, and then adding them, and taking the square root of the result to obtain the straight-line path length between the forklift and the corresponding charging pile, and executing the above calculation steps one by one for all combinations of forklifts and all target charging piles, and finally forming a data table containing all distance values, according to which the straight-line distances between all forklifts and each pile are obtained.

[0015] Preferably, the two-dimensional coordinate position of the forklift in the current storage yard in S4 is provided in real time by a positioning module on the forklift.

[0016] Preferably, the positioning module in S4 includes a laser radar, an inertial navigation system or UWB.

[0017] Preferably, after obtaining the straight-line distances between all forklifts and the piles in S4, a closest distance matching is performed on each forklift, and the target pile closest to it is preferentially selected as the final scheduling object, and a matching result is generated.

[0018] Preferably, in the process of performing the nearest neighbor matching between forklifts and charging piles in S4, if there are two or more forklifts with the same charging pile at the same closest distance, the one with the higher priority is selected according to the priority score of each forklift to be paired with the charging pile.

[0019] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0020] The dynamic allocation method of charging piles in the scenario of multi-forklift collaborative operation generates the estimated arrival time of forklifts at each charging pile based on the current traffic flow prediction results. According to the estimated arrival time, the actual waiting time and corresponding priority weight of each forklift are calculated. Based on the priority weight, the charging queue order of the forklifts is dynamically adjusted and the adjusted queue order is sent to the target forklift to guide it to complete the charging pile allocation. The arrival time of the forklifts at each charging pile is dynamically evaluated to avoid resource waste and scheduling delays caused by static queuing strategies. Based on the comprehensive priority scoring mechanism, forklifts with low remaining power and urgent tasks are given higher priority. Charging priority, thereby alleviating the problem of unreasonable resource allocation caused by "first come, not necessarily the most urgent", and probabilistically sorting the forklifts requesting charging, realizing real-time optimization of the queue order, improving the overall operation flow efficiency of the charging area, while taking into account efficiency, avoiding low-priority forklifts from being unable to charge for a long time, enhancing the fairness and stability of the system, automatically calculating the shortest navigation path between forklifts and charging piles and performing nearest neighbor allocation, realizing intelligent dispatching and path guidance of charging tasks, effectively reducing the workload of dispatchers, and adjusting the charging queue order according to the current traffic flow forecast results to solve the problems of congestion and long waiting time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] As shown in the figure, the present invention provides a technical solution: a dynamic allocation method for charging piles in a multi-forklift collaborative operation scenario, the method comprising S1, based on the current traffic flow prediction result, generating an estimated arrival time of a forklift at each charging pile, including using a map system to obtain the path distance from the current position of the forklift to each charging pile, giving the current actual average speed of each path based on traffic sensor data or historical traffic flow, calculating the estimated arrival time of the forklift to each charging pile, and storing all the arrival times of each forklift in a matrix for subsequent analysis; S2, according to the estimated arrival time, calculating the actual waiting time and corresponding priority weight of each forklift, including counting the number of forklifts that are currently or are expected to arrive at the point for each charging pile, setting the number of forklifts that are currently or are expected to arrive at the point at the same time for the pile. The maximum number of charges that can be processed is a constant, and a priority score is calculated for each forklift; S3. Based on the priority weight, the forklift charging queue order is dynamically adjusted, including collecting the priority score of each forklift, calculating the sum of the priority scores of all forklifts, calculating the probability of each forklift being dispatched and being paid attention to, and sorting the forklifts from high to low according to the probability of each forklift being dispatched and paid attention to as the new round of charging queue order; S4. The adjusted queue order is sent to the target forklift to guide it to complete the charging pile allocation, including obtaining the current position coordinates of all forklifts, obtaining the position coordinates of all target charging piles, calculating the straight-line distance between each pair of forklifts and piles, performing nearest neighbor matching on the current queue order, determining the final allocation relationship, and sending the allocation result to the corresponding forklift through the network interface.

