A method and system for division of labor of intelligent robots

Abstract

The application discloses a kind of division of labor cooperation method and system of intelligent robot, it is related to robot control technical field, the method includes: obtaining job task demand and robot group information, execute path planning and calculate path energy consumption distribution, construct job task atlas;Job cooperation constraint set is generated based on atlas, and the constraint is combined to divide robot group into work body and energy support body, form initial division of labor scheme set;Simulate division of labor cooperation in virtual simulation environment, evaluate cooperation adaptability;Select the highest division of labor scheme of cooperation adaptability as final execution scheme, and issue to robot group execution job scheduling.Further achieve the technical effect of improving task execution efficiency, enhancing adaptability and job radius, optimizing the self-adaptability of division of labor and energy efficiency ratio in complex environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot control, in particular to a division and cooperation method and system of intelligent robots. BACKGROUND

[0002] In the field of intelligent robots, robot clusters are increasingly widely used in various complex tasks. However, most of the current robot cooperation methods adopt a homogeneous job model, that is, all robots perform the same or similar tasks, lacking differentiated division of labor for robot roles. In addition, in the aspect of path planning, the existing technology usually only considers geographical path information, ignoring the impact of energy consumption distribution on task execution, which leads to insufficient refinement of task allocation, and the relationship between robot energy consumption and task adaptation cannot be fully considered, thereby affecting the task accessibility, task adaptability and overall efficiency of the robot cluster. SUMMARY

[0003] The present application provides a division and cooperation method and system of intelligent robots to solve the technical problems of single division and cooperation mechanism, disconnection between path planning and energy consumption management, and insufficient adaptability to complex tasks in the prior art, and achieves the technical effects of improving task execution efficiency, enhancing adaptability and job radius, and optimizing division and adaptability and energy efficiency ratio in complex environments.

[0004] In a first aspect, the present application provides a division and cooperation method of intelligent robots, wherein the division and cooperation method of intelligent robots comprises:

[0005] Obtaining task demand information of a target job task and robot group information, and performing job path planning according to the task demand information and the robot group information.

[0006] Corresponding to the path energy consumption distribution of the job path planning result, and combining the path energy consumption distribution, the job path planning result, the task demand information and the robot group information, a job task graph is constructed.

[0007] According to the job task graph, intuitive job cooperation constraints and indirect job cooperation constraints are generated, and the output is a job cooperation constraint set. The robot group is randomly divided into a job body and a power support body in iteration, and an initial division scheme set is obtained.

[0008] Based on the job task graph, the job cooperation constraints and the initial division scheme set, a virtual simulation environment is established, a division and cooperation simulation is performed, and a cooperation adaptability evaluation is performed based on the division and cooperation simulation result.

[0009] Selecting the division scheme with the highest cooperation adaptability as the target execution scheme and issuing it to the robot group for job scheduling.

[0010] In an implementation, the task demand information comprises at least task area information, work object information and task timeliness information, and the robot group information comprises at least robot quantity, robot intrinsic parameter and corresponding robot energy consumption model.

[0011] In an implementation, a path energy consumption distribution corresponding to the work path planning result is generated, and a work task graph is constructed by combining the path energy consumption distribution, the work path planning result, the task demand information and the robot group information, comprising:

[0012] A path point sequence is generated based on the work path planning result.

[0013] An energy consumption function model of each robot is extracted from the robot group information.

[0014] An energy consumption characteristic value of each path point in the path point sequence is calculated based on geographic space data, the robot energy consumption model and the task demand information.

[0015] A path energy consumption distribution is obtained based on the energy consumption characteristic value, and the path energy consumption distribution is associated with the work path planning result, and a work task graph is constructed by combining the task demand information and the robot group information.

[0016] In an implementation, intuitive work cooperation constraints and indirect work cooperation constraints are generated according to the work task graph, and output as a work cooperation constraint set, comprising:

[0017] The task demand information and the robot group information are regularized to form intuitive work cooperation constraints.

[0018] Indirect work cooperation constraints are calculated based on the work task graph and the intuitive work cooperation constraints, wherein the indirect work cooperation constraints comprise at least task guarantee constraints, endurance safety constraints, time window constraints and service capacity constraints.

[0019] The intuitive work cooperation constraints and the indirect work cooperation constraints are integrated to output the work cooperation constraint set.

[0020] In an implementation, a virtual simulation environment is established based on the work task graph, the work cooperation constraints and the initial division scheme set, and a division cooperation simulation is performed, comprising:

[0021] The work task graph is taken as a simulation target input parameter, and the work cooperation constraint set is taken as an optimization restriction condition to establish the virtual simulation environment.

[0022] According to the job speed interval extracted from the job task graph, each initial division scheme is differentially configured to obtain a set of candidate division schemes.

[0023] The set of candidate division schemes is input to the virtual simulation environment for initial screening of division strategies and effectiveness evaluation, corresponding to cleaning the set of candidate division schemes, and performing division cooperation simulation on the cleaned set of candidate division schemes.

[0024] In a feasible implementation, the division scheme with the highest cooperation fitness is selected as the target execution scheme, which is issued to the robot group for job scheduling, wherein the execution steps of the job body include:

[0025] When the remaining endurance is lower than a preset threshold α, a power supplement request information is automatically generated and broadcasted, the power supplement request information including the current position coordinates and the power supplement demand.

[0026] The current task is continued to be executed until the power supplement guarantee body arrives.

[0027] The power supplement process is completed by docking with the power supplement guarantee body, and a power supplement confirmation signal is sent after the power supplement is completed.

[0028] The current task execution process is continued, and the remaining endurance is monitored synchronously.

[0029] In a feasible implementation, the division scheme with the highest cooperation fitness is selected as the target execution scheme, which is issued to the robot group for job scheduling, wherein the execution steps of the power supplement guarantee body include:

[0030] The power supplement request information of the job body is continuously monitored.

[0031] The power supplement request information is matched in proximity to generate a power supplement task and perform path planning to the target job body.

[0032] The path planning result is used to arrive at the job body to perform power supplement operation, wherein the power supplement mode includes contact type power supplement or non-contact type power supplement.

[0033] After the power supplement is completed, the remaining endurance is detected.

