A space manipulator operation task planning method, system, device and medium

Abstract

The present invention relates to the field of space station technology, and in particular to a space manipulator operation task planning method, system, device and medium, the method comprising: obtaining six numbers of an orbit at any time, determining a full orbit position sequence table of a spacecraft on the orbit based on the six numbers; determining the distribution of the period of occlusion between the space manipulator and the full orbit position sequence table when the space manipulator resides on a base after executing a task instruction in the space station measurement and control period; obtaining a full set of basic instructions for space manipulator operation tasks, and performing planning processing based on the full set of basic instructions for space manipulator operation tasks to obtain an instruction sequence; obtaining a space manipulator operation task planning result based on the distribution and the instruction sequence. The present invention can reasonably plan the operation tasks of the space manipulator.

Description

Background Art

[0002] Mission planning is a popular research direction in the field of robotics and is the core of robots to achieve autonomous tasks. The mission planning problem is the process of obtaining action sequences and behavior sequences under multiple constraints based on the robot's mission objectives. With the deepening of space exploration, the application of robotic arms in space missions is becoming more and more extensive. Space robotic arms can replace or assist astronauts in going out of the cabin, monitor visiting spacecraft, separate, transfer and re-docking cabin segments, and perform cabin surface inspections. The diversity and complexity of robotic arm operation tasks in unstructured space environments increase the difficulty of mission planning and put forward higher requirements for the accuracy and efficiency of mission planning.

[0003] With the initial construction and continuous operation of my country's manned space station, the requirements for measurement, control and communication between the space station and the ground are becoming increasingly higher. At present, the space station adopts a space-based measurement, control and communication mode based on the relay satellite system. The geosynchronous orbit (GEO) satellite is selected as the relay node of the space station because its flight period around the earth is the same as the rotation period of the earth, and it can maintain a relatively static state with the ground base station to establish a synchronous communication link. The communication radio wave propagation mode between the space station and the relay satellite is direct wave, and communication is required without obstruction. Due to the restrictive factors of the physical structure of the space station itself, the communication system and antenna equipment can only be installed in the designated area space. When the space manipulator resides between the bases in a pre-planned configuration, it may block the transmission and reception of the communication signal between the space station and the relay satellite, causing the signal to weaken or even interrupt, affecting the arrangement of the mission instruction plan. Therefore, we need to conduct mission planning research on the space manipulator for time-varying communication obstruction scenarios. Summary of the invention

[0004] The technical problem to be solved by the present invention is to address the deficiencies of the prior art, specifically to address the problem that when a space manipulator resides between bases in a pre-planned configuration, it will block the transmission and reception of communication signals between a space station and a relay satellite. Specifically, a space manipulator operation task planning method, system, device and medium are provided, which are as follows:

[0005] 1) In the first aspect, the present invention provides a space manipulator operation task planning method, and the specific technical solution is as follows:

[0006] S1, obtaining six numbers of the orbit at any time, and determining a full orbit position sequence table of the spacecraft on the orbit based on the six numbers;

[0007] S2, determining the distribution of the period of time when the space robot arm resides on the base after executing the task instruction and there is an obstruction with the full orbit position sequence table in the space station measurement and control period;

[0008] S3, obtaining a complete set of basic instructions for space manipulator operation tasks, and performing planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence;

[0009] S4, obtaining an operation task planning result of the space robot arm based on the distribution and the instruction sequence.

[0010] The beneficial effects of a space manipulator operation task planning method provided by the present invention are as follows:

[0011] By using six numbers to determine the full orbit position sequence table, we can more accurately obtain the position information of the entire orbit at different times. Combined with the distribution of obstruction periods in the space station's measurement and control period, we can more effectively determine which periods are blocked and cannot communicate. Based on the above two pieces of information, we can reasonably plan the operating tasks of the space robot arm.

[0012] Based on the above solution, the present invention can also be improved as follows.

[0013] Furthermore, the six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

[0014] Further, the specific process of determining the full orbital position sequence table of the spacecraft on the orbit based on the six numbers is as follows:

[0015] Determining the eccentric anomaly of the orbit by using the eccentricity of the orbit and the true anomaly of the orbit;

[0016] Determining a first anomaly angle after a preset period based on the anomaly angle of the orbit;

[0017] Determining a first true anomaly after a preset period based on a geometric relationship between the first eccentric anomaly and the first true anomaly;

[0018] Replacing the true anomaly of the orbit in the six numbers with the first true anomaly to obtain new six numbers, and determining the position information of the spacecraft on the orbit after a preset period based on the new six numbers;

[0019] The first eccentric anomaly angle after the next preset period is calculated until the spacecraft completes one orbit, and the position information after each preset period is obtained, and all the position information is collected to form the full orbit position sequence table.

[0020] Further, the process of determining the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period is as follows:

[0021] Based on the configuration of the space robot arm residing on the base, the position information of all spacecraft in the full orbit position sequence table is traversed, and the position information of the spacecraft at any time and the space robot arm are used to determine the occlusion situation through geometric simulation method. When there is an occlusion situation, the start time and the end time of the occlusion are recorded, and the occlusion period is determined according to the start time and the end time of the occlusion, and the distribution of the occlusion period in the measurement and control period of the space station is determined. When there is no occlusion situation, no processing and recording are performed.

