Business service robot path planning method and system

Abstract

The invention relates to the technical field of path planning, in particular to a path planning method and system for a commercial service robot, and the method comprises the following steps: analyzing regional attributes to obtain passage levels, constructing task response coefficients to adjust a task sequence, evaluating path node conflicts, planning an execution path, recognizing obstacles, and correcting the path. And ground parameters are extracted to adjust the attitude control parameters, and robot attitude adjustment configuration is generated. According to the method, the regional function state and the real-time time are dynamically corresponded, the regional traffic attribute is adjusted in real time, the task response coefficient is constructed, the task priority is dynamically optimized, the traffic frequency and the stay duration of the cross nodes are comprehensively quantized, and the path execution priority is finely evaluated; according to the method, the obstacle displacement track and the contour feature are combined, behavior recognition and path dynamic adjustment are achieved, and the refinement and real-time performance of path planning of the robot in a dynamic scene are effectively improved by evaluating the ground structure parameter value and the interference level in real time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path planning, in particular to a commercial service robot path planning method and system. BACKGROUND

[0002] The technical field of path planning includes various technical means for realizing the optimal or feasible path planning of mobile robots in known or unknown environments to reach target positions, including key contents such as environment mapping, path search, trajectory generation and dynamic obstacle avoidance, involving laser radar, inertial navigation, global and local path planning algorithms, real-time environment perception technology and other aspects, and is widely used in mobile robots, autonomous driving, unmanned vehicle navigation and intelligent warehousing scenes. It is the basic link for service robots to realize autonomous navigation and mobile control. In practical applications, path planning needs to consider various factors such as environment topology, dynamic obstacles, human-robot coexistence, and task goals and time constraints, and realize a safe and efficient path decision-making process. Among them, the commercial service robot path planning method refers to a specific method for supporting robots to realize autonomous path decision and motion control in commercial places, covering scene map construction, target point planning, path generation method, obstacle response strategy during travel, and path dynamic correction mechanism. It specifically includes constructing an environment expression method based on a two-dimensional grid map, using a path search algorithm combined with a cost function to select a path, setting up a dynamic obstacle avoidance control logic to guide the robot to avoid moving obstacles, and using a task point priority sorting rule to realize multi-task path dynamic re-planning. In the path planning process, the region attribute identification in the scene is combined to define the non-passable region and the limited passage, and a set of path constraint rules is established to screen and adjust the path candidate results, ensuring that the generated path has spatial rationality and execution feasibility.

[0003] The traditional commercial service robot path planning technology only relies on static scene mapping and simple path search to realize robot path planning in actual operation, and cannot fully consider the dynamic changes of the function state of the service area of the commercial place over time, resulting in a lack of real-time response mechanism for temporary closed or restricted areas during the actual travel of the robot. The node traffic congestion condition and obstacle dynamic behavior characteristics are not deeply evaluated, which reduces the stability of the robot path execution and the real-time obstacle avoidance ability, and easily causes problems such as path congestion and obstacle avoidance response delay. SUMMARY

[0004] In order to solve the technical problems existing in the prior art, the embodiments of the present application provide a commercial service robot path planning method and system. The technical solution is as follows: In order to achieve the above purpose, the present application adopts the following technical scheme, a commercial service robot path planning method, comprising the following steps: S1: using the regional grid map, extracting the function state label and the corresponding effective time period of each service area, adjusting the passing attribute type field of each area by comparing with the real-time time, and generating the passing level field set; S2: using the passing level field set, constructing the task response coefficient according to the task priority level, task generation time and waiting time information of each task in the task list, comparing with the newly input task request, evaluating the adjustment demand of the task execution order, and generating the task position switching record; S3: calling the task position switching record, planning the task execution path of the robot by using the execution position of each task, extracting the passing node number in each candidate path, obtaining the passing frequency value, path intersection frequency value and average residence time value under the cross channel node, calculating the node conflict quantitative value and evaluating the execution priority of the candidate path, and generating the path planning result; S4: using the path planning result, combining the image acquisition device to identify the obstacles in the path, collecting the displacement trajectory and edge pixel set of the obstacles in the continuous period, identifying the obstacle type and correcting the path, and outputting the behavior path adjustment record.

[0005] As a further scheme of the application, the passing level field set includes a function state label, a passing attribute type field and an effective time period label, the task position switching record specifically includes a task priority level identifier, a task waiting period identifier and a task response coefficient field, and the path planning result includes a passing node number list, a cross channel node conflict quantitative value and a candidate path priority sorting structure.

[0006] As a further scheme of the application, the obtaining step of the passing level field set specifically includes: S101: obtaining the regional grid map, extracting the function state label and the effective time period of each service area, analyzing the label type and the corresponding time period content of each service area, judging the coincidence state of the current real-time time and the effective time period, and obtaining the area label activation state value; S102: based on the area label activation state value, adjusting each passing attribute type field in the service area, matching the passing attribute type of each area with the current label state, including passable, limited passing and temporary closure, adjusting the passing attribute type corresponding to the area, and generating the area passing attribute type set; S103: calling the area passing attribute type set, synchronously updating the attribute value of the corresponding area in the grid map, synchronously updating the attribute type and the grid map structure field, and establishing the passing level field set.