[0024] This embodiment aims to efficiently dispatch forklifts in multiple charging pile scenarios and constructs a dynamic charging pile allocation mechanism that integrates path planning, traffic perception, and intelligent scheduling algorithms. First, in step S1, by integrating a map navigation system (such as GIS or Baidu Map API) with traffic flow monitoring data, the current location of each forklift and the length of its path to multiple charging piles are obtained in real time. Combined with information from traffic sensors (such as geomagnetic coils, cameras, radar equipment) or historical traffic databases, the average speed of each path is estimated, and the travel time of each path is dynamically calculated based on this, thereby obtaining the estimated arrival time of each forklift to different charging piles. The estimated arrival times of all forklift-pile pairs form a two-dimensional matrix for subsequent scheduling analysis. In step S2, the system will count the number of forklifts currently or about to arrive at each charging pile based on this matrix and the real-time forklift scheduling status. The concurrent charging capacity of each pile is used as a constraint parameter, and situations where the capacity is exceeded are handled by postponing the arrangement or reallocation. The priority score calculation for each forklift can refer to the following model: Priority score = expected waiting time × distance factor × current power weight × urgency coefficient, where each weight can be adjusted according to the specific on-site operation strategy. This takes into account both fairness (waiting time) and efficiency (path cost) and demand (power / task urgency). In step S3, the system adopts a quasi-soft scheduling mechanism to normalize the priority scores of all forklifts and calculate the probability of being scheduled, that is, the relative importance of each forklift in the overall scheduling. It is then sorted in descending order according to this probability value to form a new charging queue. Compared with the simple FCFS (first come, first served) strategy, this sorting method has higher system adaptability and scheduling fairness. In step S4, the system further considers the geographical distance factor and, through the nearest neighbor priority matching principle, prioritizes the high-priority forklifts to the nearest idle charging piles to ensure their shortest path. The system uses a GPS module to collect the real-time coordinates of the forklifts, and combines this with the fixed position coordinates of the charging piles in the map to calculate the Euclidean distance or weighted distance considering path resistance for each forklift-pile pair, and uses this as the basis for matching optimization. The scheduling results are ultimately sent to the forklift control system via industrial communication protocols (such as MQTT, Modbus, CAN bus, etc.), guiding it to the target pile along the assigned path to complete the charging operation. The entire system is based on a modular design and can flexibly connect to various map services, scheduling algorithms, and communication protocols. It has good scalability and adaptability, and is suitable for deployment in multi-forklift intensive operating environments such as automated warehousing, port logistics, and large-scale manufacturing bases.

[0025] This implementation significantly optimizes the efficiency of charging resource allocation in multi-forklift environments. Dynamic traffic status perception and estimated arrival time estimation effectively improve the real-time and accuracy of allocation decisions. A priority scoring system, which comprehensively considers factors such as waiting time, path length, and pile load, helps achieve a fair and reasonable scheduling mechanism. A sorting strategy based on the probability of being noticed by the dispatcher prevents extreme waiting times or resource contention. Finally, a nearest neighbor matching algorithm ensures the shortest forklift path, reduces idle time, and improves overall system efficiency.

[0026] The specific method for calculating the estimated arrival time of the forklift to each charging pile in S1 includes using a map system or a path planning engine to obtain the shortest travel path from the current position of the forklift to the target charging pile, calculating the length of each section in the path and summing them to obtain the total path distance, calling a traffic flow sensor or a historical traffic database to obtain the real-time traffic speed of each section of the road on the path, and dividing the path distance from the current position of the forklift to the charging pile by the actual traffic speed of the current section to obtain the estimated arrival time of the forklift to each charging pile.