[0022] Further, the process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task includes:

[0023] Determine the task content, task objectives and task attributes corresponding to each instruction in the complete set of basic instructions for the space manipulator operation task;

[0024] Determine the subsystem and action corresponding to each instruction;

[0025] Based on the constraints in the task planning domain, each instruction is analyzed to obtain the prerequisites and effects corresponding to each instruction;

[0026] Build a domain description file based on the subsystem, task objectives, and task content corresponding to each instruction;

[0027] Using the PDDL planning model, the task content, task objectives, task attributes, subsystems and actions corresponding to each instruction are taken as input to obtain the instruction sequence.

[0028] Furthermore, the process of planning and processing to obtain an instruction sequence based on the complete set of basic instructions for the space manipulator operation task also includes:

[0029] Taking a space robot arm operation remote control instruction sequence set as a sample set, training the sample set by an unsupervised learning method and a self-supervised learning method to obtain a trained network model;

[0030] The trained network model is used to process the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence.

[0031] 2) In the second aspect, the present invention also provides a space manipulator operation task planning system, and the specific technical solution is as follows:

[0032] The first acquisition module is used to: acquire six numbers of the orbit at any time, and determine the full orbit position sequence table of the spacecraft on the orbit based on the six numbers;

[0033] The determination module is used to: determine the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period;

[0034] The second acquisition module is used to: acquire a complete set of basic instructions for space manipulator operation tasks, and perform planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence;

[0035] The planning module is used to obtain the operation task planning result of the space robot arm based on the distribution situation and the instruction sequence.

[0036] Based on the above solution, the present invention can also be improved as follows.

[0037] Furthermore, the six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

[0038] Further, the specific process of determining the full orbital position sequence table of the spacecraft on the orbit based on the six numbers is as follows:

[0039] Determining the eccentric anomaly of the orbit by using the eccentricity of the orbit and the true anomaly of the orbit;

[0040] Determining a first anomaly angle after a preset period based on the anomaly angle of the orbit;

[0041] Determining a first true anomaly after a preset period based on a geometric relationship between the first eccentric anomaly and the first true anomaly;

[0042] Replacing the true anomaly of the orbit in the six numbers with the first true anomaly to obtain new six numbers, and determining the position information of the spacecraft on the orbit after a preset period based on the new six numbers;

[0043] The first eccentric anomaly angle after the next preset period is calculated until the spacecraft completes one orbit, and the position information after each preset period is obtained, and all the position information is collected to form the full orbit position sequence table.

[0044] Further, the process of determining the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period is as follows:

[0045] Based on the configuration of the space robot arm residing on the base, the position information of all spacecraft in the full orbit position sequence table is traversed, and the position information of the spacecraft at any time and the space robot arm are used to determine the occlusion situation through geometric simulation method. When there is an occlusion situation, the start time and the end time of the occlusion are recorded, and the occlusion period is determined according to the start time and the end time of the occlusion, and the distribution of the occlusion period in the measurement and control period of the space station is determined. When there is no occlusion situation, no processing and recording are performed.

[0046] Further, the process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task includes:

[0047] Determine the task content, task objectives and task attributes corresponding to each instruction in the complete set of basic instructions for the space manipulator operation task;

[0048] Determine the subsystem and action corresponding to each instruction;

[0049] Based on the constraints in the task planning domain, each instruction is analyzed to obtain the prerequisites and effects corresponding to each instruction;

[0050] Build a domain description file based on the subsystem, task objectives, and task content corresponding to each instruction;

[0051] Using the PDDL planning model, the task content, task objectives, task attributes, subsystems and actions corresponding to each instruction are taken as input to obtain the instruction sequence.

[0052] Furthermore, the process of planning and processing to obtain an instruction sequence based on the complete set of basic instructions for the space manipulator operation task also includes:

[0053] Taking a space robot arm operation remote control instruction sequence set as a sample set, training the sample set by an unsupervised learning method and a self-supervised learning method to obtain a trained network model;

[0054] The trained network model is used to process the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence.

[0055] 3) In a third aspect, the present invention further provides a computer device, comprising a processor, wherein the processor is coupled to a memory, wherein at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the computer device implements a space robot arm operation task planning method as described in any of the above claims.

[0056] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein at least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by a processor so that a computer implements a space robot arm operation task planning method as described in any of the above claims.

[0057] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0059] Figure 1 A schematic diagram of a flow chart of a space manipulator task planning method embodiment of the present invention;

[0060] Figure 2 A structural framework diagram provided for an embodiment of a space manipulator operation task planning system according to an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of a space station antenna signal coverage area provided in an embodiment of a space manipulator operation task planning method according to an embodiment of the present invention;

[0062] Figure 4 A schematic diagram of two space capsules provided in an embodiment of a space manipulator task planning method according to an embodiment of the present invention;

[0063] Figure 5 A distribution diagram of the occlusion period between a manipulator and a relay satellite provided in an embodiment of a space manipulator operation task planning method of an embodiment of the present invention;

[0064] Figure 6 A schematic diagram of a basic task instruction set for manipulator operation provided in a space manipulator operation task planning method embodiment of the present invention;

[0065] Figure 7 A task planning modeling and solution flow chart provided for a space manipulator operation task planning method embodiment of the present invention;

[0066] Figure 8 A schematic diagram of a task sample library provided for a space manipulator task planning method embodiment of the present invention;

