Real-time scheduling method and system for mobile robot based on dynamic topology map

Abstract

The application relates to the technical field of robot scheduling, in particular to a mobile robot real-time scheduling method and system based on a dynamic topological map. In the system, an emergency scheduling planning scheme management module collects the transportation state information of each mobile robot in real time through a sensor, judges the current working load state of the mobile robot, combines the topological map updating result under the corresponding transportation scene at the current time and the initial material delivery target planning route of the corresponding mobile robot, generates an emergency scheduling planning scheme of the mobile robot with the abnormal working load state at the current time, and calculates the load balance comprehensive adaptive deviation value of each emergency scheduling planning scheme. The application realizes the dynamic updating of the material delivery target planning route of the mobile robot in combination with the transportation state of the mobile robot itself, introduces the concept of the load balance comprehensive adaptive deviation value, realizes the effective screening of the best emergency scheduling planning scheme of the mobile robot, and realizes the collaborative management among different mobile robots.

Description

Technical Field

[0001] This invention relates to the field of robot scheduling technology, specifically to a method and system for real-time scheduling of mobile robots based on dynamic topology maps. Background Technology

[0002] With the rapid development of logistics automation and intelligence, delivery robots are increasingly being used in warehousing, hospitals, hotels, shopping malls, and other scenarios. However, traditional robot scheduling methods in static environments have significant limitations in complex dynamic environments, failing to adapt to environmental changes and real-time demands. The combination of dynamic topology maps and real-time scheduling technology has become key to solving this problem.

[0003] Traditional scheduling methods are typically based on pre-built static maps, which cannot update environmental information (such as obstacles and path changes) in real time. This makes it difficult to cope with sudden tasks or environmental changes, leading to robots being prone to collisions or path failures in dynamic environments and resulting in low delivery efficiency. Furthermore, in existing technologies, the delivery tasks of robots transporting goods are usually manually set and executed sequentially, without taking into account the delivery status of the mobile robots themselves, and without being able to perform emergency scheduling based on actual conditions to achieve collaborative management between different mobile robots. Therefore, existing technologies have significant shortcomings. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for real-time scheduling of mobile robots based on dynamic topology maps, so as to solve the problems mentioned in the background art.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a real-time scheduling method for mobile robots based on a dynamic topology map, the method comprising:

[0006] S1. Sequentially obtain each delivery target node corresponding to each mobile robot and generate the initial delivery target planning route for the corresponding mobile robot.

[0007] S2. Upload the path captured by the mobile robot during the delivery process to the cloud control terminal. The cloud control terminal dynamically updates the topology map of the corresponding delivery scenario at the current time and provides real-time feedback to the mobile robot.

[0008] S3. Collect the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot; combine the topology map update results of the corresponding delivery scenario at the current time and the initial delivery target planning route of the corresponding mobile robot to generate an emergency scheduling plan for the mobile robot with abnormal workload status at the current time, and calculate the load balancing comprehensive adaptation deviation value of each emergency scheduling plan.

[0009] S4. Based on the load balancing adaptation deviation value of each emergency dispatch planning scheme, match the best emergency dispatch planning scheme for mobile robots under the current topology map, and provide feedback to the administrator for confirmation.

[0010] According to the above technical solution, each delivery target node in S1 corresponds to a delivery location; the initial delivery target planning route of the mobile robot is the splicing result of any two adjacent delivery target nodes in the database corresponding to each mobile robot's delivery target nodes.

[0011] According to the above technical solution, S2 includes:

[0012] S21. The cloud control terminal summarizes the path acquisition images uploaded by each mobile robot during the delivery process in real time; the path acquisition image uploaded by the i-th mobile robot during the delivery process at the j-th path is denoted as A. (i，j) ;

[0013] S22, Obtain information from the mobile robot collecting A (i，j) The location information at that time is denoted as W. (i，j) ; Extract location information from the database as W (i，j) The static topological map layer information corresponding to the surrounding unit radius area is denoted as STA. (i，j) The topology map includes a static layer and a dynamic layer;

[0014] In this invention, static layer information represents the structural information of objects with fixed positions in the topology map; dynamic layer information represents the structural information of objects with movable positions in the topology map.

[0015] S23. Identify and label A using image recognition technology. (i，j) Belongs to STA (i，j) The area, and combined with STA (i，j) The positional relationships between the static objects in A (i，j) The identification failed and it belongs to STA. (i，j) Update the marked region, and add A. (i，j) The unmarked remaining area is denoted as A. (i，j) The corresponding dynamic layer region to be identified;

[0016] S24, Extract A (i，j) The corresponding dynamic layer to be identified region, distinct from the location information in the database under the corresponding mobile robot's shooting perspective, is W. (i，j) The area of ​​the dynamic topology map layer within a unit radius of the surrounding area is denoted as SA. (i，j) And SA was identified through image recognition technology. (i，j)The database contains pre-defined forms for each object model and the location area occupied by each identified object model, resulting in A. (i，j) The corresponding dynamic layer information to be updated;

[0017] S25. The summary result of the dynamic layer information to be updated corresponding to each path captured by each mobile robot at the same time during the delivery process is used as the update result of the dynamic layer information in the topology map at the corresponding time under the corresponding delivery scenario; the combination result of the static layer information and the dynamic layer information update result of the topology map is used as the dynamic topology map at the corresponding time.