[0027] This embodiment expands the calculation logic of the estimated arrival time of the path in claim 1, and refines the segmented acquisition mechanism of the path distance and speed. The system first calls the map service interface or the path planning engine, and identifies the shortest or optimal path to each charging pile based on the current forklift's geographic coordinate information and the location of the charging pile. The path is discretized into multiple continuous sections, each of which corresponds to a spatial path with a clear geographical starting point and end point. For each section in the path, the system extracts its length value one by one, and performs a sum operation to obtain the total length of the path for the forklift to reach a specific charging pile. At the same time, the system obtains the real-time traffic speed data of each section of the path by calling the traffic flow perception network (such as video detection, radar speed measurement, geomagnetic sensor, etc.) or accessing the historical traffic speed database. These speed information may be stored in the edge server or cloud platform with the section ID as the index, and the system accesses it through API or local query. Then, for each route segment, the system divides the distance by the corresponding speed to calculate the travel time for that segment. The system then sums the travel times for all segments to arrive at the estimated total arrival time for the forklift from its current location to the designated charging station. This calculation method effectively accounts for the heterogeneity of actual traffic conditions across different road segments, enabling more accurate arrival time predictions and thus improving the accuracy of subsequent scheduling mechanisms.

[0028] This embodiment refines the calculation method of the estimated arrival time of the forklift, and uses segmented path analysis combined with real-time traffic data to improve the prediction accuracy. Specifically, the system first uses a map system or a path planning engine (such as Dijkstra algorithm or A* algorithm) to determine the path from the forklift's current position P to the next location. f To the target charging station P cThe shortest path, which consists of several consecutive segments, is recorded as:

[0029] Path = {l1, l2, ..., l n};

[0030] Among them, each path segment l i The corresponding length is d i , the passing speed is v i , then the travel time t i for:

[0031]

[0032] The total travel time of the path, that is, the estimated time T for the forklift to reach the charging station from the current location, is the sum of the time of all sections:

[0033]

[0034] Among them, the length of the path segment d i Obtained through the map API, the traffic speed v i The prediction method takes into account the differences in traffic conditions on different road sections and can dynamically reflect the current road network congestion and traffic capacity.

[0035] All calculation results will be summarized into a forklift-charging station estimated arrival time matrix:

[0036]

[0037] Where m represents the number of forklifts, k represents the number of charging piles, and T ij It represents the estimated arrival time of the i-th forklift to the j-th charging station.

[0038] Through the above formula and matrix modeling method, the system can realize multi-point arrival time prediction of large-scale forklift groups, providing efficient and accurate data support for subsequent scheduling optimization and charging pile matching.

[0039] By segmenting the route and calculating travel times for each segment, this implementation significantly improves the accuracy of forklift arrival time predictions. Compared to traditional estimates based on overall average speed, the segmented speed calculation model better reflects the dynamic changes in traffic conditions across different road sections, making it particularly suitable for complex logistics environments with traffic bottlenecks or multiple vehicles crossing paths. Furthermore, this method supports multi-source data fusion and can flexibly access a variety of traffic sensor systems and historical databases, improving the overall system's scalability and data compatibility.

[0040] The specific method for calculating the priority score in S2 includes counting the number of forklifts currently or expected to arrive at the charging pile in the queue for each charging pile, setting a standard value for each charging pile, indicating the number of forklifts that the charging pile can simultaneously accept and process at the same time, and converting the current queue number of the charging pile where the forklift is located into the ratio of the maximum number of forklifts that the charging pile can serve. This ratio is then used to derive a priority score, and this score is recorded for each forklift, and the forklifts are sorted according to the score. The forklifts with higher scores will be given priority in the subsequent scheduling to enter the queue at the front position.

[0041] In this embodiment, the calculation mechanism of the priority score is based on the dynamic comparison logic of the charging pile load and service capacity. Specifically, the system first calculates the priority score for each charging pile c j Set a service capacity upper limit C j , which represents the maximum number of forklifts that the pile can handle at the same time. Then, the system counts the current or predicted number of forklifts that will enter the queue at the pile Q j , used to measure the current load level of the charging pile.

