1. A method for planning a robot time-series task, the method comprising:

acquiring environment information of a robot working area, and performing discretization modeling on the working area to obtain an environment graph model;

sending the environment graph model to the robot so that the robot constructs a weighted switching system according to the environment graph model and the movement capability and task execution capability of the robot in a working area;

acquiring an individual time sequence task and a global cooperative time sequence task of the robot;

sending the individual time sequence task to the robot;

calculating a cooperative task sequence meeting the global cooperative time sequence task according to the global cooperative time sequence task, wherein the cooperative task sequence is composed of one or more cooperative subtasks;

constructing a cooperative subtask allocation model according to the cooperative task sequence;

calculating the cooperative subtask allocation model to obtain a cooperative subtask allocation result;

and sending the cooperative subtask allocation result to the robot, so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, then fusing the weighted switching system, and finally calculating a task execution plan by the robot individual.

2. The method of claim 1, wherein computing a collaborative task sequence that satisfies the global collaborative timing task based on the global collaborative timing task comprises:

converting the global cooperative timing task into a corresponding non-deterministic finite automaton modelAnd removeSet of all out robotsAn edge of task execution capability;

after removal, searching the non-deterministic finite automaton modelOne reception path ρ in_FAnd extracting a reception path ρ_FAnd taking the corresponding proposition sequence sigma as a cooperative task sequence which needs to be completed by multiple robots.

3. The method of claim 1, wherein constructing a collaborative subtask allocation model based on the collaborative task sequence comprises:

establishing a Boolean variable set for each robot iCollectionThe variables in (1) correspond to the cooperative subtasks to be allocated in the cooperative task sequence one by one;

based on the set of Boolean variablesConstructing a cooperative subtask allocation model;

wherein the constraint of the collaborative subtask allocation model is as follows:

(1) collaborative constraints, wherein the number and types of robots required by collaborative subtasks to be distributed in each collaborative task sequence need to be met;

(2) time constraints, each robot cannot participate in two simultaneously executed tasks at the same time;

(3) and communication constraint, namely for adjacent cooperative subtasks which need to be successively and continuously executed, if the distance between the environment areas corresponding to the adjacent cooperative subtasks which need to be continuously executed is long, so that a communication link cannot be established, requiring that the intersection set of the robot sets to which the adjacent cooperative subtasks are allocated is not empty.

4. The method of claim 1, wherein the robot constructs an updated individual time-series task according to the individual time-series task and the cooperative subtask allocation result, fuses the weighted switching system, and calculates a task execution plan by the robot individual, comprising:

according to the cooperative subtask allocation result, the robot constructs a new individual cooperative task, wherein the individual cooperative task comprises a cooperative subtask allocated to the robot i and a time sequence constraint between the cooperative subtasks;

fusing the individual cooperative task into the individual time sequence task to obtain an updated individual time sequence task;

fusing the updated individual time sequence task with the weighting switching system to construct an individual product automaton;

and according to the individual product automaton, the robot searches a shortest receiving path on the individual product automaton to obtain a task execution plan of the robot.

5. The method of claim 1, further comprising, after the individual computing a task execution plan:

(1) based on the task execution plan, the robot searches for the robot which arrives at the current cooperation subtask at the latest in all the robots participating in the current cooperation subtask through communicationThe robotAdjusting the receiving path of the robot in the individual product automaton to ensure that the robotAt an earlier stageThe time of the task reaches the working area corresponding to the current cooperative subtask, so that the waiting time of other robots is reduced;

(2) if the adjustment in the step (1) can not shorten the total task execution time, the robot searches for the robot which reaches the current cooperation subtask earliest among all the robots participating in the current cooperation subtask through communicationThe robotAdjusting the receiving path of the robot in the individual product automaton to ensure that the robotThe working area corresponding to the current cooperative subtask is reached at a later moment, so that the robot is reducedWaiting time of the robot when the current cooperative subtask is executed, wherein the total task execution time comprises the total time spent by all robots to complete all individual time sequence tasks and the global cooperative time sequence task;

(3) if the total task execution time cannot be reduced after the robot adjusts all the cooperative subtasks, the adjustment is finished; otherwise, circularly traversing all the cooperative subtasks, and carrying out the adjustment processes of (1) and (2).

