Intelligent robot path scheduling method and system for scenic spot tourists

Abstract

The invention provides an intelligent robot path scheduling method and system for scenic spot tourists, and relates to the technical field of path scheduling, and the method comprises the steps: obtaining robot position and state information and tourist service request information, and building a scenic spot electronic map; constructing a robot service capability evaluation model; path planning and scoring are carried out, and an optimal service path is selected; selecting the most suitable target service robot; sending the optimal service path to the target robot and loading the optimal service path; the motion trail of the robot is monitored in real time, and the path is dynamically adjusted if necessary. According to the method, the scenic spot service efficiency and tourist experience are improved.

Description

Technical Field

[0001] The present invention relates to path scheduling technology, and particularly to an intelligent robot path scheduling method and system for scenic area tourists. Background Art

[0002] Traditional robot scheduling methods often only consider the position information of the robot, ignoring the working status and service capabilities of the robot. This may result in assigning a robot with insufficient power or an overloaded current task volume when selecting a service robot, affecting the service quality and efficiency.

[0003] Secondly, existing path planning algorithms usually only based on the principle of the shortest distance, without fully considering the actual situation of the scenic area. This may lead to the planned path passing through congested areas or areas with sparse scenic spots, affecting the tourist experience.

[0004] Finally, the current scheduling system lacks a real-time monitoring and dynamic adjustment mechanism. Once a path blockage or an emergency occurs, it cannot respond and adjust in time, easily causing service interruption or delay. This not only reduces the service efficiency but may also affect the safety of tourists. Summary of the Invention

[0005] Embodiments of the present invention provide an intelligent robot path scheduling method and system for scenic area tourists, which can solve the problems in the prior art.

[0006] In the first aspect of the embodiments of the present invention,

[0007] An intelligent robot path scheduling method for scenic area tourists is provided, including:

[0008] Obtaining the real-time position information and working status information of multiple intelligent robots in the scenic area, and obtaining service request information sent by scenic area tourists through a mobile terminal, where the service request information includes the current position information, destination position information, and service type information of the tourists, establishing a scenic area electronic map according to the geographical information in the scenic area, dividing multiple service areas in the scenic area electronic map, and presetting multiple alternative service paths in each service area; the working status information includes the remaining power information, current task volume information, and historical service score information of the intelligent robot;

[0009] Based on the real-time position information and working status information of the intelligent robot, a robot service capability evaluation model is constructed. The robot service capability evaluation model calculates the comprehensive service capability score of the robot through weighted calculation by assigning corresponding weights to the remaining power information, current task volume information, and historical service score information respectively; according to the current position information, destination position information, and service type information of the tourist, path planning is carried out from the multiple alternative service paths, multiple candidate service paths are calculated using the A* algorithm, and the candidate service paths are scored based on path distance, congestion degree, and scenic spot distribution density, and the path with the highest score is selected as the optimal service path;

[0010] Sort the multiple intelligent robots in the scenic area according to the robot comprehensive service capability score, and select the intelligent robot with the highest score and in the same service area as the target service robot; send the optimal service path to the target service robot, and load the optimal service path in the navigation system of the target service robot; monitor the movement trajectory of the target service robot in real time through the scenic area electronic map, and when a path blockage or emergency is detected, trigger the path dynamic adjustment mechanism, recalculate the optimal service path and send the updated path information to the target service robot.

[0011] Establish a scenic area electronic map according to the geographical information in the scenic area, divide multiple service areas in the scenic area electronic map, and preset multiple alternative service paths in each service area, including:

[0012] Obtain the coordinate points of road nodes, scenic spot positions, service facility positions, and terrain feature coordinate points in the scenic area through the RTK-GPS measurement system, perform vectorization processing on the existing CAD topographic map of the scenic area to obtain vectorized terrain data, and register and fuse the vectorized terrain data with the coordinate points obtained by the RTK-GPS measurement system;

[0013] Obtain satellite remote sensing image data and perform orthorectification on the satellite remote sensing image data, overlay the orthorectified satellite remote sensing image data with the registered and fused data to obtain a geographical information dataset; construct a scenic area electronic map, and divide the scenic area electronic map into multiple service areas;

[0014] Construct a weighted directed graph model of the road network within each of the said service areas. The edge weights of the weighted directed graph model include the actual road distance, road passing capacity, road section congestion prediction value, scenic spot weight coefficient along the route, and terrain slope coefficient. Based on the weighted directed graph model, generate multiple differentiated alternative service paths within each of the said service areas, calculate the path similarity of the multiple differentiated alternative service paths, eliminate the alternative service paths with a path overlap rate exceeding a preset threshold, and verify the feasibility of the remaining alternative service paths according to the robot motion constraint conditions to obtain a set of final alternative service paths that meet the motion constraint conditions.

[0015] The method further includes:

[0016] Use the Kriging interpolation algorithm to perform spatial interpolation on the discrete sampling points in the geographic information dataset to obtain a continuous terrain surface model, use a vector data structure to store the building information and service facility information in the scenic area, and correct the error of the RTK-GPS signal through a differential correction algorithm to obtain an electronic map of the scenic area;

[0017] Divide the electronic map of the scenic area into multiple grid cells, obtain the tourist flow density information, service facility density information, and terrain feature information of each grid cell, perform weighted calculations on the tourist flow density information, service facility density information, and terrain feature information respectively according to preset weight coefficients to obtain the service demand weight of each grid cell, and use an improved K-means clustering algorithm to divide the electronic map of the scenic area into multiple service areas according to the service demand weight.

[0018] Based on the real-time position information and working status information of the intelligent robot, construct a robot service ability evaluation model. The robot service ability evaluation model calculates the comprehensive service ability score of the robot by assigning corresponding weights to the remaining battery power information, current task volume information, and historical service score information respectively through weighted calculation, including:

[0019] Use GPS and an indoor positioning system to obtain the real-time position information of the robot, perform Kalman filtering processing on the real-time position information to obtain the accurate position information of the robot, collect the battery pack voltage parameters, current parameters, and temperature parameters through a battery management system, calculate the remaining battery power information of the robot based on a battery health state evaluation algorithm, collect the number of service tasks being executed by the robot, task type information, and estimated completion time information, and obtain tourist evaluation information, task completion rate information, and service timeliness information;

[0020] Input the tourist evaluation information, task completion rate information, and service timeliness information into a time decay function for weighted processing to obtain a historical service score. Calculate a remaining battery score by performing a weighted calculation on the ratio of the remaining battery information to the preset full battery information and the predicted service mileage information. Multiply the difference between the maximum task capacity information and the current task quantity information by the task difficulty coefficient to obtain a task volume score;

[0021] Based on the historical service score, establish a fuzzy relation matrix. Use the analytic hierarchy process to determine the weights of the remaining battery dimension, the current task volume dimension, and the historical service dimension. Multiply the weights of the remaining battery dimension, the current task volume dimension, and the historical service dimension by the corresponding remaining battery score, task volume score, and historical service score respectively, and sum them up to obtain the robot's comprehensive service ability score.

[0022] Perform path planning from the multiple alternative service paths, use the A* algorithm to calculate multiple candidate service paths, and score the candidate service paths based on path distance, congestion level, and scenic spot distribution density. Select the path with the highest score as the optimal service path, including:

[0023] Construct a dynamic heuristic function that includes parameters such as the straight-line distance from the node to the target point, congestion cost parameter, and scenic spot density parameter. Collect the tourist flow data of the road section, calculate the congestion cost parameter according to the ratio of the tourist flow data of the road section to the maximum carrying capacity of the road section, collect the location information of surrounding scenic spots, and calculate the scenic spot density parameter according to the sum of the reciprocals of the distances from the node to be evaluated to each scenic spot;

[0024] Based on the dynamic heuristic function, perform multiple rounds of iterative search. In each round of iterative search, set a search taboo table and record the search tree expansion historical information. Determine the node traversal order according to the search tree expansion historical information, and generate multiple candidate service paths;

[0025] Calculate the maximum path length and the minimum path length among the multiple candidate service paths. Normalize the path length of the current candidate service path to be evaluated with the maximum path length and the minimum path length to obtain a path length score. Calculate the path congestion degree score as the ratio of the sum of the products of the congestion coefficients of each road section and the corresponding road section lengths to the total path length. Calculate the scenic spot coverage score as the ratio of the sum of the weights of the scenic spots covered by the path to the total number of scenic spots;

[0026] Collect the real-time congestion degree information of the road section, the road section traffic status information, the scenic spot activity information, and the service demand density information. When the real-time congestion degree information of the road section exceeds the preset congestion threshold, or the road section traffic status information changes, or new scenic spot activity information appears, or the service demand density information changes significantly, determine the affected path section;

[0027] Update the road network status information of the affected path section, recalculate the dynamic heuristic function of the affected path section, keep the unaffected path sections in the original path unchanged, regenerate candidate service paths for the affected path section and calculate the path length score, path congestion score, and scenic spot coverage score, perform weighted calculation on the path length score, path congestion score, and scenic spot coverage score to obtain the comprehensive path score, and select the candidate service path with the highest comprehensive path score as the optimal service path.

