Modularized robot motion unit form design and optimization method

Abstract

The invention provides a modular robot motion unit form design and optimization method, which comprises the following steps of: extracting multi-type typical robot configuration characteristics facing specific task requirements to obtain a configuration characteristic set; designing a configuration characterization rule, characterizing the configuration features as directed topological graphs, and obtaining a directed topological graph set; basic nodes capable of forming all topological graphs in the directed topological graph set are extracted, a modular robot motion unit form mapping and generating rule is formulated, and a basic form of a motion unit is obtained; representing morphological characteristics of the motion unit, and obtaining morphological characteristic optimization parameters of the motion unit; and constructing a motion unit form optimization strategy according to the motion unit form characteristic optimization parameters, and obtaining the optimal form of the motion unit through an adaptive differential evolution algorithm. According to the technical scheme provided by the embodiment of the invention, rapid design and optimization of the modular robot motion unit form can be realized.

Description

Technical Field

[0001] The present invention relates to a method for the morphological design and optimization of a modular robot motion unit, belonging to the technical field of robot design and control.

Background Art

[0002] With the continuous improvement of industrial automation, intelligent manufacturing, and special operation requirements, modular robots have become an important development direction for future robot technologies due to their advantages such as reconfigurable configuration, environmental adaptability, and low-cost maintenance. Facing complex dynamic environments such as disaster rescue, space exploration, and flexible production lines, traditional robots with fixed configurations of "special machines for special purposes" are difficult to meet diverse task requirements. Modular robots can adapt to different tasks through different combinations of motion units, have the ability of "one machine with multiple functions", and can significantly improve multi-task coverage and economic benefits.

[0003] However, the multi-task adaptability and task execution ability of modular robots highly depend on the morphological design of their motion units, including the main external shape, the number and arrangement of degrees of freedom, the number and arrangement of interfaces, etc. of the morphological characteristics of the motion units. Existing morphological design methods mostly adopt bionic design (such as the SnailBot snail-like design, the NAR alkane structure-like design, etc.) or forward design schemes of morphological combination, that is, morphological design is achieved through simple combination of morphological characteristics, which makes the multi-task adaptability of the modular robots composed of them poor. Specifically, the module units designed forward need to optimize their morphological composition in different task environments, and the switching of task environments will destroy some of the previously optimized morphologies, so that morphological optimization does not have continuity and task generalization. Therefore, how to be guided by multi-task requirements, through reverse mapping, to achieve the autonomous generation and optimization of the morphology of modular robot motion units from task requirements, so as to improve the task adaptability and generalization ability of modular robots, is still a key technical problem to be solved at the present stage.

Summary of the Invention

[0004] In view of this, the present invention provides a method for the morphological design and optimization of a modular robot motion unit, realizing the mapping generation and optimization from multi-task requirements to the morphology of the motion unit, and at the same time improving the multi-task adaptability and generalization ability of the module unit.

[0005] The present invention provides a method for the morphological design and optimization of a modular robot motion unit, including:

[0006] Step S1: Extract various types of typical robot configuration features for specific task requirements to obtain a configuration feature set;

[0007] Step S2: According to the configuration feature set, design a configuration representation rule to represent the configuration features as a directed topological graph to obtain a directed topological graph set;

[0008] Step S3 extracts the basic nodes that can form all the topological graphs in the set of directed topological graphs according to the set of directed topological graphs;

[0009] Step S4 formulates the morphological mapping and generation rules of the modular robot motion unit based on the extracted basic nodes to obtain the basic form of the motion unit;

[0010] Step S5 characterizes the morphological features of the motion unit based on the basic form of the motion unit to obtain the optimized parameters of the motion unit morphological features;

[0011] Step S6 constructs the morphological optimization strategy of the motion unit based on the optimized parameters of the motion unit morphological features to obtain the optimal form of the motion unit.

[0012] In the above method, the step S1 includes:

[0013] According to the various types of tasks that the modular robot needs to perform, sort out and analyze the configuration characteristics of typical robots to form a configuration set:

[0014] A = {a1, a2,..., a n}

[0015] where n is the number of robot configurations that can complete the various types of tasks.

