Airport terminal manned robot scheduling method and system

Abstract

The invention relates to the technical field of robot scheduling, in particular to an airport terminal manned robot scheduling method and system, and the method comprises the steps: collecting the state information of a robot in real time, and receiving the real-time demands of a passenger, the real-time demands comprising the position information of the passenger and the target position information; a bipartite graph matching model is constructed, the bipartite graph matching model comprises a passenger demand set and an idle robot set which form two subsets of a bipartite graph respectively, and by constructing an OD matrix, the path length, the passenger waiting time and the robot energy consumption serve as optimization targets. A self-adaptive Hungary order dispatching algorithm is adopted, special environment characteristics of an airport terminal are combined, an airport terminal multi-robot dispatching scene is established, a bipartite graph order dispatching matching model is constructed, the path length, the passenger waiting time and the robot energy consumption serve as optimization targets, an OD matrix of a loading task is calculated, and the loading task is dispatched. And dynamic allocation and intelligent scheduling of passenger demands and robot resources are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of robot scheduling, and in particular to a method and system for scheduling manned robots in terminal buildings. Background Art

[0002] With the rapid development of global air traffic, airports, as important transportation hubs, are facing increasing passenger traffic and service demands. In public indoor environments such as terminals, passengers often need to walk from the security checkpoint to the boarding gate. Long-distance walking not only affects the travel experience of passengers, but is also a burden, especially for the elderly, passengers with limited mobility, and passengers carrying a lot of luggage.

[0003] With the rapid development of robotics technology, manned robots are gradually being used as a new type of automated transportation tool in indoor environments such as terminal buildings. Compared with manually driven passenger cars, manned robots have the ability of autonomous navigation, intelligent decision-making and flexible dispatching, and can autonomously pick up passengers and send them to designated locations.

[0004] However, when the above technology is actually used, the needs of passengers in the terminal are dynamically changing, and the scheduling of robots needs to consider multiple factors such as path length, energy consumption, passenger waiting time, etc. The existing scheduling system has the problems of low efficiency and poor flexibility when facing dynamic and multi-objective scheduling needs, and it is difficult to meet the optimization needs in complex scenarios. Summary of the invention

[0005] The purpose of the present invention is to provide a terminal manned robot scheduling method and system to solve the problems of low efficiency and poor flexibility when facing dynamic and multi-objective scheduling needs, and the problem of difficulty in meeting optimization needs in complex scenarios.

[0006] To achieve the above object, the present invention provides the following technical solution: a method and system for dispatching a manned robot in a terminal, comprising the following steps: S1. Information collection: collect robot status information in real time and receive passengers' real-time needs, including passengers' location information and target location information; S2. Bipartite graph model construction: construct a bipartite graph matching model, which includes a passenger demand set and an idle robot set, which respectively constitute two subsets of the bipartite graph. By constructing an OD matrix, the path length, passenger waiting time, and robot energy consumption are taken as optimization targets, and the weight cost value in the bipartite graph is calculated; S3, adaptive algorithm solution: Use the adaptive Hungarian algorithm to solve the bipartite graph matching model, realize dynamic task allocation, form the optimal dispatching plan, and send the dispatching instructions to the corresponding robot; S4, robot task execution: execute the pick-up task according to the dispatch instruction, and plan the path according to the path planning algorithm, while monitoring the robot's power status in real time and dynamically adjusting the task order strategy; S5. Serial dispatching mechanism: At the end of the task completion, the robot can instantly access new dispatching tasks through the serial dispatching mechanism to reduce idle time.

[0007] Preferably, step S3 also includes constructing a three-dimensional simulation model of the terminal building and conducting round-the-clock dispatch simulation experiments to verify and optimize the dispatch strategy.

[0008] Preferably, the adaptive Hungarian algorithm in step S3 specifically forms a priority ranking for robot task assignment according to multi-objective constraints of the spatiotemporal distribution of passenger density, passenger waiting time and robot energy consumption.

[0009] Preferably, the path planning algorithm in step S4 adopts the A* algorithm, combined with the raster map inside the terminal, to calculate the optimal driving path of the robot in real time to avoid crowded areas or obstacles.

[0010] Preferably, the chain dispatching mechanism in step S5 is specifically to add the robot that is about to reach the target point to the dispatching queue in advance, and give priority to matching the new task that is closest to the target point.