[0034] When the remaining endurance is lower than a preset threshold, a return path is preferentially planned, otherwise, the power supplement request monitoring state is resumed.

[0035] In a feasible implementation, the initial division scheme set includes the configuration number and configuration ratio of the job body and the power supplement guarantee body, wherein:

[0036] The job body is used to execute a target job task.

[0037] The energy supplement body is provided with an energy supply assembly for providing energy supplement for the work body during work.

[0038] In an implementable manner, the cooperation adaptability is obtained by fusing preset multi-dimensional evaluation factors, and the multi-dimensional evaluation factors at least include total energy consumption, total path length, unit path energy consumption, cooperation cost, and effective work energy consumption ratio.

[0039] In a second aspect, the application further provides a work cooperation system of intelligent robots, wherein the work cooperation system of intelligent robots comprises:

[0040] A work information acquisition module is configured to acquire task demand information of a target work task and robot group information, and perform work path planning according to the task demand information and the robot group information.

[0041] A work task graph construction module is configured to correspondingly calculate path energy consumption distribution of work path planning results, and construct a work task graph by combining the path energy consumption distribution, work path planning results, task demand information and robot group information.

[0042] A work division scheme acquisition module is configured to generate direct work cooperation constraints and indirect work cooperation constraints according to the work task graph, output as a work cooperation constraint set, and combine the work cooperation constraint set to iteratively randomly divide the robot group into work bodies and energy supplement guarantee bodies to obtain an initial work division scheme set.

[0043] A simulation and evaluation module is configured to establish a virtual simulation environment based on the work task graph, the work cooperation constraints and the initial work division scheme set, perform work cooperation simulation, and evaluate cooperation adaptability based on work cooperation simulation results.

[0044] A scheme screening and execution module is configured to select a work division scheme with the highest cooperation adaptability as a target execution scheme, and issue the target execution scheme to the robot group for work scheduling.

[0045] The application discloses a kind of intelligent robot's division cooperation method and system, comprising: obtaining the task demand data of target operation task and corresponding robot group information, and based on the task demand data and group information, operation path planning is executed;Path energy consumption distribution calculation is carried out to operation path planning result, and the operation task graph is constructed in combination with path energy consumption distribution, operation path planning result, task demand data and robot group information;According to operation task graph, generate direct cooperation constraint and indirect cooperation constraint for depicting collaborative relationship, form operation cooperation constraint set, and based on the cooperation constraint set, robot group is iteratively randomly divided, constructs operation body and energy support body, obtains initial division scheme set;Relying on operation task graph, operation cooperation constraint set and initial division scheme set, construct virtual simulation environment, carry out division cooperation simulation, and according to simulation result, cooperation adaptation degree is evaluated;From initial division scheme set, the division scheme with highest adaptation degree is selected as target scheduling scheme, and target scheduling scheme is issued to robot group to execute operation task scheduling.The intelligent robot's division cooperation method and system disclosed in the application solve the technical problems that division cooperation mechanism is single, path planning is disconnected with energy consumption management, and complex task adaptability is insufficient, realize the technical effects of improving task execution efficiency, enhancing adaptability and operation radius, optimizing division self-adaptability and energy efficiency ratio in complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 It is a flowchart of the division cooperation method of the intelligent robot of the application.

[0047] Figure 2 It is a structure diagram of the division cooperation system of the intelligent robot of the application.

[0048] In the drawings, the components represented by each reference numeral are described as follows:

[0049] Operation information acquisition module 11, operation task graph construction module 12, division scheme acquisition module 13, simulation and evaluation module 14, scheme screening and execution module 15. DETAILED DESCRIPTION

[0050] The above technical solutions will be described in detail below in conjunction with the drawings and specific embodiments, so as to better understand the above technical solutions. Obviously, the described embodiments are only part of the embodiments of the application, not all embodiments of the application, and it should be understood that the application is not limited to the example embodiments for explaining the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the application. In addition, it should be noted that, for convenience of description, only the parts related to the application are shown in the drawings, not all.

[0051] Embodiment one, as Figure 1 A flowchart of a division and cooperation method of an intelligent robot according to the present application, wherein the division and cooperation method of the intelligent robot comprises:

[0052] S100: Obtain task demand information of a target operation task and robot group information, and perform operation path planning according to the task demand information and the robot group information.

[0053] Specifically, the target operation task refers to a specific operation task to be completed received by the intelligent robot. Optionally, the target operation task can include different types of tasks such as carrying, inspection, spraying, and cleaning. The target operation task is usually issued by a task management module or a higher-level scheduling system. The task demand information refers to detailed parameters describing the target operation task, for example, including task type, operation area, time requirement, load demand, and precision requirement, etc., which are used to guide the robot to perform the task.

[0054] Specifically, the robot group information refers to the capability, state, and position of the robot individuals participating in the cooperation, which reflects the state of the robots in the robot group. For example, it includes the load capacity, endurance state, current task state, navigation precision, and reachable area of each robot. Based on the known task demand and robot capability, the robot group can be assigned a specific task and the optimal movement and operation path can be generated, i.e., operation path planning, so that the overall task completion efficiency is optimal, resource utilization is maximized, and the task constraint conditions are met.

[0055] Specifically, first, the target operation task issued by the task management module or the higher-level scheduling system is received, and the task demand information is parsed, for example: the task type is warehouse carrying, the target area is area A to area B, and it is required to be completed within 30 minutes, and the single carrying weight is not more than 10 kg. At the same time, the robot group information of the current robot group is obtained, for example, the position, load capacity, passability, power state, and congestion of 5 robots. Then, based on the task demand and the robot group information, an obstacle avoidance path is generated for the current robot group, for example, if the robot R1 is currently located in area A and has sufficient power and a load capacity of 15 kg, it can be assigned to a heavy load path with a shorter distance, and a shortest path is planned for it to bypass obstacles from area A to area B.

[0056] Through the above process, based on the task demand and the robot capability, an efficient operation path (i.e., operation path planning result) can be generated by combining the path planning algorithm. The determined operation path provides feasible and accurate path information for subsequent analysis and simulation, facilitating efficient execution of division and cooperation decision-making.

[0057] In some embodiments, the task demand information at least includes task area information, work object information and task timeliness information, and the robot group information at least includes the number of robots, robot intrinsic parameters and corresponding robot energy consumption models.