[0016] In the above method, the step S2 includes:

[0017] S2.1 designs the configuration characterization rule f a→g:rule , specifically including:

[0018] Characterize the links of the robot as the nodes v of the directed topological graph, and construct the node set V = {v1, v2,... v m} where m is the total number of nodes;

[0019] Characterize the joints of the robot as the edges e of the directed topological graph, and construct the edge set E = {e1, e2,... e q} where q is the total number of edges;

[0020] Characterize the number of input joints connected by the robot link as the in-degree of the node of the directed topological graph, denoted as deg + (v);

[0021] Characterize the number of output joints connected by the robot link as the out-degree of the node of the directed topological graph, denoted as deg - (v);

[0022] Characterize the number of all joints connected by the robot link as the degree of the node of the directed topological graph, denoted as deg(v);

[0023] S2.2 According to the configuration representation rules, represent the configuration feature set of the robot as a directed graph topology set, which is expressed as:

[0024] G = f a→g:rule (A)

[0025] Among them, G=(V, E) represents the directed topology graph set.

[0026] In the above method, the step S3 includes:

[0027] According to the directed topology graph set, design the basic node extraction strategy f(u) as:

[0028]

[0029] According to the extraction strategy, extract the basic nodes u that can form all the topology graphs in the directed topology graph set as the basic nodes.

[0030] In the above method, the step S4 includes:

[0031] S4.1 According to the extracted basic nodes u, formulate the form mapping and generation rules f of the modular robot motion unit u→M:rule , specifically:

[0032] Map the basic node u to the body B of the motion unit;

[0033] Map the edges connected to the basic node u to the docking interface C of the motion unit. The axis of the docking interface passes through the center of the body B, and the axes of any two docking interfaces are perpendicular or collinear;

[0034] Map the in-degree deg + (u) of the basic node u to the number n of input interfaces of the motion unit I , and map the input edge where the in-degree is located to the input interface C I ;

[0035] Map the out-degree deg - (u) of the basic node u to the number n of output interfaces of the motion unit O , and map the output edge where the out-degree is located to the output interface C O ;

[0036] Map the connection attribute of two basic nodes to the connection degree of freedom DoF C , and allocate the connection degree of freedom to the output interface C O , and the axis of the DoF C coincides with the axis of the C O ;

[0037] Construct the auxiliary degree of freedom DoF BAdjust the pose relationship between each docking interface to assist the degree of freedom DoF B The axis passes through the center of the body B;

[0038] S4.2 According to the form mapping and generation rules, map the basic node u to the basic form of the modular robot motion unit, which is expressed as:

[0039] M = f u→M:rule (u)

[0040] Among them, M = (B, C I , C O , n I , n O , DoF C , DoF B ) represents the basic form of the motion unit.

[0041] In the above method, the step S5 includes:

[0042] S5.1 According to the basic form of the motion unit, characterize the outer shape of its body B as:

[0043] S ∈ {S3, S4, S6, S8, S 12 , S ∞}

[0044] Among them, S i represents a polyhedron with i faces, and S ∞ represents a sphere;

[0045] S5.2 Characterize the input interface C I and the output interface C O The axis direction vector is:

[0046]

[0047] Among them, ζ CI represents the direction vector of the axis of the input interface C I in the central base coordinate system of the body B, and ζ CO represents the direction vector of the axis of the output interface C O in the central base coordinate system of the body B, represents the two-norm of the vector , and the direction vector relationship satisfies:

[0048]

[0049] S5.3 Characterize the degrees of freedom DoF C and DoF B The axis vector and the constraint relationship are:

[0050]

[0051] Among them, represents the interface degree of freedom DoF C is the direction vector in the central base coordinate system of the body B, represents the auxiliary degree of freedom DoF C is the direction vector in the central base coordinate system of the body B, and α is the auxiliary degree of freedom DoF C is the angle between the axis and the y-axis of the central base coordinate system of the body B;

[0052] According to the characterization results of the body shape, docking interface and degrees of freedom, determine the morphological feature optimization parameter x = (s, α) of the motion unit, where s ∈ S is the shape parameter and α ∈ [-π / 2, π / 2] is the auxiliary degree of freedom DoF C is the angle between the axis and the y-axis of the central base coordinate system of the body B;

[0053] S5.4 According to the characterization results of the body shape, docking interface and degrees of freedom, combined with the screw theory, construct the position-level forward kinematic model of the motion unit as:

[0054]

[0055] Among them, k is the number of degrees of freedom in the selected motion transfer path of the motion unit, θ i is the rotation angle of the degree of freedom, is the Lie group representation of the screw coordinates, and T0 is the homogeneous transformation matrix of the initial configuration;

[0056] According to the position-level forward kinematic model, establish the velocity-level kinematic model of the motion unit as:

[0057]

[0058] Among them, is the velocity at the end of the motion transfer chain of the motion unit, is the joint velocity in its joint space, and J is its Jacobian matrix.