[0011] A system applied to the above-mentioned terminal manned robot scheduling method comprises: Dispatch service: used to receive passengers' real-time demands and build a bipartite graph matching model. The bipartite graph matching model includes a passenger demand set and an idle robot set, which respectively constitute two subsets of the bipartite graph. By constructing an OD matrix, the path length, passenger waiting time and robot energy consumption are used as optimization targets, the weight cost value in the bipartite graph is calculated, and the adaptive Hungarian algorithm is used to optimize the dispatch strategy. Robot side: according to the instructions of the dispatching service side, it executes the pick-up task, and performs path planning, energy consumption management and dynamic task allocation according to the algorithm. The robot side includes a path planning module, an energy consumption management module and a dynamic task allocation module; Passenger side: used to publish pick-up requirements, send real-time location through positioning, interact with the robot side, and receive information from the order-taking robot on the dispatch service side.

[0012] Preferably, the dispatch service terminal includes: Task scheduling module: Calculate the OD matrix based on the bipartite graph, generate a dispatch scheduling plan, update the dispatch strategy in real time, dynamically allocate robot tasks, and recalculate the optimal dispatch plan based on demand changes; Supply and demand forecasting module: Combines historical data with real-time data to predict changes in passenger demand over the next period of time and prioritizes allocating robot resources to high-demand areas; Chain dispatch module: Add robots that are about to reach the target point to the dispatch queue to improve dispatch efficiency.

[0013] Preferably, the robot terminal comprises: Path planning module: uses the A* algorithm to generate the optimal path from the current location to the target location. Combined with the raster map inside the terminal, it calculates the optimal driving path of the robot in real time to avoid crowded areas or obstacles. Energy consumption management module: monitors the power status of the robot in real time, dynamically adjusts the task ordering strategy according to the current power and driving route, and gives priority to robots with high power to perform long-distance tasks, thereby reducing task interruptions caused by insufficient power; Dynamic task allocation module: Through the chain dispatching mechanism, new dispatching tasks can be immediately connected after the task is completed to reduce idle time.

[0014] Preferably, the passenger terminal includes: Demand release module: users send real-time pick-up requests through terminal devices and specify the target location; Positioning module: provides the passenger's current location, which is used by the dispatching server to calculate the matching weight; Interactive feedback module: displays the robot's location, estimated arrival time, and driving path to the user, and provides interactive functions.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention adopts an adaptive Hungarian dispatching algorithm, combines the special environmental characteristics of the terminal, establishes a multi-robot dispatching scenario for the terminal, constructs a bipartite graph dispatching matching model, takes path length, passenger waiting time and robot energy consumption as optimization targets, calculates the OD matrix of the pick-up task, realizes dynamic allocation and intelligent dispatching of passenger demand and robot resources, ensures efficient completion of the pick-up task, and further improves the task turnover rate of the robot through a chain dispatching mechanism, significantly improves the dispatching efficiency of the multi-robot system in the terminal, reduces passenger waiting time, and ensures efficient operation of the robot and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the terminal multi-robot dispatching system architecture of the present invention; Figure 2 This is the service scenario of the manned robot in the terminal building of the present invention; Figure 3 It is a schematic diagram of a bipartite graph of the present invention; Figure 4 This is a schematic diagram of the cloud server dispatching process of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1-4 The present invention provides a technical solution: a method and system for dispatching a manned robot in a terminal, comprising the following steps: S1. Information collection: collect robot status information in real time and receive passengers' real-time needs, including passengers' location information and target location information; S2. Construction of bipartite graph model: The dispatching service regards the demand passengers and idle robots as two sets of bipartite graph, which constitute two subsets of the bipartite graph. One side is the passenger demand point, and the other side is the idle robot. The weight of each edge represents the cost of the robot's pick-up task, which is composed of the path length, passenger waiting time and energy consumption. Figure 1 As shown in the figure, by constructing the OD matrix, taking the path length, passenger waiting time and robot energy consumption as optimization targets, the dispatch service receives the passenger's pick-up demand in real time, and combines the robot's operating status to build a bipartite graph matching model based on the set of demanded passengers and the set of idle robots, and calculates the weight cost value in the bipartite graph; S3. Adaptive algorithm solution: The dispatch service uses the adaptive Hungarian algorithm to solve the bipartite graph matching model based on the multi-objective constraints of the spatiotemporal distribution of passenger density, passenger waiting time, and robot energy consumption, to achieve dynamic task allocation, prioritize robot task assignments, form the optimal dispatch solution, and send dispatch instructions to the corresponding robot. At the same time, a three-dimensional simulation model of the terminal is constructed, and all-weather dispatch simulation experiments are conducted to verify and optimize the dispatch strategy. S4, Robot task execution: The robot performs the pick-up task according to the dispatch instruction and performs path planning according to the path planning algorithm. The path planning algorithm adopts the A* algorithm, combined with the raster map inside the terminal, to calculate the robot's optimal driving path in real time, avoid crowded areas or obstacles, and monitor the robot's power status in real time to dynamically adjust the task order acceptance strategy; S5. Chain dispatch mechanism: At the end of the task completion, the robot side instantly accesses the new dispatch task through the chain dispatch mechanism. Specifically, the robot that is about to reach the target point is added to the dispatch queue in advance, and the new task that is closest is matched first to reduce idle time.