[0018] S26. The motion control terminal feeds back the obtained dynamic topology map to each mobile robot in real time.

[0019] In this invention, the topology map is divided into a static layer and a dynamic layer in order to achieve dynamic management of the topology map by keeping the static layer information unchanged and updating the dynamic layer information. This method can effectively reduce the amount of data updates to the topology map information. At the same time, keeping the static layer information unchanged and combining the positional relationships between the objects in the static layer can provide positioning for the dynamic layer information, ensuring the accuracy of the updated dynamic topology map.

[0020] According to the above technical solution, the transportation status information of the mobile robot in S3 includes the remaining battery power, the delivery target node to be arrived, the weight of the item to be delivered corresponding to each delivery target node, and the route segments not traversed in the initial delivery target planning route of the corresponding mobile robot.

[0021] The workload coefficient for the i-th mobile robot at the current time is calculated using the following formula:

[0022]

[0023] Where Hi represents the workload coefficient of the i-th mobile robot at the current time; Yi represents the remaining battery power in the delivery status information of the i-th mobile robot at the current time; L (i,n) L represents the distance between the starting point and the nth delivery target node to be traversed within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; (i,n-1) This represents the distance between the starting point and the (n-1)th delivery target node to be traversed within the untraveled route segment of the initial delivery target planned by the i-th mobile robot at the current time; when n=1, then L is determined. (i,n-1) The value of is 0; Ni represents the total number of delivery target nodes to be traversed in the route segment not yet traversed in the initial delivery target planning route of the i-th mobile robot at the current time; G (i,k)G represents the weight of the item to be delivered, corresponding to the k-th delivery target node, within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; where k∈[0, Ni-1], and when k=0, G (i,k) The value is 0; GZ represents the sum of the weights of the items to be delivered for the target node to be executed by the i-th mobile robot at the current time. Indicates the load capacity of the mobile robot within historical data. The average value of the ratio between power consumption and distance traveled while driving;

[0024] If Hi is less than or equal to the preset value, the current workload state of the mobile robot is determined to be abnormal; otherwise, the current workload state of the mobile robot is determined to be normal.

[0025] According to the above technical solution, in the process of generating an emergency scheduling plan for a mobile robot with an abnormal workload status at the current time in S3, the summary set of delivery target nodes to be reached by the mobile robot with an abnormal workload status at the current time is extracted; this set is denoted as the emergency adaptation option set for the corresponding mobile robot; different emergency adaptation schemes are obtained, each of which is a set consisting of one or more elements in the emergency adaptation option set for the corresponding mobile robot. The workload status of the corresponding mobile robot is normal after removing delivery tasks from the emergency adaptation scheme, and the workload status of the corresponding mobile robot is abnormal when not all delivery tasks from the emergency adaptation scheme are removed.

[0026] The specific steps to obtain the corresponding emergency dispatch planning schemes under different emergency adaptation solutions are as follows:

[0027] S31. Obtain the location of the first delivery target node belonging to the emergency adaptation scheme in the initial delivery target planning route of the mobile robot with abnormal workload status, and record it as the scheduling reference node.

[0028] S32. The route segment from the starting point of the untraveled route segment in the initial delivery target planning route of the mobile robot with the corresponding abnormal workload status to the previous delivery target node based on the scheduling reference node is recorded as the scheduling adaptation intersection segment.

[0029] S33. Within the untraveled route segment of the initial delivery target planning route of the mobile robot with normal workload status, any delivery target node that is the same as the delivery target node in the scheduling adaptation intersection segment is recorded as an emergency scheduling transfer node; the corresponding mobile robot with abnormal workload status completes the transfer of the goods to be delivered corresponding to the corresponding emergency adaptation scheme at the emergency scheduling transfer node; the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme update the delivery target planning route, and obtain different emergency scheduling planning schemes for the mobile robot with abnormal workload status at the current time. In the obtained emergency scheduling planning scheme, the workload status of the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme are both normal based on the updated delivery target planning route.

[0030] The updated delivery target planning route for a mobile robot with an abnormal workload is the result of splicing the corresponding planned road segments in the database for any two adjacent delivery target nodes among the unreached delivery target nodes after removing the corresponding emergency adaptation scheme.