[0059] In the above method, the step S6 includes:

[0060] According to the morphological feature optimization parameter of the motion unit, design the morphological optimization model of the motion unit:

[0061]

[0062] Among them, σ(x) is the maximum stress of the motion unit shape under a certain morphological feature combination, [σ] is the limit stress of the motion unit shape material, f E (x) = E f +E m +E σTaking [objective function] as the objective function, according to the kinematic model of the velocity level of the motion unit, the expressions of various performance functions are derived as follows:

[0063]

[0064] Among them, E f represents the motion flexibility index, E m represents the motion isotropy index, E σ represents the mechanical isotropy index, ||J|| F represents the Frobenius norm of the Jacobian matrix J, J + represents the pseudo-inverse of J, λ max (J) is the maximum singular value of J, λ min (J) is the minimum singular value of J, σ(x) is the stress value on the shape, σ avg (x) is the mean value of the stress on the shape, σ(x) and σ avg (x) are both obtained through finite element simulation, ||σ(x) - σ avg (x)||2 represents the two-norm of the difference between the stress value σ(x) and the stress mean value σ avg (x).

Description of the Drawings

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative and laborious efforts, other drawings can also be obtained based on these drawings.

[0066] Figure 1 is a schematic flow chart of the morphological design and optimization method of the modular robot motion unit provided by the embodiment of the present invention;

[0067] Figure 2 is a process diagram of the basic node mapping and generation of the directed topological graph in the embodiment of the present invention;

[0068] Figure 3 is the initial morphological diagram of the modular robot motion unit in the embodiment of the present invention;

[0069] Figure 4 is the morphological parameter characterization diagram of the modular robot motion unit in the embodiment of the present invention;

[0070] Figure 5 is the iterative optimization curve of the morphological parameters of the modular robot motion unit in the embodiment of the present invention;

[0071] Figure 6 is the schematic diagram of the optimal morphology of the modular robot motion unit in the embodiment of the present invention;

[0072] Figure 7 It is a schematic diagram of the process of a modular robot composed of motion units in an embodiment of the present invention executing a long sequence of multi-type tasks for truss assembly.

Specific Embodiment

[0073] To better understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0074] It should be clear that the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work belong to the scope of protection of the present invention.

[0075] An embodiment of the present invention provides a method for the form design and optimization of a modular robot motion unit, as Figure 1 shown, which is a schematic flow diagram of the form design and optimization of a modular robot motion unit provided by an embodiment of the present invention. The method includes the following steps:

[0076] Step A1: Extract various types of typical robot configuration features for specific task requirements to obtain a configuration feature set.

[0077] Specifically, according to the various types of tasks that the modular robot needs to execute, analyze and sort out the configuration features of typical robots to form a configuration set:

[0078] A = {a1, a2,..., a n}

[0079] where n is the number of robot configurations for executing the various types of tasks.

[0080] Step A2: According to the configuration feature set, design a configuration representation rule to represent the configuration features as a directed topological graph to obtain a set of directed topological graphs, which specifically includes the following steps:

[0081] Step A2.1: According to the configuration feature set, design a configuration representation rule f a→g:rule , which specifically includes:

[0082] Represent the link of the robot as a node v of the directed topological graph, and construct a node set V = {v1, v2,... v m}, where m is the total number of nodes;

[0083] Represent the robot joint as an edge e of the directed topological graph, and construct an edge set E = {e1, e2,... e q}, where q is the total number of edges;

[0084] Characterize the number of input joints connected by a robotic link as the in-degree of a node in a directed topological graph, denoted as deg + (v);

[0085] Characterize the number of output joints connected by a robotic link as the out-degree of a node in a directed topological graph, denoted as deg - (v);

[0086] Characterize the number of all joints connected by a robotic link as the degree of a node in a directed topological graph, denoted as deg(v);

[0087] Step A2.2 According to the configuration characterization rules, characterize the configuration feature set of the robot as a directed graph topological set, which is expressed as:

[0088] G = f a→g:rule (A)

[0089] where G = (V, E) represents the directed topological graph set.

[0090] Step A3 According to the directed topological graph set, extract the basic nodes that can form all the topological graphs in the directed topological graph set.