[0067] Fig. 9A schematic diagram of a UIE prompt learning network framework based on a pre-trained model provided in an embodiment of a space manipulator task planning method according to an embodiment of the present invention;

[0068] Fig.10 A command event logic network diagram provided for a space manipulator task planning method embodiment of the present invention;

[0069] Fig.11 A schematic diagram of a space manipulator task planning process taking into account communication shielding provided in an embodiment of a space manipulator task planning method according to an embodiment of the present invention;

[0070] Fig.12 A schematic diagram of the computer device structure provided for a space manipulator task planning method embodiment of the present invention;

[0071] Fig.13 A schematic diagram of the manipulator DH coordinate system provided in an embodiment of a space manipulator task planning method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0072] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0073] like Figure 1 As shown, a space manipulator operation task planning method according to an embodiment of the present invention comprises the following steps:

[0074] S1, obtaining six numbers of the orbit at any time, and determining a full orbit position sequence table of the spacecraft on the orbit based on the six numbers;

[0075] S2, determining the distribution of the period of time when the space robot arm resides on the base after executing the task instruction and there is an obstruction with the full orbit position sequence table in the space station measurement and control period;

[0076] S3, obtaining a complete set of basic instructions for space manipulator operation tasks, and performing planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence;

[0077] S4, obtaining an operation task planning result of the space robot arm based on the distribution and the instruction sequence.

[0078] The beneficial effects of a space manipulator operation task planning method provided by the present invention are as follows:

[0079] By using six numbers to determine the full orbit position sequence table, we can more accurately obtain the position information of the entire orbit at different times. Combined with the distribution of obstruction periods in the space station's measurement and control period, we can more effectively determine which periods are blocked and cannot communicate. Based on the above two pieces of information, we can reasonably plan the operating tasks of the space robot arm.

[0080] S1, obtaining the six numbers of the orbit at any time, and determining the full orbit position sequence table of the spacecraft on the orbit based on the six numbers. The specific process is:

[0081] S11, obtain the six numbers of the orbit at any moment;

[0082] The six numbers include: semi-major axis a e , eccentricity e, orbit inclination i, right ascension of ascending node Ω, perigee angular distance ω e and the true anomaly θ e ;

[0083] S12, calculate the preset period T according to the six numbers p After that, the position vector of the spacecraft in the geocentric inertial coordinate system is obtained, and then the full orbit position sequence table is obtained;

[0084] Preset period T p It is defined according to the properties of the space station itself and is not limited here;

[0085] Geocentric inertial coordinate system S e , defined as follows:

[0086] The origin O of the coordinate system is the center of mass of the earth, the x-axis is in the equatorial plane and points to the vernal equinox, the z-axis points to the North Pole along the earth's rotation axis, and the y-axis is in the equatorial plane and follows the right-hand rule.

[0087] The space orbit coordinate system S corresponds to the geocentric inertial coordinate system. o , defined as follows:

[0088] The origin of the coordinate system is the center of mass of the earth. The x-axis is along the line connecting the center of the earth and the spacecraft, pointing to the spacecraft. The z-axis is perpendicular to the orbital plane, pointing in the direction of the orbital momentum vector of the spacecraft. The y-axis is in the orbital plane and follows the right-hand rule.

[0089] Assume that after the spacecraft orbit is established, there is only the true anomaly angle θ e The position vector of the spacecraft in the geocentric inertial coordinate system is determined as follows:

[0090] 1. From the eccentric anomaly E0, eccentricity e and true anomaly θ e The geometric relationship (1) is used to calculate the first eccentric anomaly angle E0:

[0091]

[0092] 2. Calculate T from Kepler's time equation (2) p The first near corner point E after time t :

[0093]

[0094] Where μ is the Earth's gravitational constant.

[0095] 3. Calculate T from the geometric relationship between the near corner point and the true near corner point (3) p The first true pericentric point after time θ t :

[0096]

[0097] Calculate T from the updated six orbital numbers p The position vector r of the spacecraft in the geocentric inertial coordinate system after time e :

[0098] The position vector of the spacecraft in the space orbit coordinate system is r0 = [r0 0 0]

[0099]

[0100] The positive semi-focal chord p in the formula can be calculated by the orbital momentum formula: p = a e (1-e 2 ).

[0101] The transformation relationship from the geocentric inertial coordinate system to the space orbit coordinate system can be obtained by the coordinate transformation matrix:

[0102]

[0103] In the formula represents the geocentric inertial coordinate system S e To the space orbit coordinate system S o Transformation matrix; Rot(·,*) represents the rotation matrix of the * angle around the · axis; z represents the z-axis corresponding to the geocentric inertial coordinate system, and x represents the x-axis corresponding to the geocentric inertial coordinate system.

[0104] The position vector in the geocentric inertial coordinate system can be calculated

[0105] S13, repeating the process of S12 until the prediction of the position information of the entire track at all times is completed, and forming a full track position sequence table according to the position vectors corresponding to different preset periods;

[0106] The full track position sequence table includes the position vector and the time corresponding to the position vector, that is, the preset cycle number.

[0107] S2, the specific process of determining the distribution of the period of time when the space robot arm resides on the base after executing the task instruction and there is an obstruction in the full orbit position sequence table in the space station measurement and control period is:

[0108] S21, determining the state of the space manipulator when the space manipulator stays on the base after executing the current task instruction;

[0109] The base is a component that provides installation location and structural support for the robotic arm system on the space station, and is distributed in multiple locations in each compartment of the space station. It carries and fixes the joints and arm segments of the robotic arm, ensuring that the robotic arm can perform various movements and operations in space.

