A method for allocating AGV tasks in logistics plants based on residual power prediction

Abstract

The present invention relates to the technical field of logistics task allocation and analysis, and specifically to a method for allocating AGV task endurance in a logistics plant based on residual power prediction, comprising the following steps: scanning cargo handover points in a logistics task path, detecting whether a wireless charging module is deployed at the handover point and obtaining its charging power value, and generating a set of rechargeable handover points; calculating the expected amount of charging required at the rechargeable handover point based on the dwell time and charging power value of the rechargeable handover point, and generating a multi-task performance gain value in combination with the remaining power of the AGV; selecting an AGV whose multi-task performance gain value exceeds a preset gain threshold to allocate logistics tasks, and inserting the rechargeable handover point into the task sequence. The present invention prevents high-power AGVs from being assigned low-yield charging tasks, thereby improving the economy and energy utilization of task allocation; in conditions such as task peaks or battery aging, a dynamic gain threshold adjustment mechanism is used to ensure the stability of the candidate set and significantly reduce the risk of AGV task interruption.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics task allocation and analysis, and in particular to a method for allocating AGV task endurance in a logistics plant based on residual power prediction. Background Art

[0002] With the development of intelligent manufacturing and industrial automation, automated guided vehicles (AGVs) have been widely used in logistics plants to achieve efficient cargo transfer between different workstations. Current AGV scheduling systems primarily make scheduling decisions based on task priority, path length, or current idle status during task allocation, with little consideration given to the impact of the AGV's battery status on its ability to continuously execute tasks. This can lead to the following problems in complex working conditions:

[0003] Traditional systems typically use periodic centralized charging strategies or set fixed thresholds for scheduling charging. They are unable to flexibly respond to factors such as dynamic task flows, uneven charging point layouts, and battery aging, resulting in AGVs frequently interrupting tasks or passively returning to charging areas, affecting overall logistics efficiency.

[0004] Most existing solutions rely on static databases to mark the locations of fixed charging piles, lack the ability to dynamically perceive "potential charging opportunity points" along the path, and ignore rechargeable resources such as wireless charging modules deployed at intersections, resulting in low system resource utilization.

[0005] The scheduling model is fragmented, and it is difficult to coordinate tasks and charging behaviors. The traditional scheduling algorithm separates task path planning from charging operations, and fails to evaluate the impact of charging behavior on the task path cost, resulting in inefficient phenomena such as "high detour and low charging" and "high power plug-in charging", and even causing scheduling conflicts and path blockages.

[0006] The lack of a power value assessment mechanism makes charging benefits difficult to quantify: existing methods only use the absolute amount of charging as the evaluation basis, ignoring the relative value differences of AGV's charging behavior at different remaining power stages, and cannot achieve refined and economical task-charging coordinated optimization. Summary of the Invention

[0007] The present invention provides a logistics plant AGV task endurance allocation method based on residual power prediction. It is a new AGV task allocation method that integrates power state prediction, path structure perception, charging behavior scheduling and environmental constraint modeling, improves task endurance and charging efficiency, and realizes intelligent energy scheduling for complex plant logistics systems.

[0008] A method for allocating AGV task endurance in a logistics plant based on residual power prediction includes the following steps:

[0009] S1, charging handover point identification: Scan the cargo handover points in the logistics task path, detect whether the handover points are deployed with wireless charging modules and obtain their charging power values, and generate a set of charging handover points;

[0010] S2, performance gain calculation: Based on the residence time and charging power value of the rechargeable handover point, the expected amount of charging performed at the rechargeable handover point is calculated, and the multi-task performance gain value is generated in combination with the remaining power of the AGV;

[0011] S3, gain priority allocation: select AGVs whose multi-task efficiency gain value exceeds the preset gain threshold to allocate logistics tasks, and insert the rechargeable handover point into the task sequence.