[0042] For each forklift f i , the system is based on its target charging pile c j , calculate its load ratio R ij :

[0043]

[0044] The larger the ratio, the higher the queue pressure the forklift will face, and therefore it should be given a higher priority in the scheduling system. i It can be modeled through inverse relationships, common forms such as:

[0045]

[0046] Among them, ∈ is a very small positive number (such as 0.01), which is used to avoid the situation where the denominator is 0 and adjust the scoring scale. The system will set the score S of all forklifts i The scores are recorded and sorted from high to low to generate a new round of dispatch priority order. This mechanism ensures that forklifts with greater queue pressure are ranked higher in the order, thereby reducing overall system congestion.

[0047] By introducing a priority scoring mechanism based on load ratio, this implementation can dynamically reflect the real-time congestion situation of each charging station and rationally guide the distribution of forklift traffic, thereby optimizing resource utilization efficiency. This method is simple and efficient, facilitating rapid system execution. Furthermore, through the adjustment of the scoring model, it can flexibly adapt to the different on-site charging station capacity and operation rhythm. Furthermore, this mechanism has good fairness and scalability, and can be integrated into more complex scheduling algorithms or combined with other factors (such as power urgency, task priority, etc.) for multi-factor scheduling optimization.

[0048] In S2, a standard value is set for each charging pile. The specific determination method includes marking the maximum number of interfaces for connecting forklifts at the same time on each charging pile device when it leaves the factory, which is the standard value set for the charging pile.

[0049] The standard value C of the "maximum service capacity of the charging pile" in this embodiment is j The data source is clearly defined, that is, it is set according to the number of physical interfaces provided by the charging pile during the design and manufacturing stage. When each charging pile device leaves the factory, its product technical manual or device nameplate usually indicates the maximum number of forklifts that can be connected concurrently (for example, the maximum number of output ports is 4, which means that up to 4 forklifts can be charged simultaneously). During the deployment or initialization phase, the system reads this data and records it as the standard service capability of the charging pile, namely:

[0050] C j =MaxInterfaceCount j ;

[0051] Among them, C j Indicates charging pile c j The standard value of MaxInterfaceCount j is the maximum number of charging ports. This parameter is then used in the priority scoring mechanism to calculate the load ratio. The denominator of the value. Because this value is based on actual hardware capabilities, it offers advantages such as strong stability, high operability, and the absence of dynamic estimation. By using the maximum number of interfaces provided by the charging pile as the basic capacity parameter in the scheduling system, the scheduling logic is aligned with the actual pile capacity, preventing irrational resource allocation due to estimation errors. This also facilitates rapid system adaptation when new equipment is connected or replaced, ensuring excellent engineering feasibility and ease of maintenance.

[0052] This method simplifies parameter initialization and improves the consistency and reliability of system configuration by directly using the factory-marked maximum number of interfaces as the standard value. This eliminates the need for complex data modeling or on-site measurements to accurately determine the service capacity of each charging station, facilitating rapid system deployment. Furthermore, since the standard value originates from the equipment production process, it is authoritative and unified, facilitating subsequent maintenance and upgrade management. This setting method also effectively supports modular and standardized expansion of the system, enhancing the engineering practicality of the forklift dispatching platform.

[0053] The specific method for calculating the probability of each forklift being selected for scheduling in S3 includes collecting the priority scores calculated for each forklift in S2, summing up the priority scores of all the above forklifts, and dividing the priority score of each forklift by the total score just obtained to obtain a percentage value. This percentage value represents the probability of the forklift being selected in the current round of scheduling. According to the probability of attention of each forklift, they are sorted from high to low. The sorting result will be used as the order of the charging queue in the next round. The forklifts with higher probabilities are ranked first and get priority in charging arrangements.