[0091] Specifically, according to the directed topological graph set, design the basic node extraction strategy f(u) as:

[0092]

[0093] According to the extraction strategy, extract the basic node u that can form all the topological graphs in the directed topological graph set as the basic node.

[0094] Specifically, according to the methods provided in steps A1 to A3 of the embodiment of the present invention, taking the four tasks (T) of load grasping and assembly, multi-arm coordinated assembly operation, load handling, and compliant operation of in-cabin auxiliary experimental instruments as examples, conduct a characterization analysis experiment from task requirements to the basic node u. The characterization analysis experiment is as Figure 2 shown. First, based on the task requirements, sort out the typical spatial robot configurations for performing these four tasks at the current stage, which mainly include a single-arm configuration a1 (such as the CMM robotic arm used for performing single-point load handling and assembly of an on-orbit large experimental platform), a two-arm configuration a2 (such as the JERMS robotic arm mainly used for performing coordinated assembly operation tasks), a three-branch configuration a3 (such as the SPDM robotic arm mainly used for load handling and fine assembly tasks), and a multi-branch configuration a4 (such as the Skybot F-850 robot mainly used for performing flexible operation tasks of in-cabin auxiliary experimental instruments), to form the following configuration set:

[0095] A = {a1, a2, a3, a4}

[0096] According to the configuration characterization rule fa→g:rule , characterize the configuration set A as a set of directed graph topologies:

[0097]

[0098] According to the extraction strategy f(u):

[0099]

[0100] Extract all the basic nodes in the set of directed graph topologies, and finally obtain the basic node u with an in-degree of 1, an out-degree of 3, and a total degree of 4.

[0101] Step A4 formulates the morphological mapping and generation rules of the modular robot motion unit based on the extracted basic nodes to obtain the basic form of the motion unit, which specifically includes the following steps:

[0102] Step A4.1 formulates the morphological mapping and generation rule f of the modular robot motion unit based on the extracted basic node u u→M:rule , specifically:

[0103] The basic node u is mapped to the body B of the motion unit;

[0104] The edges connected to the basic node u are mapped to the docking interfaces C of the motion unit. The axis of the docking interface passes through the center of the body B, and the axes of any two docking interfaces are perpendicular or collinear to each other;

[0105] The in-degree deg + (u) is mapped to the number n of input interfaces of the motion unit I , and the input edges where the in-degree is located are mapped to the input interfaces C I ;

[0106] Map the out-degree deg of the basic node u - (u) to the number n of output interfaces of the motion unit O , and the output edges where the out-degree is located are mapped to the output interfaces C O ;

[0107] Map the connection attribute of two basic nodes to the connection degree of freedom DoF C , and the allocation of the connection degree of freedom is given to the output interface C O , the DoF C axis coincides with the C O axis;

[0108] Construct an auxiliary degree of freedom DoF B to adjust the pose relationship between each docking interface. The axis of the auxiliary degree of freedom DoF B passes through the center of the body B;

[0109] Step A4.2 maps the basic node u to the basic form of the modular robot motion unit according to the form mapping and generation rules, which is expressed as:

[0110] M = f u→M:rule (u)

[0111] where M = (B, C I , C O , n I , n O , DoF C , DoF B ) represents the basic form of the motion unit.

[0112] Specifically, according to the method provided in step A4 of the embodiment of the present invention, the basic node u is mapped to the basic form of the modular robot motion unit, and the basic form of the mapped motion unit is as Figure 3 shown. The mapped motion unit has a total of 4 docking interfaces, and the axes of the 4 docking interfaces all pass through the center of the body B. Among them, 3 are output interfaces C Oi (i = 1, 2, 3), and 1 is an input interface C I1 , and there is a rotational degree of freedom DoF Ci (i = 1, 2, 3) on the output interface, and at the same time has an auxiliary degree of freedom DoF B , and the axes of the interface degree of freedom DoF Ci (i = 1, 2, 3) and the auxiliary degree of freedom DoF B also pass through the center of the body B. At the same time, the angle between the axis of the auxiliary degree of freedom DoF C and the y-axis of the central base coordinate system of the body B is α.

[0113] Step A5 characterizes the morphological features of the motion unit based on the basic form of the motion unit, and obtains the optimized parameters of the morphological features of the motion unit.