[0007] As a further aspect of the present invention, the step of obtaining the task location switching record specifically includes: S201: Obtain the access level field set, collect the task priority level, task generation time and task waiting time of each task in the robot task list, calculate the urgency of each task, construct the task response coefficient, and generate a task urgency sequence. S202: Based on the task urgency sequence, collect the newly input task request parameters, compare the urgency of the new input task with the task urgency sequence, determine the priority order of the new task request in the existing task structure, adjust the execution order of tasks in the task list, and obtain the task priority adjustment result. S203: Based on the task priority adjustment results, update the order of each task in the robot task list, filter tasks whose execution order has changed and position change information, and establish a task position switching record.

[0008] As a further aspect of the present invention, the specific formula for calculating the urgency of each task is as follows: ; Calculate the task urgency level; in, For the first The urgency value of each task. For the first Task priority level for each task For the first The normalized value of the waiting time for each task. For the first The complexity factor of each task For the first The constraint level value of the region to which each task belongs. For the current number The operating load factor of each robot Indicates the task number index. This indicates the robot's ID index.

[0009] As a further aspect of the present invention, the step of obtaining the path planning result specifically includes: S301: Call the task location switching record, obtain the execution location of each task in the task list, analyze the path distribution of each task on the service area map, plan the path of each task, extract the access node number in each path, and generate a path node number sequence. S302: According to the path node number sequence, a cross-channel node is detected, the passing frequency, path intersection frequency and average residence time of the cross-channel node are collected, the node conflict quantitative value is calculated, the passing pressure of each cross-channel node is quantified, and a node passing pressure numerical set is established. The specific formula for calculating the node conflict quantitative value is: ; The node conflict quantitative coefficient is calculated. Wherein, is the conflict quantitative coefficient of the mth cross-channel node, is the passing frequency of the mth cross-channel node, is the path intersection frequency of the mth cross-channel node, is the average value of the path intersection frequency of all cross-channel nodes, is the average residence time normalization value of the mth cross-channel node, is the number of robots passing through the mth node, is the index number of the cross-channel node. S303: Based on the node passing pressure numerical set, the node passing pressures of multiple candidate paths are compared, the priority order of each path is adjusted in combination with the task priority order, and a path planning result is obtained.

[0010] As a further scheme of the application, the behavior path adjustment record acquisition step specifically comprises: S401: Using the path planning result, using an image acquisition device, monitoring and identifying obstacles in the robot path range, detecting the spatial displacement coordinate points and contour edge pixel set of the obstacles in the continuous sampling period, and recording the continuous spatial displacement and contour shape change of each obstacle, generating an obstacle shape change index; S402: Based on the obstacle shape change index, analyzing the displacement direction trend and contour shape change amplitude of the obstacle, evaluating the stability and spatial position continuity of the obstacle, grouping the behavior characteristics of the obstacle in the time sequence, and obtaining an obstacle behavior classification parameter; S403: According to the obstacle behavior classification parameter, judging the type of the obstacle, including static, intermittent movement, continuous dynamic, and adjusting the task execution path of the robot according to the type of the obstacle, correcting the path node position and sequence, and establishing a behavior path adjustment record.

[0011] As a further scheme of the application, the method further comprises: S5: adjusting the ground structure parameter values of each grid in the grid map by using the behavior path adjustment record, including the damping material parameter, the ground inclination parameter and the vibration mark state value, evaluating the ground disturbance level of each path node, and adjusting the posture adjustment control parameter of the robot to generate a robot posture adjustment configuration; The robot posture adjustment configuration includes a posture control frequency parameter, a response cycle configuration table and a ground disturbance level classification.

[0012] As a further scheme of the present application, the obtaining step of the robot posture adjustment configuration is specifically: S501: adjusting the ground structure parameter values of each grid in the path-covered grid map by using the behavior path adjustment record, including the damping material parameter, the ground inclination parameter and the vibration mark state value, to obtain a path node ground parameter set; S502: based on the path node ground parameter set, analyzing and calculating the ground disturbance level of each path node to generate a ground disturbance level sequence; S503: according to the ground disturbance level sequence, mapping the disturbance level to the posture control configuration table, matching the control cycle and amplitude of each node, adjusting the posture adjustment control parameter of the robot, and obtaining a robot posture adjustment configuration.

[0013] On the other hand, a commercial service robot path planning system is provided, which is applied to the commercial service robot path planning method, and the system includes: A function area state processing module uses the area grid map to extract the function state label and the corresponding effective time period of each service area, compares with the real-time time, adjusts the passing attribute type field of each area, and generates a passing level field set; An execution task order adjustment module uses the passing level field set, constructs a task response coefficient according to the task priority level, the task generation time and the waiting time information of each task in the task list, compares with the newly input task request, evaluates the adjustment demand of the task execution order, and generates a task position switching record; A path passing pressure evaluation module calls the task position switching record, plans the task execution path of the robot by using the execution position of each task, extracts the passing node number in each candidate path, obtains the passing frequency value, the path intersection frequency value and the average stay time value under the cross channel node, calculates the node conflict quantitative value and evaluates the execution priority of the candidate path, and generates a path planning result; A path obstacle dynamic identification module uses the path planning result, identifies the obstacles in the path in combination with the image acquisition device, acquires the displacement trajectory and the edge pixel set of the obstacles in the continuous period, identifies the obstacle type and corrects the path, and outputs a behavior path adjustment record; The posture adjustment configuration matching module extracts ground structure parameter values of each grid in the grid map, including damping material parameters, ground inclination parameters and vibration mark state values, evaluates the ground interference level of each path node, and adjusts the posture adjustment control parameters of the robot, to generate a robot posture adjustment configuration, by using the behavior path adjustment record.