[0031] When updating the delivery target planning route for a mobile robot with a normal workload, if all delivery target nodes in the corresponding emergency adaptation scheme are delivery target nodes that were not reached in the delivery target planning route of the mobile robot with a normal workload before the current time update, then the delivery order of the delivery target nodes and the corresponding delivery target planning route remain unchanged. If there are delivery target nodes in the corresponding emergency adaptation scheme that are not reached in the delivery target planning route of the mobile robot with a normal workload before the current time update, then the delivery target nodes that are not reached in the delivery target planning route of the mobile robot with an abnormal workload before the current time update are added to the end of the sequence of delivery target nodes to be reached, thus generating the delivery target planning route for the corresponding mobile robot.

[0032] According to the above technical solution, the calculation formula for the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme in S3 is as follows:

[0033] P g =|HD1 g μ (g,1) -HD2 g μ (g,2) |+ξ·LC g

[0034] Among them, P gThis represents the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme; the mobile robot with an abnormal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the first reference object; the mobile robot with a normal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the second reference object; HD1 g This indicates the workload coefficient of the first reference object after the delivery target planning route is updated; HD2 g Indicates the workload coefficient of the second reference object after the delivery target planning route is updated; μ (g，1) This represents the road condition difficulty coefficient in the updated topology map update result for the first reference object's planned delivery route in the current time and corresponding delivery scenario; μ (g，2) This indicates the road condition difficulty coefficient corresponding to the updated delivery target route for the second reference object in the current time's corresponding delivery scenario's topology map update result; the road condition difficulty coefficient is obtained by querying the maximum value of the cumulative traffic anomaly deviation corresponding to the dynamic layer information in each preset area of ​​the dynamic topology map in the corresponding time in the database within a preset form; the cumulative traffic anomaly deviation corresponding to the dynamic layer information in the preset area is equal to the sum of the products between the area occupied by each identified object model in the dynamic layer information of the preset area and the corresponding preset weight coefficient of the identified object model; LC g ξ represents the absolute value of the distance difference between the current location of the first and second reference objects in the updated corresponding delivery target planning route and the emergency dispatch transfer node; ξ represents the preset conversion factor.

[0035] When matching the best emergency dispatch plan for mobile robots based on the current topology map, the emergency dispatch plan with the smallest load balancing comprehensive adaptation deviation value is taken as the best emergency dispatch plan for mobile robots based on the current topology map.

[0036] A real-time scheduling system for mobile robots based on a dynamic topology map, the system comprising: a target information acquisition module, a topology map dynamic update module, an emergency scheduling planning scheme management module, and an emergency scheduling scheme screening and feedback module;

[0037] The target information acquisition module sequentially acquires each delivery target node corresponding to each mobile robot and generates the initial delivery target planning route for the corresponding mobile robot.

[0038] The topology map dynamic update module uploads the path captured by the mobile robot during the delivery process to the cloud control terminal. The cloud control terminal dynamically updates the topology map of the corresponding delivery scenario at the current time and provides real-time feedback to the mobile robot.

[0039] The emergency dispatch planning scheme management module collects the delivery status information of each mobile robot in real time through sensors, determines the current workload status of the mobile robot, combines the topology map update results of the corresponding delivery scenario at the current time with the initial delivery target planning route of the corresponding mobile robot, generates an emergency dispatch planning scheme for the mobile robot with abnormal workload status at the current time, and calculates the load balancing comprehensive adaptation deviation value of each emergency dispatch planning scheme.

[0040] The emergency dispatch scheme screening and feedback module matches the best emergency dispatch plan for mobile robots based on the load balancing comprehensive adaptation deviation value of each emergency dispatch plan scheme, and sends the result back to the administrator for confirmation.

[0041] According to the above technical solution, the topology map dynamic update module includes a data upload unit and a topology map update feedback unit;

[0042] The data uploading unit uploads the path captured by the mobile robot during the delivery process to the cloud control terminal;

[0043] The topology map update feedback unit controls the cloud control terminal to dynamically update the topology map of the corresponding transportation scenario at the current time and provides real-time feedback to the mobile robot.

[0044] According to the above technical solution, the emergency dispatch planning scheme management module includes a load status determination unit, an emergency dispatch planning unit, and an adaptation deviation calculation unit;

[0045] The load status determination unit collects the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot.

[0046] The emergency dispatch planning unit combines the updated topology map results of the corresponding delivery scenario at the current time with the initial delivery target planning route of the corresponding mobile robot to generate an emergency dispatch planning scheme for the mobile robot whose workload status is abnormal at the current time.

[0047] The adaptation deviation calculation unit calculates the load balancing comprehensive adaptation deviation value for each emergency dispatch planning scheme.

[0048] According to the above technical solution, the emergency dispatch scheme screening and feedback module includes an optimal emergency dispatch planning scheme screening unit and a feedback confirmation unit;

[0049] The optimal emergency dispatch planning scheme selection unit matches the optimal emergency dispatch planning scheme for the mobile robot based on the load balancing comprehensive adaptation deviation value of each emergency dispatch planning scheme.

[0050] The feedback confirmation unit will provide the administrator with the best emergency dispatch plan for mobile robots based on the current topology map for confirmation.