[0028] Real-time monitor the movement trajectory of the target service robot through the scenic area electronic map. When a path blockage or emergency is detected, trigger the path dynamic adjustment mechanism, recalculate the optimal service path, and send the updated path information to the target service robot, including:

[0029] Construct a movement trajectory based on the real-time position information of the service robot, calculate the speed change rate of the movement trajectory, the deviation distance from the planned path, and the residence duration of the positioning point. When the speed change rate exceeds the preset speed threshold, or the deviation distance exceeds the preset distance threshold, or the residence duration exceeds the preset duration threshold, trigger anomaly detection;

[0030] Collect lidar scan data and road control information of the scenic area management system, identify fixed obstacles based on the lidar scan data, determine temporarily closed sections based on the road control information, collect the operating parameters of the key components of the service robot and tourist help signals, and determine whether there are equipment failures or emergency help requests;

[0031] When a fixed obstacle or a temporarily closed section is detected, calculate the obstacle influence range, search for alternative detour paths within the obstacle influence range, calculate the detour cost and waiting cost of the alternative detour paths, and select the alternative detour path with the minimum cost as the local adjustment plan;

[0032] When an equipment failure or an emergency help signal is detected, determine the position of the nearest safe avoidance point, plan the shortest evacuation path from the current position of the service robot to the position of the safe avoidance point, and generate an emergency avoidance instruction;

[0033] Convert the alternative detour path or the shortest evacuation path corresponding to the local adjustment plan into a sequence of path points, add speed parameters and steering parameters, generate a path execution time sequence table, and send the path execution time sequence table to the service robot in segments through a communication channel, receive the instruction confirmation information returned by the service robot, and record the deviation information and anomaly information during the path execution process.

[0034] In the second aspect of the embodiments of the present invention,

[0035] Provide an intelligent robot path scheduling system for scenic area tourists, including:

[0036] The first unit is used to obtain the real-time location information and working status information of multiple intelligent robots in the scenic area, and obtain the service request information sent by scenic area tourists through mobile terminals. The service request information includes the current location information, destination location information and service type information of the tourists. An electronic map of the scenic area is established according to the geographical information in the scenic area. Multiple service areas are divided in the electronic map of the scenic area, and multiple alternative service paths are preset in each service area; the working status information includes the remaining battery information, current task volume information and historical service score information of the intelligent robot;

[0037] The second unit is used to construct a robot service ability evaluation model based on the real-time location information and working status information of the intelligent robot. The robot service ability evaluation model obtains the comprehensive service ability score of the robot by performing weighted calculation on the remaining battery information, current task volume information and historical service score information by assigning corresponding weights respectively; according to the current location information, destination location information and service type information of the tourist, path planning is carried out from the multiple alternative service paths, the A* algorithm is used to calculate multiple candidate service paths, and the candidate service paths are scored based on path distance, congestion degree and scenic spot distribution density, and the path with the highest score is selected as the optimal service path;

[0038] The third unit is used to sort multiple intelligent robots in the scenic area according to the comprehensive service ability score of the robot, and select the intelligent robot with the highest score and located in the same service area as the target service robot; send the optimal service path to the target service robot, and load the optimal service path in the navigation system of the target service robot; monitor the movement track of the target service robot in real time through the electronic map of the scenic area, and when a path blockage or emergency is detected, trigger a path dynamic adjustment mechanism, recalculate the optimal service path and send the updated path information to the target service robot.

[0039] In the third aspect of the embodiments of the present invention,

[0040] An electronic device is provided, including:

[0041] A processor;

[0042] A memory for storing instructions executable by the processor;

[0043] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0044] In the fourth aspect of the embodiments of the present invention,

[0045] Provided is a computer-readable storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the foregoing method is implemented.

[0046] The beneficial effects of this application are as follows:

[0047] 1. Improve the accuracy and efficiency of intelligent robot scheduling:

[0048] By obtaining the real-time location information and working status information of multiple intelligent robots in the scenic area, and combining with the robot service ability evaluation model, the most suitable target service robot can be selected more accurately. This method takes into account multiple factors such as the remaining battery power of the robot, the current task volume, and the historical service score, thereby improving the accuracy and efficiency of the scheduling decision-making and ensuring the best quality service for tourists.

[0049] 2. Optimize the service path planning and enhance the tourist experience:

[0050] By establishing an electronic map of the scenic area and presetting multiple alternative service paths, calculating the candidate service paths in combination with the A* algorithm, and scoring considering factors such as path distance, congestion degree, and scenic spot distribution density, the optimal service path can be selected. This method can not only improve the service efficiency, but also provide a more comfortable and convenient tour experience for tourists, avoiding unnecessary congestion and detours.

[0051] 3. Implement dynamic path adjustment to cope with emergencies:

[0052] By real-time monitoring the movement trajectory of the target service robot and triggering the path dynamic adjustment mechanism when detecting path blockage or emergency, the optimal service path can be recalculated in time. This real-time adjustment ability greatly improves the flexibility and adaptability of the system, can effectively cope with emergencies in the scenic area, ensure the continuity and reliability of the service, and further improve the tourist satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flowchart of the intelligent robot path scheduling method for scenic area tourists according to an embodiment of the present invention;

[0054] Figure 2 It is a structural diagram of the intelligent robot path scheduling system for scenic area tourists according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0056] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0057] Figure 1 It is a flowchart of an intelligent robot path scheduling method for scenic area visitors according to an embodiment of the present invention. As Figure 1 shown, the method includes:

[0058] S101. Obtain the real-time position information and working status information of multiple intelligent robots in the scenic area, and obtain the service request information sent by scenic area visitors through a mobile terminal. The service request information includes the current position information, destination position information, and service type information of the visitors. Establish a scenic area electronic map based on the geographical information in the scenic area, divide multiple service areas in the scenic area electronic map, and preset multiple alternative service paths in each service area; the working status information includes the remaining battery information, current task volume information, and historical service score information of the intelligent robot.

[0059] S102. Based on the real-time position information and working status information of the intelligent robot, construct a robot service ability evaluation model. The robot service ability evaluation model obtains the comprehensive service ability score of the robot by performing weighted calculation on the remaining battery information, current task volume information, and historical service score information by assigning corresponding weights respectively; according to the current position information, destination position information, and service type information of the visitor, perform path planning from the multiple alternative service paths, calculate multiple candidate service paths using the A* algorithm, and score the candidate service paths based on the path distance, congestion degree, and scenic spot distribution density, and select the path with the highest score as the optimal service path.

[0060] S103. Sort multiple intelligent robots in the scenic area according to the comprehensive service ability scores of the robots, and select the intelligent robot with the highest score and in the same service area as the target service robot; send the optimal service path to the target service robot, and load the optimal service path into the navigation system of the target service robot; monitor the movement trajectory of the target service robot in real time through the scenic area electronic map, and when a path blockage or emergency is detected, trigger a path dynamic adjustment mechanism, recalculate the optimal service path, and send the updated path information to the target service robot.