6. The method of claim 5, wherein the robotAdjusting the receiving path of the self in the individual product automaton, comprising:

the robotSearching the individual product automatonThe cooperative node in (1);

according to the obtained cooperative nodes, for each current cooperative subtask, the robotRandom ergodic individual product automatonThe robot keeps for each node q to be selected in the cooperative nodes belonging to the current cooperative subtaskToUnchanged, slave nodeInitially, find through candidate node q and arriveThe shortest path of the receiving node is used as the candidate path corresponding to the candidate node qWhereinRepresenting the shortest receive path searched on the individual product automaton,to representThe cooperative node corresponding to a cooperative subtask previous to the current cooperative subtask,to representThe first node of (a);

for the candidate path corresponding to each candidate node qThe robotConsider whether it satisfies the following two conditions: (1)whether to reduce latency at the cooperative subtask; (2)whether to reduce the total task execution time; candidate route meeting the two conditionsWill be updated to new

7. The method of claim 1, after the robot individual calculates the task execution plan, further comprising:

negating the cooperative subtask allocation result, adding the negation into a constraint of the cooperative subtask allocation model, and solving to obtain a next cooperative subtask allocation result;

and sending the next cooperative subtask allocation result to the robot, so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the next cooperative subtask allocation result, and then fusing the weighted switching system to calculate a task execution plan by the robot individual.

8. The method of claim 7, after solving for a next cooperative subtask assignment, further comprising:

screening the next cooperative subtask allocation result, and sending the screened next cooperative subtask allocation result to the robot;

wherein the screening is specifically: if it isSo thatThe current cooperative subtask allocation result can be directly negated, and then added into the constraint of the cooperative subtask allocation model, and the next cooperative subtask allocation result is calculated;

wherein the aboveThe result of the cooperative subtask allocation obtained in the mth iteration calculation and the nth iteration calculation is respectively obtained, whereinMeans thatEach element in the composition has a value greater thanMiddle pairThe corresponding element value.

9. A robot time series mission planning apparatus, the method comprising:

the first acquisition module is used for acquiring environment information of a robot work area;

the modeling module is used for carrying out discretization modeling on the working area to obtain an environment graph model;

the first sending module is used for sending the environment graph model to the robot so that the robot can construct a weighted switching system according to the environment graph model and the movement and task execution capacity of the robot in a working area;

the second acquisition module is used for acquiring an individual time sequence task and a global cooperative time sequence task of the robot;

the second sending module is used for sending the individual time sequence task to the robot;

the first calculation module is used for calculating a cooperative task sequence meeting the global cooperative time sequence task according to the global cooperative time sequence task, wherein the cooperative task sequence is composed of one or more cooperative subtasks;

the construction module is used for constructing a cooperative subtask distribution model according to the cooperative task sequence;

the second calculation module is used for calculating the cooperative subtask allocation model to obtain a cooperative subtask allocation result;

and the third sending module is used for sending the cooperative subtask allocation result to the robot so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, then the weighted switching system is fused, and finally the individual calculates a task execution plan.

10. An electronic device, comprising:

one or more processors;

a memory for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-8.

Background

The task planning problem widely exists in social life, such as the fields of data acquisition, continuous observation, intelligent logistics and the like. In recent years, more and more researches are focused on the task planning problem of the robot under the constraint of complex time sequences, so as to solve the more complex task requirements in real production. The task planning method based on the model checking theory can describe more complex task timing constraints such as task sequence constraints, safety obstacle avoidance constraints, response constraints and the like by means of a formal language such as a Linear Temporal Logic (LTL) language, so that the method is widely concerned and researched.

The prior method is difficult to effectively treat the following technical problems: (1) the problem that the robot has an individual time sequence task and a global cooperative time sequence task at the same time is difficult to effectively solve; (2) unnecessary waiting time caused by different arrival times when the optimization multi-robot executes the cooperative task is not considered; (3) most of the existing methods assume that the collaborative task allocation relation of the robot is known, and the privacy information of the individual time sequence tasks of the robot is not considered to be protected.