[0019] The present invention also provides a terminal manned robot dispatching system, comprising: Dispatch service: used to receive passengers' real-time demands and build a bipartite graph matching model. The bipartite graph matching model includes a passenger demand set and an idle robot set, which respectively constitute two subsets of the bipartite graph. By constructing an OD matrix, the path length, passenger waiting time and robot energy consumption are used as optimization targets, the weight cost value in the bipartite graph is calculated, and the adaptive Hungarian algorithm is used to optimize the dispatch strategy. The dispatch service includes: Task scheduling module: Calculate the OD matrix based on the bipartite graph, generate a dispatch scheduling plan, update the dispatch strategy in real time, dynamically allocate robot tasks, and recalculate the optimal dispatch plan based on demand changes; Supply and demand forecasting module: Combines historical data with real-time data to predict changes in passenger demand over the next period of time and prioritizes allocating robot resources to high-demand areas; Chain dispatch module: Add robots that are about to reach the target point to the dispatch queue to improve dispatch efficiency.

[0020] Robot side: according to the instructions of the dispatching service side, it executes the pick-up task, and performs path planning, energy consumption management and dynamic task allocation according to the algorithm. The robot side includes a path planning module, an energy consumption management module and a dynamic task allocation module; The robot side includes: Path planning module: uses the A* algorithm to generate the optimal path from the current location to the target location. Combined with the raster map inside the terminal, it calculates the optimal driving path of the robot in real time to avoid crowded areas or obstacles. Energy consumption management module: monitors the power status of the robot in real time, dynamically adjusts the task ordering strategy according to the current power and driving route, and gives priority to robots with high power to perform long-distance tasks, thereby reducing task interruptions caused by insufficient power; Dynamic task allocation module: Through the chain dispatching mechanism, new dispatching tasks can be immediately connected after the task is completed to reduce idle time.

[0021] Passenger side: used to publish pick-up requirements, send real-time location through positioning, interact with the robot side, and receive information from the order-taking robot on the dispatching service side; The passenger side includes: Demand release module: users send real-time pick-up requests through terminal devices and specify the target location; Positioning module: provides the passenger's current location, which is used by the dispatching server to calculate the matching weight; Interactive feedback module: displays the robot's location, estimated arrival time, and driving path to the user, and provides interactive functions.

[0022] The present invention also provides a specific implementation method, including: For the dispatching space of manned robots in the terminal, the feasible domain of manned robots will be the area from the security checkpoint to each boarding gate, providing convenient boarding services for passengers. When studying the feasible domain of manned robots, the main consideration is the area from the security checkpoint to each boarding gate, which is composed of various corridors in the terminal. Figure 2 As shown in the business flow chart of the flight, if the security checkpoint is the starting point and the boarding gates are the end points, the main manned robot serves in the isolation area (security checkpoint-boarding gate area), and sets up service information such as urban road network and U-turn landmarks in the isolation area. The main body is mainly formed by the transportation and return of passengers between the security checkpoint and the boarding gate, and sometimes there are services for picking up passengers in the middle. The main starting point for passengers to take the bus is mainly the security check exit, that is, the waiting area for transportation. Other starting points can be spread throughout the isolation area, and the target body for transportation to arrive is mainly the boarding gates spread throughout the isolation area. Of course, it can also be specified according to the destination location confirmed by the passengers in need, such as shopping and dining. Therefore, when there is a demand for cars in the isolation area space, it can be considered to implement a platform dispatching model to provide transportation services for passengers with global needs.