[0014] The technical scheme provided by the embodiment of the application has at least the following beneficial effects: By dynamically corresponding the regional function state and the real-time time, adjusting the regional traffic attribute in real time, constructing the task response coefficient and dynamically optimizing the task priority order, and comprehensively quantifying the traffic frequency and the residence time length of the intersection node, the path execution priority is finely evaluated, the behavior recognition and path dynamic adjustment are realized by combining the obstacle displacement trajectory and the contour feature, and the ground structure parameter value and the interference level are evaluated in real time, so that the fineness and real-time performance of the path planning of the robot in the dynamic scene are effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical scheme in the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0016] Figure 1 The workflow diagram of the application; Figure 2 The system flowchart of the application. DETAILED DESCRIPTION

[0017] The technical scheme in the application will be described below with reference to the drawings.

[0018] In the embodiments of the application, the words such as "example", "for example" are used to represent as an example, illustration or description. Any embodiment or design scheme described as "example" in the application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0019] In the embodiments of the present application, the terms "image" and "picture" can be used interchangeably, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0020] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the distinction is not emphasized.

[0021] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, the following will be described in detail with reference to the drawings and specific embodiments.

[0022] Please refer to Figure 1 The present application provides a technical solution, a business service robot path planning method, comprising the following steps: S1: using the area grid map, extracting the function state label and the corresponding effective time period of each service area, adjusting the passage attribute type field of each area by comparing with the real-time time, and generating the passage level field set; S2: using the passage level field set, constructing the task response coefficient according to the task priority level, task generation time and waiting time information of each task in the task list, comparing with the newly input task request, evaluating the adjustment demand of the task execution order, and generating the task position switching record; S3: calling the task position switching record, using the execution position of each task to plan the task execution path of the robot, extracting the passage node number in each candidate path, obtaining the passage frequency value, path intersection frequency value and average stay time value under the cross channel node, calculating the node conflict quantitative value and evaluating the execution priority of the candidate path, and generating the path planning result; S4: using the path planning result, identifying the obstacles in the path by combining the image acquisition device, collecting the displacement trajectory and edge pixel set of the obstacles in the continuous period, identifying the obstacle type and correcting the path, and outputting the behavior path adjustment record; S5: using the behavior path adjustment record, extracting the ground structure parameter value of each grid in the grid map, including the damping material parameter, ground inclination parameter and vibration mark state value, evaluating the ground interference level of each path node, and adjusting the posture adjustment control parameter of the robot, and generating the robot posture adjustment configuration.

[0023] The passage level field set includes a function state label, a passage attribute type field, a valid time period label, a task position switching record specifically a task priority level identifier, a task waiting time period identifier, and a task response coefficient field. The path planning result includes a passage node number list, a cross channel node conflict quantization value, and a candidate path priority sorting structure. The behavior path adjustment record specifically refers to an obstacle type label, a path segment position update identifier, and a contour recognition feature set. The robot posture adjustment configuration includes a posture control frequency parameter, a response cycle configuration table, and a ground disturbance level classification.

[0024] The obtaining step of the passage level field set specifically includes: S101: Obtain a regional grid map, extract the function state label and the valid time period of each service area, analyze the label type and the corresponding time period content of each service area, determine the coincidence state of the current real-time time and the valid time period, and obtain the regional label activation state value. When obtaining the regional grid map, first, read the spatial range coordinates of all service areas according to the map data structure, and then find the function state label of each service area marked with the number one by one, such as passable, restricted passage, and temporary closure, collect the valid time period corresponding to each label, set the time period as 8:00 am to 5:00 pm, 5:00 pm to 10:00 pm, and 10:00 pm to 8:00 am the next day, and represent the above time periods as , , , divide the label types into (passable), (restricted passage), (temporary closure), and then analyze the relationship between the label type of each area and the time period content thereof, compare the label type of each service area with the time period bound thereto in turn, determine whether the current system real-time time, such as 9:30, falls within , and determine whether the current time is 9:30, 17:30, or 23:00 through the conditional determination formula (wherein is the real-time time), if the current time is 9:30, it belongs to , and the label type of the service area is determined as , passable, if the current time is 17:30, it belongs to , and the corresponding label type is switched to , restricted passage, if the current time is 23:00, it is , and it is determined as , temporary closure, and the label activation state value of each service area is determined through the determination formula and the label mapping relationship, specifically by generating an activation state set for all service areas with the number , for example, for the first floor of the shopping mall area numbered , the label is time period is 8:00-17:00, if the current time is 16:30, then the area determination is in the active state, and the tag active state value is set to 1, if the current time is 18:00, the area active state value is set to 0, and the active state judgment of all area tags is completed step by step to obtain the area tag active state value.

[0025] S102: Based on the area tag active state value, adjust each passage attribute type field in the service area, match the passage attribute type of each area with the current tag state, including passable, restricted passage, temporary closure, adjust the passage attribute type corresponding to the area, and generate an area passage attribute type set; Based on the area tag active state value, adjust each passage attribute type field in all service areas in turn, for the active state value of each service area, call the attribute type field , , respectively represent passable, restricted passage, and temporary closure, when the active state value is 1, the passage attribute type field of the area is directly set to passable, when the active state value is 0 and the time falls within the restricted passage time period, the field is adjusted to restricted passage, if the time period is temporary closure, the field is adjusted to temporary closure, according to the above process, traverse each area number, and execute the field adjustment process in turn, for example, assuming the area is in the restricted passage time period at system time 19:00, and its active state value is 0, the attribute field is adjusted to restricted passage, for the area , the current time is 23:15, which falls within the temporary closure period, and the field is adjusted to temporary closure, after all areas are adjusted according to the process, extract all area numbers and their corresponding passage attribute type fields to form an area passage attribute type set.