[0054] This implementation aims to enable the scheduling system to probabilistically sort the charging arrangements among multiple forklifts by normalizing the priority scores, thereby building a more flexible and differentiated weight-controlled charging queue mechanism. Specifically, the system first summarizes the priority score S of each forklift in the previous stage (i.e., step S2). i , then calculate the total priority score S total :

[0055]

[0056] Where m is the total number of forklifts currently involved in the dispatch. Next, for each forklift f i , the probability of its scheduling being paid attention to is P i It is given by:

[0057]

[0058] The probability value P i It reflects the importance of the forklift in the whole scheduling sequence. i All forklifts are sorted from high to low, and the sorted results form the next round of charging waiting queues. Because the sorting is based on probability weights, the system can prioritize high-scoring forklifts while appropriately adjusting the scheduling positions of other forklifts, preventing resource bias caused by extreme sorting. This mechanism allows for more fine-grained scheduling control in actual operation, supporting various strategies such as rotation, polling, and non-preemptive scheduling. It also provides data support for the further introduction of machine learning scheduling or real-time feedback scheduling.

[0059] This method provides a scheduling and sorting mechanism based on normalized priority scores, ensuring that each forklift receives a "probability of attention" commensurate with its demand intensity based on its comprehensive evaluation score. This mechanism not only ensures fair allocation but also improves the interpretability and controllability of the scheduling strategy. Through probabilistic sorting, the system avoids the resource monopoly issues that can arise from traditional hard sorting and can adapt to actual operating conditions such as load fluctuations and temporary task insertions, improving the dynamic responsiveness of the overall scheduling system.

[0060] The specific calculation method for calculating the straight-line distance between each pair of forklifts and charging piles in S4 includes reading the two-dimensional coordinate position of each forklift in the current storage site, reading the fixed coordinate values of all charging piles on the site map, and for each pair of forklifts and target charging piles, subtracting the values of the forklift position and the pile position in two directions respectively to obtain the difference in the horizontal and vertical directions, squaring the differences in the two directions just obtained, and then adding them together, and taking the square root of the result to obtain the straight-line path length between the forklift and the corresponding charging pile. The above calculation steps are performed one by one for all combinations of forklifts and all target charging piles, and finally a data table containing all distance values is formed according to the straight-line distances between all forklifts and each pile.

[0061] This implementation is used to accurately calculate the two-dimensional Euclidean distance between the forklift and each target charging pile during the charging pile allocation phase, thereby providing a data basis for subsequent matching optimization. First, the system obtains the current position coordinates of each forklift in the storage yard from the forklift positioning system, recorded as (x f ,y f ), and at the same time call the map database or preset parameters to obtain the fixed installation coordinates (x c ,y c ).

[0062] Then, for each forklift-charging station pair, the system performs the following calculation steps:

[0063] Calculate the horizontal distance difference: Δx = x f -x c ;

[0064] Calculate the vertical distance difference: Δy = y f -y c ;

[0065] Perform the square sum operation: D 2 =Δx 2 +Δy 2 ;

[0066] Taking the square root gives the straight-line distance:

[0067] The above calculation results are organized into a distance matrix between the forklift and all charging piles:

[0068]

[0069] Among them, m is the number of forklifts, k is the number of charging piles, D ij represents the straight-line distance between the i-th forklift and the j-th charging station. This matrix will serve as the basic input for the nearest neighbor matching algorithm (such as minimum distance matching or Hungarian algorithm).

[0070] By employing a two-dimensional Euclidean distance calculation method, this implementation efficiently and accurately obtains the minimum physical path length between a forklift and a charging station. Compared to complex distance estimation methods based on path planning engines, this method is computationally simple and resource-efficient, making it suitable for scenarios involving the parallel scheduling of large fleets of forklifts and multiple charging stations. Furthermore, the resulting matrix facilitates rapid integration with scheduling results, enabling the generation of optimal pairing solutions with low latency. This method is also highly versatile and scalable, making it widely applicable to various warehouse automation and unmanned vehicle scheduling systems.