[0058] Specifically, the task demand information refers to a data set for describing the specific content and constraint conditions of the to-be-executed task, including three types of key elements: task area information, i.e., the geographical or spatial location range of the task, such as two-dimensional map coordinates, three-dimensional space area or specific functional block, for guiding robot path planning and scheduling; work object information, referring to the operation target involved in the task, such as carrying articles, detecting area, cleaning ground, etc., usually including the type, number, position and operation requirement of the object; and task timeliness information, referring to the time constraint of the task, such as the earliest start time, the latest completion time or the execution duration, etc., for realizing time-sensitive scheduling or priority sorting.

[0059] Specifically, the robot group information is a data set for describing the current schedulable robot capability and state, including: the number of robots, i.e., the total number of robot individuals currently available for executing the task; robot intrinsic parameters, referring to the inherent performance parameters of the robot, such as maximum speed, load capacity, battery capacity, sensor type, etc.; and robot energy consumption model, referring to a mathematical model for describing the energy consumption law of the robot when executing different tasks, usually based on variables such as speed, load, path length, etc.

[0060] S200: Corresponding to the path energy consumption distribution calculated for the work path planning result, and combining the path energy consumption distribution, the work path planning result, the task demand information and the robot group information, a work task graph is constructed.

[0061] Specifically, according to the path information obtained by path planning and the energy consumption model of the robot group and other data, the energy consumption of the robot on each path during task execution can be calculated, and then the path energy consumption distribution is generated. The path energy consumption distribution reflects the specific energy consumption of the robot group on the work path planning result, which is used to quickly call the energy consumption characteristic parameters in subsequent simulation analysis, thereby improving the efficiency.

[0062] Specifically, the construction of the work task graph is to combine the calculated path energy consumption distribution with the work path planning result, the task demand information and the robot group information, to form a structured model that comprehensively reflects various key elements and their mutual relationships in the task execution process, for clearly and intuitively expressing the task information, robot related information and path constraint information involved in the division of labor and cooperation, facilitating quick search and call in subsequent simulation.

[0063] Through the above process, the energy consumption of the robot operation path can be finely modeled, and at the same time, by constructing the operation task graph, the dispersed task information, robot capability and path performance data can be structured and integrated, the global understanding and management capability of the task execution process is improved, which is used for subsequent simulation.

[0064] In some embodiments, a path energy consumption distribution corresponding to the operation path planning result is calculated, and an operation task graph is constructed by combining the path energy consumption distribution, the operation path planning result, task demand information and robot group information, comprising:

[0065] A path point sequence is generated based on the operation path planning result; an energy consumption function model of each robot is extracted from the robot group information; an energy consumption characteristic value of each path point in the path point sequence is calculated by combining geographic space data, the robot energy consumption model and the task demand information; a path energy consumption distribution is obtained by fitting based on the energy consumption characteristic value, and the path energy consumption distribution is associated with the operation path planning result, and an operation task graph is constructed by combining the task demand information and the robot group information.

[0066] Specifically, the path point sequence is a series of ordered path points based on the operation path planning result, wherein each path point contains position information (such as two-dimensional or three-dimensional coordinates), a timestamp, a speed or an attitude, etc., for discretely describing the running track of the robot on the path, facilitating subsequent analysis and calculation. The energy consumption function model is a mathematical model for describing the energy consumption relationship of the robot per unit time or per unit distance under different running states (such as speed, load, terrain, etc.), which is usually an empirical formula or a function fitted based on experimental data.

[0067] Specifically, the geographic space data is used to reflect the actual three-dimensional information of the operation path planning result, and exemplary includes information such as the terrain type, slope, obstacle distribution and surface material of the area where the path point is located. The factors contained in the geographic space data will significantly affect the energy consumption performance of the robot on the path.

[0068] Specifically, the energy consumption characteristic value refers to the energy consumption estimate value calculated by combining the motion state of the robot and the environmental information at a specific path point, which is exemplary in joules / m. The path energy consumption distribution is a continuous energy consumption change curve formed by fitting the energy consumption characteristic values of all path points, which is used to approximately describe the energy consumption trend of the robot when operating along the operation path planning result.

[0069] Specifically, the operation task graph is a task knowledge graph represented in a graph structure by fusing the path energy consumption distribution, the operation path planning result, the task demand information and the robot group information, which is used to provide a structured information basis for subsequent steps.

[0070] Specifically, first, based on the path planning result, a path point sequence is generated, for example, according to a preset division granularity of 0.5 m, the path from the starting point A to the ending point B of the path planning result is divided into a path point every 0.5 m, and a path point sequence P = {p1, p2,..., pn} is obtained. Each path point contains its spatial coordinates and expected speed. Subsequently, the energy consumption function model of each robot is extracted from the robot group information, for example, assuming that all robots in the robot group are of the same model R1, the energy consumption model can be represented as:

[0071] E(v, w, θ) = αv 2 + βw + γsin(θ);

[0072] where v is the speed, w is the load, θ is the slope angle, and α, β, γ are empirical parameters (which can be determined based on experiments).

[0073] Specifically, then, in combination with geographic spatial data and task demand information, the energy consumption characteristic value of each point in the path point sequence is calculated. For example, at path point p3, the slope is 10°, the speed is 1.2 m / s, and the load is 8 kg. Substituting into the corresponding energy consumption model gives an energy consumption characteristic value of 9.45 J / km. Similarly, the path point energy consumption sequence E = {0.3, 0.35, 0.45,..., 0.25} is obtained.

[0074] Further, the obtained path point energy consumption sequence is fitted by a curve fitting method such as spline interpolation or Gaussian fitting, to generate a path energy consumption distribution curve. The energy consumption distribution is then structured and integrated with the path point sequence, task demand information, and robot group information to construct a job task graph. For example, the nodes in the graph can include path segments, robots, task objects, energy consumption values, etc. The edges represent execution, passing, consumption, and other semantic relationships.

[0075] Through the above process, the energy consumption distribution of the robot during task execution is accurately obtained, and combined with other key information in the task execution process, a job task graph is generated, providing detailed and accurate data support and calling basis for subsequent job collaboration constraint generation, division scheme formulation, simulation, etc. It helps to improve the overall efficiency and task execution effect.