[0051] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0052] (1) The objects in the topology map of this invention are divided into static and dynamic layers. By keeping the static layer information unchanged and updating the dynamic layer information, dynamic management of the topology map is achieved. This method not only reduces the amount of data update and data transmission burden of the topology map, but also allows the static layer information to provide positioning for the dynamic layer information, ensuring the accuracy of the updated dynamic topology map.

[0053] (2) In the delivery process, the present invention can dynamically update the delivery target planning route of the mobile robot and dynamically obtain the emergency scheduling planning scheme of the mobile robot with abnormal workload status by combining the mobile robot's own delivery status. It also introduces the concept of load balancing comprehensive adaptation deviation value to effectively screen the best emergency scheduling planning scheme of the mobile robot, thereby achieving emergency scheduling of the mobile robot in the case of abnormal workload status and realizing collaborative management between different mobile robots. Attached Figure Description

[0054] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0055] Figure 1 This is a schematic diagram of the structure of the real-time scheduling system for mobile robots based on a dynamic topology map according to the present invention;

[0056] Figure 2 This is a flowchart illustrating the real-time scheduling method for mobile robots based on a dynamic topology map according to the present invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Please see Figures 1-2 The present invention provides a technical solution: such as Figure 1As shown, this embodiment provides a real-time scheduling system for mobile robots based on a dynamic topology map. The system includes: a target information acquisition module, a topology map dynamic update module, an emergency scheduling planning scheme management module, and an emergency scheduling scheme screening and feedback module.

[0059] The target information acquisition module sequentially acquires each delivery target node corresponding to each mobile robot and generates the initial delivery target planning route for the corresponding mobile robot.

[0060] The topology map dynamic update module includes a data upload unit and a topology map update feedback unit;

[0061] The data uploading unit uploads the path captured by the mobile robot during the delivery process to the cloud control terminal;

[0062] The topology map update feedback unit controls the cloud control terminal to dynamically update the topology map of the corresponding transportation scenario at the current time and provides real-time feedback to the mobile robot.

[0063] The emergency dispatch planning scheme management module includes a load status determination unit, an emergency dispatch planning unit, and an adaptation deviation calculation unit.

[0064] The load status determination unit collects the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot.

[0065] The emergency dispatch planning unit combines the updated topology map results of the corresponding delivery scenario at the current time with the initial delivery target planning route of the corresponding mobile robot to generate an emergency dispatch planning scheme for the mobile robot whose workload status is abnormal at the current time.

[0066] The adaptation deviation calculation unit calculates the load balancing comprehensive adaptation deviation value for each emergency dispatch planning scheme;

[0067] The emergency dispatch plan screening and feedback module includes an optimal emergency dispatch plan screening unit and a feedback confirmation unit.

[0068] The optimal emergency dispatch planning scheme selection unit matches the optimal emergency dispatch planning scheme for the mobile robot based on the load balancing comprehensive adaptation deviation value of each emergency dispatch planning scheme.

[0069] The feedback confirmation unit will provide the administrator with the best emergency dispatch plan for mobile robots based on the current topology map for confirmation.

[0070] like Figure 2 As shown, this embodiment provides a real-time scheduling method for mobile robots based on a dynamic topology map, the method including:

[0071] S1. Sequentially obtain each delivery target node corresponding to each mobile robot and generate the initial delivery target planning route for the corresponding mobile robot.

[0072] In S1, each delivery target node corresponds to a delivery location; the initial delivery target planning route of the mobile robot is the result of splicing the corresponding planned road segments of any two adjacent delivery target nodes in the database for each mobile robot.

[0073] S2. Upload the path captured by the mobile robot during the delivery process to the cloud control terminal. The cloud control terminal dynamically updates the topology map of the corresponding delivery scenario at the current time and provides real-time feedback to the mobile robot.

[0074] S2 includes:

[0075] S21. The cloud control terminal summarizes the path acquisition images uploaded by each mobile robot during the delivery process in real time; the path acquisition image uploaded by the i-th mobile robot during the delivery process at the j-th path is denoted as A. (i，j) ;

[0076] S22, Obtain information from the mobile robot collecting A (i，j) The location information at that time is denoted as W. (i，j) ; Extract location information from the database as W (i，j) The static topological map layer information corresponding to the surrounding unit radius area is denoted as STA. (i，j) The topology map includes a static layer and a dynamic layer;

[0077] In this embodiment, static layer information represents the structural information of objects with fixed positions in the topology map; dynamic layer information represents the structural information of objects with movable positions in the topology map.

[0078] S23. Identify and label A using image recognition technology. (i，j) Belongs to STA (i，j) The area, and combined with STA (i，j) The positional relationships between the static objects in A (i，j) The identification failed and it belongs to STA. (i，j) Update the marked region, and add A. (i，j) The unmarked remaining area is denoted as A. (i，j) The corresponding dynamic layer region to be identified;

[0079] S24, Extract A (i，j) The corresponding dynamic layer to be identified region, distinct from the location information in the database under the corresponding mobile robot's shooting perspective, is W. (i，j)The area of ​​the dynamic topology map layer within a unit radius of the surrounding area is denoted as SA. (i，j) And SA was identified through image recognition technology. (i，j) The database contains pre-defined forms for each object model and the location area occupied by each identified object model, resulting in A. (i，j) The corresponding dynamic layer information to be updated;

[0080] In this embodiment, the Mask R-CNN model is used to identify objects and their corresponding positions in the dynamic layer to achieve dynamic updating of the topology map.