[0110] The state of the space robot arm includes but is not limited to a retracted state and an extended state.

[0111] S22, determining, by means of a geometric simulation method, the occlusion period during which occlusion occurs in the current state of the manipulator and the distribution of the occlusion period in the measurement and control period of the space station;

[0112] The geometric simulation method is:

[0113] The area covered by the space station antenna signal is assumed to be an obstacle entity. When the space robotic arm causes communication obstruction, the space robotic arm entity collides with the imaginary obstacle entity, thereby converting the space robotic arm communication obstruction problem into a collision detection problem between the two entities. This is the core idea of ​​the geometric simulation method.

[0114] like Figure 3 As shown in (a), the coverage area of ​​the space station antenna signal is approximately a combination of a cylinder and a cone. The mathematical model of the hypothetical obstacle entity is established as follows:

[0115] Obstacle={f(x,y,z)x 2 +y 2 =r1 2 ,0#zd;x 2 +y 2 =(za) 2 ,a#zh}

[0116] The arm of the space robot can be simplified into a cylinder, the radius of the cylinder is the maximum radius of the connecting rod, and the height is the length of the connecting rod. However, the collision detection between cylinders must also consider the collision of the two bottom surfaces of the cylinder, which is very complicated. Therefore, the robot body is enveloped by a capsule body, so that the distance from the capsule envelope surface to the central axis section of the space robot arm is uniform. Considering the maximum physical size of the space robot in the straight state, the space robot will not be blocked by the upper part of the signal coverage area, so the communication cone can be simplified to a cylinder, that is, Figure 3(b) The radius of the cylinder is the radius of the cylinder at the bottom of the communication cone plus a certain safety margin. It is also encapsulated by a capsule. After processing, the display of the two capsules in space is as follows Figure 4 shown.

[0117] Since the distance from the envelope capsule to the central axis segment is uniform, the collision detection between capsules can be simplified to the positional relationship judgment between the central axis segments. The space manipulator is simplified to an open-loop rigid bar chain, each rigid bar represents a space manipulator link, and only represents the central axis of the corresponding space manipulator link. The communication link between the space station and the relay satellite is represented by a line segment between the two antennas. Based on the simplified model, the collision between the two entities of the manipulator and the imaginary obstacle is converted into the positional relationship judgment between the straight line segment of the link and the line segment connecting the centers of the two antennas.

[0118] Known space manipulator DH parameters θ i ,d i ,a i ,α i (i=1,2,…,7), the schematic diagram of establishing the robot arm DH coordinate system is as follows Fig.13 , through the method of homogeneous coordinate transformation, the mathematical model of each connecting rod straight line segment is established as follows:

[0119]

[0120] Where, T A ,T B ,T C ,T D ,T E ,T F ,T G They represent the homogeneous pose transformation matrices at the first, second, third, fourth, fifth, sixth, and seventh joints respectively. The coordinates of points B, C, D, E, F, and G are B = [x b ,y b ,z b ]=[T B (1,4),T B (2,4),T B (3,4)],C=[x c ,y c ,z c ]=[T C (1,4),T C (2,4),T C (3,4)],D=[x d ,y d ,z d ]=[T D (1,4),T D (2,4),T D(3,4)],E=[x e ,y e ,z e ]=[T E (1,4),T E (2,4),T E (3,4)],F=[x f ,y f ,z f ]=[T F (1,4),T F (2,4),T F (3,4)],G=[x g ,y g ,z g ]=[T G (1,4),T G (2,4),T G (3,4)],

[0121] Among them, θ i Represents the joint angles, which are variables, d i Denotes the connecting rod offset, α i represents the torsion angle of each connecting rod, a i Indicates the length of each link, and these three are constants. Rot(·,*) is the rotation transformation matrix, · indicates the coordinate axis around which the rotation is made, * indicates the angle of rotation around the corresponding axis, Trans(x,y,z) is the translation transformation matrix, and the vector [x,y,z] indicates the size and direction of the translation.

[0122] Then the equations of line BC and line CD are as follows:

[0123]

[0124]

[0125] The same is true for the equations of the lines DE, EF, and FG.

[0126] The location of the space station antenna is known [x t1 ,y t1 ,z t1 ] and the relay satellite antenna position [x t2 ,y t2 ,z t2 ], then the simplified mathematical equation of the communication link is

[0127] Assume that the shortest distances between the straight line segment between the two antennas and the straight line segment representing the connecting rod of the space robot are {d1, d2, d3, d4, ...}. i≥r+R,i=1,2,…,7. If so, the two entities will not collide and the manipulator will not cause communication obstruction. Where R represents the maximum radial radius of the space manipulator, and r represents the radius of the cylinder that envelops the communication cone.

[0128] Usually, after the mission command is issued, the motion planning software will pre-plan a set of action sequence start and end configurations. After the mission is completed, the space manipulator stays on the base in the pre-planned end configuration. Assuming that the space station manipulator stays on base a, the positions of the space station and each relay satellite at different times can be obtained according to the full orbit position sequence table. The space station manipulator traverses the orbit position in the planned configuration, and the geometric simulation method is used to determine the communication obstruction of the space station manipulator in the full orbit position. From the start of the obstruction to the end of the obstruction is a period of obstruction, and the distribution of the manipulator communication obstruction period in the space station measurement and control period is obtained.