Disclosure of Invention

The embodiment of the invention aims to provide a robot time sequence task planning method and device and electronic equipment, so as to solve the problems that in the related technology, a robot has an individual time sequence task and a global cooperative time sequence task at the same time, and consider the technical problems of protecting the privacy of the individual time sequence task of the robot in the task planning process.

According to a first aspect of embodiments herein, there is provided a method comprising: acquiring environmental information of a robot work area; carrying out discretization modeling on the working area to obtain an environment graph model; sending the environment graph model to the robot so that the robot constructs a weighted switching system according to the environment graph model and the movement capability and task execution capability of the robot in a working area; acquiring an individual time sequence task and a global cooperative time sequence task of the robot; sending the individual time sequence task to the robot; calculating a cooperative task sequence meeting the global cooperative time sequence task according to the global cooperative time sequence task, wherein the cooperative task sequence is composed of one or more cooperative subtasks; constructing a cooperative subtask allocation model according to the cooperative task sequence; calculating the cooperative subtask allocation model to obtain a cooperative subtask allocation result; and sending the cooperative subtask allocation result to the robot, so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, then fusing the weighted switching system, and finally calculating a task execution plan by the robot individual.

According to a second aspect of embodiments of the present application, there is provided an apparatus comprising: the first acquisition module is used for acquiring environment information of a robot work area; the modeling module is used for carrying out discretization modeling on the working area to obtain an environment graph model; the first sending module is used for sending the environment graph model to the robot so that the robot can construct a weighted switching system according to the environment graph model and the movement and task execution capacity of the robot in a working area; the second acquisition module is used for acquiring an individual time sequence task and a global cooperative time sequence task of the robot; the second sending module is used for sending the individual time sequence task to the robot; the first calculation module is used for calculating a cooperative task sequence meeting the global cooperative time sequence task according to the global cooperative time sequence task, wherein the cooperative task sequence is composed of one or more cooperative subtasks; the construction module is used for constructing a cooperative subtask distribution model according to the cooperative task sequence; the second calculation module is used for calculating the cooperative subtask allocation model to obtain a cooperative subtask allocation result; and the third sending module is used for sending the cooperative subtask allocation result to the robot so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, then the weighted switching system is fused, and finally the robot individually calculates a task execution plan.

According to a third aspect of embodiments of the present application, there is provided an electronic apparatus, including: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the steps of the robot sequence task planning method of any of the first aspects.

The technical scheme provided by the embodiment of the application can have the following beneficial effects:

by adopting a mixed task planning framework combining centralized type and distributed type, the problem that the existing method is difficult to effectively process that the robot has an individual time sequence task and a collaborative time sequence task at the same time is solved, the deployment scene of multiple robots is expanded, and the efficiency is improved; the task execution plan adjusting strategy based on the automaton model is provided, so that unnecessary waiting time when the robot actually executes the collaborative sequential task is shortened, and the time efficiency of completing the task is improved; the individual planning is adopted to independently solve the task execution plan of the individual robot, the limitation of higher complexity of the existing method is overcome, and the individual time sequence task privacy information of the robot is effectively protected while the calculated amount is dispersed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.

FIG. 1 is a flow chart illustrating a method of robot time-series mission planning in accordance with an exemplary embodiment.

FIG. 2 illustrates a solution to a time-series mission plan in a simulation experiment involving 3 robots at a time, according to an exemplary embodiment.

FIG. 3 illustrates a comparison of optimization quality versus solution time for the task execution plan adjustment process at different robot counts and work area scales, according to an exemplary embodiment.

Fig. 4 illustrates an optimization effect of a task execution plan adjustment process on each cooperative subtask allocation result and a screening efficiency of a screening strategy on the cooperative subtask allocation result in a simulation experiment including 30 robots according to an exemplary embodiment.

FIG. 5 illustrates a schematic diagram of a robot time-series mission planning apparatus, according to an exemplary embodiment.

Detailed Description

In order to explain the technical means of the present invention, the following description will be given by way of specific examples.