[0081] S25. The summary result of the dynamic layer information to be updated corresponding to each path captured by each mobile robot at the same time during the delivery process is used as the update result of the dynamic layer information in the topology map at the corresponding time under the corresponding delivery scenario; the combination result of the static layer information and the dynamic layer information update result of the topology map is used as the dynamic topology map at the corresponding time.

[0082] S26. The motion control terminal feeds back the obtained dynamic topology map to each mobile robot in real time.

[0083] S3. Collect the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot; combine the topology map update results of the corresponding delivery scenario at the current time and the initial delivery target planning route of the corresponding mobile robot to generate an emergency scheduling plan for the mobile robot with abnormal workload status at the current time, and calculate the load balancing comprehensive adaptation deviation value of each emergency scheduling plan.

[0084] The delivery status information of the mobile robot in S3 includes the remaining battery power, the delivery target node to be arrived at, the weight of the item to be delivered corresponding to each delivery target node, and the route segments not traversed in the initial delivery target planning route of the corresponding mobile robot.

[0085] The workload coefficient for the i-th mobile robot at the current time is calculated using the following formula:

[0086]

[0087] Where Hi represents the workload coefficient of the i-th mobile robot at the current time; Yi represents the remaining battery power in the delivery status information of the i-th mobile robot at the current time; L (i,n) L represents the distance between the starting point and the nth delivery target node to be traversed within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; (i,n-1)This represents the distance between the starting point and the (n-1)th delivery target node to be traversed within the untraveled route segment of the initial delivery target planned by the i-th mobile robot at the current time; when n=1, then L is determined. (i,n-1) The value of is 0; Ni represents the total number of delivery target nodes to be traversed in the route segment not yet traversed in the initial delivery target planning route of the i-th mobile robot at the current time; G (i,k) G represents the weight of the item to be delivered, corresponding to the k-th delivery target node, within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; where k∈[0, Ni-1], and when k=0, G (i,k) The value is 0; GZ represents the sum of the weights of the items to be delivered for the target node to be executed by the i-th mobile robot at the current time. Indicates the load capacity of the mobile robot within historical data. The average value of the ratio between power consumption and distance traveled while driving;

[0088] If Hi is less than or equal to the preset value, the current workload state of the mobile robot is determined to be abnormal; otherwise, the current workload state of the mobile robot is determined to be normal.

[0089] In the process of generating an emergency scheduling plan for a mobile robot with an abnormal workload status at the current time in S3, the summary set of delivery target nodes to be reached by the mobile robot with an abnormal workload status at the current time is extracted; this set is denoted as the emergency adaptation option set for the corresponding mobile robot; different emergency adaptation schemes are obtained, each of which is a set consisting of one or more elements in the emergency adaptation option set for the corresponding mobile robot. The workload status of the corresponding mobile robot is normal after removing delivery tasks from the emergency adaptation scheme, and the workload status of the corresponding mobile robot is abnormal when not all delivery tasks from the emergency adaptation scheme are removed.

[0090] The specific steps to obtain the corresponding emergency dispatch planning schemes under different emergency adaptation solutions are as follows:

[0091] S31. Obtain the location of the first delivery target node belonging to the emergency adaptation scheme in the initial delivery target planning route of the mobile robot with abnormal workload status, and record it as the scheduling reference node.

[0092] S32. The route segment from the starting point of the untraveled route segment in the initial delivery target planning route of the mobile robot with the corresponding abnormal workload status to the previous delivery target node based on the scheduling reference node is recorded as the scheduling adaptation intersection segment.

[0093] S33. Within the untraveled route segment of the initial delivery target planning route of the mobile robot with normal workload status, any delivery target node that is the same as the delivery target node in the scheduling adaptation intersection segment is recorded as an emergency scheduling transfer node; the corresponding mobile robot with abnormal workload status completes the transfer of the goods to be delivered corresponding to the corresponding emergency adaptation scheme at the emergency scheduling transfer node; the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme update the delivery target planning route, and obtain different emergency scheduling planning schemes for the mobile robot with abnormal workload status at the current time. In the obtained emergency scheduling planning scheme, the workload status of the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme are both normal based on the updated delivery target planning route.

[0094] The updated delivery target planning route for a mobile robot with an abnormal workload is the result of splicing the corresponding planned road segments in the database for any two adjacent delivery target nodes among the unreached delivery target nodes after removing the corresponding emergency adaptation scheme.

