A fast task planning method and system under spatial robot arm motion constraint

Abstract

The application discloses a kind of quick task planning method and system under the movement constraint of space mechanical arm, continuous variable solving is divided into short-time continuous variable solving and long-time continuous variable solving, short-time continuous variable solving is carried out first, and long-time continuous variable solving is carried out using action sequence planning result, can reduce state search space, reduce the time of action sequence planning, while guaranteeing the correctness and executability of action sequence to some extent, reduce the possibility of re-planning, improve the overall efficiency of planning;According to the action sequence obtained by planning, long-time continuous variable solving can reduce the number of long-time continuous variable solving, improve the overall efficiency of planning.Compared with prior art, the application not only can ensure the executability of action sequence obtained by planning, but also has higher planning efficiency.

Description

Technical Field

[0001] This invention belongs to the field of space robotic arm technology, and relates to a rapid task planning method and system for space robotic arms under motion constraints. Background Technology

[0002] Mission planning for a space robotic arm refers to the planning of its motion sequence based on the requirements of on-orbit operations. By executing the planned motion sequence, the space robotic arm can complete the given task. The motion sequence obtained through mission planning enables the space robotic arm to achieve the desired mission objectives while meeting constraints. A mature and reliable mission planning system not only ensures the smooth execution of on-orbit operations but also improves mission efficiency and enhances mission completion quality in terms of time and other indicators. This is of great significance for the long-term operation of spacecraft in the resource-scarce environment of space.

[0003] The core issue of task planning lies in how to apply intelligent planning methods to plan the sequence of actions given task information, guiding the robotic arm to complete the task. Traditional robotic arm task planning assumes that each action in the sequence has a motion planning solution. Under this assumption, task planning and motion planning are independent and performed separately. However, this assumption is practically nonexistent. Due to the lack of motion information, the task planning process often ignores some motion constraints, such as obstacle constraints and kinematic constraints. These constraints may render the action sequence unexecutable, leading to the task's failure. Therefore, to ensure the executability of the action sequence and enable the robotic arm to successfully complete the task, motion constraints need to be introduced into task planning, i.e., task planning under motion constraints.

[0004] For task planning problems under motion constraints, researchers have extended the representation of classical planning problems based on the solution approach of joint task and motion planning, calling it the Task and Motion Planning (TAMP) problem. Compared to classical planning representations, TAMP extends the description of continuous variables, enabling the representation of motion constraints in robotic arm systems. Since classical planning methods can only solve discrete planning problems and cannot plan systems containing continuous variables, a planning method specifically for TAMP problems is needed. To address this, Dornhege proposed the concept of semantic attachments, extending the Planning Domain Definition Language (PDDL) to enable it to call external kinematics modules. This allows for motion planning during the action sequence search process and determination of action feasibility, ultimately achieving task planning under motion constraints. While semantic attachments guarantee that the planned action sequence satisfies the motion constraints, the motion planning during the search process consumes significant computation time. Furthermore, it still operates within a finite domain, failing to address planning problems in infinite domains of continuous space. Building on Dornhege's research, and addressing the issues of poor hardware conditions in space, limited onboard computer storage, and insufficient computing power that could impact mission planning, Coles extended the classic planner to enable it to process temporal information with lower memory consumption, achieving lightweight processing of temporal mission planning. However, it still suffers from long computation times.

[0005] Task planning problems under motion constraints can be divided into two parts: solving for continuous variables and planning action sequences, such as... Figure 1As shown. In response, Garrett proposed the PDDLStream representation method based on the PDDL representation method. By representing the solution of continuous variables as a Stream, the planning problem can be extended based on the results of the continuous variable solution, thus enabling task planning under motion constraints. Based on the PDDLStream representation method, Garrett proposed incremental and optimistic algorithms. The incremental algorithm is a planning method that first solves the continuous variables and then plans the action sequence. By solving the continuous variables, it obtains the continuous variable objects and predicates that satisfy the motion constraints and extends the planning problem to further plan the action sequence that satisfies the motion constraints. Although the incremental algorithm can achieve task planning under motion constraints, due to the lack of task information guidance, its continuous variable solution process has a certain degree of randomness and may perform multiple continuous variable solutions unrelated to the action sequence. Therefore, the solution efficiency of this method decreases as the problem complexity increases. Unlike incremental algorithms, optimistic algorithms are a planning method that first plans the action sequence and then solves the continuous variables. It only solves the continuous variables within the planned action sequence, which can reduce the number of times the continuous variables are solved and improve the efficiency of solving the continuous variables. However, since it may perform multiple action sequence planning, its action sequence planning time is relatively long.