[0026] S103: Call the area passage attribute type set to synchronize the attribute value of the corresponding area in the grid map, synchronize the field value of the adjusted attribute type and the grid map structure field, and establish a passage level field set; The system calls the set of regional access attribute types to synchronously update the attribute values ​​of each region in the raster map structure. It iterates through each grid point in the raster map, finding the adjusted regional access attribute type field based on the service area number corresponding to the grid point. This attribute type field is then synchronously assigned to the corresponding attribute value in the raster map, completing the one-to-one mapping between regional attributes and raster fields. During this process, the service area to which each grid point belongs is confirmed through a number mapping table. Conflicting grid points are assigned priority during updates, determined by the tag activation status value. If a grid point is located on the boundaries of two regions simultaneously, the regional attribute field with the higher activation status value is used. For example, if the number is... The area and number are The regions have overlapping boundaries, and the grid points are numbered. belong The region activation status value is 1. If the region activation state value is 0, then Grid synchronization The area access attribute type has the following fields: Passage is permitted. Finally, the attribute types of all areas and their corresponding rasters are synchronized, the attribute correspondence between all areas and rasters is summarized, and a passability level field set is established.

[0027] The specific steps for obtaining task location switching records are as follows: S201: Obtain the access level field set, collect the task priority level, task generation time and task waiting time of each task in the robot task list, calculate the urgency of each task, construct the task response coefficient, and generate a task urgency sequence. The specific formula for calculating the urgency of each task is as follows: ; Calculate the task urgency level; in, For the first The urgency value of each task. For the first Task priority level for each task For the first The normalized value of the waiting time for each task. For the first The complexity factor of each task For the first The constraint level value of the region to which each task belongs. For the current number The operating load factor of each robot Indicates the task number index. This indicates the robot's ID index.

[0028] formula: ; Formula details and formula calculation derivation process: The formula is used to calculate the task urgency value in robot task scheduling, and the result is used for subsequent task sorting and execution priority determination.

[0029] Parameter meaning and setting value: The task priority level of the jth task is set by the task allocation system according to the actual business needs, and the value interval is generally 1 to 5. The larger the priority value, the more important the task. The task priority level allocated by the system is 4, indicating that the task needs to be processed in priority; The normalized value of the waiting time of the jth task is set to 26 minutes, and the maximum waiting time is 42 minutes. The normalized processing obtains ; The complexity factor of the jth task is obtained by analyzing the number of turns and the number of road segment switches, and the task complexity factor is set to 3, reflecting the complexity of the route to be executed; The constraint level value of the service area corresponding to the task is set to 2, indicating that there is partial traffic restriction in the area; The current robot running load coefficient is set to 1.5, the number of tasks allocated to the current robot is 3, and the average number of tasks of all robots is 2; Substitute the parameters into the formula to calculate: ; ; ; The result 1.372 indicates the current urgency of the task. By sorting with the urgency values of other tasks in the task list, it can be determined whether the task is prioritized and executed in this scheduling period. The larger the value, the higher the scheduling order of the task under the current constraint, load and complexity conditions. Finally, the result will be used to decide the robot to select the task first, and guide the subsequent execution path and resource allocation.

[0030] S202: According to the task urgency sequence, collect the new input task request parameters, compare the urgency of the new input task with the task urgency sequence, judge the priority order of the new task request in the existing task structure, adjust the execution order of the tasks in the task list, and obtain the task priority adjustment result; According to the task urgency sequence, the new input task request parameters are collected, the priority level, generation time, waiting time and other attribute fields of the new task are extracted, the urgency of the new task is calculated according to the formula in paragraph 1, and the new task number is set as , and the urgency thereof is , and is compared with the original task urgency sequence in turn, and the sorting algorithm is used to sort all task urgency values from high to low, analyze the priority order of each task after the new task is inserted, adjust the execution order of all tasks in the task list, and screen whether the order of the task has changed due to the insertion of the new task. For example, the original task urgency sequence is , and the calculation result of the new task is 35, so the new task should be ranked second, and the order of the second and subsequent tasks is moved backward in turn. After adjustment, the task list order is , the task execution list is rearranged according to the order, the order change information of each task is recorded, and the task priority adjustment result is obtained.

[0031] S203: Based on the task priority adjustment result, update the arrangement order of each task in the robot task list, screen the tasks and position change information whose execution order has changed, and establish the task position switching record; Based on the task priority adjustment result, update the arrangement order of each task in the robot task list, write the adjusted task list into the system task scheduling table, traverse all task numbers, screen out the task numbers whose execution order has changed before and after adjustment, and extract the position change information thereof to form a task number, original position and new position triple, construct a change information table, for example, if the original task A is changed from the third position to the second position, it is recorded as A,3,2, and the B task is changed from the first position to the fourth position, it is recorded as B,1,4. Merge and count all the task numbers and corresponding position information that have changed, finally summarize all the task numbers whose order has changed and their position relationship before and after, and establish the task position switching record.