[0023] Collect robot status information in real time to obtain coordinate information of dispatchable manned robots in indoor environments, and present it as a supply point in space. When the manned robot receives a command from the cloud, it will go to a certain location to pick up passengers, or stop and pick up passengers when they wave on the road. When the demanding passenger initiates the use demand as a mobile terminal, it will be presented as a demand point in the space. When the passenger sees the robot and gestures to get on, the manned robot will move to the passenger and remind the passenger to get on.

[0024] Construct two sets of manned robots and required passengers, where the sets are , , forming two mutually disjoint subsets. After constructing the two sets, an edge is defined to match the points in the two sets in real time and assign a weighted cost value , construct a bipartite graph mathematical model, so the matching of manned robots and required passengers is a bipartite graph matching problem as follows Figure 3 shown.

[0025] According to the above bipartite graph relationship, the path cost of each manned robot to visit passengers with global demand will be calculated based on the indoor raster map.

[0026] The global raster map is represented by 0-1 (0 is the free area and 1 is the obstacle area), constructed as a 0-1 matrix of the raster map, and saved to the RP (Route Planning) server.

[0027] The A* path planning algorithm is used to calculate step by step from the starting point to the end point, and a multi-threaded access path planning is formed to obtain the estimated pick-up path cost from the global multi-manned robot to the target passenger.

[0028] Assume that at 2 seconds, the number of idle manned robots is (Penetration calculation), the number of passengers required is , the number of path visits is , showing a linear growth relationship.

[0029] Assuming demand for passengers , where the location information is , manned robot in , the location information is . Use the path planning algorithm to calculate the manned robot to each passenger in need The path cost is Therefore, the manned robot itself is taken as the source point and the passenger in need is taken as the end point to construct the pick-up OD matrix M As shown:

[0030] Therefore, when constructing a manned robot to calculate the current state of passengers with access needs, if only the pick-up task distance is considered, the estimated average waiting time for individual passengers with access needs is:

[0031] in Represents the average speed of the manned robot.

[0032] Introducing environmental road condition factors Expressing passenger density The impact on the speed of the manned robot is obtained, and according to the Greenshields model, the impact of passenger density on the speed of the manned robot at the terminal and the estimated arrival time are obtained, and the relevant formula is established as follows:

[0033]

[0034]

[0035]

[0036] in: : The number of passengers that can reach the quarantine area at each time period.

[0037] : The area accessible to passengers in the quarantine zone.

[0038] : Passenger density.

[0039] : The passenger saturation density of the area accessible by passengers in the isolation zone, that is, the passenger density obtained when the peak hour flow enters the isolation zone throughout the year.

[0040] : Maximum normal travel speed of manned robot.

[0041] :Each time period i The manned robot drives to the j The travel speed of each passenger.

[0042] : Estimated arrival time of the manned robot.

[0043] : The remaining mileage of the manned robot during its current transportation journey.

[0044] The number of idle manned robots in the terminal is , the number of passengers required is , and each pick-up task is a one-to-one shared ride service. Representing the needs of passengers from all regions, A manned robot representing a global number.

[0045] If manned robot With demand passengers The path cost obtained by the manned robot visiting the required passenger is , build its access path matrix M On this basis, the environmental factors of individual manned robots will be considered, including the pick-up path, passenger waiting time and robot energy consumption, that is, the path-time-energy scheduling model, and the three indicators will be considered and integrated into the weights , get the fitness function value As the matching weight value, the bipartite graph is constructed to carry the task OD matrix (Orignation-Destination) , which can be expressed as:

[0046] The principle of minimizing the fitness value of the terminal's global manned robot access and pick-up mission is established to obtain the global manned robot pick-up dispatch matrix, and the dispatch association process of the manned robot and the passenger in need is modeled. Its objective function is is the independent variable, and The corresponding state variable matrix is ​​obtained is the dispatch association matrix, let its dispatch association matrix be M ,at this time .

[0047] The target fitness function is:

[0048] Constraints: in: m : is the total number of idle manned robots.

[0049] n : is the number of passengers required.