[0061] The present invention provides an intelligent robot path scheduling method for scenic area tourists, and the specific implementation manner is as follows:

[0062] First, the system obtains the real-time position information and working status information of multiple intelligent robots in the scenic area. The real-time position information can be obtained through the built-in GPS module of the robot, and the accuracy can reach the meter level. The working status information includes remaining battery information, current task volume information, and historical service score information. The remaining battery information is monitored in real time by the robot battery management system, accurate to 1%. The current task volume information records the number of tasks that the robot is executing and to be executed. The historical service score information is based on the evaluation of the robot service by tourists, using a 5-point system.

[0063] At the same time, the system obtains the service request information sent by scenic area tourists through the mobile terminal. The service request information includes the current position information, destination position information, and service type information of the tourists. Tourists can initiate a service request through the scenic area APP, and the system automatically obtains the current GPS position of the tourists. The tourists also need to select the destination position and the required service type, such as tour guide, voice explanation, luggage handling, etc.

[0064] The system establishes an electronic map according to the geographical information of the scenic area, including elements such as roads, buildings, and scenic spots. Multiple service areas are divided on the electronic map, such as the east area, west area, central area, etc. Multiple alternative service paths are preset in each service area, considering factors such as scenic spot distribution and road conditions, and are pre-planned and input into the system by the scenic area management personnel.

[0065] Next, the system constructs a robot service capacity evaluation model based on the real-time position information and working status information of the intelligent robot. This model assigns weights to the remaining battery information, current task volume information, and historical service score information respectively for weighted calculation to obtain the comprehensive service capacity score of the robot. Specifically, the weight of the remaining battery information is 0.4, the weight of the current task volume information is 0.3, and the weight of the historical service score information is 0.3. For example, if a robot has a remaining battery of 80%, a current task volume of 2, and a historical service score of 4.5, then its comprehensive service capacity score is: 0.4 * 80 + 0.3 * ((10 - 2) / 10 * 100) + 0.3 * 90 = 80.6 points.

[0066] According to the current location information, destination location information, and service type information of the tourist, the system performs path planning from multiple alternative service paths. First, the A* algorithm is used to calculate multiple candidate service paths. The A* algorithm combines the advantages of best-first search and heuristic search and can quickly find the optimal path. Then, the candidate service paths are scored based on path distance, congestion level, and scenic spot distribution density. The shorter the path distance, the higher the score; the lower the congestion level, the higher the score; the higher the scenic spot distribution density, the higher the score. The weights of each factor are 0.4, 0.3, and 0.3 respectively. The path with the highest score is selected as the optimal service path.

[0067] The system sorts multiple intelligent robots in the scenic area according to the comprehensive service capacity scores of the robots, and selects the intelligent robot with the highest score and located in the same service area as the target service robot. For example, if a tourist is located in the east area and requests service, the system will select the robot with the highest comprehensive service capacity score in the east area as the target service robot.

[0068] The system sends the optimal service path to the target service robot and loads this path in the robot's navigation system. The robot navigation system uses laser SLAM technology. By scanning the surrounding environment with a lidar, combining odometer information and IMU data, it constructs an environmental map in real time and locates its own position, and accurately navigates to the destination.

[0069] The system monitors the movement trajectory of the target service robot in real time through the scenic area electronic map. When a path blockage or emergency is detected, such as road construction, crowded people flow, etc., the path dynamic adjustment mechanism is triggered. The system recalculates the optimal service path, considering the current road conditions and the robot's position, and uses the dynamic programming algorithm to quickly generate a new optimal path. The system sends the updated path information to the target service robot in real time, and the robot immediately adjusts the navigation direction after receiving the new path.

[0070] The beneficial effects of the present invention are mainly reflected in the following three aspects:

[0071] 1. The service efficiency and quality of intelligent robots are improved. By constructing a robot service ability evaluation model, considering factors such as remaining battery power, current task volume, and historical service scores, the most suitable robot is selected to provide services, ensuring the stability and reliability of service quality.

[0072] 2. The service path planning is optimized, enhancing the tourist experience. The A* algorithm combined with a multi-factor scoring mechanism is used to select the optimal service path, taking into account factors such as path distance, congestion level, and scenic spot distribution density, which can not only improve efficiency but also allow tourists to appreciate more scenic spots.

[0073] 3. The real-time dynamic adjustment of the path is realized, improving the flexibility and adaptability of the system. By monitoring the robot's movement trajectory in real time, path blockages or emergencies are detected and responded to in a timely manner, and the path is dynamically adjusted, ensuring the continuity and reliability of the service and greatly enhancing the intelligence level of the system.

[0074] In an optional implementation manner, an electronic map of the scenic area is established according to the geographical information in the scenic area. Multiple service areas are divided in the electronic map of the scenic area, and multiple alternative service paths are preset in each service area, including:

[0075] The coordinate points of road nodes, scenic spot locations, service facility locations, and terrain feature points in the scenic area are obtained through the RTK-GPS measurement system. The existing CAD topographic map of the scenic area is vectorized to obtain vectorized terrain data, and the vectorized terrain data is registered and fused with the coordinate points obtained by the RTK-GPS measurement system;

[0076] Satellite remote sensing image data is obtained and orthorectified for the satellite remote sensing image data. The orthorectified satellite remote sensing image data is superimposed on the registered and fused data to obtain a geographical information dataset; an electronic map of the scenic area is constructed, and the electronic map of the scenic area is divided into multiple service areas;

[0077] A weighted directed graph model of the road network in each service area is constructed. The edge weights of the weighted directed graph model include actual road distance, road traffic capacity, predicted value of road section congestion, weight coefficient of passing scenic spots, and terrain slope coefficient. Based on the weighted directed graph model, multiple differentiated alternative service paths are generated in each service area. The path similarity of the multiple differentiated alternative service paths is calculated, and the alternative service paths with a path overlap rate exceeding a preset threshold are removed. The remaining alternative service paths are verified for feasibility according to the robot's movement constraint conditions to obtain a set of final alternative service paths that meet the movement constraint conditions.

[0078] The present invention provides a method for constructing an electronic map of a scenic area based on geographical information and dividing service areas. The specific implementation manner is as follows:

[0079] First, various coordinate point information within the scenic area is obtained through the RTK-GPS measurement system. The RTK-GPS measurement system includes a reference station and several mobile stations. The reference station is fixedly installed at a known coordinate point within the scenic area, and the mobile stations are carried by surveyors to conduct measurements within the scenic area. The surveyors move along the scenic area roads and record a road node coordinate point every 10 meters; record the scenic spot location coordinate points at the center position of each scenic spot; record the service facility location coordinate points at the entrance positions of service facilities such as restaurants, toilets, and ticket offices within the scenic area; record the terrain feature coordinate points for terrain feature points such as mountains, rivers, and lakes within the scenic area. For example, a total of 5000 road node coordinate points, 50 scenic spot location coordinate points, 100 service facility location coordinate points, and 200 terrain feature coordinate points are recorded within a certain scenic area.

[0080] Next, the existing CAD topographic maps of the scenic area are vectorized. First, the CAD topographic maps are scanned to obtain raster images. Then, image processing software is used to perform preprocessing such as binarization and thinning on the raster images. After that, an automatic vectorization algorithm is used to extract the lines and polygon contours in the images and convert them into vector data formats. Finally, manual checking and editing are carried out to correct the errors in the automatic vectorization process to obtain accurate vectorized topographic data.

[0081] Then, the vectorized topographic data is registered and fused with the coordinate points obtained by RTK-GPS measurement. First, several homologous points are selected in the vectorized topographic data and GPS coordinate points, such as the scenic area entrances and exits, and major scenic spots. The coordinate transformation parameters between the two sets of data are calculated using these homologous points. Then, the vectorized topographic data is subjected to coordinate transformation according to the calculated parameters to make it consistent with the GPS coordinate system. Finally, the coordinate points measured by GPS are superimposed on the transformed vector topographic data to obtain the fused accurate geographic data.