[0032] The acquisition step of the path planning result is specifically: S301: Call the task position switching record to obtain the execution position of each task in the task list, analyze the path distribution of each task on the service area map, plan the path of each task, extract the passing node number in each path, and generate a path node number sequence;

[0033] When the task position switching record is called, first, the adjustment order information of each task in the task list is read, and the target point thereof on the service area map is associated, and the actual execution starting point and ending point of each task are determined in sequence, and the spatial coordinate information in the service area map is taken as the basic data of the path node, for each task, all reachable path segments from the current position of the robot to the target service point are determined by sequentially scanning the area map grid data, the grid node numbers passed in each path are collected, the node numbers are arranged in sequence according to the actual route, and all node numbers of each path form a path number chain, and a plurality of tasks form a plurality of path number chain tables, for the case that there are a plurality of robots or a plurality of tasks in the scene, the tasks currently allocated to each robot need to be independently split, and the path node number sequence of each task is independently archived, for example, if the task of a robot numbered R1 in a shopping center is to deliver food from the dining area to the second floor conference room, the corresponding path node numbers in the map grid are 101, 102, 103, 104, etc., and all task paths need to be extracted in this way to form a complete path node number sequence set, which is convenient for subsequent path interaction and passing node analysis, the whole process guarantees that each task has an independent path node number chain, which is convenient for subsequent statistics and conflict judgment, and finally a path node number sequence is generated.

[0034] S302: According to the path node number sequence, the intersection channel node is detected, the passing frequency, the path intersection times and the average residence time of the intersection channel node are collected, the node conflict quantitative value is calculated, the passing pressure of each intersection channel node is quantified, and the node passing pressure numerical set is established; The specific formula for calculating the node conflict quantitative value is: ; The node conflict quantitative coefficient is calculated. Wherein, is the conflict quantitative coefficient of the mth intersection channel node, is the passing frequency of the mth intersection channel node, is the path intersection times of the mth intersection channel node, is the average value of the path intersection times of all intersection channel nodes, is the average residence time normalization value of the mth intersection channel node, is the number of robots passing through the mth node, is the index serial number of the intersection channel node.

[0035] Formula: ; Formula explanation and formula calculation derivation process: The formula is used to calculate the conflict quantization coefficient of a single intersection node, and the result is used to generate the node traffic pressure numerical set, which affects the judgment and sorting of the node traffic priority in subsequent path planning; Parameter meaning and setting value: The node traffic frequency is the number of robots passing through the node within the path planning statistical period, which is collected in real time by the background log counting and navigation control system, and is set to ; The node path intersection frequency is obtained by counting the events of the node being intersected by different robot paths within the same time period, and is set to ; The average value of the path intersection frequencies of all intersection nodes is set to ; The average stay time normalization value of the node is the ratio of the average stay time of different robots at the same node to the maximum average stay time of all intersection nodes, and the maximum average stay time is set to 45 seconds. If the average stay time of the node is 32 seconds, then ; The number of robots on the current path is detected in real time by the navigation system, and the instantaneous statistical robot number of the node is set to 4. Substitute the parameters into the formula to calculate: ; The result 4.5686 indicates that the node is in a high traffic pressure state within the statistical period, reflecting that the node needs to be given a lower priority or avoided in the current task scheduling. The node traffic pressure numerical set is formed by collecting the results of all nodes, which is used for global path conflict screening and node risk sorting.

[0036] S303: Based on the node traffic pressure numerical set, compare the node traffic pressures of multiple candidate paths, combine the task priority order, adjust the priority order of each path, and obtain the path planning result. Based on the node traffic pressure numerical set, the candidate paths of all tasks are systematically compared in terms of traffic pressure, the pressure level of each node on the path is counted one by one, the pressure values of all nodes contained in each path are added up to form the overall traffic pressure statistics of the path, then different paths are given weights in combination with the priority order of the tasks, the path selection of the task with high priority is allowed to appropriately increase the pressure threshold, the response demand of the task is preferentially met, for the paths with similar pressure, the task with high priority is preferentially assigned, the candidate paths of the same task are sorted, the path with low pressure value is preferentially ranked, the path of the task with high priority is ranked in front when the pressure values are the same, the system finally outputs the optimal number and sorting sequence of the paths corresponding to all tasks, synchronizes the results to the task scheduling center and pushes them to the robot execution end, ensures that the path execution and scheduling sorting are consistent, and finally obtains the path planning result.

[0037] The obtaining step of the behavior path adjustment record is specifically: S401: using the path planning result, using an image acquisition device, monitoring and identifying obstacles in the path range of the robot, detecting the spatial displacement coordinate points and contour edge pixel set of the obstacles in the continuous sampling period, and recording the continuous spatial displacement and contour shape change of each obstacle, and generating an obstacle shape change index; Using the path planning result, first, the task execution path and the current position data of the robot are called, the monitoring area of the image acquisition device is automatically set according to the path sequence, the system cyclically scans all obstacle targets within three meters in front of the robot in each sampling period, records the spatial displacement coordinate points of the obstacles at each sampling time in turn, and captures the edge pixel set of the obstacles under each frame of image, by sequentially saving the coordinate points of the same obstacle in multiple continuous sampling periods, extracting the obstacle motion path line segment and tracking the change of the obstacle center coordinate in real time, the external contour shape is extracted by using the pixel point counting method, the contour shape parameters of the obstacles in different sampling periods are compared, the change amount of the contour area, the circumference and the like are analyzed, if the contour area change is greater than 15% of the area of the previous frame, the obstacle is recorded as a shape active state, for example, if the obstacle coordinates of the first frame are (2.3, 4.1), the second frame is (2.6, 4.5), and the third frame is (2.8, 4.9), the edge pixel set is 950, 1200 and 1400 respectively, the spatial displacement coordinate points are sequentially increased, and the edge pixels are also increased, combined with the above data, the obstacle motion trajectory and contour change are saved, the system takes the spatial path displacement distance and the contour area change of the obstacle in three frames as the shape change basic data of the target, and finally archives all the obstacle records in all sampling periods one by one to generate an obstacle shape change index.