[0076] S300: According to the job task graph, intuitive job collaboration constraints and indirect job collaboration constraints are generated, output as a job collaboration constraint set, and the robot group is iteratively randomly divided into a job body and a power support body based on the job collaboration constraint set, obtaining an initial division scheme set.

[0077] Specifically, the direct job cooperation constraint refers to the explicit constraint condition derived directly from the task demand information and the robot group information, for example: "heavy load task can only be executed by robots with load capacity greater than 10 kg", "the task must be completed within 10 minutes" and the like. The regularization is the unified coding and standardized processing of task and robot information of different sources and different formats, so as to have comparability and calculability, for example, the task time limit is uniformly converted into an acceptable time window, and the robot capability attribute is converted into a standard capability vector.

[0078] Specifically, the indirect job cooperation constraint is an implicit cooperation restriction obtained by analyzing the association relationship between various entities in the job task graph and the direct job cooperation constraint reasoning, including but not limited to:

[0079] Task guarantee constraint, such as the task must be completed by a robot with specific ability or historical experience; endurance safety constraint, such as the robot must ensure that the remaining power is higher than a certain threshold in the task path to prevent failure halfway; time window constraint, such as the task can only be executed within a certain time period, and the robot needs to arrive within the window; service capacity constraint, such as the robot serves N targets or carries M units of material in one task at most. The generated job cooperation constraint set is the set of integrating the above direct and indirect constraints, which is used as the input condition set of task scheduling and path optimization.

[0080] In some embodiments, the initial division scheme set includes the configuration number and configuration ratio of the job body and the energy guarantee body, wherein:

[0081] The job body is used to execute the target job task; the energy guarantee body is configured with an energy supply component for providing energy supply for the job body during the job process.

[0082] Specifically, the job body and the energy guarantee body are two subsets of the robot group determined according to the cooperative role and energy support function, wherein the job body is responsible for directly executing the job task, such as inspection, carrying, collection, etc.; the energy guarantee body is used to provide energy support for the job body, such as power supply, fuel supply, task replacement, etc. Through the cooperation of the job body and the energy guarantee body, the service radius and the job stability of the robot group can be improved efficiently and at low cost without improving the performance of the single robot.

[0083] Specifically, under the premise of meeting part of the cooperative constraints in the obtained job cooperation constraint set, a heuristic or probabilistic method is used to randomly divide the robot group to explore various possible cooperation strategies, and an initial division scheme set is generated, i.e., a combination scheme of multiple job bodies and energy supplement guarantee bodies generated through the above division process, as a candidate solution space for subsequent optimization and scheduling. For example, the job bodies and energy supplement guarantee bodies can be configured as 2:1, and the total number is 30.

[0084] Through the above process, a diversified initial division scheme set can be generated based on cooperative constraints, providing a rich candidate solution for subsequent task scheduling optimization and strategy selection.

[0085] In some embodiments, intuitive job cooperation constraints and indirect job cooperation constraints are generated according to the job task graph, and the output is a job cooperation constraint set, which includes:

[0086] The task demand information and the robot group information are regularized to form intuitive job cooperation constraints; indirect job cooperation constraints are calculated based on the job task graph and the intuitive job cooperation constraints, wherein the indirect job cooperation constraints at least include: task guarantee constraints, endurance safety constraints, time window constraints, and service capacity constraints; and the intuitive job cooperation constraints and the indirect job cooperation constraints are integrated to output the job cooperation constraint set.

[0087] Specifically, first, the task demand information and the robot group information are regularized. For example, the task T1 requires "to complete the inspection of 3 target points within 15 minutes", which can be regularized as a vector [task type: inspection; time window: 0, 15; service capacity: 3] to generate intuitive job cooperation constraints. Then, based on the aforementioned constructed job task graph, indirect cooperation constraints are further inferred. For example, the graph shows that the path energy consumption of T1 is 25J, and the remaining power of R1 is 30J, so the endurance safety constraint "R1 should safely complete T1 and return to the base station" can be derived. Finally, all intuitive and indirect constraints can be integrated to form a job cooperation constraint set as the input of the subsequent steps.

[0088] Through the above process, direct usable cooperation constraints can be generated, and various implicit constraint conditions can be inferred through the structured representation and semantic association of the task graph, enriching the understanding of the cooperation relationship between tasks and robots. Further, the rationality and adaptability of the scheduling strategy are improved, and problems such as task conflict, resource waste, or robot failure caused by ignoring implicit constraints are avoided.

[0089] S400: Based on the job task graph, the job cooperation constraints, and the initial division scheme set, a virtual simulation environment is established, a division and cooperation simulation is performed, and a cooperation adaptation degree evaluation is performed based on the division and cooperation simulation result.

[0090] Specifically, the virtual simulation environment is a digital platform built in a computer or an edge computing platform, etc. for simulating the collaborative work of the robot group in the actual task environment. Through the geographical information, task nodes, path structure, energy consumption model, etc. contained in the task graph, the virtual simulation environment can reconstruct a real or approximately real work scene, so that the robot behavior, path planning, energy consumption change, etc. can be digitally simulated and evaluated.

[0091] Specifically, in the above virtual simulation environment, the initial division scheme (i.e. the scheme in which the robots are divided into work bodies and energy supplement guarantee bodies) can be simulated respectively to simulate the collaborative behavior of each robot in the task execution process, including task allocation, path execution, energy supplement scheduling, etc. and then through the analysis of the evaluation of the task completion rate, energy consumption efficiency, task conflict rate, energy supplement success rate, etc. in the simulation process, the adaptation effect of multiple initial division schemes under specific task graphs and collaboration constraints is quantified to provide a basis for subsequent optimization.

[0092] Through the above process, the collaboration effect of differentiated division strategies can be quickly evaluated without actually deploying robots, thereby reducing the trial and error cost before actual deployment. Among them, the virtual simulation environment can provide a highly controllable and repeatable test platform; the collaboration adaptation evaluation result provides a quantitative feedback for the subsequent optimization algorithm, and attention is paid to the closed loop of simulation-evaluation-optimization to improve the collaboration efficiency, task completion rate and stability of the robot group.