[0071] In S4, the two-dimensional coordinate position of the forklift in the current storage site is provided in real time by the positioning module on the forklift. In this embodiment, real-time positioning technology is introduced in the distance calculation link between the forklift and the charging pile to ensure the accuracy and timeliness of the forklift position information used. Specifically, each forklift is equipped with a positioning module, which can realize real-time collection of its current location according to different technology types. These positioning modules can be GNSS (such as GPS or Beidou), UWB (ultra-wideband), RFID positioning systems, or indoor high-precision positioning technologies such as visual SLAM (simultaneous localization and mapping). The positioning module sends data containing the current spatial position of the forklift to the scheduling system regularly or on demand. After receiving the information, the system parses it into a standard two-dimensional coordinate form. In this way, without relying on manual measurement or preset static points, the system can automatically and dynamically obtain the precise position of the forklift in the storage environment. This positioning information is directly used in the straight-line distance calculation step in S4 to form a coordinate pair with the fixed position of the charging pile, thereby realizing path evaluation based on real-time data. Since forklifts may move frequently and their paths may change dynamically in storage sites, the use of real-time positioning can significantly improve the timeliness and accuracy of scheduling responses, avoiding pairing errors or waste of resources caused by using outdated location information.

[0072] The positioning module in S4 includes lidar, inertial navigation system or UWB. This embodiment clarifies the key technical solutions used by the positioning module on the forklift to provide real-time two-dimensional coordinates, including lidar (LiDAR), inertial navigation system (INS) and ultra-wideband (UWB) positioning system. These three technologies each have their own advantages and can be flexibly combined or deployed independently according to different operating environments and accuracy requirements. LiDAR scans the surrounding environment through a high-speed rotating laser beam to build a high-precision point cloud map, and matches it with the preset map based on environmental characteristics to achieve centimeter-level positioning accuracy. It is suitable for indoor storage environments with clear spatial structures and a large number of reflective surfaces, and has extremely high spatial resolution capabilities. The inertial navigation system mainly relies on gyroscopes and accelerometers to collect and integrate data such as the acceleration and angular velocity of the forklift in real time to obtain relative displacement information. The inertial navigation system has self-contained characteristics, does not rely on external signals, and performs excellent performance in high-frequency dynamic response. It is suitable for application scenarios with severe signal shielding or extremely high real-time requirements. The UWB positioning system uses low-power broadband pulse signals, deploys multiple anchor points (base stations) on the site, and utilizes time difference of arrival (TDOA) or time of arrival (TOA) measurement methods to achieve high-precision two-dimensional or even three-dimensional positioning. UWB signals have strong anti-interference capabilities and are suitable for complex electromagnetic environments and high-density storage operation areas. In actual applications, the system can select the most appropriate technical solution based on positioning accuracy requirements, deployment environment characteristics, and cost considerations. For example, LiDAR is suitable for automatic forklift systems with high visual navigation requirements, UWB is suitable for multi-forklift collaboration scenarios with many devices but irregular spaces, and INS is suitable for fast-moving forklifts with high dynamics or partial occlusion. The three can also work together to form a fusion navigation system to improve the overall robustness and environmental adaptability of the system.

[0073] By employing LiDAR, inertial navigation systems, or UWB as the core technologies for the forklift positioning module, the system achieves high-precision, high-stability real-time positioning capabilities, greatly improving the accuracy of distance assessment and path planning between the forklift and the charging station. At the same time, the compatible design that supports multiple positioning technologies enhances the system's deployment flexibility and adapts to a variety of application environments, from open warehouses to enclosed workshops. Furthermore, the combined use of different positioning methods helps overcome the limitations of a single system, which is susceptible to factors such as occlusion and drift, thereby improving the system's robustness and continuity and adapting to more complex dynamic operating scenarios.

[0074] After obtaining the straight-line distances between all forklifts and each pile in S4, the closest distance matching is performed on each forklift, and the target pile closest to it is preferentially selected as the final scheduling object, and the pairing result is generated.