[0038] S402: Based on the obstacle shape change index, the displacement direction trend and the profile shape change amplitude of the obstacle are analyzed, the stability degree and the spatial position continuity of the obstacle are evaluated, the behavior characteristics of the obstacle in the time sequence are grouped, and the obstacle behavior classification parameters are obtained; Based on the obstacle shape change index, the spatial displacement trend of each obstacle is analyzed, the spatial coordinate point difference of the obstacle at consecutive time is calculated as the displacement velocity vector, the motion direction of the obstacle is determined according to the difference value of three frames of data, if the spatial difference of each frame is greater than 0.3 meters, it is determined that the motion state is, if it is less than 0.1 meters, it is determined that the static state is, combined with the profile area change amplitude, if the area change in three frames is within twenty pixels, the shape change is low, more than fifty pixels is high, when evaluating the stability degree of the obstacle, the spatial position change and the profile change are two dimensions, the stability degree can be divided into three categories: high stability, low stability and instability, the spatial position continuity uses the Euclidean distance continuity between coordinate points as the judgment standard, if the point-to-point distance of three consecutive frames is within 0.1 meters, it is determined that the continuity is high, otherwise it is low, the behavior characteristics of the obstacle in the time sequence are grouped according to the above classification method, assuming that all indicators of target A in the sampling time period are high, it is classified as unstable active group, all indicators of target B are low, it is classified as stable group, finally all obstacle behavior states are summarized, and the obstacle behavior classification parameters are obtained.

[0039] S403: According to the obstacle behavior classification parameters, the type of the obstacle is judged, including static, intermittent movement and continuous dynamic, and the task execution path of the robot is adjusted according to the type of the obstacle, the path node position and sequence are corrected, and the behavior path adjustment record is established; According to the obstacle behavior classification parameters, the behavior type of each obstacle is judged in turn, the system sets the static type as the target with high stability degree and high position continuity, the intermittent movement type as the target with medium position continuity, the profile area fluctuates and the spatial displacement changes intermittently, and the continuous dynamic type as the target with all high behavior indicators, the path adjustment logic is set for each type of obstacle, the static type is directly set to bypass in the path planning, the intermittent movement type is set to temporarily wait strategy for the path segment, and the continuous dynamic type is set to recalculate the path bypass in the path planning in advance, the node coordinates and sequence of the robot task execution path are adjusted according to the type of the obstacle, for example, if a continuous dynamic obstacle is detected at path node A, node A in the path chain table is temporarily removed, and is replaced by bypassing nodes B and C and returning to A, if a static obstacle is encountered, the system directly sets node A to an unreachable state, finally, the adjustment position, sequence change, obstacle type classification of all path nodes are recorded, and are merged into multiple field information such as task number, adjustment node, adjustment sequence and obstacle type, to establish the behavior path adjustment record.

[0040] The obtaining step of the robot posture adjustment configuration is specifically as follows: S501: Extract the damping material parameter, ground inclination parameter and vibration mark state value of each grid in the path covered grid map by using the behavior path adjustment record to obtain the path node ground parameter set; By using the behavior path adjustment record, first, the coordinates of all adjusted nodes in the path chain table are retrieved, the grid map number covered by each node number is found, the attribute field of the grid map is called one by one, the damping material parameter, ground inclination parameter and vibration mark state value of each grid are extracted, the damping material parameter is distinguished by a preset parameter table, the damping value of metal material is 0.6, the damping value of carpet material is 0.3, the ground inclination parameter is based on sensor data, the unit is angle, the inclination parameter is, for example, 5°, 10°, etc., the vibration mark state value is read from the vibration detection field bound by each node in the map, 0 is no vibration, 1 is vibration section, for all nodes covered by the path, the damping material parameter, ground inclination parameter and vibration mark state of each node are sequentially summarized to form a complete path node ground parameter table, for example, the damping material of path node A is metal, the parameter is 0.6, the ground inclination is 8°, and the vibration state is 1, the damping material of node B is carpet, the parameter is 0.3, the ground inclination is 3°, and the vibration state is 0, all node parameters are arranged in order according to the path sequence to obtain the path node ground parameter set.

[0041] S502: Based on the path node ground parameter set, analyze and calculate the ground interference level of each path node to generate a ground interference level sequence; Based on the path node ground parameter set, interval analysis is performed on the three parameters of each path node, the damping material parameter is divided into three types of metal, ceramic tile and carpet, the damping parameter higher than 0.5 is determined as a high damping node, and the damping parameter lower than 0.3 is determined as a low damping node, the ground inclination parameter is divided into three types of flat, gentle slope and steep slope, the vibration state value is directly determined as 0 for non-vibration area and 1 for vibration area, the system combines and weights the three parameters of each node, the interference level of each node is equal to the damping level weight plus the ground inclination level weight plus the vibration state weight, the damping weight is set to 2, the inclination weight is 1, and the vibration state weight is 2, for example, if node A is high damping, gentle slope and vibration area, its interference level is 2+1+2=5, node B is low damping, flat and non-vibration area, its interference level is 0+0+0=0, all nodes on the path are traversed in sequence, the ground interference level of all nodes is calculated, a ground interference level sequence is formed according to the path node number, and all node interference levels are uniformly arranged in sequence form to generate a ground interference level sequence.