[0114] Specifically, the morphological features of the motion unit are as Figure 4 shown. The specific characterization steps are as follows:

[0115] Step A5.1 characterizes the outer shape of its body B based on the basic form of the motion unit as:

[0116] S ∈ {S3, S4, S6, S8, S 12 , S ∞}

[0117] where S i represents a polyhedron with i faces, and S ∞ represents a sphere;

[0118] Step A5.2 characterizes the input interface C I and the output interface CO The axial direction vector is:

[0119]

[0120] where ζ CI represents the direction vector of the axis of the input interface C I in the central base coordinate system of the body B, and ζ CO represents the direction vector of the axis of the output interface C O in the central base coordinate system of the body B, represents the two-norm of the vector, and the relationship of the direction vectors satisfies:

[0121]

[0122] Step A5.3 characterizes the degrees of freedom DoF of the motion unit C and DoF B The axis vector and the constraint relationship are:

[0123]

[0124] where represents the direction vector of the interface degree of freedom DoF C in the central base coordinate system Σ0 of the body B. Specifically, there are three interface degrees of freedom, which are respectively characterized as represents the direction vector of the auxiliary degree of freedom DoF C in the central base coordinate system Σ0 of the body B, and α is the angle between the axis of the auxiliary degree of freedom DoF C and the y-axis of the central base coordinate system Σ0 of the body B;

[0125] According to the characterization results of the body shape, docking interface and degrees of freedom, determine the morphological feature optimization parameter x = (s, α) of the motion unit, where s ∈ S is the shape parameter and α ∈ [-π / 2, π / 2] is the auxiliary degree of freedom DoF C the angle between the axis and the y-axis of the central base coordinate system y of the body B;

[0126] Step A5.4 According to the characterization results of the body shape, docking interface and degrees of freedom, combined with screw theory, construct the position-level forward kinematics model of the motion unit as:

[0127]

[0128] where k is the number of degrees of freedom in the selected motion transfer path of the motion unit, and θ i is the rotation angle of the degree of freedom, is the Lie group representation of the screw coordinates, and T0 is the homogeneous transformation matrix of the initial configuration;

[0129] Based on the position-level forward kinematics model, a kinematics model at the velocity level of the motion unit is established as follows:

[0130]

[0131] Wherein, is the end velocity of the motion unit, is the joint velocity in the joint space, and J is the Jacobian matrix.

[0132] Step A6 optimizes the parameters according to the morphological characteristics of the motion unit, constructs a morphological optimization strategy for the motion unit, and obtains the optimal morphology of the motion unit.

[0133] Specifically, the parameters are optimized according to the morphological characteristics of the motion unit, and a morphological optimization model is designed:

[0134]

[0135] Wherein, σ(x) is the maximum stress on the outer shape of the motion unit under a certain combination of morphological characteristics, [σ] is the ultimate stress of the outer shape material of the motion unit, and f E (x) = E f + E m + E σ is the objective function. According to the kinematics model at the velocity level of the motion unit, the expressions of various performance functions are derived as follows:

[0136]

[0137] Wherein, E f represents the motion flexibility index, E m represents the motion isotropy index, E σ represents the mechanical isotropy index, ||J|| F represents the Frobenius norm of the Jacobian matrix J, J + represents the pseudo-inverse of J, λ max (J) is the maximum singular value of J, λ min (J) is the minimum singular value of J, σ(x) is the stress value on the outer shape, σ avg (x) is the mean value of the stress on the outer shape, σ(x) and σ avg (x) are both obtained through finite element simulation, and ||σ(x) - σ avg (x)||2 represents the two-norm of the difference between the stress value σ(x) and the stress mean value σ avg (x).

[0138] Specifically, according to the method provided in step A6 of the embodiment of the present invention, the optimization model of the motion unit is constructed. On this basis, the optimization model of the motion unit is optimized and solved by improving the adaptive differential evolution algorithm, and the solution results are as followsFigure 5 As shown. After 50 iterations of solution, the algorithm converges. The optimal morphological parameters of the modular robot motion unit are shown in Table 1. Its optimal auxiliary degree-of-freedom angle is 48.7°, and the optimal shape is spherical.

[0139] Table 1 Optimal Morphological Parameters of the Motion Unit

[0140]

[0141] Generally, to reduce the processing difficulty and increase the interchangeability of parts, the morphology of the modular robot motion unit is as regular and symmetric as possible. Therefore, the auxiliary degree-of-freedom angle value is approximated to the nearest regular angle (such as 0°, 30°, 45°, 60°, 90°). In the embodiment of the present invention, the auxiliary degree-of-freedom angle value of the motion unit is taken as 45°. When the auxiliary degree-of-freedom angle value is 45°, the docking interfaces C O2 and C O3 are equivalent, then the degree of freedom of the docking interface C O3 can be removed, and the final formed morphology of the motion unit is as Figure 6 shown.