[0012] Optionally, the S1 specifically includes:

[0013] S11, parse the coordinates of key nodes in the logistics task path and extract the location identifiers of all cargo handover points;

[0014] S12, sending a wireless detection signal to the physical node corresponding to the location identifier, and when receiving a response signal including a charging feature code, determining that a wireless charging module is deployed;

[0015] S13, decoding the charging power value and available time period from the response signal;

[0016] S14, aggregating information of all intersections determined to have wireless charging modules deployed therein to generate a set of chargeable intersections.

[0017] Optionally, the S12 specifically includes:

[0018] Treat each cargo handover point as a physical node and send wireless detection signals to the physical nodes in turn. The detection signal carries identity and request information, including the identity number of the AGV initiating the detection, the current detection timestamp, and the detection instruction code used to trigger the response;

[0019] A detection signal is sent. After the physical node receives the detection signal, if the physical node has a wireless charging module deployed, it returns a structured response signal. If the response signal meets the judgment criteria, the physical node is determined to be a chargeable junction point with a wireless charging module deployed, and its power is decoded and verified.

[0020] Optionally, the element structure of the chargeable junction point set is (location identifier, charging power value, available time period).

[0021] Optionally, the S2 specifically includes:

[0022] S21, calculating the expected amount of replenishment: obtaining the residence time and charging power value of each charging junction in the charging junction point set, and calculating the expected amount of replenishment based on the transmission efficiency coefficient;

[0023] S22, determining a gain reference value: generating a charging gain coefficient based on the expected amount of charge and the current remaining amount of the AGV;

[0024] S23, calculating the path optimization amount: obtaining the original task path length when charging is not embedded, subtracting the new path length after the charging junction is embedded, and generating a path shortening rate;

[0025] S24, synthesized gain value: When the charging gain coefficient is greater than the gain critical value, the path shortening rate is added to the charging gain coefficient according to the dynamic weight factor, and the multi-task performance gain value is output.

[0026] Optionally, the expected amount of replenishment is calculated as:

[0027] ;

[0028] in, Indicates that the AGV is at a rechargeable handover point The amount of electricity replenished, Indicates the estimated time spent at the handover point. is the effective charging power of the rechargeable junction, is the transmission efficiency coefficient.

[0029] Optionally, the path shortening rate is calculated as:

[0030] ;in, Indicates the path shortening rate, which indicates the relative saving percentage brought by path optimization. represents the total length of the task path when no charging point is embedded, represents the adjusted path length after embedding the charging point.

[0031] Optionally, S2 further includes a transmission efficiency coefficient compensation rule:

[0032] Specifically, if there are interference sources or different ground materials around the charging junction, the transmission efficiency coefficient will be dynamically adjusted according to the interference sources or different ground materials. .

[0033] Optionally, the S3 specifically includes:

[0034] S31, determining a basic gain threshold based on the factory task backlog rate and battery type, calculating a floating compensation value based on the AGV historical task interruption rate, and generating a preset gain threshold;

[0035] S32, candidate AGV screening: selecting AGVs whose multi-task performance gain values ​​are greater than a preset gain threshold to form a high-gain candidate set;

[0036] S33, time-space conflict resolution: Detect the estimated time window of each AGV in the high-gain candidate set to arrive at the rechargeable handover point. When the time window has a high overlap with the available time period of the rechargeable handover point, generate the valid insertion point coordinates; specifically, obtain the estimated time interval of each candidate AGV to arrive at the handover point. , and the available charging period at the junction For comparison, calculate the time window overlap:

[0037] ;

[0038] If the overlap satisfies: ;

[0039] The intersection point is marked as a valid insertion point and its coordinates are recorded Enter the set of feasible insertion points .

[0040] In the above: Indicates the estimated time when the AGV will arrive at the rechargeable handover point. Indicates the time when the AGV is expected to leave the rechargeable handover point. Indicates the available start time of the charging module at this junction. Indicates the end time of the available charging module at the junction point. Indicates the amount of overlap between the actual available charging period of the AGV and the available period of the handover point. Represents the two-dimensional coordinates of the rechargeable junction point j

[0041] S34, task sequence reconstruction: embed the effective insertion point coordinates into the original task path according to the charging period requirements, generate a new task sequence including charging behavior and assign it to the corresponding AGV.