[0093] In some embodiments, a virtual simulation environment is established based on the task graph, the task collaboration constraints and the initial division scheme set, corresponding to the execution of division collaboration simulation, comprising:

[0094] The task graph is taken as the simulation target input parameter, and the task collaboration constraint set is taken as the optimization restriction condition to establish the virtual simulation environment; according to the work speed interval extracted from the task graph, each initial division scheme is differentiated to obtain a set of candidate division schemes; the set of candidate division schemes is input into the virtual simulation environment for division strategy preliminary screening and effectiveness evaluation, corresponding to cleaning the set of candidate division schemes, and performing division collaboration simulation on the cleaned set of candidate division schemes.

[0095] Specifically, the work speed interval refers to the preset robot execution speed range for different task nodes or path segments in the task graph, which is limited by factors such as terrain type, task complexity, task urgency, etc. For example, in rugged terrain or narrow passages, the robot needs to execute the task at a lower speed to ensure safety; while in flat areas or emergency task scenarios, the robot can be allowed to work at a higher speed.

[0096] Specifically, the differential configuration is a process of individualizing the robot behavior parameters in the initial division scheme according to the above-mentioned work speed interval, so as to obtain various alternative division schemes. The division strategy preliminary screening and effectiveness evaluation are rapid screening of the alternative division schemes before simulation, which are used to eliminate schemes that obviously do not meet the constraint conditions or have too low performance indicators, so as to reduce the simulation calculation overhead.

[0097] Specifically, first, the work task graph is taken as the simulation target input parameter, the task nodes, path structure, energy consumption model, work speed interval and other information are extracted, and the work cooperation constraint set is taken as the optimization restriction condition to construct a virtual simulation environment. For example, assuming that for a task graph containing 20 task points, 5 types of terrain and 10 robots (R1-R10), the work speed interval allowed by the 10 robots in different terrains can be extracted, such as grassland 0.5-1.0 m / s, cement road 1.0-1.5 m / s, and slope 0.3-0.8 m / s.

[0098] Specifically, then, the task execution speed of the robots in each scheme in the initial division scheme set is differentially configured in combination with the random number generator and the obtained work speed interval, thereby generating a plurality of variant schemes to form an alternative division scheme set. Then, the above-mentioned alternative division scheme set is input into the virtual simulation environment, and the division strategy preliminary screening is performed through the rapid evaluation indexes configured based on the work task graph, such as the predicted task completion time, the predicted total energy consumption, the task coverage rate and the like, to eliminate schemes that obviously do not meet the constraint conditions, such as the predicted task completion rate being lower than 80% and the energy consumption exceeding the upper limit of the power.

[0099] Further, the high-potential schemes retained after cleaning are further input into the virtual simulation environment for fine-grained division and cooperation simulation, simulating the collaborative work process of the robots under different speed configurations, recording and analyzing key indicators such as the total energy consumption, the total mileage of the robot group, the actual energy consumption per path, the number of path conflicts, and the stability, and generating the division and cooperation simulation results.

[0100] Through the above process, the simulation efficiency can be improved, and redundant simulation of a large number of inefficient division schemes is avoided, and each alternative scheme is more realistic and feasible. Among them, the differential configuration enhances the adaptability to complex heterogeneous task environments, facilitates obtaining more realistic schemes, and thus provides the robot group with more targeted and effective cooperation strategies.

[0101] In some embodiments, the cooperation adaptation degree is obtained by fusing preset multi-dimensional evaluation factors, and the multi-dimensional evaluation factors at least include total energy consumption, total path length, unit path energy consumption, division and cooperation cost, and effective work energy consumption ratio.

[0102] Specifically, the cooperation adaptability is a dimensionless numerical index for quantifying the overall coordination effect and execution efficiency of the alternative division of labor scheme under the specific task graph and cooperation constraints, which can reflect the rationality of cooperation between robots and the efficiency of task completion; the multi-dimensional evaluation factor is a basic evaluation parameter constituting the cooperation adaptability, which covers multiple angles to comprehensively quantify the advantages and disadvantages of the division of labor scheme. Specifically, it includes:

[0103] The total energy consumption, i.e., the total energy consumed by all robots in the process of completing the task, reflects the overall energy consumption; the total path length, i.e., the cumulative length of the path walked by all robots in the execution of the task, is used to evaluate the path planning efficiency; the unit path energy consumption, i.e., the energy consumed per unit path length, measures the coordination between energy consumption and path planning; the division of labor cooperation cost refers to the resource consumption cost caused by task division, cooperation scheduling, communication coordination, etc.; the effective work energy consumption ratio, i.e., the proportion of energy consumption for actual effective work such as resource transportation in the total energy consumption, is used to evaluate the efficiency of energy use.

[0104] S500: Select the division of labor scheme with the highest cooperation adaptability as the target execution scheme and issue it to the robot group for job scheduling.

[0105] Specifically, the target execution scheme is the scheme with the highest comprehensive cooperation adaptability index score among all the alternative schemes after simulation evaluation and cleaning, which has the optimal comprehensive performance in task completion efficiency, energy consumption control, path optimization and cooperation coordination, and meets the preferences and requirements of the target manufacturer on multi-dimensional evaluation factors, and can be used for actual deployment of division of labor and scheduling configuration.

[0106] Specifically, the target execution scheme contains specific parameters such as task allocation, path planning, time scheduling of each robot, which can be sent to each robot terminal in the form of instructions or task packages through the communication module, so that the robot group can start and execute the job task according to the target execution scheme.

[0107] Through the above process, the finally deployed division of labor scheme can be ensured to be optimal under multi-dimensional performance indicators, so as to realize efficient collaborative work of the robot group in complex task environment, improve the stability and energy efficiency ratio of task completion, reduce the risk of resource waste and task failure caused by unreasonable division of labor, and enhance the application value and intelligent decision-making ability of the robot group in actual scenarios.

[0108] In some embodiments, the division of labor scheme with the highest cooperation adaptability is selected as the target execution scheme, which is issued to the robot group for job scheduling, wherein the execution steps of the job body include:

[0109] Monitoring the remaining endurance, when below the preset threshold a, automatically generating and broadcasting the energy supplement request information, including the current position coordinates and the energy supplement demand; continue to execute the current task until the energy supplement guarantee body arrives; interface with the energy supplement guarantee body, complete the energy supplement process, and send the energy supplement confirmation signal after the energy supplement is completed; continue the current task execution process, and monitor the remaining endurance.