[0006] Space robotic arm systems face two types of motion constraints: kinematic constraints and obstacle avoidance trajectory constraints. To ensure successful mission completion, task planning under these constraints is essential. Existing methods for task planning under motion constraints in space robotic arms can obtain action sequences that satisfy the constraints, but they struggle to guarantee rapid planning. The efficiency of task planning directly impacts the efficiency of the space robotic arm in completing on-orbit operations. Therefore, a rapid task planning method under motion constraints has significant theoretical and practical engineering value. Summary of the Invention

[0007] The purpose of this invention is to solve the problems in the prior art and provide a rapid task planning method and system for space robotic arms under motion constraints.

[0008] To achieve the above objectives, the present invention employs the following technical solution:

[0009] In a first aspect, the present invention provides a rapid task planning method for a space robotic arm under motion constraints, comprising the following steps:

[0010] Construct a task planning model for a spatial robotic arm under motion constraints, including the planning domain, planning problem, and Stream module;

[0011] Based on the initial state of the planning problem and the input parameters of the Stream module, determine which Stream modules can be called;

[0012] Based on the short-term Stream instance invoked, generate a robotic arm configuration object and predicates that satisfy kinematic constraints; based on the long-term Stream instance invoked, generate an optimistic object and predicates.

[0013] Add the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem to obtain the extended planning problem;

[0014] Using the planning domain and the extended planning problem as input, the FastDownward planner is used to plan the action sequence, resulting in an action sequence that includes optimistic objects.

[0015] Based on the action sequence obtained from the plan, backtrack the optimistic objects to obtain a long-term Stream instance set that generates all optimistic objects;

[0016] The Stream instances in the long-lived Stream instance collection are called sequentially, and the corresponding optimistic objects in the successfully called Stream instances are assigned values ​​until all optimistic objects are assigned values. Then the action sequence is output to complete the planning.

[0017] Secondly, the present invention provides a rapid task planning system under motion constraints for a space robotic arm, comprising:

[0018] The model building module is used to build a task planning model under the motion constraints of a space robotic arm, including the planning domain, planning problem, and Stream module;

[0019] The first calculation module is used to determine which Stream module can be called based on the initial state of the planning problem and the input parameters of the Stream module.

[0020] The object and predicate generation module is used to generate robotic arm configuration objects and predicates that satisfy kinematic constraints based on the short-term Stream instance being called; and to generate optimistic objects and predicates based on the long-term Stream instance being called.

[0021] The second calculation module is used to add the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem to obtain the extended planning problem.

[0022] The sequence planning module is used to perform action sequence planning using the FastDownward planner, taking the planning domain and the extended planning problem as input, to obtain an action sequence that includes optimistic objects.

[0023] The object backtracking module is used to backtrack the optimistic objects in the action sequence obtained from the plan, and obtain a long-term Stream instance collection that generates all optimistic objects;

[0024] The object assignment module is used to sequentially call the Stream instances in the long-term Stream instance collection, assign values ​​to the corresponding optimistic objects in the successfully called Stream instances, until all optimistic objects have been assigned values, and then output the action sequence to complete the planning.

[0025] Thirdly, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0026] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described above.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] This invention improves planning efficiency while obtaining motion sequences that satisfy motion constraints, achieving rapid task planning under motion constraints for a spatial robotic arm. This invention divides the solution of continuous variables into short-time continuous variable solution and long-time continuous variable solution. By first solving the short-time continuous variables and then using the motion sequence planning results to solve the long-time continuous variables, the state search space can be reduced, the motion sequence planning time can be decreased, and the correctness and executability of the motion sequence can be guaranteed to a certain extent, reducing the possibility of replanning and improving the overall planning efficiency. Solving the long-time continuous variables based on the planned motion sequence can reduce the number of long-time continuous variable solutions, improving the overall planning efficiency. Compared with existing methods, this invention not only ensures the executability of the planned motion sequence but also has higher planning efficiency. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A schematic diagram illustrating the solution process for the task planning problem under motion constraints.

[0031] Figure 2This is a flowchart of the method of the present invention.

[0032] Figure 3 This is a schematic diagram of the system of the present invention.