[0129] In the above text, motion planning software is: system software that mainly integrates basic algorithm modules such as forward and inverse kinematics solution and collision detection of multi-degree-of-freedom manipulators and core algorithm modules of path planning and trajectory optimization. First, the operator needs to describe the task to be performed by the manipulator. After receiving the task description, the motion planning software generates the final trajectory through the kinematic module and the path planning and trajectory optimization algorithm module. After that, the software will generate control instructions and send them to the manipulator controller to actually execute the movement of the manipulator.

[0130] like Figure 5 It shows the distribution of the shielding period between the space station antenna and the three relay satellites A, B, and C when the robotic arm is at base a. After deducting it from the overall communication period, the final communication period is [T start ,T a1 ]∪[T a2 ,T c1 ]∪[T c2 ,T a3 ]∪[T a4 ,T b3 ]∪[T b4 ,T c3 ]∪[T c4 ,T end ].

[0131] S3, the specific process of obtaining the complete set of basic instructions for space manipulator operation tasks, and performing planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence is as follows:

[0132] The construction process of the basic instruction set of space manipulator operation tasks is as follows:

[0133] The space manipulator operation task is a complex task involving multiple system events. Figure 6, the mission instructions are classified using a tree structure. The complete set of mission instructions includes instructions for the astronaut system, camera system, energy system, GNC system, robotic arm system, etc. Each subsystem instruction set is composed of basic system instructions.

[0134] There are M systems that may be involved in the space manipulator operation task, and each system can execute N operating instructions. i (i=1,…,M), all subsystem instruction sets together constitute the basic instruction set of space manipulator operation tasks, then there are N instructions in the basic instruction set of space manipulator operation tasks, N=N 1 +N 2 +…+N M .

[0135] The process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task is as follows:

[0136] Assuming that task 1 involves n basic instructions, there are C possible instruction sequences. n N However, due to the logical timing relationship between the space manipulator operation instructions and various related system event instructions, it is necessary to screen out the appropriate instruction sequence from all possible instruction sequences.

[0137] Two instruction sequence solving methods are proposed below, one is the traditional method based on modeling and the other is a learning method based on knowledge extraction.

[0138] (1) Instruction event sequencing method based on modeling solution

[0139] Aiming at the problem of space manipulator operation task planning, the task process and constraint conditions of the manipulator operation are analyzed, the planning domain definition language is used to model the domain knowledge and planning problems of space manipulator operation task planning, and then an effective algorithm is used to solve the planning problem.

[0140] Planning Domain Definition Language (PDDL) is a standardized formal language for describing problems in the mission planning domain. In PDDL, a planning problem can be described as a two-tuple:

[0141] D=<Domain,Problem>

[0142] Among them, Domain is a domain description file, and Problem is a problem description file. The domain description file includes object types (types), predicates (predicates) and activities (action), and the problem description file includes objects (objects), initial state (init) and planning goals (goal).

[0143] Among them, the object types include: robotic arm system, camera system, power system, camera system and other subsystems.

[0144] Predicates include: logical representation of each subsystem's capabilities, current state, and target state.

[0145] Actions include: all actions of each subsystem and their defined properties.

[0146] like Figure 7 As shown, the task content, objectives, attributes, subsystems and actions of the space manipulator are determined; according to the constraints in the task planning domain, the prerequisites and effects of the actions are analyzed, the domain files and problem files in the PDDL planning model are established, the model is solved using an effective algorithm, the solution is confirmed, and it is determined whether the solution is reasonable. If it is reasonable, the planning result is output and the planning result is used as an instruction sequence. If it is unreasonable, the task content, objectives, attributes, subsystems and actions are reconfirmed.

[0147] Among them, the task planning domain refers to the part used to describe and define the abstract environment and problem domain of the task planning problem, which is mainly composed of state description, action description, initial state, target state, etc.

[0148] According to the task content, target, attributes, subsystems and actions of the space manipulator, the initial state of the task, target state, subsystem objects and their states, actions and their attributes can be analyzed.

[0149] The same domain file can correspond to multiple question files;

[0150] Integrate the task initial status, target status, and subsystem objects into a problem description document;

[0151] The preconditions of the actions and the effects they produce are analyzed in the following way:

[0152] For example, before the robot transfers, it is necessary to confirm the current position of the robot, the feasible path, and whether the robot is ready for movement. After the transfer action is executed, the robot reaches the target position. For the robot transfer action, the prerequisites are: the robot is in the initial position, the robot is ready for movement, and there is a feasible path for the robot. The effect is: the robot reaches the target position.

[0153] In order to facilitate the understanding of the domain description file and the problem description file, the space manipulator transport load task is used as an example for explanation: the manipulator is initially parked at the base JZ01, and the load is initially located near JZ02. The manipulator needs to move from JZ02 to the base JZ03 to install the load. After the installation is completed, the manipulator is parked at the base JZ03, and the load is installed near JZ03. As shown below, Table 1 is a list of object types, Table 2 is a list of predicates and their descriptions, Table 3 is an activity and its description, and Table 4 is a problem file.