[0095] When updating the delivery target planning route for a mobile robot with a normal workload, if all delivery target nodes in the corresponding emergency adaptation scheme are delivery target nodes that were not reached in the delivery target planning route of the mobile robot with a normal workload before the current time update, then the delivery order of the delivery target nodes and the corresponding delivery target planning route remain unchanged. If there are delivery target nodes in the corresponding emergency adaptation scheme that are not reached in the delivery target planning route of the mobile robot with a normal workload before the current time update, then the delivery target nodes that are not reached in the delivery target planning route of the mobile robot with an abnormal workload before the current time update are added to the end of the sequence of delivery target nodes to be reached, thus generating the delivery target planning route for the corresponding mobile robot.

[0096] The formula for calculating the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme in S3 is as follows:

[0097] P g =|HD1 g μ (g,1) -HD2 g μ (g,2) |+ξ·LC g

[0098] Among them, P gThis represents the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme; the mobile robot with an abnormal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the first reference object; the mobile robot with a normal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the second reference object; HD1 g This indicates the workload coefficient of the first reference object after the delivery target planning route is updated; HD2 g Indicates the workload coefficient of the second reference object after the delivery target planning route is updated; μ (g，1) This represents the road condition difficulty coefficient in the updated topology map update result for the first reference object's planned delivery route in the current time and corresponding delivery scenario; μ (g，2) This indicates the road condition difficulty coefficient corresponding to the updated delivery target route for the second reference object in the current time's corresponding delivery scenario's topology map update result; the road condition difficulty coefficient is obtained by querying the maximum value of the cumulative traffic anomaly deviation corresponding to the dynamic layer information in each preset area of ​​the dynamic topology map in the corresponding time in the database within a preset form; the cumulative traffic anomaly deviation corresponding to the dynamic layer information in the preset area is equal to the sum of the products between the area occupied by each identified object model in the dynamic layer information of the preset area and the corresponding preset weight coefficient of the identified object model; LC g ξ represents the absolute value of the distance difference between the current location of the first and second reference objects in the updated corresponding delivery target planning route and the emergency dispatch transfer node; ξ represents the preset conversion factor.

[0099] S4. Based on the load balancing adaptation deviation value of each emergency dispatch planning scheme, match the best emergency dispatch planning scheme for mobile robots under the current topology map, and provide feedback to the administrator for confirmation.

[0100] When matching the best emergency dispatch plan for mobile robots based on the current topology map, the emergency dispatch plan with the smallest load balancing comprehensive adaptation deviation value is taken as the best emergency dispatch plan for mobile robots based on the current topology map.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real-time scheduling method for mobile robots based on dynamic topology maps, characterized in that, The method includes: S1. Sequentially obtain each delivery target node corresponding to each mobile robot and generate the initial delivery target planning route for the corresponding mobile robot. S2. Upload the path captured by the mobile robot during the delivery process to the cloud control terminal. The cloud control terminal dynamically updates the topology map of the corresponding delivery scenario at the current time and provides real-time feedback to the mobile robot. S3. Collect the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot; combine the topology map update results of the corresponding delivery scenario at the current time and the initial delivery target planning route of the corresponding mobile robot to generate an emergency scheduling plan for the mobile robot with abnormal workload status at the current time, and calculate the load balancing comprehensive adaptation deviation value of each emergency scheduling plan. The delivery status information of the mobile robot in S3 includes the remaining battery power, the delivery target node to be arrived at, the weight of the item to be delivered corresponding to each delivery target node, and the route segments not traversed in the initial delivery target planning route of the corresponding mobile robot. The workload coefficient for the i-th mobile robot at the current time is calculated using the following formula: ； Where Hi represents the workload coefficient of the i-th mobile robot at the current time; Yi represents the remaining battery power in the delivery status information of the i-th mobile robot at the current time; L (i,n) L represents the distance between the starting point and the nth delivery target node to be traversed within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; (i,n-1) This represents the distance from the starting point to the (n-1)th delivery target node to be traversed within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; when n=1, then L is determined. (i,n-1) The value of is 0; Ni represents the total number of delivery target nodes to be traversed in the route segment not yet traversed in the initial delivery target planning route of the i-th mobile robot at the current time; G (i,k) G represents the weight of the item to be delivered, corresponding to the k-th delivery target node, within the route segment not yet traversed in the initial delivery target route planned by the i-th mobile robot at the current time; where k∈[0, Ni-1], and when k=0, G (i,k) The value is 0; GZ represents the sum of the weights of the items to be delivered for the target node to be executed by the i-th mobile robot at the current time. Indicates the load capacity of the mobile robot within historical data. The average value of the ratio between power consumption and distance traveled while driving; If Hi is less than or equal to the preset value, the current workload state of the mobile robot is determined to be abnormal; otherwise, the current workload state of the mobile robot is determined to be normal. S4. Based on the load balancing adaptation deviation value of each emergency dispatch planning scheme, match the best emergency dispatch planning scheme for mobile robots under the current topology map, and provide feedback to the administrator for confirmation.