[0042] Optionally, the preset gain threshold :

[0043] ;

[0044] in, is the preset gain threshold, As the basic gain threshold, set according to the battery type, represents the backlog rate coefficient, is the interruption rate coefficient, Indicates the current task backlog rate, Indicates the total number of tasks currently to be assigned. Indicates the number of AGVs currently available. Indicates the task interruption rate in the past 7 days, Indicates the total number of AGV mission interruptions in the past 7 days. Indicates the total number of tasks performed by the AGV in the past 7 days.

[0045] Beneficial effects of the present invention:

[0046] The present invention introduces a wireless detection response + measured load verification mechanism into the AGV task path, dynamically identifies the intersection points of the deployed wireless charging modules and evaluates their actual power output, and realizes accurate modeling of charging capabilities in complex factory environments such as metal shelves and electromagnetic interference. Compared with the traditional solution that relies on static databases, it automatically eliminates faulty charging points and improves the accuracy of resource availability identification. The dynamic transmission efficiency compensation mechanism can adjust the efficiency coefficient in real time according to the shelf occlusion rate and the ground material, effectively avoiding charging estimation deviations; constructs a triple set including position, power, and available time, and provides multi-dimensional charging capacity constraint support for task scheduling.

[0047] This invention calculates the ratio of the expected charge capacity to the current remaining charge of the AGV to generate a "charging gain coefficient" and sets a dynamic activation threshold. Only when the gain is significant does path optimization factor into the final performance gain value. This prevents high-charge AGVs from being assigned low-yield charging tasks, improving the economic efficiency and energy utilization of task allocation. During peak task loads or battery aging, a dynamic gain threshold adjustment mechanism ensures the stability of the candidate set and significantly reduces the risk of AGV task interruption.

[0048] This invention calculates the overlap between the AGV's predicted arrival time and the available charging window at the handover point. A matching condition of ≥80% of the available window is set to generate valid insertion points, ensuring that charging does not interfere with the main task. Task path reconstruction logic embeds charging activities according to available window periods, resolving task-charging conflicts caused by traditional decoupling of charging scheduling from logistics routing. This approach also avoids insertion failures and task delays in scenarios with low overlap or high latency, improving on-time task completion rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 This is a schematic diagram of charging power value verification according to an embodiment of the present invention;

[0051] Figure 2A schematic diagram of transmission efficiency coefficient compensation according to an embodiment of the present invention;

[0052] Figure 3 Schematic diagram of a method flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Those skilled in the art may also implement some known technologies in other alternative ways. The accompanying drawings are only for describing the embodiments in more detail and are not intended to limit the present invention in any specific way.

[0054] like Figure 1-Figure 3 As shown, a method for allocating AGV task endurance in a logistics plant based on residual power prediction includes the following steps:

[0055] S1, charging handover point identification: Scan the cargo handover points in the logistics task path, detect whether the handover points are deployed with wireless charging modules and obtain their charging power values, and generate a set of charging handover points;

[0056] S2, performance gain calculation: Based on the residence time and charging power value of the rechargeable handover point, the expected amount of charging performed at the rechargeable handover point is calculated, and the multi-task performance gain value is generated in combination with the remaining power of the AGV;

[0057] S3, gain priority allocation: select AGVs whose multi-task efficiency gain value exceeds the preset gain threshold to allocate logistics tasks, and insert the rechargeable handover point into the task sequence.

[0058] S1 specifically includes:

[0059] S11, Path node parsing and junction point extraction: Given a task path topology , where each path node Contains the following fields: ;in, is a two-dimensional coordinate, Indicates whether it is a cargo delivery point.