[0075] After calculating the straight-line distances between forklifts and all charging piles, this implementation introduces a "nearest neighbor matching" strategy to perform the final pairing of forklifts and charging piles. The system constructs a distance matrix between forklifts and charging piles through the preceding calculation steps. This matrix contains the Euclidean distance values between all forklifts and each charging pile. Based on this matrix, the scheduling algorithm selects the smallest value from the distance list for each forklift, corresponding to the charging pile closest to it. After traversing all forklifts, this process results in a set of minimum distance matching pairs. Each forklift selects only the nearest charging pile as its target charging point. If multiple forklifts select the same charging pile, the system assigns that pile based on priority (e.g., probability of attention or rating). Forklifts that fail to successfully match are either relegated to the next round of scheduling or selected to the next closest charging pile. Finally, the system organizes all successfully matched forklift-charging pile pairs into a matching result table and, through the control module, issues scheduling instructions to the corresponding forklifts, directing them to the corresponding charging pile to perform charging operations. This matching method is based on a greedy strategy and has the characteristics of low computational complexity and fast response speed. It is particularly suitable for scenarios with high scheduling frequency and a large number of forklifts. It can quickly generate preliminary allocation plans without sacrificing performance.

[0076] The closest distance matching strategy significantly reduces the idle time and path costs required for forklifts to reach their target piles, effectively saving energy, reducing operation time, and alleviating the risk of on-site traffic congestion. This strategy is simple to implement and highly real-time, enabling rapid response to environmental changes and forklift status updates. Furthermore, combined with prior priority scoring and attention probability calculation, this approach maintains efficiency while also balancing scheduling fairness and overall system load balancing.

[0077] In S4, when performing the nearest neighbor matching of forklifts and charging piles, if there are two or more forklifts with the same closest distance to the same charging pile, the one with higher priority is selected to be paired with the charging pile based on the priority score of each forklift. In this embodiment, in order to solve the resource contention problem that may occur during the nearest neighbor matching of forklifts and charging piles, the system introduces a conflict judgment mechanism based on priority scores. When performing the nearest neighbor pairing of forklifts and charging piles through the distance matrix, if it is detected that there are two or more forklifts with the same straight-line distance equal to the same charging pile, the system will compare the priority scores of these forklifts. The priority score is determined by the previous step S2 and reflects the resource allocation priority of each forklift in the current scheduling round. In the event of a distance conflict, the forklift with the highest score will be assigned to the charging pile first, and other forklifts with lower scores will retry matching based on their next nearest target pile, or enter the next round of scheduling. By introducing scheduling priority as a decision-making basis, this method enhances the fairness, stability, and controllability of the system while maintaining the rapid response of nearest neighbor pairing. It effectively prevents pairing confusion caused by equidistant competition and improves resource allocation efficiency and operational orderliness in a multi-forklift collaborative environment. Introducing priority scores as a decision-making basis for conflict judgment effectively improves the stability and scheduling rationality of the system in a high-density scheduling environment. By prioritizing service requests from high-scoring forklifts, the scheduling system can better respond to the charging needs of mission-critical or high-demand equipment, improving overall scheduling quality and resource utilization efficiency. At the same time, it avoids the random processing caused by equidistant matching, enhancing the predictability and business controllability of the scheduling mechanism.

[0078] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic allocation method for charging piles in a multi-forklift collaborative operation scenario, characterized in that: The method comprises: S1. Generate the estimated arrival time of the forklift at each charging station based on the current traffic flow prediction results. This includes using the mapping system to obtain the path distance from the forklift's current location to each charging station, giving the current actual average speed of each path based on traffic sensor data or historical traffic flow, calculating the estimated arrival time of the forklift at each charging station, and storing all arrival times of each forklift in a matrix for subsequent analysis. S2. Calculate the actual waiting time and corresponding priority weight of each forklift based on the estimated arrival time. This includes counting the number of forklifts currently or expected to arrive at each charging station, setting the maximum number of charges that the station can handle simultaneously as a constant, and calculating a priority score for each forklift. S3. Dynamically adjust the charging queue order of forklifts based on priority weights, including collecting the priority score of each forklift, calculating the sum of the priority scores of all forklifts, calculating the probability of each forklift being dispatched and being noticed, and sorting the forklifts from high to low according to the probability of dispatching and being noticed, which will be used as the queue order for the next round of charging; S4. Send the adjusted queue order to the target forklift to guide it to complete the charging pile allocation, including obtaining the current position coordinates of all forklifts, obtaining the position coordinates of all target charging piles, calculating the straight-line distance between each pair of forklifts and piles, performing nearest neighbor matching on the current queue order, determining the final allocation relationship, and sending the allocation result to the corresponding forklift through the network interface.

2. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 1 is characterized by: The specific method for calculating the estimated arrival time of the forklift to each charging pile in S1 includes using a map system or a path planning engine to obtain the shortest travel path from the current position of the forklift to the target charging pile, calculating the length of each section in the path and summing them to obtain the total path distance, calling a traffic flow sensor or a historical traffic database to obtain the real-time traffic speed of each section of the road on the path, and dividing the path distance from the current position of the forklift to the charging pile by the actual traffic speed of the current section to obtain the estimated arrival time of the forklift to each charging pile.

3. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 1 is characterized by: The specific method for calculating the priority score in S2 includes, for each charging pile, counting the number of forklifts currently or expected to arrive at the charging pile in the queue, setting a standard value for each charging pile, indicating the number of forklifts that the charging pile can simultaneously accept and process at the same time, and converting the current queue number of the charging pile where the forklift is located into the ratio of the maximum number of forklifts that the charging pile can serve, and then using this ratio to derive a priority score, recording this score for each forklift, and sorting them according to the score. Forklifts with higher scores will be given priority in subsequent scheduling to enter the front of the queue.

4. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 3 is characterized by: The specific determination method of setting a standard value for each charging pile in S2 includes marking the maximum number of interfaces for connecting forklifts at the same time on each charging pile device when it leaves the factory, which is the standard value set for the charging pile.

5. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 1 is characterized by: The specific method for calculating the probability of each forklift being scheduled in S3 includes collecting the priority score calculated for each forklift in S2, summing up the priority scores of all the above forklifts, and calculating their sum. The priority score of each forklift is divided by the total score just obtained to obtain a percentage value. This percentage value represents the probability of the forklift being selected in the current round of scheduling. According to the probability of attention of each forklift, they are sorted from high to low. The sorting result will be used as the order of the charging queue in the next round. The forklifts with higher probabilities are ranked first and get priority in charging arrangements.

6. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 1, characterized in that: The specific calculation method for calculating the straight-line distance between each pair of forklifts and charging piles in S4 includes reading the two-dimensional coordinate position of each forklift in the current storage site, reading the fixed coordinate values of all charging piles on the site map, and for each pair of forklifts and target charging piles, subtracting the values of the forklift position and the pile position in two directions respectively to obtain the difference in the horizontal and vertical directions, squaring the differences in the two directions just obtained, and then adding them together, and taking the square root of the result to obtain the straight-line path length between the forklift and the corresponding charging pile. The above calculation steps are performed one by one for all combinations of forklifts and all target charging piles, and finally a data table containing all distance values is formed, based on the straight-line distances between all forklifts and each pile.

7. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 6, characterized in that: In S4, the two-dimensional coordinate position of the forklift in the current storage yard is provided in real time by the positioning module on the forklift.

8. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 7, characterized in that: The positioning module in the S4 includes a laser radar, an inertial navigation system or UWB.

9. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 6, characterized in that: After obtaining the straight-line distances between all forklifts and each pile in S4, the closest distance matching is performed on each forklift, and the target pile closest to it is preferentially selected as the final scheduling object, and a matching result is generated.

10. The method for dynamically allocating charging piles in a multi-forklift collaborative operation scenario according to claim 9, characterized in that: In the process of performing the nearest neighbor matching between forklifts and charging piles in S4, if there are two or more forklifts with the same closest distance to the same charging pile, the one with the higher priority is selected according to the priority score of each forklift to be paired with the charging pile.

Images