2. The real-time scheduling method for mobile robots based on a dynamic topology map according to claim 1, characterized in that: In S1, each delivery target node corresponds to a delivery location; the initial delivery target planning route of the mobile robot is the result of splicing the corresponding planned road segments of any two adjacent delivery target nodes in the database for each mobile robot.

3. The real-time scheduling method for mobile robots based on a dynamic topology map according to claim 1, characterized in that: S2 includes: S21. The cloud control terminal summarizes the path acquisition images uploaded by each mobile robot during the delivery process in real time; the path acquisition image uploaded by the i-th mobile robot during the delivery process at the j-th path is denoted as A. （i，j） ; S22, Obtain information from the mobile robot collecting A （i，j） The location information at that time is denoted as W. （i，j） ; Extract location information from the database as W （i，j） The static topological map layer information corresponding to the surrounding unit radius area is denoted as STA. （i，j） The topology map includes a static layer and a dynamic layer; S23. Identify and label A using image recognition technology. （i，j） Belongs to STA （i，j） The area, and combined with STA （i，j） The positional relationships between the static objects in A （i，j） The identification failed and it belongs to STA. （i，j） Update the marked region, and add A. （i，j） The unmarked remaining area is denoted as A. （i，j） The corresponding dynamic layer region to be identified; S24, Extract A （i，j） The corresponding dynamic layer to be identified region, distinct from the location information in the database under the corresponding mobile robot's shooting perspective, is W. （i，j） The area of ​​the dynamic topology map layer within a unit radius of the surrounding area is denoted as SA. （i，j） And SA was identified through image recognition technology. （i，j） The database contains pre-defined forms for each object model and the location area occupied by each identified object model, resulting in A. （i，j） The corresponding dynamic layer information to be updated; S25. The summary result of the dynamic layer information to be updated corresponding to each path captured by each mobile robot at the same time during the delivery process is used as the update result of the dynamic layer information in the topology map at the corresponding time under the corresponding delivery scenario; the combination result of the static layer information and the dynamic layer information update result of the topology map is used as the dynamic topology map at the corresponding time. S26. The motion control terminal feeds back the obtained dynamic topology map to each mobile robot in real time.

4. The real-time scheduling method for mobile robots based on a dynamic topology map according to claim 1, characterized in that: In the process of generating an emergency scheduling plan for a mobile robot with an abnormal workload status at the current time in S3, the summary set of delivery target nodes to be reached by the mobile robot with an abnormal workload status at the current time is extracted; this set is denoted as the emergency adaptation option set for the corresponding mobile robot; different emergency adaptation schemes are obtained, each of which is a set consisting of one or more elements in the emergency adaptation option set for the corresponding mobile robot. The workload status of the corresponding mobile robot is normal after removing delivery tasks from the emergency adaptation scheme, and the workload status of the corresponding mobile robot is abnormal when not all delivery tasks from the emergency adaptation scheme are removed. The specific steps to obtain the corresponding emergency dispatch planning schemes under different emergency adaptation solutions are as follows: S31. Obtain the location of the first delivery target node belonging to the emergency adaptation scheme in the initial delivery target planning route of the mobile robot with abnormal workload status, and record it as the scheduling reference node. S32. The route segment from the starting point of the untraveled route segment in the initial delivery target planning route of the mobile robot with the corresponding abnormal workload status to the previous delivery target node based on the scheduling reference node is recorded as the scheduling adaptation intersection segment. S33. Within the untraveled route segment of the initial delivery target planning route of the mobile robot with normal workload status, any delivery target node that is the same as the delivery target node in the scheduling adaptation intersection segment is recorded as an emergency scheduling transfer node; the corresponding mobile robot with abnormal workload status completes the transfer of the goods to be delivered corresponding to the corresponding emergency adaptation scheme at the emergency scheduling transfer node; the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme update the delivery target planning route, and obtain different emergency scheduling planning schemes for the mobile robot with abnormal workload status at the current time. In the obtained emergency scheduling planning scheme, the workload status of the mobile robots with normal workload status and the mobile robots with abnormal workload status that transfer the goods to be delivered corresponding to the corresponding emergency adaptation scheme are both normal based on the updated delivery target planning route. The updated delivery target planning route for a mobile robot with an abnormal workload is the result of splicing the corresponding planned road segments in the database for any two adjacent delivery target nodes among the unreached delivery target nodes after removing the corresponding emergency adaptation scheme. When updating the delivery target planning route of a mobile robot with normal workload, if the delivery target nodes in the corresponding emergency adaptation plan are all delivery target nodes that have not been reached in the delivery target planning route of the mobile robot with normal workload before the current time update, then the delivery order of the delivery target nodes and the corresponding delivery target planning route shall remain unchanged. If there are delivery target nodes in the corresponding emergency adaptation plan that are not in a normal workload state and have not been reached in the delivery target planning route before the current time update, then the delivery target nodes that are not in an abnormal workload state and have not been reached in the delivery target planning route before the current time update will be added to the end of the sequence of delivery target nodes to be reached, and the delivery target planning route of the corresponding mobile robot will be generated.