[0060] Extract all junction points: ; Represents a set of handover locations (including only nodes marked as "handover" in the task path), For the The location identifier of the junction point, Indicates the intersection The two-dimensional space coordinates of Representation node The type field has a value of handover (indicating the cargo handover point). Represents a path collection The path nodes, It is the set of all nodes in the AGV task path.

[0061] S12, wireless detection signal transmission and response verification:

[0062] S121, each junction Considered as a physical node , send wireless detection signals to these physical nodes in turn ;

[0063] S122, Composition of wireless detection signal: detection signal It contains identity and request information. Its structure consists of three parts, totaling 16 bytes:

[0064] AGV ID: the identity number of the AGV that initiated the detection;

[0065] Time Stamp: Current detection timestamp, used for synchronization and anti-replay;

[0066] Probe Request_Code: Specific probe command code used to trigger a response.

[0067] The signal is sent via a communication frequency of 125 kHz to improve anti-interference and adapt to the complex electromagnetic environment of industrial plants.

[0068] S123, format and content of response signal: physical node After receiving the detection signal, if the node is equipped with a wireless charging module, it will return a structured response signal. , contains three parts of information:

[0069] Charge Code: Indicates whether the physical node has wireless charging function;

[0070] Power Code: indicates the current module charging power code;

[0071] Time Slot: The available time period of the charging module.

[0072] S124, judgment standard: If the Charge Code field value in the response signal is 0xA5, the system determines that the node is a chargeable connection point deployed with a wireless charging module, and subsequently performs power decoding and verification on it.

[0073] S13, decoding and dynamic verification of charging power and time period:

[0074] Nominal power decoding: , Indicates the intersection The nominal charging power value decoded from the response signal, represents a decoding function that extracts a numeric value from an encoded field, Indicates the field encoding indicating the power level in the response signal;

[0075] Dynamic power verification: Apply a standard simulated load to the rechargeable junction (Known resistance value), measure the measured voltage , calculate the real output power : ;

[0076] Power deviation judgment: , For the intersection The power deviation percentage indicates the difference between the measured power and the nominal value; if , then the nominal value is accepted; otherwise, the measured mean value is used instead:

[0077] , is the deviation tolerance threshold, which is set to 5%;

[0078] Available time period analysis: .

[0079] S14, rechargeable junction set Construction: For each cargo delivery point that meets the charging conditions , generate triple structure: ; Aggregate to generate a set of rechargeable junction points:

[0080] .

[0081] S2 specifically includes:

[0082] S21, calculation of expected replenishment capacity: for each rechargeable junction , according to the residence time , charging power and transmission efficiency coefficient Calculate the expected amount of power replenishment at this point :

[0083] ;

[0084] in, Indicates that the AGV is at a rechargeable handover point The amount of electricity replenished, Indicates the estimated time spent at the handover point. is the effective charging power of the rechargeable junction, is the wireless energy transmission efficiency coefficient, ranging from 0.82 to 0.90, depending on the site environment.

[0085] S22, charging gain coefficient calculation: the above charging power and the current remaining power of the AGV are calculated. Divide to generate the charging gain coefficient : ;

[0086] in, Indicates the relative power gain of charging behavior to the current AGV, Indicates the amount of charge brought by a certain rechargeable junction point. Indicates the current remaining power of the AGV.

[0087] S23, path shortening rate calculation: compare the original task path length Path length after embedding the rechargeable junction , and get the path optimization value (i.e. path shortening rate):

[0088] ;in, Indicates the path shortening rate, which indicates the relative saving percentage brought by path optimization. represents the total length of the task path when no charging point is embedded, represents the adjusted path length after embedding the charging point.

[0089] S24, multi-task performance gain value synthesis: when (set gain activation threshold), the path shortening rate is dynamically weighted according to the congestion state Weighted superposition to generate the final multi-task performance gain value :

[0090] ;

[0091] in, Indicates the final multi-task performance gain value, which is used as the basis for task priority allocation. It represents the weight factor of path shortening, which is adjusted depending on the congestion situation in the factory area (the value range is 0.2-0.8). Indicates the gain activation threshold. The default value is 0.3 (i.e. the amount of power to be replenished must reach 30% of the remaining power).