Example one

Fig. 1 is a flowchart of a robot time-series task planning method according to an embodiment of the present invention. Referring to fig. 1, the robot time-series mission planning method may include the steps of:

step S01: the method comprises the steps of obtaining environment information of a robot working area, carrying out discretization modeling on the working area, and obtaining an environment graph model.

Specifically, the environment information of the work area includes an initial position of the robot in the work area, a position of a task to be completed in the work area, and the like. The present application does not limit the specific manner of acquiring the environmental information. It is noted that the terminal needs to establish a communication connection with the robot before starting planning.

Specifically, the environment map model is defined asWhereinIs a node set corresponding to a task area in a work area; ε isA collection of connection relationships between all nodes in the set. Robot setIn-environment graph modelIn autonomous operation, each robotHaving task execution capability c_jE to Cap and can provide corresponding functional service in the working area, wherein Cap: ═ c_j}_{j∈*1,...,|Cap|}}Is all robot tasksA set of execution capabilities. Modeling distributed tasks in a work area as a tupleWherein: pi^tsIs a unique atomic proposition corresponding to task ts;is the environment area corresponding to the task ts;indicating to complete the task ts (or to make an atom proposition π^tsTrue), need to be satisfiedAt least is provided withEach having task execution capability c_jIs cooperatively accomplished, hereA subscript set of robot task performance capabilities required to complete task ts. The task performance capabilities may include, but are not limited to, technical action performance capabilities such as grab, move, operate, and the like.

Step S02: and sending the environment graph model to the robot so that the robot constructs a weighted switching system according to the environment graph model and the movement and task execution capacity of the robot in the working area.

In particular, the task performance capabilities include, but are not limited to, technical actions such as grabbing, moving, manipulating, and the like. The weighted switching system of robot i is represented asWherein the content of the first and second substances,is a set of nodes, each node and workerMaking task areas in the areas correspond to each other;represents the initial position of robot i;is a node pair set, wherein a feasible path meeting the kinematic constraint of the robot exists between two nodes contained in each node pair;is a weight function of the node transfer;is a collection of atomic propositions, one corresponding to the completion of a certain task in the work area;is a label function, and the label of the node corresponds to the task execution capacity of the robot in the task area corresponding to the node.

Step S03: and acquiring an individual time sequence task and a global cooperative time sequence task of the robot.

Specifically, each robot is set according to the user task requirementsThe individual time series task LTL expression isWhereinConsists of atomic propositions and sequential logical operators corresponding to individual sequential tasks,all atomic propositions contained in|I_ts|＝1,And the robotHere, theIndicating all have task execution capability c_jA set of robots of (1); in addition, the expression of the global cooperative time sequence task LTL which needs to be completed by all robots in a cooperative mode is set as phi, and all atomic propositions contained in the phi form a set The task ct has larger task amount and more complex function requirements, and needs multiple robots with different task execution capabilities to complete cooperatively. For example, the individual time series task LTL expression of the robot is set as Setting the LTL expression of the global cooperative time sequence task as WhereinAndrespectively representing individual time-sequential tasks ts₁And a collaborative timing task ct₁The corresponding atom propositions and other atom propositions are defined in the same way. In addition, the time-series task ct is coordinated₁、ct₂、ct₃、ct₄Respectively requiring 1, 3, 2 and 2 robots to participate in the completion. It should be understood that the present application does not limit the specific manner of acquiring the individual timing tasks and the global coordinated timing tasks of the robot.

Step S04: and sending the individual time sequence task to the robot.

The specific transmission mode for transmitting the individual time sequence task to the robot is not limited in the application.

Step S05: and calculating a cooperative task sequence meeting the global cooperative time sequence task according to the global cooperative time sequence task, wherein the cooperative task sequence is composed of one or more cooperative subtasks.

In an example, the step S05 may specifically include:

step S51: converting the global cooperative time sequence task LTL expression phi into a corresponding non-deterministic finite automaton model(Nonderteristic finish Automation, NFA) according toNode transfer condition in (1), removeSet of all out robotsThe edge of the capability range. In particular, traverse surveyFor each branch condition, extracting the minimum positive atom subject set satisfying the branch conditionJudging whether the existing robot set can simultaneously make the atom titles in the minimum positive atom title set true, if so, keeping the corresponding edges; else, removing the edge from the non-deterministic finite automaton modelIs removed.