[0042] S503: According to the ground interference level sequence, the interference level is mapped to the attitude control configuration table, the control period and amplitude of each node are matched, the robot attitude adjustment control parameters are adjusted, and the robot attitude adjustment configuration is obtained; According to the ground interference level sequence, the interference level of each node is mapped to the attitude control configuration table, and the system sets different control period and amplitude parameters for different interference level nodes. For nodes with interference level greater than 4, the attitude control period is set to output adjustment instructions every 0.1 seconds, and the amplitude parameter is 0.8. For nodes with interference level in the interval of 2-4, the control period is every 0.3 seconds, and the amplitude parameter is 0.5. For nodes with interference level of 0-1, the control period is every 0.5 seconds, and the amplitude parameter is 0.2. All parameters are archived according to node number, matched one by one, and the system automatically writes the parameter configuration into the robot execution end control table. For example, node A on the path is 5, configured with 0.1 second period and 0.8 amplitude, node B is 0, configured with 0.5 second period and 0.2 amplitude, and the attitude control parameters of all nodes are formed into a table and output, and finally the robot attitude adjustment configuration is obtained.

[0043] Please refer to Figure 2 A commercial service robot path planning system, the commercial service robot path planning system is used to execute the above-mentioned commercial service robot path planning method, the system comprises: A function area state processing module extracts the function state label and the corresponding effective time period of each service area using the area grid map, adjusts the passage attribute type field of each area by comparing with the real-time time, and generates a passage level field set; An execution task ordering adjustment module uses the passage level field set to construct a task response coefficient according to the task priority level, task generation time and waiting time information of each task in the task list, compares it with the newly input task request, evaluates the adjustment demand of the task execution order, and generates a task position switching record; A path passage pressure evaluation module uses the execution position of each task to plan the task execution path of the robot, extracts the passage node number in each candidate path, obtains the passage frequency value, path intersection frequency value and average stay time value under the cross channel node, calculates the node conflict quantitative value and evaluates the execution priority of the candidate path, and generates a path planning result; A path obstacle dynamic identification module uses the path planning result to identify obstacles in the path in combination with the image acquisition device, acquires the displacement trajectory and edge pixel set of the obstacle in the continuous period, identifies the obstacle type and corrects the path, and outputs a behavior path adjustment record; The posture adjustment configuration matching module extracts ground structure parameter values of each grid in the grid map, including damping material parameters, ground inclination parameters and vibration mark state values, by using the behavior path adjustment record, evaluates the ground interference level of each path node, and adjusts the posture adjustment control parameters of the robot to generate a robot posture adjustment configuration.

[0044] The above-described embodiments can be implemented in whole or in part by software, hardware (such as a circuit), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the flow or function described in the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid state disk.

[0045] It should be understood that the term "and / or" herein merely describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents that the front and rear associated objects are in an "or" relationship, but can also represent an "and / or" relationship, which can be understood according to the context before and after.

[0046] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of these items, including any combination of single item or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0047] It should be understood that the size of the sequence number of each process described above does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the application.

[0048] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0049] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0050] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0051] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0052] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0053] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0054] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A business service robot path planning method, characterized in that, The method comprises: S1: using the regional grid map, extracting the function state label and the corresponding effective time period of each service area, adjusting the passing attribute type field of each area by comparing with the real-time time, and generating the passing level field set; S2: using the passing level field set, constructing the task response coefficient according to the task priority level, task generation time and waiting time information of each task in the task list, comparing with the newly input task request, evaluating the adjustment demand of the task execution order, and generating the task position switching record; S3: calling the task position switching record, planning the task execution path of the robot by using the execution position of each task, extracting the passing node number in each candidate path, obtaining the passing frequency value, path intersection frequency value and average residence time value under the cross channel node, calculating the node conflict quantitative value and evaluating the execution priority of the candidate path, and generating the path planning result; S4: using the path planning result, identifying the obstacles in the path by combining the image acquisition device, collecting the displacement trajectory and edge pixel set of the obstacles in the continuous period, identifying the obstacle type and correcting the path, and outputting the behavior path adjustment record.

2. The business service robot path planning method of claim 1, wherein, The passing level field set comprises a function state label, a passing attribute type field and an effective time period label, the task position switching record specifically comprises a task priority level identifier, a task waiting time period identifier and a task response coefficient field, and the path planning result comprises a passing node number list, a cross channel node conflict quantitative value and a candidate path priority sorting structure. The behavior path adjustment record specifically refers to an obstacle type label, a path segment position update identifier and a contour recognition feature set.

3. The business service robot path planning method of claim 1, wherein, The obtaining step of the passing level field set specifically comprises: S101: obtaining the regional grid map, extracting the function state label and the effective time period of each service area, analyzing the label type and the corresponding time period content of each service area, judging the coincidence state of the current real-time time and the effective time period, and obtaining the regional label activation state value; S102: based on the regional label activation state value, adjusting each passing attribute type field in the service area, matching the passing attribute type of each area with the current label state, including passable, limited passing and temporary closure, adjusting the passing attribute type corresponding to the area, and generating the regional passing attribute type set; S103: calling the regional passing attribute type set, synchronously updating the attribute value of the corresponding area in the grid map, synchronously updating the attribute type and the grid map structure field, and establishing the passing level field set.