[0033] Figure 4 This is a flowchart of the rapid task planning method for the spatial robotic arm under motion constraints according to the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0035] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0036] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0037] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0038] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0039] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0040] The present invention will now be described in further detail with reference to the accompanying drawings:

[0041] See Figure 2 This invention discloses a rapid task planning method for a space robotic arm under motion constraints, comprising the following steps:

[0042] S1 constructs a task planning model under the motion constraints of a space robotic arm, including a planning domain, a planning problem, and a Stream module; the construction of the task planning model under the motion constraints of the space robotic arm includes:

[0043] Step 1-1: Construction of the Task Planning Domain

[0044] Based on the model characteristics and task features of the space robotic arm, the objects, predicates and actions in the planning system are defined using the PDDL representation method, thus completing the construction of the planning domain;

[0045] Step 1-2: Task Planning Problem Description

[0046] For a given on-orbit operation task, the initial state and target state of the planning problem are determined based on the objects and predicates in the planning domain and described using the PDDL representation method;

[0047] Steps 1-3: Stream representation for solving continuous variables

[0048] Based on the kinematic model and motion planning algorithm of the space robot, the PDDLStream representation method is used to establish the Stream representation form of the inverse kinematics solution and obstacle avoidance motion planning of the space robot.

[0049] S2 determines the callable Stream modules based on the initial state of the planning problem and the input parameters of the Stream modules; the determination of the callable Stream modules is performed using the following method:

[0050] Determine the callable Stream modules and generate short-term Stream instances and long-term Stream instances; the short-term Stream instance is an inverse kinematics solution Stream instance, and the long-term Stream instance is an obstacle avoidance motion planning Stream instance.

[0051] S3 generates a robotic arm configuration object and predicate that satisfy kinematic constraints based on the called short-term Stream instance; it also generates an optimistic object and predicate based on the called long-term Stream instance; the optimistic object is a new object without specific values ​​generated by calling the long-term Stream instance without solving for continuous variables.

[0052] S4 adds the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem, resulting in the extended planning problem;

[0053] S5 takes the planning domain and the extended planning problem as input, and uses FastDownward to perform action sequence planning, resulting in an action sequence that includes optimistic objects.

[0054] S6 backtracks the optimistic objects in the action sequence obtained from the plan to obtain a long-term Stream instance set that generates all optimistic objects;

[0055] S7 sequentially calls the Stream instances in the long-term Stream instance collection, assigns values ​​to the corresponding optimistic objects in the successfully called Stream instances, until all optimistic objects have been assigned values, and then outputs the action sequence to complete the planning.

[0056] The process of sequentially calling Stream instances in the long-lived Stream instance collection and assigning values ​​to the corresponding optimistic objects in the successfully called Stream instances includes:

[0057] The Stream instances in the backtracked Stream instance set are called sequentially. If the Stream instance fails to generate a new object and predicate, i.e. the continuous variable solution fails, the Stream module that can be called is returned. If the Stream instance is called successfully, the optimistic object corresponding to the continuous variable solution result is assigned a value.

[0058] like Figure 3 As shown, this embodiment of the invention discloses a rapid task planning system under motion constraints for a space robotic arm, comprising:

[0059] The model building module is used to build a task planning model under the motion constraints of a space robotic arm, including the planning domain, planning problem, and Stream module;

[0060] The first calculation module is used to determine which Stream module can be called based on the initial state of the planning problem and the input parameters of the Stream module.

[0061] The object and predicate generation module is used to generate robotic arm configuration objects and predicates that satisfy kinematic constraints based on the short-term Stream instance being called; and to generate optimistic objects and predicates based on the long-term Stream instance being called.

[0062] The second calculation module is used to add the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem to obtain the extended planning problem.

[0063] The sequence planning module is used to perform action sequence planning using classical planning methods, taking the planning domain and the extended planning problem as input, to obtain an action sequence that includes optimistic objects.

[0064] The object backtracking module is used to backtrack the optimistic objects in the action sequence obtained from the plan, and obtain a long-term Stream instance collection that generates all optimistic objects;

[0065] The object assignment module is used to sequentially call the Stream instances in the long-term Stream instance collection, assign values ​​to the corresponding optimistic objects in the successfully called Stream instances, until all optimistic objects have been assigned values, and then output the action sequence to complete the planning.