[0154] Table 1

[0155] Object types describe Arm Robotic Arm Power Power System Load Load Base Base Camera Camera system

[0156] Table 2

[0157] Predicates describe (arm_at?a–Arm?b-Base) The robotic arm is on the base (power_avail?p-Power) Power available (arm_on?a-Arm) Power on the robot arm (arm_off?a-Arm) Robotic arm power off (state_set) Status Settings (load_near?l-Load?b-Base) The load is near the base (arm_empty?a-Arm) The robot end effector is empty (holding?l-load?a-arm) The robot grabs the load (vaild_path?base_from?base_to) Viable Path (camera_state_set?c-Camera) Camera status settings (arm_motion_avail) The robotic arm can move (arm_preparation_ready) The robot arm is ready

[0158] Table 3

[0159] Action describe Motion_preparation Pre-exercise preparation Platform_status_set Platform status settings Move Robotic arm transfer Capture Capture Load Mounting Mounting load

[0160] Table 4

[0161]

[0162] An effective algorithm means that each step of the algorithm should be able to be effectively executed and obtain a certain result; in this solution, the effective algorithms are: FF (Fast-Forward) and Metric-FF, etc.

[0163] Fast-Forward is a classic heuristic search planner that combines a heuristic function and a cost function to find the optimal solution in the state space. It evaluates the state of the problem domain based on domain-specific information to guide the search process. During the search process, FF uses domain knowledge to reduce the search space. It uses the PDDL description of the planning problem to obtain information about the problem structure to better guide the search direction. At the same time, Fast-Forward uses different pruning strategies to reduce the size of the search space and the set of feasible solutions. This helps to improve the search efficiency and speed up the solution process. Finally, the solution obtains the sequence planning result.

[0164] The Metric-FF algorithm is developed based on the FF (Fast-Forward) algorithm and is a planner that combines heuristic search and relaxed graph planning.

[0165] The process of confirming whether the obtained solution is reasonable is as follows: manually analyze and judge the planning sequence, simulate the robot arm action sequence in the simulation system, and judge whether a collision will occur during the actual movement. At the same time, other factors such as task priority and emergencies are taken into consideration to judge the rationality of the planning results and further adjust the planning results.

[0166] (2) Command event orchestration method based on domain knowledge learning

[0167] At present, the space station has successfully completed a number of robotic arm operation tasks in orbit and established a task sample library based on instruction sequences (such as Figure 8 ), the systems involved in each task sample may be different, and the number of basic instructions involved may also be different.

[0168] The UIE prompt learning network based on the pre-trained model is selected to learn the existing mature task instruction sequence categories (i.e. instruction event relationship categories) and each instruction content (i.e. instruction events) in the training set. For the task instruction sequence to be analyzed, the automatic extraction of knowledge can be realized based on the above-mentioned trained network, and the instruction event relationship categories and instruction events of the sequence can be obtained at the same time, which respectively constitute the logical relationship between instructions and instruction contents of the diagram, and finally form a large-scale logical network diagram based on semantic knowledge. The specific description is as follows:

[0169] First, combined with the characteristics of space manipulator operations, under the premise that the command events have been pre-defined, the command event relationship set is designed. The command event relationship set is shown in Table 5;

[0170] Table 5

[0171]

[0172] Then, in view of the characteristics of few labeled samples and low resources, the UIE generation method of the domain knowledge pre-training model for space robot arm operations is adopted. The space robot arm operation remote control instruction sequence set is used as the sample set, and the instruction sequence category (instruction event relationship) and instruction content (instruction event) learning network training and optimization are performed through unsupervised and self-supervised learning methods, such as mask-based language model training. Then, based on the trained network model, the knowledge of the set of remote control instruction sequences for robot arm operations is automatically extracted to obtain the instruction sequence category and instruction content for the entire process of robot arm operations, that is, instruction relations and instruction set events. Based on the instruction relations and instruction set events in the above task instruction sequences, a large-scale logical network diagram is constructed through graph database technology. The learning network framework is such as Fig. 9 The final logical network diagram is shown as Fig.10 shown.

[0173] S4, the process of obtaining the operation task planning result of the space manipulator based on the distribution and the instruction sequence is as follows:

[0174] Since all robot actions and related system events have duration, it is necessary to perform timing planning on the sequence planning solution obtained in STEP4, that is, the activity instructions cannot be scheduled outside the communication period. Let the start time of the i-th activity be The duration is t during , the communication period can be expressed as The timing constraint model is expressed as:

[0175]

[0176]

[0177] When the execution time of the i-th activity has the above-mentioned timing constraint conflict, the activity will be delayed to the next communication period that meets the constraint, and the task of the activity will be delayed by the same time.

[0178] Fig.11 This is a schematic diagram of the space manipulator task planning process considering communication occlusion. When there are new tasks to be executed during the planning cycle, the manipulator task planning sends task instructions within a specific measurement and control communication period based on the analysis and prediction of communication occlusion. The task instructions are a series of confirmed reasonable activity instruction sequences.

[0179] Furthermore, the six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

[0180] Further, the specific process of determining the full orbital position sequence table of the spacecraft on the orbit based on the six numbers is as follows:

[0181] Determining the eccentric anomaly of the orbit by using the eccentricity of the orbit and the true anomaly of the orbit;

[0182] Determining a first anomaly angle after a preset period based on the anomaly angle of the orbit;

[0183] Determining a first true anomaly after a preset period based on a geometric relationship between the first eccentric anomaly and the first true anomaly;

[0184] Replacing the true anomaly of the orbit in the six numbers with the first true anomaly to obtain new six numbers, and determining the position information of the spacecraft on the orbit after a preset period based on the new six numbers;

[0185] The first eccentric anomaly angle after the next preset period is calculated until the spacecraft completes one orbit, and the position information after each preset period is obtained, and all the position information is collected to form the full orbit position sequence table.