[0082] Then, high-resolution satellite remote sensing images covering the scenic area are obtained and orthorectification processing is carried out on them. First, the ground control point information within the scenic area is collected, including the coordinate points measured by GPS and the digital elevation model. Then, a strict photogrammetric model is established to calculate the exterior orientation elements of the images. After that, the digital elevation model is used to correct the terrain of the images to eliminate the image deformation caused by terrain undulation. Finally, the mosaicking process is carried out on the corrected images to obtain an orthophoto map covering the entire scenic area.

[0083] Overlay the orthorectified satellite remote sensing images with the data after registration and fusion above to obtain a complete scenic area geographic information dataset. When overlaying, first unify the two groups of data to the same coordinate system and projection. Then, based on the vector data, overlay the orthoimage as the background layer. Check the overlay result to ensure that the positions of all elements match. Finally, form a comprehensive geographic information dataset that includes vector data such as terrain, roads, scenic spots, and facilities, as well as a high-definition image background.

[0084] Based on the above geographic information dataset, construct an electronic map of the scenic area. First, design a map symbol system and use different symbols to represent different types of geographic elements. Then, perform map generalization, selecting and simplifying geographic elements according to the display scale of the electronic map. After that, conduct map color scheme design and select a color scheme suitable for the characteristics of the scenic area. Finally, add map decoration elements such as a north arrow, scale, and legend.

[0085] Divide the constructed electronic map of the scenic area into multiple service areas. First, divide the scenic area into several relatively independent areas according to factors such as the topographic and geomorphic characteristics, scenic spot distribution, and tourist flow. Then, consider the distribution of service facilities and service radius within each area and optimize the division results. Finally, form multiple service areas that cover the entire scenic area and have relatively complete functions. For example, a certain scenic area can be divided into 5 service areas such as a mountain tour area, a lake leisure area, and a historical and cultural area.

[0086] For each divided service area, construct a weighted directed graph model of the road network. First, extract the topological structure of the road network within the area, with road intersections as the nodes of the graph and road segments as the edges. Then, assign multi-dimensional weights to each edge: 1) the actual road distance, calculated based on GPS measurement data; 2) the road passing capacity, evaluated according to factors such as road width and pavement type; 3) the predicted value of road section congestion, predicted based on historical passenger flow data and spatio-temporal distribution models; 4) the weight coefficient of scenic spots passed through, evaluated according to the importance and attractiveness of the scenic spots; 5) the terrain slope coefficient, calculated based on the digital elevation model to obtain the average slope of the road section.

[0087] Based on the constructed weighted directed graph model, generate multiple differentiated alternative service paths within each service area. First, determine the start and end points of the path, such as from the service area entrance to the main scenic spot. Then, use an improved Dijkstra algorithm, considering multi-dimensional edge weights, to calculate multiple different paths. By adjusting the proportion of each weight, paths that focus on different factors can be obtained, such as the shortest distance path, the path covering the most scenic spots, and the path with the gentlest slope. For each pair of start and end points, generate 10 - 20 differentiated paths.

[0088] Calculate the path similarity for multiple generated differential alternative service paths. Using the path overlap rate index, calculate the proportion of the length of the sections passed by two paths in the total length. If the overlap rate of two paths exceeds a preset threshold (such as 80%), it is considered that the path similarity is too high. Traverse all path pairs, eliminate the paths with an overlap rate exceeding the threshold, and retain the set of paths with greater differences.

[0089] Finally, according to the motion constraint conditions of the service robot, verify the feasibility of the remaining alternative service paths. Considering parameters such as the maximum climbing ability and minimum turning radius of the robot, check each path section by section to see if it meets the requirements. Make local adjustments or directly eliminate the paths that do not meet the constraint conditions. Finally, obtain a set of alternative service paths that meet both the differential requirements and the robot's motion constraints.

[0090] The beneficial effects of the method of the present invention are mainly reflected in the following three aspects:

[0091] 1. Through multi-source data fusion and high-precision measurement, an accurate and reliable scenic area electronic map is constructed, providing high-quality basic data for subsequent path planning, and improving the accuracy and reliability of path planning.

[0092] 2. Adopting a weighted directed graph model with multiple weights, comprehensively considering various factors such as distance, traffic capacity, congestion degree, scenic spot attraction, terrain slope, etc., can generate more differential service paths that meet the actual needs, and improve the flexibility and adaptability of path planning.

[0093] 3. Through path similarity calculation and verification of robot motion constraints, a set of alternative paths that are both different and feasible is selected, ensuring both the diversity of path selection and the executability of the paths in actual applications, and improving service efficiency and quality.

[0094] In an optional implementation manner, the method further includes:

[0095] Use the Kriging interpolation algorithm to perform spatial interpolation on the discrete sampling points in the geographic information dataset to obtain a continuous terrain surface model, use a vector data structure to store the building information and service facility information in the scenic area, and correct the error of the RTK-GPS signal through a differential correction algorithm to obtain the scenic area electronic map;

[0096] Divide the scenic area electronic map into multiple grid cells, obtain the tourist flow density information, service facility density information, and terrain feature information of each grid cell, perform weighted calculations on the tourist flow density information, service facility density information, and terrain feature information respectively according to preset weight coefficients to obtain the service demand weight of each grid cell, and use an improved K-means clustering algorithm to divide the scenic area electronic map into multiple service areas according to the service demand weight.

[0097] The method further includes the following steps:

[0098] First, perform spatial interpolation on the discrete sampling points in the geographic information dataset to obtain a continuous terrain surface model. Specifically, use the Kriging interpolation algorithm to perform spatial interpolation on the sampling points. The Kriging interpolation algorithm is based on the variogram theory and structural analysis, takes into account the spatial positions and attribute values of the sample points, and can obtain the best unbiased estimation result. For example, for 1000 elevation sampling points in a scenic area, a grid digital elevation model (DEM) with a resolution of 5m×5m can be generated through Kriging interpolation to achieve a continuous expression of the terrain.

[0099] Next, use a vector data structure to store the building information and service facility information in the scenic area. The vector data structure includes geometric elements such as points, lines, and surfaces, and can accurately express the positions and shapes of spatial entities. For example, represent the point-like service facilities such as restaurants and toilets in the scenic area with point elements, represent the roads with line elements, represent the building outlines with surface elements, and store the corresponding attribute information.

[0100] Then, perform error correction on the RTK-GPS signal through a differential correction algorithm to obtain the scenic area electronic map. RTK-GPS can provide centimeter-level positioning accuracy, but there are still error sources such as atmospheric delay and multipath effects. The differential correction algorithm uses the observation data of a reference station with known coordinates to correct the errors of the rover station, which can further improve the positioning accuracy. For example, deploy 5 reference stations in the scenic area, collect 24-hour continuous observation data, and the positioning error can be controlled within 1-2 cm through differential processing.

[0101] Next, divide the scenic area electronic map into multiple grid cells. An equally spaced grid division method can be used to divide the scenic area into 10m×10m square grid cells. For each grid cell, obtain its tourist flow density information, service facility density information, and terrain feature information. The tourist flow density can be statistically obtained through mobile phone signaling data, the service facility density can be calculated according to the number of facility points, and the terrain features can include information such as elevation and slope.

[0102] Then, the tourist flow density information, service facility density information, and terrain feature information are weighted and calculated according to the preset weight coefficients to obtain the service demand weight of each grid cell. For example, the weight coefficients of tourist flow density, service facility density, and terrain feature can be set to 0.5, 0.3, and 0.2 respectively. For a certain grid cell, assuming that its normalized tourist flow density is 0.8, service facility density is 0.6, and terrain feature is 0.4, then the service demand weight of this cell is 0.8×0.5 + 0.6×0.3 + 0.4×0.2 = 0.68.

[0103] Finally, the improved K-means clustering algorithm is used to divide the scenic area electronic map into multiple service areas according to the service demand weights. The traditional K-means algorithm is prone to falling into a local optimal solution. The improved algorithm improves the clustering effect by optimizing the selection of the initial clustering center, introducing a simulated annealing strategy, etc. During the clustering process, grid cells with similar service demand weights are divided into the same service area. For example, the scenic area can be divided into 5 service areas, and the grid cells within each area have similar service demand characteristics.