[0066] Example

[0067] This embodiment provides a rapid task planning method under motion constraints of a space robotic arm. Before planning, it is necessary to establish a task planning model under motion constraints of the space robotic arm, which mainly includes:

[0068] (1) Construction of the task planning domain

[0069] Based on the model characteristics and task features of the space robotic arm, the objects, predicates, and actions in the planning system are defined using the PDDL representation method, thus completing the construction of the planning domain.

[0070] (2) Task planning problem description

[0071] For a given on-orbit operation task, the initial state and target state of the planning problem are determined based on the objects and predicates in the planning domain and described using the PDDL representation method.

[0072] (3) Stream representation for solving continuous variables

[0073] Based on the kinematic model and motion planning algorithm of the space robot, the PDDLStream representation method is used to establish the Stream representation form of the inverse kinematics solution and obstacle avoidance motion planning of the space robot.

[0074] Through the above modeling process, given a space robotic arm system and on-orbit operation tasks, the planning domain, planning problem, and Stream module of the planning system can be obtained. This invention uses these three parts as input, and divides the Stream into long-time Stream and short-time Stream based on the solution time of continuous variables. Then, it employs methods such as solving the continuous variables first and then planning the action sequence, and planning the action sequence first and then solving the continuous variables, respectively, to achieve rapid task planning under the motion constraints of the space robotic arm. The planning flowchart is shown below. Figure 4 As shown, the specific planning process is as follows:

[0075] Step 1: Based on the initial state in the task planning problem description and the input parameters in the Stream module, determine the callable Stream modules and generate the corresponding Stream instances. Define the inverse kinematics solution Stream instance as a short-time Stream instance and the obstacle avoidance motion planning Stream instance as a long-time Stream instance;

[0076] Step 2: Call the generated short-time Stream instance, i.e. the inverse kinematics solution Stream instance, to generate a robotic arm configuration object and predicates that satisfy the kinematic constraints;

[0077] Step 3: For the generated long-term Stream instance, namely the obstacle avoidance motion planning Stream instance, call it without solving for continuous variables to generate new objects and predicates. The new objects do not have specific values ​​and are called "optimistic objects".

[0078] Step 4: Add the objects and predicates generated by the short-time Stream instance and the "optimistic objects" and predicates generated by the long-time Stream instance to the initial state of the planning problem, so that the planning problem contains continuous variable objects and predicates that satisfy the motion constraints;

[0079] Step 5: Using the planning domain and the expanded planning problem as input, perform action sequence planning using classical planning methods to obtain an action sequence that includes "optimistic objects".

[0080] Step 6: Based on the action sequence obtained from the plan, backtrack on the "optimistic objects" to obtain a long-term Stream instance set that generates all "optimistic objects";

[0081] Step 7: Initiate calls to the Stream instances in the backtracked Stream instance set sequentially. If a Stream instance fails to generate a new object and predicate, indicating a failure in solving the continuous variables, return to Step 1. If the Stream instance call succeeds, assign values ​​to the "optimistic objects" corresponding to the continuous variable solution results. If all "optimistic objects" are assigned values, output the action sequence, indicating successful planning. If the given number of loops or time limit is exceeded, planning fails.

[0082] To verify the speed of this invention in solving the task planning problem under the motion constraints of a space robot, this embodiment conducts simulation experiments on an on-orbit operation task of a space robot using the incremental algorithm under the PDDLStream representation method, the optimistic algorithm, and this invention. The experimental results are shown in Table 1:

[0083] Table 1 Comparison of simulation experiment results

[0084]

[0085] As shown in the table, the incremental algorithm has the longest planning time but the shortest action sequence planning time. This indicates that the "blindness" in solving continuous variables consumes a lot of computation time, but the continuous variable objects obtained will also reduce the search range of the action sequence and the possibility of replanning, thereby improving the efficiency of action sequence planning. The optimistic algorithm has the longest action sequence planning time but the shortest continuous variable solution time. This indicates that solving continuous variables for "optimistic objects" within the action sequence can greatly reduce the number of continuous variable solutions, thereby reducing the continuous variable solution time. However, replanning of the action sequence will increase the action sequence planning time. The overall planning time of this invention is the shortest. According to the algorithm principle, its action sequence planning is performed after short-time continuous variable solutions. When including some continuous variable objects, the search range of the action sequence and the probability of replanning will decrease. Therefore, its action sequence time will be shorter than that of the optimistic algorithm. Since the initial continuous variable solution has short-time characteristics, the time increase from short-time continuous variable solutions is shorter than the time reduction from action sequence planning. Therefore, the overall planning time will be reduced. The simulation results verified the advantages of the present invention. For the task planning problem under the motion constraints of a space robotic arm, the present invention has higher planning efficiency.