[0186] Further, the process of determining the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period is as follows:

[0187] Based on the configuration of the space robot arm residing on the base, the position information of all spacecraft in the full orbit position sequence table is traversed, and the position information of the spacecraft at any time and the space robot arm are used to determine the occlusion situation through geometric simulation method. When there is an occlusion situation, the start time and the end time of the occlusion are recorded, and the occlusion period is determined according to the start time and the end time of the occlusion, and the distribution of the occlusion period in the measurement and control period of the space station is determined. When there is no occlusion situation, no processing and recording are performed.

[0188] Further, the process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task includes:

[0189] Determine the task content, task objectives and task attributes corresponding to each instruction in the complete set of basic instructions for the space manipulator operation task;

[0190] Determine the subsystem and action corresponding to each instruction;

[0191] Based on the constraints in the task planning domain, each instruction is analyzed to obtain the prerequisites and effects corresponding to each instruction;

[0192] Build a domain description file based on the subsystem, task objectives, and task content corresponding to each instruction;

[0193] Using the PDDL planning model, the task content, task objectives, task attributes, subsystems and actions corresponding to each instruction are taken as input to obtain the instruction sequence.

[0194] Furthermore, the process of planning and processing to obtain an instruction sequence based on the complete set of basic instructions for the space manipulator operation task also includes:

[0195] Taking a space robot arm operation remote control instruction sequence set as a sample set, training the sample set by an unsupervised learning method and a self-supervised learning method to obtain a trained network model;

[0196] The trained network model is used to process the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence.

[0197] like Figure 2 As shown, the present invention also provides a space manipulator operation task planning system 200, and the specific technical solution is as follows:

[0198] The first acquisition module 210 is used to: acquire six numbers of the orbit at any time, and determine a full orbit position sequence table of the spacecraft on the orbit based on the six numbers;

[0199] The determination module 220 is used to: determine the distribution of the occlusion period between the space robot arm and the full orbit position sequence table when the space robot arm resides on the base after executing the task instruction in the space station measurement and control period;

[0200] The second acquisition module 230 is used to: acquire a complete set of basic instructions for space manipulator operation tasks, and perform planning processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence;

[0201] The planning module 240 is used to obtain the operation task planning result of the space robot arm based on the distribution situation and the instruction sequence.

[0202] Furthermore, the six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

[0203] Further, the specific process of determining the full orbital position sequence table of the spacecraft on the orbit based on the six numbers is as follows:

[0204] Determining the eccentric anomaly of the orbit by using the eccentricity of the orbit and the true anomaly of the orbit;

[0205] Determining a first anomaly angle after a preset period based on the anomaly angle of the orbit;

[0206] Determining a first true anomaly after a preset period based on a geometric relationship between the first eccentric anomaly and the first true anomaly;

[0207] Replacing the true anomaly of the orbit in the six numbers with the first true anomaly to obtain new six numbers, and determining the position information of the spacecraft on the orbit after a preset period based on the new six numbers;

[0208] The first eccentric anomaly angle after the next preset period is calculated until the spacecraft completes one orbit, and the position information after each preset period is obtained, and all the position information is collected to form the full orbit position sequence table.

[0209] Further, the process of determining the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period is as follows:

[0210] Based on the configuration of the space robot arm residing on the base, the position information of all spacecraft in the full orbit position sequence table is traversed, and the position information of the spacecraft at any time and the space robot arm are used to determine the occlusion situation through geometric simulation method. When there is an occlusion situation, the start time and the end time of the occlusion are recorded, and the occlusion period is determined according to the start time and the end time of the occlusion, and the distribution of the occlusion period in the measurement and control period of the space station is determined. When there is no occlusion situation, no processing and recording are performed.

[0211] Further, the process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task includes:

[0212] Determine the task content, task objectives and task attributes corresponding to each instruction in the complete set of basic instructions for the space manipulator operation task;

[0213] Determine the subsystem and action corresponding to each instruction;

[0214] Based on the constraints in the task planning domain, each instruction is analyzed to obtain the prerequisites and effects corresponding to each instruction;

[0215] Build a domain description file based on the subsystem, task objectives, and task content corresponding to each instruction;

[0216] Using the PDDL planning model, the task content, task objectives, task attributes, subsystems and actions corresponding to each instruction are taken as input to obtain the instruction sequence.

[0217] Furthermore, the process of planning and processing to obtain an instruction sequence based on the complete set of basic instructions for the space manipulator operation task also includes:

[0218] Taking a space robot arm operation remote control instruction sequence set as a sample set, training the sample set by an unsupervised learning method and a self-supervised learning method to obtain a trained network model;

[0219] The trained network model is used to process the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence.

[0220] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0221] It should be noted that the beneficial effects of a space manipulator operation task planning system provided in the above embodiment are the same as the beneficial effects of the space manipulator operation task planning method described above, which will not be described in detail here. In addition, when the system provided in the above embodiment realizes its functions, it only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be described in detail here.