[0104] The beneficial effects of this method include:

[0105] 1. Through Kriging interpolation and vector data structure, the precise expression of the terrain and facilities in the scenic area is realized, providing high-quality basic data for subsequent analysis.

[0106] 2. The differential correction algorithm is adopted to improve the RTK-GPS positioning accuracy. Combined with the grid division method, the refined management of the spatial information in the scenic area is realized.

[0107] 3. Based on multi-source data fusion and the improved K-means clustering algorithm, the intelligent division of the service areas in the scenic area is realized, providing a scientific basis for the management and service optimization of the scenic area.

[0108] In an alternative embodiment, based on the real-time position information and working status information of the intelligent robot, a robot service ability evaluation model is constructed. The robot service ability evaluation model obtains the comprehensive service ability score of the robot by weighted calculation of the remaining power information, current task volume information, and historical service score information by assigning corresponding weights respectively, including:

[0109] The real-time position information of the robot is obtained by using GPS and the indoor positioning system. The obtained real-time position information is processed by Kalman filtering to obtain the accurate position information of the robot. The battery management system is used to collect the voltage parameters, current parameters and temperature parameters of the battery pack. Based on the battery health state assessment algorithm, the remaining power information of the robot is calculated. The number of service tasks being executed by the robot, the task type information and the estimated completion time information are collected. The tourist evaluation information, the task completion rate information and the service timeliness information are obtained;

[0110] The tourist evaluation information, the task completion rate information and the service timeliness information are input into a time decay function for weighted processing to obtain the historical service score. The ratio of the remaining power information to the preset full power information is weighted with the predicted serviceable mileage information to obtain the remaining power score. The difference between the maximum task load capacity information and the current task quantity information is multiplied by the task difficulty coefficient to obtain the task quantity score;

[0111] Based on the historical service score, a fuzzy relation matrix is established. The analytic hierarchy process is used to determine the weights of the remaining power dimension, the current task quantity dimension and the historical service dimension. The weights of the remaining power dimension, the current task quantity dimension and the historical service dimension are multiplied by the corresponding remaining power score, task quantity score and historical service score respectively and summed up to obtain the comprehensive service ability score of the robot.

[0112] In order to build a robot service ability evaluation model based on the real-time position information and working state information of the intelligent robot, this embodiment provides a specific implementation method.

[0113] First, the real-time position information of the robot is obtained by using GPS and the indoor positioning system. GPS is mainly used for outdoor positioning, with an accuracy of up to 5 - 10 meters. The indoor positioning system uses technologies such as Bluetooth beacons, WiFi or UWB, with an accuracy of up to 0.1 - 1 meter. The original position data obtained by GPS and the indoor positioning system are fused, and then the Kalman filtering algorithm is used to process the fused position data to eliminate noise interference and improve the positioning accuracy. For example, the original GPS positioning result is (116.3912°E, 39.9075°N), the indoor positioning result is (116.3914°E, 39.9076°N), and after Kalman filtering processing, a more accurate position (116.3913°E, 39.9075°N) is obtained.

[0114] Secondly, parameters such as the voltage, current, and temperature of the battery pack are collected in real time through the Battery Management System (BMS). Suppose the collected data is: voltage 12.6V, current 5A, temperature 25°C. Based on these parameters, a neural network-based State of Health (SOH) evaluation algorithm for the battery is used to calculate that the remaining battery capacity is 80%, that is, the remaining power is 16Ah (rated capacity 20Ah).

[0115] Next, collect the service task information that the robot is currently performing, including the number of tasks, types, and estimated completion times. For example, there are currently 2 tasks, namely food delivery and luggage transportation, and the estimated completion times are 15 minutes and 20 minutes respectively. At the same time, obtain historical service data, including tourist evaluations (4.8 points out of 5), task completion rate (98%), and service timeliness (completed 2 minutes ahead of schedule on average).

[0116] Input the above historical service data into a time decay function for weighted processing. For example, use the exponential decay function f(t) = e^(-λt), where λ is the decay coefficient (such as 0.1) and t is the time interval (unit: days). Suppose the most recent evaluation was 3 days ago, then the weight is e^(-0.1*3) = 0.74. After weighted processing, the historical service score is 4.5 points out of 5.

[0117] When calculating the remaining power score, the ratio of the remaining power of 16Ah to the full power of 20Ah, which is 0.8, is weighted with the predicted service mileage of 10km. Suppose the weights are 0.6 and 0.4 respectively, then the remaining power score is 0.8*0.6 + 10 / 15*0.4 = 0.75 (full score 1 point).

[0118] When calculating the task volume score, assume that the maximum task load of the robot is 5, the current number of tasks is 2, and the difference is 3. The task difficulty coefficients are 0.8 for food delivery and 1.2 for luggage transportation. Then the task volume score is 3*(0.8 + 1.2) / 2 = 3 (full score 5 points).

[0119] Based on the historical service score of 4.5 points, establish a fuzzy relationship matrix. Use the Analytic Hierarchy Process to determine the weights of each dimension: the weight of the remaining power dimension is 0.3, the weight of the current task volume dimension is 0.3, and the weight of the historical service dimension is 0.4. Multiply the weights of each dimension by the corresponding scores and sum them up: 0.3*0.75 + 0.3*3 / 5 + 0.4*4.5 / 5 = 0.81 (full score 1 point).

[0120] The beneficial effects of this method are mainly reflected in three aspects:

[0121] 1. Improve the accuracy of robot service ability evaluation. By integrating multi-source data, such as GPS, indoor positioning, battery parameters, and historical service data, etc., comprehensively consider various factors affecting the robot's service ability, making the evaluation results more objective and comprehensive.

[0122] 2. Enhance the adaptability and robustness of the evaluation model. Use algorithms such as Kalman filtering to process the original data, effectively reducing noise interference. Introduce a time decay function to automatically adjust the influence of historical data over time, ensuring that the model can timely reflect the latest state of the robot.

[0123] 3. Improve the interpretability and operability of the evaluation results. Determine the weights of each dimension through the analytic hierarchy process, making the evaluation process more transparent. The final comprehensive service ability score is intuitive and clear, facilitating managers to quickly understand the robot's state and make corresponding decisions.

[0124] In an optional implementation manner, for path planning from the multiple alternative service paths, use the A* algorithm to calculate multiple candidate service paths, and score the candidate service paths based on path distance, congestion degree, and scenic spot distribution density, and select the path with the highest score as the optimal service path, including:

[0125] Construct a dynamic heuristic function including parameters such as the straight-line distance from the node to the target point, congestion cost parameter, and scenic spot density parameter, collect the tourist flow data of the road section, calculate the congestion cost parameter according to the ratio of the tourist flow data of the road section to the maximum carrying capacity of the road section, collect the location information of surrounding scenic spots, and calculate the scenic spot density parameter according to the sum of the reciprocals of the distances from the node to be evaluated to each scenic spot;

[0126] Based on the dynamic heuristic function, perform multiple rounds of iterative search. In each round of iterative search, set a search taboo table and record the search tree expansion historical information, determine the node traversal order according to the search tree expansion historical information, and generate multiple candidate service paths;

[0127] Calculate the maximum path length and minimum path length among the multiple candidate service paths, normalize the path length of the current candidate service path to be evaluated with the maximum path length and minimum path length to obtain the path length score, obtain the path congestion degree score by taking the ratio of the sum of the products of the congestion coefficients of each road section and the corresponding road section lengths to the total path length, and obtain the scenic spot coverage score by taking the ratio of the sum of the weights of the scenic spots covered by the path to the total number of scenic spots;

[0128] Collect real-time congestion degree information, road section traffic status information, scenic spot activity information, and service demand density information of the road section. When the real-time congestion degree information of the road section exceeds the preset congestion threshold, or the traffic status information of the road section changes, or new scenic spot activity information appears, or the service demand density information changes significantly, determine the affected path section;

[0129] Update the road network status information of the affected path section, recalculate the dynamic heuristic function of the affected path section, keep the unaffected path sections in the original path unchanged, regenerate candidate service paths for the affected path section, calculate the path length score, path congestion score, and scenic spot coverage score, and perform weighted calculation on the path length score, path congestion score, and scenic spot coverage score to obtain the path comprehensive score. Select the candidate service path with the highest path comprehensive score as the optimal service path.