Step S52: searching the robot modelOne reception path ρ in_FReception path ρ_FSatisfying the automaton modelAn initial node ofStarting and ending at a receiving nodeAccording to the reception path ρ_FObtaining a proposition sequence sigma (1) sigma (2) … sigma (L-1) sigma (L) by the transfer condition of the adjacent nodes, and satisfying the proposition sequence sigma (1) sigma (2) … sigma (L-1) sigma (L)That is, σ (i) is a value satisfying the node transition condition δ (ρ)_F(i),ρ_F(i + 1)). The proposition sequence sigma is used as a cooperative task sequence which needs to be completed by multiple robots.

In addition, in implementations, the path may be based on the receive path ρ_FThe corresponding cooperative task sequence sigma is divided into S independent subtask sequences sigma by the decomposition node¹；σ²；…；σ^S. The independent asynchronous execution of these sub-task sequences may ensure that global cooperative timing task requirements may be met without executing in order in the cooperative task sequence σ. Wherein the reception path ρ is judged_FThe method for judging whether the node is a decomposition node is as follows: adjustment is taken asThe execution sequence of two proposition sequences divided by the front node is checked to see whether the proposition sequence after the execution sequence is adjusted can be inIn which a receive path is generated (i.e. starting from the robot model)At the initial node of the robot model, ending at the robot modelThe receiving node) of (1), if yes, the current node is a decomposition node; otherwise it is not a decomposition node.

Step S06: and constructing a cooperative subtask allocation model according to the cooperative task sequence.

In an example, the step S06 may specifically include:

step S61: definition ofσ^k(m) is the kth subtask sequence σ^kThe m-th element of (1), whereinIs σ^kThe jth element of (m). For each subtaskDefining a new symbolWherein Definition ofIs a set of boolean variables for robot i,true, indicating that robot i is assigned to perform a cooperative subtask

Step S62: constructing the cooperative subtask allocation model based on SMT (Satiscapability Module tools) based on the Boolean variable set, wherein the cooperative subtask allocation model satisfies the following constraints:

(1) and (4) collaboratively constraining. The number and type of robots required for each subtask needs to be met. For each sigma^kRequireIs satisfied. Where 1(·) is a label function, 1(true) is 1, and 1(false) is 0.

(2) And (4) time constraint. Each robot cannot participate simultaneously in two tasks that need to be performed simultaneously. I.e. for each sigma^k，Is satisfied.

(3) And (4) communication constraint. For some adjacent subtasks that need to be executed successively, if the environmental regions corresponding to the subtasks are far apart from each other, so that a communication link cannot be established, the communication limitation can be overcome by the following constraints. For each sigma^k，Needs to be satisfied wherein

And step S07, calculating the cooperative subtask allocation model to obtain a cooperative subtask allocation result.

Specifically, solving cooperative subtask allocation knots using a z3 SMT solverFruitAnd sent to the robot.

In practical implementation, if the solution time is sufficient, all feasible cooperative subtask allocation results can be iteratively traversed, and finally the allocation result with the highest time efficiency is returned. In a specific embodiment, each time the z3 solver is called, the result of the last solution is used as the basisDefining expressionsAnd updates f to

In addition, in the process of iteratively traversing all feasible cooperative subtask allocation results, the following screening strategy can be used for rapidly screening non-optimal feasible solutions, so that invalid calculation is avoided: if it is So thatF can be directly updated toAnd the cooperative subtask allocation result corresponding to the current feasible solution does not need to be sent to the robotAnd respectively obtaining feasible solutions in the mth iterative computation and the nth iterative computation. Here, theMeans that

And step S08, sending the cooperative subtask allocation result to the robot, so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, then fusing the weighted switching system, and finally calculating a task execution plan by the robot individual.