[0222] like Fig.12 As shown, a computer device 300 of an embodiment of the present invention includes a processor 320, the processor 320 is coupled to a memory 310, and the memory 310 stores at least one computer program 330, and the at least one computer program 330 is loaded and executed by the processor 320, so that the computer device 300 implements any one of the above-mentioned space manipulator operation task planning methods, specifically:

[0223] The computer device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors 320 (Central Processing Units, CPU) and one or more memories 310, wherein the one or more memories 310 store at least one computer program 330, and the at least one computer program 330 is loaded and executed by the one or more processors 320, so that the computer device 300 implements a space manipulator operation task planning method provided in the above embodiment. Of course, the computer device 300 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output, and the computer device 300 may also include other components for realizing device functions, which will not be repeated here.

[0224] A computer-readable storage medium according to an embodiment of the present invention stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that a computer implements a space robot arm operation task planning method according to any one of the above claims.

[0225] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0226] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes any of the above-mentioned space manipulator operation task planning methods.

[0227] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0228] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0229] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0230] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A space manipulator task planning method, characterized in that: include: S1, obtaining six numbers of the orbit at any time, and determining a full orbit position sequence table of the spacecraft on the orbit based on the six numbers; S2, determining the distribution of the period of time when the space robot arm resides on the base after executing the task instruction and there is an obstruction with the full orbit position sequence table in the space station measurement and control period; S3, obtaining a complete set of basic instructions for space manipulator operation tasks, and performing planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence; S4, obtaining an operation task planning result of the space robot arm based on the distribution and the instruction sequence.

2. A space manipulator task planning method according to claim 1, characterized in that: The six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

3. A space manipulator task planning method according to claim 2, characterized in that: The specific process of determining the full orbital position sequence table of the spacecraft on the orbit based on the six numbers is as follows: Determining the eccentric anomaly of the orbit by using the eccentricity of the orbit and the true anomaly of the orbit; Determining a first anomaly angle after a preset period based on the anomaly angle of the orbit; Determining a first true anomaly after a preset period based on a geometric relationship between the first eccentric anomaly and the first true anomaly; Replacing the true anomaly of the orbit in the six numbers with the first true anomaly to obtain new six numbers, and determining the position information of the spacecraft on the orbit after a preset period based on the new six numbers; The first eccentric anomaly angle after the next preset period is calculated until the spacecraft completes one orbit, and the position information after each preset period is obtained, and all the position information is collected to form the full orbit position sequence table.

4. A space manipulator task planning method according to claim 1, characterized in that: The process of determining the distribution of the period of time when the space manipulator stays on the base after executing the mission instruction and there is an obstruction in the full orbit position sequence table in the space station measurement and control period is as follows: Based on the configuration of the space robot arm residing on the base, the position information of all spacecraft in the full orbit position sequence table is traversed, and the position information of the spacecraft at any time and the space robot arm are used to determine the occlusion situation through geometric simulation method. When there is an occlusion situation, the start time and the end time of the occlusion are recorded, and the occlusion period is determined according to the start time and the end time of the occlusion, and the distribution of the occlusion period in the measurement and control period of the space station is determined. When there is no occlusion situation, no processing and recording are performed.

5. A space manipulator task planning method according to claim 1, characterized in that: The process of planning and processing to obtain the instruction sequence based on the complete set of basic instructions for the space manipulator operation task includes: Determine the task content, task objectives and task attributes corresponding to each instruction in the complete set of basic instructions for the space manipulator operation task; Determine the subsystem and action corresponding to each instruction; Based on the constraints in the task planning domain, each instruction is analyzed to obtain the prerequisites and effects corresponding to each instruction; Build a domain description file based on the subsystem, task objectives, and task content corresponding to each instruction; Using the PDDL planning model, the task content, task objectives, task attributes, subsystems and actions corresponding to each instruction are taken as input to obtain the instruction sequence.

6. A space manipulator task planning method according to claim 1, characterized in that: The process of planning and processing based on the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence also includes: Taking a space robot arm operation remote control instruction sequence set as a sample set, training the sample set by an unsupervised learning method and a self-supervised learning method to obtain a trained network model; The trained network model is used to process the complete set of basic instructions for the space manipulator operation task to obtain an instruction sequence.

7. A space manipulator task planning system, characterized in that: include: The first acquisition module is used to: acquire six numbers of the orbit at any time, and determine the full orbit position sequence table of the spacecraft on the orbit based on the six numbers; The determination module is used to: determine the distribution of the period of time when the space manipulator, after executing the task instruction, resides on the base and has an obstruction with the full orbit position sequence table in the space station measurement and control period; The second acquisition module is used to: acquire a complete set of basic instructions for space manipulator operation tasks, and perform planning and processing based on the complete set of basic instructions for space manipulator operation tasks to obtain an instruction sequence; The planning module is used to obtain the operation task planning result of the space robot arm based on the distribution situation and the instruction sequence.

8. A space manipulator task planning system according to claim 7, characterized in that: The six numbers include: the semi-major axis of the orbit, the eccentricity of the orbit, the inclination of the orbit, the right ascension of the ascending node of the orbit, the angular distance of the perigee of the orbit, and the true anomaly of the orbit.

9. A computer device, characterized in that: The computer device includes a processor, the processor is coupled to a memory, at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the computer device implements a space robot arm operation task planning method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: At least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by the processor so that the computer implements a space robot arm operation task planning method as described in any one of claims 1 to 6.

Images