[0130] When implementing the planning of multiple candidate service paths and the selection of the optimal path based on the A* algorithm, it is first necessary to construct a dynamic heuristic function that includes the straight-line distance parameter from the node to the target point, the congestion cost parameter, and the scenic spot density parameter. The construction process of this heuristic function is as follows:

[0131] First, calculate the straight-line distance from each node to the target point as the basic heuristic estimate value. Then, collect the real-time tourist flow data of each road section, compare it with the maximum carrying capacity of the road section, and calculate the congestion cost parameter. For example, if the current tourist flow of a road section is 1000 people per hour and the maximum carrying capacity is 2000 people per hour, its congestion cost parameter can be set to 0.5. Next, collect the location information of surrounding scenic spots, calculate the distance from the node to be evaluated to each scenic spot, and take the sum of their reciprocals as the scenic spot density parameter. For example, if there are 5 scenic spots within 1 km and 3 scenic spots within 2 km around a certain node, its scenic spot density parameter can be calculated as 5 / 1 + 3 / 2 = 6.5. Finally, perform weighted combination of these three parameters to obtain the comprehensive heuristic function value.

[0132] After the construction of the heuristic function is completed, start multiple rounds of iterative search to generate candidate paths. In each round of search, set a search taboo table to avoid repeated access to the expanded nodes. At the same time, record the expansion history of the search tree, including information such as the access times and expansion order of each node. According to this historical information, the traversal priority of the nodes can be dynamically adjusted, and the nodes with fewer access times are preferentially expanded to increase the path diversity. Through multiple rounds of iteration, multiple different candidate service paths are finally generated.

[0133] For the generated candidate paths, a comprehensive score is required to select the optimal path. First, the maximum path length and the minimum path length of all candidate paths are calculated, and the length of each path is normalized based on this to obtain the path length score. For example, if the shortest path length is 10km and the longest is 15km, and a candidate path is 12km long, then its path length score is (15-12) / (15-10)=0.6. Then calculate the average congestion of the path, sum the product of the congestion coefficient of each section and the length of the corresponding section, and then divide it by the total path length. For example, if the total length of a path is 10km, the congestion coefficient of the 5km section is 0.3, and the congestion coefficient of the other 5km section is 0.7, then the congestion score of the path is (0.3*5+0.7*5) / 10=0.5. Finally, count the number of scenic spots covered by the path and their weights, and calculate the scenic spot coverage score. For example, if a path covers 5 scenic spots with a total weight of 8, and the total number of scenic spots in the scenic area is 10 with a total weight of 15, then its scenic spot coverage score is 8 / 15≈0.53.

[0134] In practical applications, it is necessary to monitor the road network status in real time and adjust the optimal path in time. The specific approach is to collect information such as the real-time congestion of the road section, traffic status, scenic spot activities and service demand density. When this information changes significantly, such as the congestion of a road section exceeds the preset threshold of 0.8, or an important activity is added to a scenic spot, it is necessary to determine the affected path section. For the affected section, update its road network status information and recalculate the dynamic heuristic function value. Keep the unaffected sections in the original path unchanged, and only regenerate candidate paths for the affected sections. Then recalculate the path length score, congestion score and scenic spot coverage score according to the above method, and perform weighted calculation to obtain the comprehensive score. For example, weights of 0.3, 0.4 and 0.3 can be used to correspond to these three scores respectively, and finally select the path with the highest comprehensive score as the updated optimal service path.

[0135] Through the above method, dynamic planning and optimization selection of multiple candidate service paths can be achieved. This method has the following beneficial effects:

[0136] First, by constructing a dynamic heuristic function containing multi-dimensional parameters, we can comprehensively consider multiple factors such as path distance, congestion and distribution of scenic spots, thereby improving the scientificity and rationality of path planning.

[0137] Secondly, using multiple rounds of iterative search and recording search history information can effectively increase the diversity of candidate paths, avoid falling into local optimal solutions, and improve the globality of the optimal path.

[0138] Finally, by real-time monitoring of the road network status and dynamically adjusting the affected path sections, it is ensured that the optimal service path can adapt to changes in actual conditions in a timely manner, and has strong real-time and robustness.

[0139] In an alternative embodiment, the movement trajectory of the target service robot is monitored in real time through the scenic area electronic map. When a path blockage or an emergency is detected, a path dynamic adjustment mechanism is triggered, and the steps of recalculating the optimal service path and sending the updated path information to the target service robot include:

[0140] Construct a movement trajectory based on the real-time position information of the service robot, calculate the rate of change of speed of the movement trajectory, the deviation distance from the planned path, and the residence duration of the positioning point. When the rate of change of speed exceeds the preset speed threshold, or the deviation distance exceeds the preset distance threshold, or the residence duration exceeds the preset duration threshold, an anomaly detection is triggered;

[0141] Collect lidar scan data and road control information of the scenic area management system. Identify fixed obstacles based on the lidar scan data, determine temporarily closed sections according to the road control information, collect the operating parameters of the key components of the service robot and tourist help signals, and determine whether there is equipment failure or emergency help;

[0142] When a fixed obstacle or a temporarily closed section is detected, calculate the obstacle influence range, search for alternative detour paths within the obstacle influence range, calculate the detour cost and waiting cost of the alternative detour paths, and select the alternative detour path with the minimum cost as the local adjustment plan;

[0143] When an equipment failure or an emergency help signal is detected, determine the position of the nearest safe avoidance point, plan the shortest evacuation path from the current position of the service robot to the position of the safe avoidance point, and generate an emergency avoidance instruction;

[0144] Convert the alternative detour path or the shortest evacuation path corresponding to the local adjustment plan into a sequence of path points, add speed parameters and steering parameters, generate a path execution time sequence table, and send the path execution time sequence table to the service robot in segments through a communication channel. Receive the instruction confirmation information returned by the service robot, and record the deviation information and anomaly information during the path execution process.

[0145] In the movement control system of the scenic area service robot, the movement trajectory of the target service robot is monitored in real time through the electronic map, and a dynamic adjustment mechanism is triggered when a path blockage or an emergency is detected. The specific implementation method is as follows:

[0146] First, the system constructs a movement trajectory based on the real-time position information of the service robot. Specifically, the real-time coordinates of the robot in the scenic area are obtained through positioning sensors such as GPS and inertial navigation, and a sequence of coordinate points is recorded at intervals of 0.1 second. The adjacent coordinate points are connected to form a movement trajectory curve. Then, three parameters, namely, the rate of change of speed of the movement trajectory, the deviation distance from the planned path, and the residence duration of the positioning point, are calculated.

[0147] The method for calculating the rate of change of speed is as follows: take 5 adjacent coordinate points, calculate the speed between each two points, and divide the difference between the maximum speed and the minimum speed by the time interval. The method for calculating the deviation distance is as follows: project the actual trajectory points onto the planned path and take the Euclidean distance between the projected points and the actual points. The method for calculating the residence duration is as follows: count the cumulative time that the coordinate points stay within a range of 5 meters.

[0148] The system pre-sets three thresholds: the threshold for the rate of change of speed is 2 m / s², the threshold for the deviation distance is 10 meters, and the threshold for the residence duration is 120 seconds. When any parameter exceeds the corresponding threshold, the system triggers an anomaly detection process.

[0149] Next, the system collects lidar scan data and road control information from the scenic area management system. The lidar scans the surrounding environment at a frequency of 10 Hz to obtain point cloud data. The point cloud data is segmented into independent targets through a clustering segmentation algorithm and compared with a pre-established environmental map to identify newly added fixed obstacles. At the same time, the system receives road control information released by the scenic area management system in real time, including the start and end coordinates and the closure time of the closed section.

[0150] The system also collects the operating parameters of the key components of the service robot, such as the motor speed, battery power, communication signal strength, etc. When any parameter is lower than the preset threshold, it is determined as a device failure. In addition, the system receives help signals input by tourists through the robot's touch screen or voice.