[0142] According to the method provided by the embodiment of the present invention, the design and optimization from task requirements to the morphology of the modular robot motion unit are realized. At the same time, based on the motion unit with the optimal morphology, an on-orbit truss task execution efficiency verification experiment is carried out. The verification results are as Figure 7 shown. From Figure 7 (a)-(i), it can be seen that through the modular robot motion unit designed and optimized by the embodiment of the present invention, various configuration transformations can be successfully completed, and at the same time, various types of long-sequence tasks such as double-arm grasping, single-arm grasping, double-arm collaborative assembly, three-branch load-bearing crawling, and precise compliant assembly can be realized. This long-sequence and various types of tasks completely cover the execution actions of the initial four types of tasks, that is, the morphology of the motion unit designed and optimized by the embodiment of the present invention has the adaptability to the foregoing tasks, so the effectiveness of this method is proved.

[0143] The technical solution of the embodiment of the present invention has the following beneficial effects:

[0144] In the technical solution of the embodiment of the present invention, for the four types of typical task requirements for on-orbit services, by extracting various types of typical robot configuration features of these four types of task requirements, a configuration feature set including single-chain, double-branch, triple-branch, and multi-branch is obtained. By designing configuration representation rules, a set of directed topological graphs is obtained. By extracting the basic nodes that can form all the topological graphs in the set of directed topological graphs, a basic node with a degree of 4 is obtained. By formulating the form mapping and generation rules of the modular robot motion unit, the basic form of the motion unit is obtained. By characterizing the form features of the motion unit, the form feature optimization parameters of the motion unit are obtained, and by constructing the form optimization strategy of the motion unit, the optimal form of the motion unit is obtained. According to this technical solution, the rapid design and optimization of the form of the modular robot motion unit can be realized. The designed module unit can not only reliably complete various types of tasks in the long sequence of truss assembly, but also autonomously realize topological structure transformation during the switching process of multiple tasks, thus ensuring the smooth progress of the overall task.

[0145] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the scope of protection of the present invention.

[0146] The content not detailedly described in the specification of the present invention belongs to the well-known technology of those skilled in the art.

Claims

1. A method for the morphological design and optimization of a modular robot motion unit, characterized in that The method includes the following steps: Step S1: Extract various types of typical robot configuration features for specific task requirements to obtain a configuration feature set. Step S2: According to the configuration feature set, design a configuration representation rule to represent the configuration features as a directed topological graph, and obtain a set of directed topological graphs. Step S3: According to the set of directed topological graphs, extract the basic nodes that can form all the topological graphs in the set of directed topological graphs. Step S4: According to the extracted basic nodes, formulate a morphological mapping and generation rule for the modular robot motion unit to obtain the basic form of the motion unit. Step S5: According to the basic form of the motion unit, represent the morphological features of the motion unit to obtain the optimized parameters of the motion unit morphological features. Step S6: According to the optimized parameters of the motion unit morphological features, construct a morphological optimization strategy for the motion unit to obtain the optimal form of the motion unit.

2. The method according to claim 1, characterized in that The said Step S1 includes: Based on various types of tasks that the modular robot needs to execute, sort out and analyze the configuration features of typical robots to form a configuration set: A = {a1, a2,..., a n} where n is the number of robot configurations for executing the various types of tasks.

3. The method according to claim 1, wherein The said Step S2 includes: S2.1 Design the configuration representation rule f according to the set of configuration features a→g:rule , which specifically includes: Characterize the links of the robot as the nodes v of a directed topological graph, and construct the node set V = {v1, v2, … v m}, where m is the total number of nodes; Characterize the robot joints as the edges e of a directed topological graph, and construct an edge set E = {e1, e2, … e q}, where q is the total number of edges; Characterize the number of input joints connected to a robotic link as the in-degree of a node in a directed topological graph, denoted as deg + (v); Characterize the number of output joints connected by a robotic link as the out-degree of a node in a directed topological graph, denoted as deg - (v); Represent the number of all joints connected by the robot link as the degree of the node of the directed topological graph, denoted as deg(v). S2.2 According to the configuration representation rule, represent the configuration feature set of the robot as a set of directed graph topologies, which is expressed as: G = f a→g:rule (A) where G=(V,E) represents the set of directed topological graphs.