[0086] A computer device is provided according to an embodiment of the present invention. This computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various method embodiments described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above.

[0087] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.

[0088] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device may include, but is not limited to, a processor and memory.

[0089] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0090] The memory can be used to store the computer program and / or module, and the processor implements various functions of the computer device by running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory.

[0091] If the modules / units integrated into the computer device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0092] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rapid task planning method under motion constraints for a spatial robotic arm, characterized in that, Includes the following steps: A task planning model for a spatial robotic arm under motion constraints is constructed, including the planning domain, the planning problem, and the Stream module; the specific method is as follows: Step 1: Construction of the Task Planning Domain Based on the model characteristics and task features of the space robotic arm, the objects, predicates and actions in the planning system are defined using the PDDL representation method, thus completing the construction of the planning domain; Step 2: Task Planning Problem Description For a given on-orbit operation task, the initial state and target state of the planning problem are determined based on the objects and predicates in the planning domain and described using the PDDL representation method; Step 3: Stream representation for solving continuous variables Based on the kinematic model and motion planning algorithm of the space robot, the PDDLStream representation method is used to establish the Stream representation form of the inverse kinematics solution and obstacle avoidance motion planning of the space robot. Based on the initial state of the planning problem and the input parameters of the Stream module, determine which Stream modules can be called; the specific method is as follows: Determine the callable Stream modules and generate short-term Stream instances and long-term Stream instances; the short-term Stream instance is an inverse kinematics solution Stream instance, and the long-term Stream instance is an obstacle avoidance motion planning Stream instance. Based on the invoked short-term Stream instance, generate a robotic arm configuration object and predicate that satisfy kinematic constraints; based on the invoked long-term Stream instance, generate an optimistic object and predicate; the optimistic object is a new object without specific values ​​generated by invoking the long-term Stream instance without solving for continuous variables. Add the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem to obtain the extended planning problem; Using the planning domain and the extended planning problem as input, the FastDownward planner is used to plan the action sequence, resulting in an action sequence that includes optimistic objects. Based on the action sequence obtained from the plan, backtrack the optimistic objects to obtain a long-term Stream instance set that generates all optimistic objects; The process involves sequentially calling the Stream instances in the long-lived Stream instance collection, assigning values ​​to the corresponding optimistic objects in the successfully called Stream instances, and so on, until all optimistic objects have been assigned values. Then, the action sequence is output to complete the planning. The specific method is as follows: The Stream instances in the backtracked Stream instance set are called sequentially. If the Stream instance fails to generate a new object and predicate, i.e. the continuous variable solution fails, the Stream module that can be called is returned. If the Stream instance is called successfully, the optimistic object corresponding to the continuous variable solution result is assigned a value.

2. A rapid task planning system for implementing the method of claim 1 under motion constraints of a spatial robotic arm, characterized in that, include: The model building module is used to build a task planning model under the motion constraints of a space robotic arm, including the planning domain, planning problem, and Stream module; The first calculation module is used to determine which Stream module can be called based on the initial state of the planning problem and the input parameters of the Stream module. The object and predicate generation module is used to generate robotic arm configuration objects and predicates that satisfy kinematic constraints based on the short-term Stream instance being called; and to generate optimistic objects and predicates based on the long-term Stream instance being called. The second calculation module is used to add the robotic arm configuration object and predicate generated by the short-time Stream instance and the optimistic object and predicate generated by the long-time Stream instance to the initial state of the planning problem to obtain the extended planning problem. The sequence planning module is used to perform action sequence planning using the FastDownward planner, taking the planning domain and the extended planning problem as input, to obtain an action sequence that includes optimistic objects. The object backtracking module is used to backtrack the optimistic objects in the action sequence obtained from the plan, and obtain a long-term Stream instance collection that generates all optimistic objects; The object assignment module is used to sequentially call the Stream instances in the long-term Stream instance collection, assign values ​​to the corresponding optimistic objects in the successfully called Stream instances, until all optimistic objects have been assigned values, and then output the action sequence to complete the planning.

3. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in claim 1.

4. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in claim 1.

Images