[0151] When a fixed obstacle or a temporarily closed section is detected, the system first calculates the obstacle influence range. Taking the obstacle as the center, expand 5 meters around as the influence range. Then search for alternative detour paths within the influence range and use the A* algorithm to generate multiple feasible paths. For each alternative path, calculate the detour cost and the waiting cost.

[0152] The detour cost takes into account factors such as the increase in detour distance and the complexity of the road conditions, and the waiting cost takes into account factors such as the estimated waiting time and the impact on tourist satisfaction. The two costs are weighted and summed, and the alternative path with the minimum total cost is selected as the local adjustment plan. For example, when a certain obstacle blocks the original path, the system generates 3 alternative paths. Among them, path 1 has an increased detour distance of 50 meters, a lower complexity, and a waiting time of about 5 minutes; path 2 has an increased detour distance of 30 meters, a medium complexity, and a waiting time of about 8 minutes; path 3 has an increased detour distance of 10 meters, a higher complexity, and a waiting time of about 12 minutes. After calculation, the total cost of path 2 is the minimum and is selected as the local adjustment plan.

[0153] When a device failure or an emergency help signal is detected, the system determines the location of the nearest safe avoidance point. The safe avoidance points are pre-marked in the electronic map of the scenic area, including charging stations, maintenance points, medical points, etc. The system uses the Dijkstra algorithm to plan the shortest evacuation path from the current position of the service robot to the nearest safe avoidance point and generates an emergency avoidance instruction.

[0154] Finally, the system converts the alternative detour path or the shortest evacuation path corresponding to the local adjustment plan into a sequence of path points. The specific method is as follows: Take a sampling point every 5 meters on the path and record its coordinates. Then add speed parameters and steering parameters to each path point. The speed parameters are set according to the road section type. For example, the speed is 20 km / h on flat sections and 10 km / h at turning points. The steering parameters are calculated based on the positional relationship of three adjacent path points.

[0155] The system integrates the sequence of path points, speed parameters, and steering parameters to generate a path execution timing table. Considering the communication bandwidth limitation, the timing table is divided into several data packets, each packet containing the information of 10 path points, and is sent to the service robot in segments through the wireless communication channel. The system receives the instruction confirmation information returned by the service robot and records the deviation information and abnormal information during the path execution for subsequent optimization.

[0156] The technical solution has the following beneficial effects:

[0157] 1. It improves the path planning intelligence and adaptability of the service robot. By real-time monitoring the motion trajectory and dynamically adjusting the path, the robot can flexibly respond to various emergencies in the scenic area, improving service efficiency and safety.

[0158] 2. It enhances the system's anomaly detection and emergency handling capabilities. Through multi-source data fusion and multi-dimensional parameter analysis, potential risks can be detected in a timely manner, and reasonable countermeasures can be quickly formulated, improving the reliability and robustness of the system.

[0159] 3. It optimizes the decision-making process of path planning. By comprehensively considering various factors to calculate the path cost and introducing a safe avoidance mechanism, while ensuring service quality, it also takes into account the safety of the robot itself, achieving a balance between efficiency and safety.

[0160] Figure 2 The following is a schematic structural diagram of the intelligent robot path scheduling system for scenic area tourists according to the embodiment of the present invention, as Figure 2 shown, the system includes:

[0161] The first unit is used to obtain the real-time location information and working status information of multiple intelligent robots in the scenic area, and obtain the service request information sent by the scenic area tourists through the mobile terminal. The service request information includes the current location information, destination location information and service type information of the tourists. An electronic map of the scenic area is established according to the geographical information in the scenic area. Multiple service areas are divided in the electronic map of the scenic area, and multiple alternative service paths are preset in each service area; the working status information includes the remaining battery information, current task volume information and historical service score information of the intelligent robot;

[0162] The second unit is used to construct a robot service ability evaluation model based on the real-time location information and working status information of the intelligent robot. The robot service ability evaluation model obtains the comprehensive service ability score of the robot through weighted calculation by assigning corresponding weights to the remaining battery information, current task volume information and historical service score information respectively; according to the current location information, destination location information and service type information of the tourist, path planning is carried out from the multiple alternative service paths, multiple candidate service paths are calculated by using the A* algorithm, and the candidate service paths are scored based on the path distance, congestion degree and scenic spot distribution density, and the path with the highest score is selected as the optimal service path;

[0163] The third unit is used to sort the multiple intelligent robots in the scenic area according to the comprehensive service ability score of the robot, and select the intelligent robot with the highest score and in the same service area as the target service robot; send the optimal service path to the target service robot, and load the optimal service path into the navigation system of the target service robot; monitor the movement track of the target service robot in real time through the electronic map of the scenic area, and when a path blockage or emergency is detected, trigger a path dynamic adjustment mechanism, recalculate the optimal service path and send the updated path information to the target service robot.

[0164] In the third aspect of the embodiments of the present invention,

[0165] An electronic device is provided, including:

[0166] A processor;

[0167] A memory for storing instructions executable by the processor;

[0168] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0169] In the fourth aspect of the embodiments of the present invention,

[0170] Provided is a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the foregoing method.

[0171] The present invention may be a method, an apparatus, a system, and / or a computer program product. The computer program product may include a computer-readable storage medium having thereon computer-readable program instructions for performing various aspects of the present invention.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent robot path scheduling method for scenic area tourists, characterized in that, Including: Obtaining the real-time location information and working status information of multiple intelligent robots in the scenic area, and obtaining the service request information sent by scenic area tourists through mobile terminals. The service request information includes the current location information, destination location information and service type information of the tourists. An electronic map of the scenic area is established according to the geographical information in the scenic area. Multiple service areas are divided in the electronic map of the scenic area, and multiple alternative service paths are preset in each service area; The working status information includes the remaining battery information, current task volume information and historical service score information of the intelligent robot; Based on the real-time location information and working status information of the intelligent robot, a robot service ability evaluation model is constructed. The robot service ability evaluation model obtains the comprehensive service ability score of the robot through weighted calculation by assigning corresponding weights to the remaining battery information, current task volume information and historical service score information respectively; According to the current location information, destination location information and service type information of the tourist, path planning is carried out from the multiple alternative service paths, multiple candidate service paths are calculated by using the A* algorithm, and the candidate service paths are scored based on the path distance, congestion degree and scenic spot distribution density, and the path with the highest score is selected as the optimal service path; Sort the multiple intelligent robots in the scenic area according to the comprehensive service ability score of the robot, and select the intelligent robot with the highest score and located in the same service area as the target service robot; Send the optimal service path to the target service robot, and load the optimal service path into the navigation system of the target service robot; Real-time monitor the movement track of the target service robot through the electronic map of the scenic area. When a path blockage or emergency is detected, trigger the path dynamic adjustment mechanism, recalculate the optimal service path and send the updated path information to the target service robot.

2. The method according to claim 1, wherein Establishing an electronic map of the scenic area according to the geographical information in the scenic area, dividing multiple service areas in the electronic map of the scenic area, and presetting multiple alternative service paths in each service area includes: Obtain the road node coordinate points, scenic spot location coordinate points, service facility location coordinate points and terrain feature coordinate points in the scenic area through the RTK-GPS measurement system, vectorize the existing CAD topographic map of the scenic area to obtain vectorized terrain data, and register and fuse the vectorized terrain data with the coordinate points obtained by the RTK-GPS measurement system; Obtain satellite remote sensing image data and orthorectify the satellite remote sensing image data, and superimpose the orthorectified satellite remote sensing image data with the registered and fused data to obtain a geographical information dataset; Construct an electronic map of the scenic area, and divide the electronic map of the scenic area into multiple service areas; Construct a weighted directed graph model of the road network within each of the service areas. The edge weights of the weighted directed graph model include the actual road distance, road traffic capacity, predicted value of road congestion, weight coefficient of scenic spots passed by, and terrain slope coefficient. Based on the weighted directed graph model, generate multiple differentiated alternative service paths within each of the service areas, calculate the path similarity of the multiple differentiated alternative service paths, eliminate the alternative service paths with a path overlap rate exceeding a preset threshold, and verify the feasibility of the remaining alternative service paths according to the robot motion constraint conditions to obtain a set of final alternative service paths that meet the motion constraint conditions.