4. The business service robot path planning method of claim 3, wherein, The obtaining step of the task position switching record specifically comprises: S201: obtaining the passing level field set, collecting the task priority level, task generation time and task waiting time of each task in the robot task list, calculating the urgency of each task, constructing the task response coefficient, and generating the task urgency sequence; S202: According to the task emergency degree sequence, the task request parameter of the new input is collected, the emergency degree of the new input task is compared with the task emergency degree sequence, the priority order of the new task request in the existing task structure is judged, the execution order of the task in the task list is adjusted, and a task priority adjustment result is obtained; S203: Based on the task priority adjustment result, the arrangement order of each task in the robot task list is updated, the tasks and position change information whose execution order changes are screened, and a task position switching record is established.

5. The business service robot path planning method of claim 4, wherein, The specific formula for calculating the emergency degree of each task is: ； Calculate the task emergency degree value; in, For the first The urgency value of each task. For the first Task priority level for each task For the first The normalized value of the waiting time for each task. For the first The complexity factor of each task For the first The constraint level value of the region to which each task belongs. For the current number The operating load factor of each robot Indicates the task number index. This indicates the robot's ID index.

6. The business service robot path planning method of claim 4, wherein, The acquisition step of the path planning result is specifically: S301: Call the task position switching record to obtain the execution position of each task in the task list, analyze the path distribution of each task on the service area map, plan the path of each task, extract the passing node number in each path, and generate a path node number sequence; S302: According to the path node number sequence, the intersection channel node is detected, the passing frequency, path intersection frequency and average residence time of the intersection channel node are collected, the node conflict quantitative value is calculated, the passing pressure of each intersection channel node is quantified, and a node passing pressure numerical set is established; The specific formula for calculating the node conflict quantitative value is: ； Calculate the node conflict quantitative coefficient; wherein, is a conflict quantization coefficient of the mth intersection node, is a passing frequency of the mth intersection node, is a path intersection number of the mth intersection node, is an average value of path intersection numbers of all intersection nodes, is a mean residence time normalization value of the mth intersection node, is a number of robots passing through the mth node by the current path, is an index serial number of the intersection node. S303: Based on the node passing pressure numerical set, compare the node passing pressure of multiple candidate paths, adjust the priority order of each path in combination with the task priority order, and obtain the path planning result.

7. The business service robot path planning method of claim 6, wherein, The acquisition step of the behavior path adjustment record is specifically: S401: Use the path planning result to monitor and identify obstacles in the robot path range by using an image acquisition device, detect the spatial displacement coordinate points and contour edge pixel set of the obstacles in the continuous sampling period, and record the continuous spatial displacement and contour shape change of each obstacle to generate an obstacle shape change index; S402: Based on the obstacle shape change index, analyze the displacement direction trend and contour shape change amplitude of the obstacle, evaluate the stability and spatial position continuity of the obstacle, group the behavior characteristics of the obstacle in the time sequence, and obtain an obstacle behavior classification parameter; S403: According to the obstacle behavior classification parameter, judge the type of the obstacle, including static, intermittent movement and continuous dynamic, and adjust the task execution path of the robot according to the type of the obstacle, correct the path node position and order, and establish a behavior path adjustment record.

8. The business service robot path planning method of claim 1, wherein, The method further comprises: S5: Use the behavior path adjustment record to extract the ground structure parameter value of each grid in the grid map, including the damping material parameter, ground inclination parameter and vibration mark state value, evaluate the ground interference level of each path node, and adjust the posture adjustment control parameter of the robot to generate a robot posture adjustment configuration; The robot posture adjustment configuration includes posture control frequency parameter, response cycle configuration table and ground interference level classification.

9. The business service robot path planning method of claim 8, wherein, The acquisition step of the robot posture adjustment configuration is specifically: S501: Extract the damping material parameter, ground inclination parameter and vibration mark state value of each grid in the path covered grid map using the behavior path adjustment record to obtain the path node ground parameter set; S502: Based on the path node ground parameter set, analyze and calculate the ground interference level of each path node to generate a ground interference level sequence; S503: According to the ground interference level sequence, map the interference level to the attitude control configuration table, match the control period and amplitude of each node, adjust the robot attitude adjustment control parameter, and obtain the robot attitude adjustment configuration.

10. A commercial service robot path planning system, characterized by, The system is used to realize the business service robot path planning method of any one of claims 1-9, and the system comprises: A functional area state processing module uses the area grid map to extract the function state label and the corresponding effective time period of each service area, compares with the real-time time, adjusts the passage attribute type field of each area, and generates a passage level field set; An execution task ordering adjustment module uses the passage level field set, constructs a task response coefficient according to the task priority level, task generation time and waiting time information of each task in the task list, compares with the newly input task request, evaluates the adjustment demand of the task execution order, and generates a task position switching record; A path passage pressure evaluation module calls the task position switching record, plans the task execution path of the robot using the execution position of each task, extracts the passage node number in each candidate path, obtains the passage frequency value, path intersection frequency value and average stay time value under the cross channel node, calculates the node conflict quantitative value and evaluates the execution priority of the candidate path, and generates a path planning result; A path obstacle dynamic identification module uses the path planning result, identifies the obstacles in the path in combination with the image acquisition device, acquires the displacement trajectory and edge pixel set of the obstacles in the continuous period, identifies the obstacle type and corrects the path, and outputs the behavior path adjustment record; An attitude adjustment configuration matching module uses the behavior path adjustment record to extract the ground structure parameter value of each grid in the grid map, including the damping material parameter, ground inclination parameter and vibration mark state value, evaluates the ground interference level of each path node, adjusts the attitude adjustment control parameter of the robot, and generates the robot attitude adjustment configuration.

Images