[0050] : Whether the busy manned robot accepts new tasks. When the manned robot is too low in power Or there is a task in progress, then ; When the manned robot has normal power and has no tasks to perform, .

[0051] : The current state of the idle manned robot.

[0052] when =0, it means that the i-th idle manned robot will not be assigned to the j-th passenger in need; when = 1, the i-th idle manned robot is assigned to the j passengers in need, Can only be 0 or 1.

[0053] :For i An idle manned robot will pick up the j The path cost of a passenger who needs to travel.

[0054] f : is the fitness value and of the global manned robot visiting the target passenger.

[0055] : are dynamic weight values, corresponding to the pick-up route, passenger waiting time, and manned robot pick-up energy consumption.

[0056] : The running speed of the manned robot, which is affected by environmental factors.

[0057] : Output rated voltage for manned robot.

[0058] : Output rated current for manned robot.

[0059] : is the output dispatch matrix.

[0060] The system uses historical data and real-time data, combined with flight schedules and passenger flow trends, to analyze changes in passenger demand at different times in the terminal, predict passenger demand in the terminal in the future, and adjust the robot distribution in advance to ensure that there are sufficient robot resources in peak areas and optimize the scheduling strategy.

[0061] The system can monitor the changes in passenger demand and the working status of robots in the terminal in real time. When new passenger demand appears, the system will immediately update the bipartite graph and recalculate the optimal dispatching strategy according to the algorithm. When the robot completes the current task, it will be assigned the next task in advance according to its current position and power status to ensure seamless connection between tasks and reduce the robot's idle time. Figure 4 shown.

[0062] In order to ensure the efficient operation of the robot in a complex environment, the system uses the A* algorithm for path planning. Combined with the raster map of the terminal, the robot can avoid crowded areas and obstacles to ensure the selection of the optimal path. The specific implementation steps are as follows: Grid map generation: The interior of the terminal is divided into multiple small grids, and each grid point represents a possible location of the robot. By setting different weight values, the system can mark obstacles, crowded areas, etc., so that the robot can avoid these areas when planning the path.

[0063] Path calculation: The A* algorithm searches for the shortest path from the current robot position to the target point, calculates the optimal path among multiple possible paths, and generates a driving route for the robot to follow. At the same time, considering the dynamic flow of people in the terminal, the system can adjust the path in real time during the robot's driving process to avoid congested areas.

[0064] Energy consumption monitoring and management: The system monitors the energy consumption of each robot in real time. When the robot is running low on power, the system will prioritize shorter tasks to ensure that the robot can successfully complete the task and return to the charging station. The energy consumption management module can also arrange a charging path for the robot in advance based on the power status.

[0065] Through the three-dimensional simulation model, the actual operation of the robots in the terminal is simulated, and the effectiveness of the dispatching algorithm is verified under different conditions such as the distribution of passenger demand and the number of robots. Through a large number of simulation experiments, the system can detect possible operation bottlenecks in advance and optimize the dispatching strategy in actual applications to ensure the efficient operation of the system in a complex environment.

[0066] Adopting the adaptive Hungarian dispatching algorithm and combining the special environmental characteristics of the terminal, a multi-robot dispatching scenario for the terminal is established. By constructing a bipartite graph dispatching matching model, taking path length, passenger waiting time and robot energy consumption as optimization targets, the OD matrix of the pick-up task is calculated to achieve dynamic allocation and intelligent dispatching of passenger demand and robot resources. At the same time, it can also predict passenger demand in the terminal in real time in the future and adjust the robot distribution in advance to ensure that there are enough robot resources in the peak area. Through the three-dimensional simulation model, the all-weather dispatching simulation experiment is carried out. The system can detect possible operation bottlenecks in advance, optimize the dispatching strategy, ensure the efficient completion of the pick-up task, and further improve the task turnover rate of the robot through the chain dispatching mechanism, which significantly improves the dispatching efficiency of the multi-robot system in the terminal, reduces the waiting time of passengers, ensures the efficient operation and resource utilization of the robot, and is suitable for public indoor places such as terminals and railway stations with dense traffic.