3. The method according to claim 2, characterized in that, The method further includes: Using the Kriging interpolation algorithm to perform spatial interpolation on the discrete sampling points in the geographic information dataset to obtain a continuous terrain surface model, storing the building information and service facility information in the scenic area using a vector data structure, and correcting the error of the RTK-GPS signal through a differential correction algorithm to obtain an electronic map of the scenic area; Divide the electronic map of the scenic area into multiple grid cells, obtain the tourist flow density information, service facility density information, and terrain feature information of each grid cell, perform weighted calculations on the tourist flow density information, service facility density information, and terrain feature information respectively according to preset weight coefficients to obtain the service demand weight of each grid cell, and use an improved K-means clustering algorithm to divide the electronic map of the scenic area into multiple service areas according to the service demand weight.

4. The method according to claim 1, wherein Based on the real-time position information and working status information of the intelligent robot, construct a robot service ability evaluation model. The robot service ability evaluation model calculates the comprehensive service ability score of the robot by respectively assigning corresponding weights to the remaining battery power information, current task quantity information, and historical service score information and performing weighted calculations, including: Use GPS and an indoor positioning system to obtain the real-time position information of the robot, perform Kalman filtering processing on the real-time position information to obtain the accurate position information of the robot, collect the battery pack voltage parameters, current parameters, and temperature parameters through a battery management system, calculate the remaining battery power information of the robot based on a battery health state evaluation algorithm, collect the number of service tasks being executed by the robot, task type information, and estimated completion time information, and obtain tourist evaluation information, task completion rate information, and service timeliness information; Input the tourist evaluation information, task completion rate information, and service timeliness information into a time decay function for weighted processing to obtain a historical service score, perform weighted calculations on the ratio of the remaining battery power information to the preset full battery power information and the predicted serviceable mileage information to obtain a remaining battery power score, and multiply the difference between the maximum task load capacity information and the current task quantity information by the task difficulty coefficient to obtain a task quantity score; A fuzzy relation matrix is established based on the historical service scores. The analytic hierarchy process is used to determine the weights of the remaining battery level dimension, the current task volume dimension, and the historical service dimension. The weights of the remaining battery level dimension, the current task volume dimension, and the historical service dimension are respectively multiplied by the corresponding remaining battery level scores, task volume scores, and historical service scores, and the sum is calculated to obtain the comprehensive service ability score of the robot.

5. The method according to claim 1, characterized in that, Path planning is performed from the multiple alternative service paths. The A* algorithm is used to calculate multiple candidate service paths, and the candidate service paths are scored based on the path distance, congestion level, and scenic spot distribution density. Selecting the path with the highest score as the optimal service path includes: Construct a dynamic heuristic function that includes parameters such as the straight-line distance from the node to the target point, the congestion cost parameter, and the scenic spot density parameter. Collect the traffic flow data of the road segments, calculate the congestion cost parameter according to the ratio of the traffic flow data of the road segment to the maximum carrying capacity of the road segment, collect the location information of the surrounding scenic spots, and calculate the scenic spot density parameter according to the sum of the reciprocals of the distances from the node to be evaluated to each scenic spot; Based on the dynamic heuristic function, perform multiple rounds of iterative search. In each round of iterative search, set a search taboo table and record the historical information of the search tree expansion. Determine the node traversal order according to the historical information of the search tree expansion, and generate multiple candidate service paths; Calculate the maximum path length and the minimum path length among the multiple candidate service paths. Normalize the path length of the current candidate service path to be evaluated with the maximum path length and the minimum path length to obtain the path length score. Calculate the path congestion degree score as the ratio of the sum of the products of the congestion coefficients of each road segment and the corresponding road segment lengths to the total path length. Calculate the scenic spot coverage score as the ratio of the sum of the weights of the scenic spots covered by the path to the total number of scenic spots; Collect the real-time congestion degree information of the road segments, the road segment traffic status information, the scenic spot activity information, and the service demand density information. When the real-time congestion degree information of the road segment exceeds the preset congestion threshold, or the road segment traffic status information changes, or new scenic spot activity information appears, or the service demand density information changes significantly, determine the affected path section; Update the road network status information of the affected path section, recalculate the dynamic heuristic function of the affected path section, keep the unaffected path sections in the original path unchanged, regenerate candidate service paths for the affected path section and calculate the path length score, path congestion degree score, and scenic spot coverage score. Perform a weighted calculation on the path length score, path congestion degree score, and scenic spot coverage score to obtain the path comprehensive score. Select the candidate service path with the highest path comprehensive score as the optimal service path.

6. The method according to claim 1, wherein Real-time monitor the movement trajectory of the target service robot through the scenic area electronic map. When a path blockage or an emergency is detected, trigger the path dynamic adjustment mechanism, recalculate the optimal service path, and send the updated path information to the target service robot, including: Construct a motion trajectory based on the real-time position information of the service robot, calculate the rate of change of speed, the deviation distance from the planned path, and the residence duration of the positioning point of the motion trajectory. When the rate of change of speed exceeds the preset speed threshold, or the deviation distance exceeds the preset distance threshold, or the residence duration exceeds the preset duration threshold, trigger anomaly detection; Collect lidar scan data and road control information of the scenic area management system, identify fixed obstacles based on the lidar scan data, determine temporarily closed sections based on the road control information, collect the operating parameters of the key components of the service robot and tourist help signals, and determine whether there are equipment failures or emergency help requests; When a fixed obstacle or a temporarily closed section is detected, calculate the obstacle influence range, search for alternative detour paths within the obstacle influence range, calculate the detour cost and waiting cost of the alternative detour paths, and select the alternative detour path with the minimum cost as the local adjustment plan; When a device failure or an emergency help signal is detected, determine the position of the nearest safe avoidance point, plan the shortest evacuation path from the current position of the service robot to the position of the safe avoidance point, and generate an emergency avoidance instruction; Convert the alternative detour path or the shortest evacuation path corresponding to the local adjustment plan into a sequence of path points, add speed parameters and steering parameters, generate a path execution time sequence table, and send the path execution time sequence table to the service robot in segments through a communication channel, receive the instruction confirmation information returned by the service robot, and record the deviation information and anomaly information during the path execution process.

7. An intelligent robot path scheduling system for scenic area tourists, which is used to implement the method described in any one of the foregoing claims 1-6, and is characterized in that, It includes: The first unit is used to obtain the real-time position information and working status information of multiple intelligent robots in the scenic area, and obtain the service request information sent by scenic area tourists through mobile terminals. The service request information includes the current position information, destination position information, and service type information of tourists. Establish a scenic area electronic map based on the geographical information in the scenic area, divide multiple service areas in the scenic area electronic map, and preset multiple alternative service paths in each service area; the working status information includes the remaining battery information, current task volume information, and historical service score information of the intelligent robot; The second unit is used to construct a robot service ability evaluation model based on the real-time position information and working status information of the intelligent robot. The robot service ability evaluation model obtains the comprehensive service ability score of the robot by performing weighted calculations on the remaining battery information, current task volume information, and historical service score information by assigning corresponding weights respectively; according to the current position information, destination position information, and service type information of the tourist, perform path planning from the multiple alternative service paths, use the A* algorithm to calculate multiple candidate service paths, and score the candidate service paths based on the path distance, congestion degree, and scenic spot distribution density, and select the path with the highest score as the optimal service path; The third unit is used to sort multiple intelligent robots in the scenic area according to the comprehensive service ability scores of the robots, select the intelligent robot with the highest score and in the same service area as the target service robot; send the optimal service path to the target service robot, and load the optimal service path into the navigation system of the target service robot; monitor the movement trajectory of the target service robot in real time through the scenic area electronic map, and when a path blockage or an emergency is detected, trigger a path dynamic adjustment mechanism, recalculate the optimal service path and send the updated path information to the target service robot.

8. An electronic device, characterized in that, It includes: a processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, the method according to any one of claims 1 to 6 is implemented.

Images