[0110] Specifically, the job body refers to a robot individual that executes specific production, service or task behavior, has the ability of movement, perception, execution, etc., and is the main role that undertakes actual work.

[0111] Specifically, in the job execution process, the job body will monitor its remaining endurance in real time, such as obtaining the current battery percentage, drivable distance, etc. through the battery management system (BMS). When it is detected that the remaining endurance is lower than the set threshold a, the energy supplement request information will be automatically generated, such as the current position coordinates (x = 120, y = 85) and the energy supplement demand of 800Wh, and the information will be broadcasted through the wireless communication module (such as Wi-Fi, 5G or V2X). Thus, the nearest available energy supplement guarantee body can be dispatched to the job body location (i.e. the current position coordinates). During this period, the job will continue to execute the current task and update the job body position until the energy supplement guarantee body arrives, ensuring that the task does not interrupt.

[0112] Further, after the energy supplement guarantee body arrives, the energy transmission process is completed through the positioning and interfacing mechanism. After the energy supplement is completed, the job body will actively send the energy supplement completion confirmation signal for log recording and scheduling optimization, continue the current task process and continuously monitor the endurance state, and enter the next round of energy supplement judgment logic.

[0113] Through the above process, the job body can realize dynamic energy supplement without interrupting task execution, avoid the risk of passive shutdown due to robot energy depletion, and enhance the stability and autonomy in long-time and large-scale task execution.

[0114] In some embodiments, the division of labor scheme with the highest cooperation adaptability is selected as the target execution scheme, which is issued to the robot group for job scheduling. The execution steps of the energy supplement guarantee body include:

[0115] Continuously monitoring the energy supplement request information of the job body; according to the energy supplement request information, performing near matching to generate an energy supplement task and execute path planning to the target job body; according to the path planning result, arriving at the job body to perform energy supplement operation, wherein the energy supplement method includes contact type energy supplement or non-contact type energy supplement; detecting the remaining endurance after the energy supplement is completed; when the remaining endurance is lower than the preset threshold, preferentially planning a return path, otherwise, restoring the energy supplement request monitoring state.

[0116] Specifically, the energy supplement guarantee body is a mobile energy support unit specially used for providing energy supplement for the working body, and has functions of autonomous navigation, path planning, energy transmission and task communication, and can select the most suitable energy supplement guarantee body to respond according to the spatial distance between the energy supplement guarantee body and the working body sending the request, the path cost, the current task load and other factors.

[0117] Specifically, the contact type energy supplement includes fuel gun connection, cable connection, energy pack replacement and the like, and the non-contact type energy supplement includes inductive wireless charging and the like; the preset threshold is a lower limit of the power, and is used for triggering the energy supplement or the return trip and the like.

[0118] Specifically, when the energy supplement guarantee body is in a standby state, the energy supplement guarantee body is set to continuously listen to the energy supplement request information broadcast in the network, and once the request of the working body is received, the shortest path priority algorithm is used for the nearest matching according to the position, the remaining energy, the task state and other parameters of the current energy supplement guarantee body, and the optimal energy supplement guarantee body is distributed to respond to the task. After the matching is successful, the energy supplement task is generated for the energy supplement guarantee body, and the path planning component is started to plan the optimal path from the current position of the energy supplement guarantee body to the target working body, and the energy supplement guarantee body is navigated to the target working body along the path, and the energy transmission is completed according to the interface type of the working body by selecting the contact type or the non-contact type energy supplement

[0119] Further, after the energy supplement is completed, the energy supplement guarantee body detects the remaining endurance power, and if the remaining endurance power is lower than the preset threshold, the return trip path is planned to return to the energy supplement station or the charging pile; if the energy supplement guarantee body still has sufficient power, the energy supplement guarantee body reenters the listening state to be ready to respond to the next energy supplement request. Preferably, the preset threshold can be set based on the endurance dynamic evaluation model and the distance from the base station (i.e. the starting point of the robot group or the path of the mother station), for example, 1.2 times of the remaining endurance value obtained by the endurance dynamic evaluation model is set as the preset threshold.

[0120] Through the above process, the energy supplement guarantee body can realize dynamic, accurate and low-delay energy support for the working body, enhance the operation continuity and adaptability of the robot group, avoid the task interruption caused by the energy depletion of the working body, and improve the self-adaptation and cooperation ability of the entire robot group in the complex task environment.

[0121] In summary, the division and cooperation method of the intelligent robot provided by the application has the following technical effects:

[0122] The task demand data of the target work task and the corresponding robot group information are acquired, work path planning is performed based on the task demand data and the group information, path energy consumption distribution calculation is performed on the work path planning result, the work task graph is constructed in combination of the path energy consumption distribution, the work path planning result, the task demand data and the robot group information, the direct cooperation constraint and the indirect cooperation constraint for depicting the cooperation relationship are generated according to the work task graph, the work cooperation constraint set is formed, the robot group is iteratively and randomly divided based on the cooperation constraint set, the work body and the energy support body are constructed, and the initial division scheme set is obtained; the virtual simulation environment is constructed relying on the work task graph, the work cooperation constraint set and the initial division scheme set, the division and cooperation simulation is carried out, and the cooperation adaptation degree evaluation is carried out according to the simulation result; the division scheme with the highest adaptation degree is selected from the initial division scheme set as the target scheduling scheme, and the target scheduling scheme is issued to the robot group to execute the work task scheduling, so that the technical effects of improving the task execution efficiency, enhancing the adaptability and work radius, and optimizing the division self-adaptability and energy efficiency ratio in the complex environment are realized.

[0123] In the embodiment two, the flowchart of the division and cooperation method of the intelligent robot can be implemented by the structure as shown in the figure. Figure 2 is a structural schematic diagram of a division and cooperation system of an intelligent robot. For example, Figure 1 The flowchart of the division and cooperation method of the intelligent robot can be implemented by the structure as shown in the figure. Figure 2

[0124] Based on the same idea as the division and cooperation method of the intelligent robot in the embodiment, the division and cooperation system of the intelligent robot comprises:

[0125] The work information acquisition module 11 is configured to acquire the task demand information of the target work task and the robot group information, and perform work path planning according to the task demand information and the robot group information.