[0067] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0068] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for dispatching a manned robot in a terminal building, characterized in that: The following steps are involved: S1. Information collection: collect robot status information in real time and receive passengers' real-time needs, including passengers' location information and target location information; S2. Bipartite graph model construction: construct a bipartite graph matching model, which includes a passenger demand set and an idle robot set, which respectively constitute two subsets of the bipartite graph. By constructing an OD matrix, the path length, passenger waiting time, and robot energy consumption are taken as optimization targets, and the weight cost value in the bipartite graph is calculated; S3, adaptive algorithm solution: Use the adaptive Hungarian algorithm to solve the bipartite graph matching model, realize dynamic task allocation, form the optimal dispatching plan, and send the dispatching instructions to the corresponding robot; S4, robot task execution: execute the pick-up task according to the dispatch instruction, and plan the path according to the path planning algorithm, while monitoring the robot's power status in real time and dynamically adjusting the task order strategy; S5. Serial dispatching mechanism: At the end of the task completion, the robot can instantly access new dispatching tasks through the serial dispatching mechanism to reduce idle time.

2. The terminal manned robot dispatching method according to claim 1, characterized in that: The step S3 also includes constructing a three-dimensional simulation model of the terminal building and conducting round-the-clock dispatch simulation experiments to verify and optimize the dispatch strategy.

3. The terminal manned robot dispatching method according to claim 2, characterized in that: The adaptive Hungarian algorithm in step S3 specifically forms a priority ranking for robot task assignment according to the multi-objective constraints of the spatiotemporal distribution of passenger density, passenger waiting time and robot energy consumption.

4. The terminal manned robot dispatching method according to claim 3, characterized in that: The path planning algorithm in step S4 adopts the A* algorithm, combined with the raster map inside the terminal, to calculate the optimal driving path of the robot in real time to avoid crowded areas or obstacles.

5. The terminal manned robot dispatching method according to claim 4, characterized in that: The chain dispatching mechanism in step S5 is specifically to add the robot that is about to reach the target point to the dispatching queue in advance, and give priority to matching the new task that is closest.

6. A system applied to the terminal manned robot dispatching method according to any one of claims 1 to 5, characterized in that: include: Dispatch service: used to receive passengers' real-time demands and build a bipartite graph matching model. The bipartite graph matching model includes a passenger demand set and an idle robot set, which respectively constitute two subsets of the bipartite graph. By constructing an OD matrix, the path length, passenger waiting time and robot energy consumption are used as optimization targets, the weight cost value in the bipartite graph is calculated, and the adaptive Hungarian algorithm is used to optimize the dispatch strategy. Robot side: according to the instructions of the dispatching service side, it executes the pick-up task, and performs path planning, energy consumption management and dynamic task allocation according to the algorithm. The robot side includes a path planning module, an energy consumption management module and a dynamic task allocation module; Passenger side: used to publish pick-up requirements, send real-time location through positioning, interact with the robot side, and receive information from the order-taking robot on the dispatch service side.

7. The system according to claim 6, characterized in that: The dispatch service end includes: Task scheduling module: Calculate the OD matrix based on the bipartite graph, generate a dispatch scheduling plan, update the dispatch strategy in real time, dynamically allocate robot tasks, and recalculate the optimal dispatch plan based on demand changes; Supply and demand forecasting module: Combines historical data with real-time data to predict changes in passenger demand over the next period of time and prioritizes allocating robot resources to high-demand areas; Chain dispatch module: Add robots that are about to reach the target point to the dispatch queue to improve dispatch efficiency.

8. The system according to claim 7, characterized in that: The robot end comprises: Path planning module: uses the A* algorithm to generate the optimal path from the current location to the target location. Combined with the raster map inside the terminal, it calculates the optimal driving path of the robot in real time to avoid crowded areas or obstacles. Energy consumption management module: monitors the power status of the robot in real time, dynamically adjusts the task ordering strategy according to the current power and driving route, and gives priority to robots with high power to perform long-distance tasks, thereby reducing task interruptions caused by insufficient power; Dynamic task allocation module: Through the chain dispatching mechanism, new dispatching tasks can be immediately connected after the task is completed to reduce idle time.

9. The system according to claim 8, characterized in that: The passenger terminal includes: Demand release module: users send real-time pick-up requests through terminal devices and specify the target location; Positioning module: provides the passenger's current location, which is used by the dispatching server to calculate the matching weight; Interactive feedback module: displays the robot's location, estimated arrival time, and driving path to the user, and provides interactive functions.

Images