[0092] It also includes wireless energy transmission efficiency compensation rules: If there are interference sources such as metal shelves or different ground materials around the intersection, you can set it according to the following rules :

[0093] If the shelf metal coverage is greater than 50%, then ;

[0094] Otherwise, if the floor is concrete, ;

[0095] Otherwise default .

[0096] S3 specifically includes:

[0097] S31, preset gain threshold: determine the basic gain threshold based on the factory task backlog rate and AGV battery type , and combined with the AGV historical task interruption rate to generate a floating compensation amount, thus forming a preset gain threshold :

[0098] ;

[0099] in, is the preset gain threshold (used to screen high-performance AGVs), The basic gain threshold is set according to the battery type. The NMC type is 0.45, and the LFP type is 0.40. Indicates the backlog rate coefficient, the recommended value is 0.15, is the interruption rate coefficient, the recommended value is 0.08, Indicates the current task backlog rate, Indicates the total number of tasks currently to be assigned. Indicates the number of AGVs currently available. Indicates the task interruption rate in the past 7 days, Indicates the total number of AGV mission interruptions in the past 7 days. Indicates the total number of tasks performed by the AGV in the past 7 days.

[0100] S32, candidate AGV screening: screening all multi-task performance gain values Greater than the dynamic threshold And the current remaining power Greater than the safety lower limit of power AGVs, constitute a high-gain candidate set :

[0101] ;

[0102] in, is a set of high-gain candidate AGVs, Indicates the The multi-task performance gain value of each AGV (derived from S2), Indicates the The remaining power of each AGV, Indicates the safety bottom line of power, set at 20% of rated capacity.

[0103] S33, time-space conflict resolution and insertion point identification: For each candidate AGV, obtain its estimated time interval to the handover point , and the available charging period at the junction For comparison, calculate the time window overlap:

[0104] ;

[0105] If the overlap satisfies: ;

[0106] The intersection point is marked as a valid insertion point and its coordinates are recorded Enter the set of feasible insertion points .

[0107] In the above: Indicates the estimated time when the AGV will arrive at the rechargeable handover point. Indicates the time when the AGV is expected to leave the rechargeable handover point. Indicates the available start time of the charging module at this junction. Indicates the end time of the available charging module at the junction point. Indicates the amount of overlap between the actual available charging period of the AGV and the available period of the handover point. Represents the two-dimensional coordinates of the chargeable junction point j.

[0108] S34, Task sequence reconstruction: embed the effective insertion points into the original task path to construct a new task sequence including charging behavior , the embedding rules are as follows:

[0109] Charging start time: , Indicates the final scheduled charging start time;

[0110] Charging duration: , Indicates the total time the AGV plans to stay at the handover point. Indicates the actual duration of the charging behavior that can be executed;

[0111] Path structure modification diagram:

[0112] Original path: New Path: (A is the preceding task, B is the charging insertion point, and C is the subsequent task);

[0113] New Task Sequence It is assigned to the corresponding AGV and written into the task scheduling table.