5. The real-time scheduling method for mobile robots based on a dynamic topology map according to claim 4, characterized in that: The formula for calculating the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme in S3 is as follows: ; Among them, P g This represents the load balancing comprehensive adaptation deviation value of the g-th emergency dispatch planning scheme; the mobile robot with an abnormal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the first reference object; the mobile robot with a normal workload status before the transfer and handover corresponding to the g-th emergency dispatch planning scheme is recorded as the second reference object; HD1 g This indicates the workload coefficient of the first reference object after the delivery target planning route is updated; HD2 g This represents the workload coefficient of the second reference object after the delivery target planning route is updated; μ （g，1） This represents the road condition difficulty coefficient in the updated topology map update result for the first reference object's planned delivery route in the current time and corresponding delivery scenario; μ （g，2） This indicates the road condition difficulty coefficient corresponding to the updated delivery target route for the second reference object in the current time's corresponding delivery scenario's topology map update result; the road condition difficulty coefficient is obtained by querying the maximum value of the cumulative traffic anomaly deviation corresponding to the dynamic layer information in each preset area of ​​the dynamic topology map in the corresponding time in the database within a preset form; the cumulative traffic anomaly deviation corresponding to the dynamic layer information in the preset area is equal to the sum of the products between the area occupied by each identified object model in the dynamic layer information of the preset area and the corresponding preset weight coefficient of the identified object model; LC g ξ represents the absolute value of the distance difference between the current location of the first and second reference objects in the updated corresponding delivery target planning route and the emergency dispatch transfer node; ξ represents the preset conversion factor. When matching the best emergency dispatch plan for mobile robots based on the current topology map, the emergency dispatch plan with the smallest load balancing comprehensive adaptation deviation value is taken as the best emergency dispatch plan for mobile robots based on the current topology map.

6. A real-time scheduling system for mobile robots based on dynamic topology maps, employing the real-time scheduling method for mobile robots based on dynamic topology maps as described in any one of claims 1-5, characterized in that: The system includes: a target information acquisition module, a topology map dynamic update module, an emergency dispatch planning scheme management module, and an emergency dispatch scheme screening and feedback module; The target information acquisition module sequentially acquires each delivery target node corresponding to each mobile robot and generates the initial delivery target planning route for the corresponding mobile robot. The topology map dynamic update module uploads the path captured by the mobile robot during the delivery process to the cloud control terminal. The cloud control terminal dynamically updates the topology map of the corresponding delivery scenario at the current time and provides real-time feedback to the mobile robot. The emergency dispatch planning scheme management module collects the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot; combined with the topology map update result of the corresponding delivery scenario at the current time and the initial delivery target planning route of the corresponding mobile robot, it generates an emergency dispatch planning scheme for the mobile robot with abnormal workload status at the current time, and calculates the load balancing comprehensive adaptation deviation value of each emergency dispatch planning scheme. The emergency dispatch scheme screening and feedback module matches the best emergency dispatch plan for mobile robots based on the load balancing comprehensive adaptation deviation value of each emergency dispatch plan scheme, and feeds it back to the administrator for confirmation.

7. The real-time scheduling system for mobile robots based on a dynamic topology map according to claim 6, characterized in that: The topology map dynamic update module includes a data upload unit and a topology map update feedback unit; The data uploading unit uploads the path captured by the mobile robot during the delivery process to the cloud control terminal; The topology map update feedback unit controls the cloud control terminal to dynamically update the topology map of the corresponding transportation scenario at the current time and provides real-time feedback to the mobile robot.

8. The real-time scheduling system for mobile robots based on a dynamic topology map according to claim 6, characterized in that: The emergency dispatch planning scheme management module includes a load status determination unit, an emergency dispatch planning unit, and an adaptation deviation calculation unit. The load status determination unit collects the delivery status information of each mobile robot in real time through sensors to determine the current workload status of the mobile robot. The emergency dispatch planning unit combines the updated topology map results of the corresponding delivery scenario at the current time with the initial delivery target planning route of the corresponding mobile robot to generate an emergency dispatch planning scheme for the mobile robot whose workload status is abnormal at the current time. The adaptation deviation calculation unit calculates the load balancing comprehensive adaptation deviation value for each emergency dispatch planning scheme.

9. The real-time scheduling system for mobile robots based on a dynamic topology map according to claim 6, characterized in that: The emergency dispatch plan screening and feedback module includes an optimal emergency dispatch plan screening unit and a feedback confirmation unit. The optimal emergency dispatch planning scheme selection unit matches the optimal emergency dispatch planning scheme for the mobile robot based on the load balancing comprehensive adaptation deviation value of each emergency dispatch planning scheme. The feedback confirmation unit will provide the administrator with the best emergency dispatch plan for mobile robots based on the current topology map for confirmation.

Images