[0126] The work task graph construction module 12 is configured to calculate the path energy consumption distribution of the work path planning result, and construct the work task graph in combination of the path energy consumption distribution, the work path planning result, the task demand information and the robot group information.

[0127] The division scheme acquisition module 13 is configured to generate the direct work cooperation constraint and the indirect work cooperation constraint according to the work task graph, output the work cooperation constraint set, and iteratively and randomly divide the robot group into the work body and the energy support body in combination of the work cooperation constraint set, so as to obtain the initial division scheme set.

[0128] ​The simulation and evaluation module 14 is configured to establish a virtual simulation environment based on the job task graph, the job cooperation constraint and the initial division scheme set, perform a division cooperation simulation, and perform cooperation fitness evaluation based on a result of the division cooperation simulation.

[0129] The scheme screening and execution module 15 is configured to select a division scheme with the highest cooperation fitness as a target execution scheme, and issue the target execution scheme to the robot group for job scheduling.

[0130] In some embodiments, the job task graph construction module 12 comprises:

[0131] The job path planning unit is configured to obtain task demand information of a target job task and robot group information, and perform job path planning according to the task demand information and the robot group information.

[0132] The job task graph construction unit is configured to calculate a path energy consumption distribution of the job path planning result, and construct a job task graph in combination with the path energy consumption distribution, the job path planning result, the task demand information and the robot group information.

[0133] The initial division scheme set obtaining unit is configured to generate direct job cooperation constraints and indirect job cooperation constraints according to the job task graph, output the job cooperation constraints as a job cooperation constraint set, and iteratively randomly divide the robot group into a job body and an energy guarantee body in combination with the job cooperation constraint set to obtain the initial division scheme set.

[0134] The division cooperation simulation and evaluation unit is configured to establish a virtual simulation environment based on the job task graph, the job cooperation constraint and the initial division scheme set, perform a division cooperation simulation, and perform cooperation fitness evaluation based on a result of the division cooperation simulation.

[0135] The target execution scheme determination and scheduling unit is configured to select a division scheme with the highest cooperation fitness as a target execution scheme, and issue the target execution scheme to the robot group for job scheduling.

[0136] In some embodiments, the division scheme obtaining module 13 comprises:

[0137] The direct job cooperation constraint generation unit is configured to regularize the task demand information and the robot group information to form direct job cooperation constraints.

[0138] The indirect job cooperation constraint calculation unit is configured to calculate indirect job cooperation constraints based on the job task graph and the direct job cooperation constraints, wherein the indirect job cooperation constraints at least include task guarantee constraints, endurance safety constraints, time window constraints and service capacity constraints.

[0139] The job cooperation constraint set output unit is configured to integrate the direct job cooperation constraint and the indirect job cooperation constraint, and output the job cooperation constraint set.

[0140] In some embodiments, the simulation and evaluation module 14 comprises:

[0141] The virtual simulation environment establishment unit is configured to input the job task graph as a simulation target input parameter, and input the job cooperation constraint set as an optimization restriction condition, to establish the virtual simulation environment.

[0142] The alternative division scheme set configuration unit is configured to configure each initial division scheme according to the job speed interval extracted from the job task graph, to obtain an alternative division scheme set.

[0143] The division strategy simulation and evaluation unit is configured to input the alternative division scheme set into the virtual simulation environment for division strategy preliminary screening and effectiveness evaluation, to correspondingly clean up the alternative division scheme set, and to perform division cooperation simulation on the cleaned alternative division scheme set.

[0144] In some embodiments, the execution steps of the job body in the scheme screening and execution module 15 comprise:

[0145] Monitoring the remaining endurance, when the remaining endurance is lower than a preset threshold α, automatically generating and broadcasting a power supplement request information, the power supplement request information comprising a current position coordinate and a power supplement demand amount. Continuing to execute the current task until the power supplement guarantee body arrives. Docking with the power supplement guarantee body, completing the power supplement process, and sending a power supplement confirmation signal after the power supplement is completed. Continuing the current task execution process, and synchronously monitoring the remaining endurance.

[0146] In some embodiments, the execution steps of the power supplement guarantee body in the scheme screening and execution module 15 comprise:

[0147] Continuously listening to the power supplement request information of the job body. According to the power supplement request information, performing a nearest matching to generate a power supplement task and perform path planning to the target job body. Arriving at the job body according to the path planning result to perform a power supplement operation, wherein the power supplement mode comprises a contact type power supplement or a non-contact type power supplement. After the power supplement is completed, detecting the remaining endurance. When the remaining endurance is lower than a preset threshold, preferentially planning a return path, otherwise, resuming the power supplement request listening state.

[0148] In some implementations, the task demand information at least comprises task area information, job object information and task time limit information, and the robot group information at least comprises the number of robots, intrinsic parameters of the robots and corresponding robot energy consumption models.

[0149] In some implementations, the initial division scheme set includes initial division schemes including configuration numbers and configuration ratios of the work bodies and the energy supply guarantee bodies, wherein:

[0150] The work bodies are configured to perform target work tasks; and the energy supply guarantee bodies are configured with energy supply components to provide energy supply for the work bodies during work.

[0151] In some implementations, the cooperation adaptation degree is obtained by fusing preset multi-dimensional evaluation factors, and the multi-dimensional evaluation factors at least include total energy consumption, total path length, unit path energy consumption, division cooperation cost, and effective work energy consumption ratio.

[0152] It should be understood that the embodiments mentioned in the specification focus on the differences from other embodiments, and the specific embodiments in the first embodiment are also applicable to the division and cooperation system of the intelligent robot in the second embodiment. For the sake of brevity of the specification, no further expansion is made here.