[0114] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0115] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for allocating AGV mission endurance in a logistics plant based on residual power prediction, characterized in that: The following steps are involved: S1, charging handover point identification: Scan the cargo handover points in the logistics task path, detect whether the handover points are deployed with wireless charging modules and obtain their charging power values, and generate a set of charging handover points; S2, performance gain calculation: Based on the residence time and charging power value of the rechargeable handover point, the expected amount of charging performed at the rechargeable handover point is calculated, and the multi-task performance gain value is generated in combination with the remaining power of the AGV; S3, Gain priority allocation: select AGVs whose multi-task efficiency gain value exceeds the preset gain threshold to assign logistics tasks, and insert the rechargeable handover point into the task sequence; The S2 specifically includes: S21, calculating the expected amount of replenishment: obtaining the residence time and charging power value of each charging junction in the charging junction point set, and calculating the expected amount of replenishment based on the transmission efficiency coefficient; S22, determining a gain reference value: generating a charging gain coefficient based on the expected amount of charge and the current remaining amount of the AGV; S23, calculating the path optimization amount: obtaining the original task path length when charging is not embedded, subtracting the new path length after the charging junction is embedded, and generating a path shortening rate; S24, composite gain value: when the charging gain coefficient is greater than the gain threshold, the path shortening rate is added to the charging gain coefficient according to the dynamic weight factor, and the multi-task performance gain value is output; The expected amount of replenishment is calculated as: ； in, Indicates that the AGV is at a rechargeable handover point The amount of electricity replenished, Indicates the estimated time spent at the handover point. is the effective charging power of the rechargeable junction, is the transmission efficiency coefficient; The path shortening rate is calculated as: ;in, Indicates the path shortening rate, which indicates the relative saving percentage brought by path optimization. represents the total length of the task path when no charging point is embedded, represents the adjusted path length after embedding the charging point.

2. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 1 is characterized in that: Said S1 specifically includes: S11, parse the coordinates of key nodes in the logistics task path and extract the location identifiers of all cargo handover points; S12, sending a wireless detection signal to the physical node corresponding to the location identifier, and when receiving a response signal including a charging feature code, determining that a wireless charging module is deployed; S13, decoding the charging power value and available time period from the response signal; S14, aggregating information of all intersections determined to have wireless charging modules deployed therein to generate a set of chargeable intersections.

3. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 2 is characterized in that: The S12 specifically includes: Treat each cargo handover point as a physical node and send wireless detection signals to the physical nodes in turn. The detection signal carries identity and request information, including the identity number of the AGV initiating the detection, the current detection timestamp, and the detection instruction code used to trigger the response; A detection signal is sent. After the physical node receives the detection signal, if the physical node has a wireless charging module deployed, it returns a structured response signal. If the response signal meets the judgment criteria, the physical node is determined to be a chargeable junction point with a wireless charging module deployed, and its power is decoded and verified.

4. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 2 is characterized in that: The element structure of the chargeable junction point set is (location identifier, charging power value, available time period).

5. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 1 is characterized in that: The S2 also includes a transmission efficiency coefficient compensation rule: Specifically, if there are interference sources or different ground materials around the charging junction, the transmission efficiency coefficient will be dynamically adjusted according to the interference sources or different ground materials. .

6. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 1 is characterized in that: The S3 specifically includes: S31, determining a basic gain threshold based on the factory task backlog rate and battery type, calculating a floating compensation value based on the AGV historical task interruption rate, and generating a preset gain threshold; S32, candidate AGV screening: selecting AGVs whose multi-task performance gain values ​​are greater than a preset gain threshold to form a high-gain candidate set; S33, time-space conflict resolution: Detect the estimated time window of each AGV arriving at the rechargeable handover point in the high-gain candidate set. When the time window has a high degree of overlap with the available time period of the rechargeable handover point, generate the valid insertion point coordinates; S34, task sequence reconstruction: embed the effective insertion point coordinates into the original task path according to the charging period requirements, generate a new task sequence including charging behavior and assign it to the corresponding AGV.

7. The method for allocating AGV tasks in a logistics plant based on residual power prediction according to claim 6 is characterized in that: The preset gain threshold : ； in, is the preset gain threshold, As the basic gain threshold, set according to the battery type, represents the backlog rate coefficient, is the interruption rate coefficient, Indicates the current task backlog rate, Indicates the total number of tasks currently to be assigned. Indicates the number of AGVs currently available. Indicates the task interruption rate in the past 7 days, Indicates the total number of AGV mission interruptions in the past 7 days. Indicates the total number of tasks performed by the AGV in the past 7 days.

Images