In an example, the step S08 may specifically include:

step S81: the robot constructs a new LTL expression phi according to the cooperative subtask allocation result_iTo represent the cooperative subtasks assigned to robot i and the timing constraints between them. The updated individual timing task LTL expression isThe robot establishes a set And isWherein the content of the first and second substances,andall propositions in (corresponding to tasks in the work area) are sorted in order of increasing superscript k and l. Specifically, construction of phi_iThe method comprises the following steps:

(1) order constraints in each subtask sequence are modeled. For eachInitializationIs composed ofWhereinIs composed ofThe first element in (1). Then useTo iteratively replaceIn (1)WhereinAndis thatTwo adjacent elements in (1), and_m<l_m+1。

(2) defining the execution order between the subtask sequences. In order to avoid the deadlock problem caused by different execution sequences when a plurality of robots execute different subtask sequences, the execution sequence between the subtask sequences is defined as the sequence in the original collaborative task sequence sigma. For each robot i, the results obtained in (1) are comparedIn ascending order according to superscript k. From k 1 to S-1, withTo iteratively replaceWhereinIs thatThe last element in (1). In the end of this process,

step S82: for all robots, switching system wTS according to the weighting of each robot i_iAnd updated individual timing task LTL expressionsDefining an individual product automaton asWhereinIs composed ofThe corresponding NFA.Wherein:satisfy the requirement of(Π_iq_P,Π_iq′_P)∈→_iAnd isII therein_iAnd pi_FAre respectively shown as wTS_iAndmapping of the node space of (a); omega_P(q_P,q′_P)＝ω_i(Π_iq_P,Π_iq′_P)；

Step S83: the robot adopts Dijkstra algorithm and the individual product automatonSearching a shortest receiving path rho_i，ρ_iCan satisfy LTL expression at the same timeAnd limits on the mobility and task performance capabilities of the robot in the work area (the constraints are contained in wTS_iIn (1). The task execution plan of the robot i can be set by tau_i＝Π_iρ_iThus obtaining the product.

In addition, the robot may further adjust the task execution plan, so as to reduce unnecessary waiting time (caused by inconsistent arrival time of the robot at the corresponding area of the cooperative task) when the robot executes the cooperative task, specifically including:

the robot searches for cooperative nodes in the individual product automaton. For the If and only if: (1)(2)and isWherein the content of the first and second substances,is an automatic machineMiddle taskA corresponding set of the cooperative nodes.

For each cooperative subtask in the cooperative task sequence σ, a specific adjustment process is as follows:

(1) in all participating in the cooperative subtasksRobot set ofIn the selection of the latest arrivalRobot (2)In-person product automatonIn adjusting the robotSo that the robotArriving at the cooperative subtask at an earlier timeCorresponding work area, thereby reducing aggregationThe waiting time of other robots.

(2) If the adjustment in (1) cannot shorten the total task execution time, the process is startedIn search for earliest arrivalRobot (2)And in an individual product automatonIn adjusting the robotSo that the robotArriving at a cooperative subtask at a later timeCorresponding working area, thereby reducing the robotIs performing task by itselfThe latency of the clock.

In the above steps (1) and (2), the adjustment robotThe receiving path mode of (1) is specifically as follows: the robotRandom ergodic individual product automatonIn (A) belong toFor each node to be selectedRobot holdToUnchanged, slave nodeInitially, the search goes through node q and arrivesThe shortest path of the receiving node. Wherein the content of the first and second substances,indicating a reception pathUpper, nodeLast cooperative node of j^prevFor the cooperative node inThe subscripts in (1) are, for the same reason,as a cooperative subtaskIn a cooperative node ofThe subscript of (1). For each node to be selectedThe corresponding candidate pathConsider whether it satisfies the following two conditions: (1)whether or not to reduce cooperative subtasksA waiting time of (c); (2)whether the total time to complete the task is reduced. Candidate route meeting the two conditionsWill be updated to newThe robot then proceeds to perform the above adjustment process for the next collaboration subtask.

And traversing all the cooperative subtasks in sigma by the robot, adjusting the execution plan, and checking whether the total time for completing the tasks is reduced. If the number cannot be reduced, the adjustment process is ended; otherwise, the adjusting process is repeatedly executed.