[0153] It should be understood that the embodiments disclosed in the present application and the above description can enable those skilled in the art to implement the present application. Meanwhile, the present application is not limited to the above-mentioned part of the embodiments, and it should be understood that those skilled in the art can still modify the technical solutions recorded in the above embodiments or make equivalent replacement for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for division of labor and cooperation of an intelligent robot, characterized by comprising: The method comprises the following steps: obtaining task demand information and robot group information of a target task, and performing task path planning according to the task demand information and the robot group information; corresponding to the path energy consumption distribution of the task path planning result, and combining the path energy consumption distribution, the task path planning result, the task demand information and the robot group information to construct a task graph; generating direct and indirect task cooperation constraints according to the task graph, and outputting a task cooperation constraint set, and combining the task cooperation constraint set to iteratively randomly divide the robot group into a task body and an energy support body to obtain an initial division scheme set; based on the task graph, the task cooperation constraints and the initial division scheme set, a virtual simulation environment is established, corresponding to the execution of division and cooperation simulation, and based on the division and cooperation simulation result, a cooperation adaptation degree is evaluated; selecting the division scheme with the highest cooperation adaptation degree as the target execution scheme, and issuing it to the robot group for task scheduling; wherein the task demand information at least includes task area information, task object information and task time limit information, and the robot group information at least includes the number of robots, robot intrinsic parameters and corresponding robot energy consumption model; wherein corresponding to the path energy consumption distribution of the task path planning result, and combining the path energy consumption distribution, the task path planning result, the task demand information and the robot group information to construct a task graph, comprises: generating a path point sequence based on the task path planning result; extracting the energy consumption function model of each robot in the robot group information; combining geographic space data, the robot energy consumption model and the task demand information, the energy consumption characteristic value of each path point in the path point sequence is calculated; based on the energy consumption characteristic value, the path energy consumption distribution is obtained, and the path energy consumption distribution is associated with the task path planning result, and the task demand information and the robot group information are combined to construct a task graph; wherein the direct and indirect task cooperation constraints are generated according to the task graph, and output as a task cooperation constraint set, which comprises: normalizing the task demand information and the robot group information to form direct task cooperation constraints; based on the task graph and the direct task cooperation constraints, indirect task cooperation constraints are calculated, wherein the indirect task cooperation constraints at least include: task guarantee constraints, endurance safety constraints, time window constraints and service capacity constraints; integrating the direct task cooperation constraints and the indirect task cooperation constraints to output the task cooperation constraint set; wherein the cooperation adaptation degree is obtained by fusing a plurality of preset evaluation factors, and the plurality of evaluation factors at least include: total energy consumption, total path length, unit path energy consumption, division and cooperation cost, and effective task energy consumption ratio.

2. The method of claim 1, wherein the intelligent robots are divided into a plurality of groups, and the robots in each group are assigned with a specific task. based on the task graph, the task cooperation constraints and the initial division scheme set, a virtual simulation environment is established, corresponding to the execution of division and cooperation simulation, and based on the division and cooperation simulation result, a cooperation adaptation degree is evaluated; Taking the job task graph as a simulation target input parameter and the job cooperation constraint set as an optimization restriction condition, the virtual simulation environment is established; According to the job speed interval extracted from the job task graph, each initial division scheme is differentially configured to obtain a set of candidate division schemes; The set of candidate division schemes is input into the virtual simulation environment for initial screening and effectiveness evaluation of the division strategy, corresponding to cleaning the set of candidate division schemes, and performing division cooperation simulation on the cleaned set of candidate division schemes.

3. The method of claim 1, wherein the intelligent robots are divided into a plurality of groups, and the robots in each group are assigned with a specific task. Select the division scheme with the highest cooperation fitness as the target execution scheme and issue it to the robot group for job scheduling, wherein the execution steps of the job body include: Monitor the remaining endurance, and when it is lower than a preset threshold α, automatically generate and broadcast a power supplement request information, which includes the current position coordinates and the power supplement demand; Continue to execute the current task until the power supplement guarantee body arrives; Interface with the power supplement guarantee body, complete the power supplement process, and send a power supplement confirmation signal after the power supplement is completed; Continue the current task execution process and synchronously monitor the remaining endurance.

4. The method of claim 3, wherein the intelligent robot is a robot vacuum cleaner. Select the division scheme with the highest cooperation fitness as the target execution scheme and issue it to the robot group for job scheduling, wherein the execution steps of the power supplement guarantee body include: Continuously monitor the power supplement request information of the job body; According to the power supplement request information, generate a power supplement task and perform path planning to the target job body; According to the path planning result, arrive at the job body to perform power supplement operation, wherein the power supplement method includes contact type power supplement or non-contact type power supplement; After the power supplement is completed, detect the remaining endurance; When the remaining endurance is lower than a preset threshold, preferentially plan a return path, otherwise, return to the power supplement request monitoring state.

5. The method of claim 1, wherein the intelligent robot is a robot vacuum cleaner. In the set of initial division schemes, the initial division scheme includes the configuration number and configuration ratio of the job body and the power supplement guarantee body, wherein: The job body is used to execute the target job task; The power supplement guarantee body is configured with an energy supply component to provide energy supplement for the job body during the job process.

6. A division of labor system of an intelligent robot, characterized by, A division cooperation method of an intelligent robot for realizing any one of claims 1 to 5, comprising: A job information acquisition module for acquiring task demand information of a target job task and robot group information, and performing job path planning according to the task demand information and the robot group information; A job task graph construction module for corresponding calculation of path energy consumption distribution of the job path planning result, and construction of a job task graph in combination with the path energy consumption distribution, the job path planning result, the task demand information and the robot group information; A division scheme acquisition module for generating direct and indirect job cooperation constraints according to the job task graph, outputting a job cooperation constraint set, and iteratively randomly dividing the robot group into a job body and a power supplement guarantee body to obtain a set of initial division schemes in combination with the job cooperation constraint set; A job information acquisition module for acquiring task demand information of a target job task and robot group information, and performing job path planning according to the task demand information and the robot group information; A job task graph construction module for corresponding calculation of path energy consumption distribution of the job path planning result, and construction of a job task graph in combination with the path energy consumption distribution, the job path planning result, the task demand information and the robot group information; A division scheme acquisition module for generating direct and indirect job cooperation constraints according to the job task graph, outputting a job cooperation constraint set, and iteratively randomly dividing the robot group into a job body and a power supplement guarantee body to obtain a set of initial division schemes in combination with the job cooperation constraint set; The simulation and evaluation module is configured to establish a virtual simulation environment based on the job task graph, the job cooperation constraint and the initial division scheme set, perform a division and cooperation simulation, and evaluate the cooperation adaptability based on the simulation result. The scheme screening and execution module is configured to select a division scheme with the highest cooperation adaptability as a target execution scheme, and send the target execution scheme to the robot group for job scheduling.

Images