Fig. 2 is a solution result of a time-series task plan in a simulation experiment including 3 robots at a time according to an exemplary embodiment, and it can be seen that the task plan of the robots simultaneously satisfies an individual time-series task and a collaborative time-series task. In addition, the robot reduces the waiting time when the cooperative subtasks are executed through the adjusting process, and the correctness and the effectiveness of the method are verified.

Fig. 3 shows the effect of the task execution plan adjustment process on the solution efficiency and quality when multiple simulation experiments are performed in different numbers of robots and in different environmental scales, wherein each data point in the diagram is obtained by calculating the mean value and standard deviation of 10 random experiments, the rest parameters are kept unchanged in each experiment, and the regional distribution of the tasks in the working area is only changed randomly. Setting the work areas of the grid map into three scales of 20 × 20, 30 × 30 and 40 × 40, and considering the number of various robots under each environment scale, wherein the specific number of robots comprises {5,10,15,20,25 and 30 }. Setting the LTL expression of the individual time sequence task of the robot asRobot global collaborative sequential task LTL expressions such as

As can be seen from fig. 3, the task execution plan adjustment process consumes about 45% to 55% of the initial solution time (fig. 3 (a)), so that the task completion time is reduced to about 70% to 80% of the time when the adjustment process is not performed (fig. 3 (b)), thereby effectively improving the efficiency of multi-robot task cooperation. Meanwhile, the conclusion is also established under different robot numbers and environmental scales, and the applicability and the effectiveness of the task execution plan adjusting process are verified.

Fig. 4 (a) is a comparison result of task completion times of the robots performing the task execution plan adjustment process and not performing the task execution plan adjustment process under different cooperative subtask allocation results in the embodiment including 30 robots at a time. It can be seen that, for each cooperative subtask allocation result, the task execution plan adjustment process can effectively reduce the execution time loss of the task execution plan, and verify the actual optimization effect of the task execution plan adjustment process. Fig. 4 (b) shows the number of the cooperative subtask allocation results that are sieved in each round of the fast sieving policy, the total number of the cooperative subtask allocation results that need to be recorded, and the total number of the sieved cooperative subtask allocation results, which shows that the sieving policy can effectively filter the non-optimal cooperative subtask allocation results, thereby reducing the amount of computation.

Example two

Fig. 5 is a schematic diagram of a robot timing task planning apparatus according to an embodiment of the present invention. The robot time sequence task planning apparatus provided in this embodiment and the robot time sequence task planning method belong to the same inventive concept, and specifically, as shown in fig. 5, the robot time sequence task planning apparatus 200 includes: a first obtaining module 201, configured to obtain environment information of a robot work area; the modeling module 202 is used for performing discretization modeling on the working area to obtain an environment graph model; the first sending module 203 is configured to send the environment map model to the robot, so that the robot constructs a weighted switching system according to the environment map model and the movement and task execution capacity of the robot in the working area; a second obtaining module 204, configured to obtain an individual time sequence task and a global collaborative time sequence task of the robot; a second sending module 205, configured to send the individual time-series task to the robot; a first calculating module 206, configured to calculate, according to the global cooperative timing task, a cooperative task sequence that satisfies the global cooperative timing task, where the cooperative task sequence is composed of one or more cooperative subtasks; a constructing module 207, configured to construct a cooperative subtask allocation model according to the cooperative task sequence; the second calculation module 208 is configured to calculate the cooperative subtask allocation model to obtain a cooperative subtask allocation result; a third sending module 209, configured to send the cooperative subtask allocation result to the robot, so that the robot constructs an updated individual time sequence task according to the individual time sequence task and the cooperative subtask allocation result, and then merges the weighted switching system to calculate a task execution plan in an individual.

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.

Correspondingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when executed by the one or more processors, cause the one or more processors to implement a robot time-series mission planning method as described above.

Accordingly, the present application also provides a computer readable storage medium, on which computer instructions are stored, wherein the instructions, when executed by a processor, implement the robot time-series task planning method as described above.

It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.