4. The method according to claim 1, wherein The said Step S3 includes: Based on the set of directed topological graphs, design a basic node extraction strategy as follows: According to the extraction strategy, extract the basic node u that can form all the topological graphs in the set of directed topological graphs as the basic node.

5. The method according to claim 1, characterized in that, The said Step S4 includes: S4.1 Based on the extracted basic node u, formulate the morphological mapping and generation rule f of the modular robot motion unit u→M:rule , specifically: Map the basic node u to the body B of the motion unit. Map the edges connected to the basic node u to the docking interfaces C of the motion unit. The axis of the docking interface passes through the center of the body B, and the axes of any two docking interfaces are perpendicular or collinear to each other. The in-degree deg of the basic node u + is mapped to the number n of input interfaces of the motion unit I , and the input edge where the in-degree is located is mapped to the input interface C I ; Map the out-degree deg - (u) of the basic node u to the number n of output interfaces of the motion unit O , and map the output edges where the out-degree is located to the output interface C O ; Map the connection attributes of two basic nodes to the connection degree of freedom DoF C , and allocate the connection degree of freedom to the output interface C O , the said DoF C axis coincides with the C O axis; Construct auxiliary degrees of freedom (DoF) B Adjust the pose relationship between each docking interface, auxiliary degrees of freedom (DoF) B The axis passes through the center of the main body B; S4.2 According to the morphological mapping and generation rule, map the basic node u to the basic form of the modular robot motion unit, which is expressed as: M = f u→M:rule (u) where M = (B, C I , C O , n I , n O , DoF C , DoF B ) represents the basic form of the motion unit.

6. The method according to claim 1, wherein The said Step S5 includes: S5.1 According to the basic form of the motion unit, represent the outer shape of the body B as: S ∈ {S3, S4, S6, S8, S 12 , S ∞} Among them, S i represents a polyhedron with i faces, and S ∞ represents a sphere; S5.2 Characterize the input interface C of the motion unit I and the output interface C O The axis direction vector is: Among them, represents the input interface C I is the direction vector of the axis in the central base coordinate system of the body B, represents the output interface C O is the direction vector of the axis in the central base coordinate system of the body B, represents the vector is the two-norm, and the direction vector relationship satisfies: S5.3 Characterize the degrees of freedom (DoF) of the motion unit C and DoF B The axis vectors and constraint relationships are as follows: Among them, represents the interface degree of freedom DoF C is the direction vector in the central base coordinate system of the body B, represents the auxiliary degree of freedom DoF C is the direction vector in the central base coordinate system of the body B, and α is the auxiliary degree of freedom DoF C is the angle between the axis and the y-axis of the central base coordinate system of the body B; Based on the characterization results of the outer shape of the body, the docking interface, and the degrees of freedom, determine the morphological feature optimization parameter x = (s, α) of the motion unit, where s ∈ S is the shape parameter and α ∈ [-π / 2, π / 2] is the auxiliary degree of freedom DoF C The angle between the axis and the y-axis of the central base coordinate system of the body B; S5.4 According to the representation results of the body outer shape, docking interface and degrees of freedom, combined with screw theory, construct a position-level forward kinematics model of the motion unit as: where k is the number of degrees of freedom in the selected motion transfer path of the motion unit, θ i is the rotation angle of the degree of freedom, is the Lie group representation of the screw coordinates, and T0 is the homogeneous transformation matrix of the initial configuration; According to the position-level forward kinematics model, establish a velocity-level kinematics model of the motion unit as: Among them, is the end velocity of the motion transmission chain of the motion unit, is the joint velocity in its joint space, and J is its Jacobian matrix.

7. The method according to claim 1, wherein The said Step S6 includes: According to the optimized parameters of the motion unit morphological features, design a motion unit morphological optimization model: Among them, σ(x) is the maximum stress on the outer shape of the motion unit under a certain combination of morphological features, [σ] is the ultimate stress of the material of the outer shape of the motion unit, and f E (x) = E f + E m + E σ is the objective function. According to the kinematic model of the speed level of the motion unit, the expressions of various performance functions are derived as follows: Among them, E f represents the motion flexibility index, E m represents the motion isotropy index, E σ represents the mechanical isotropy index, ||J|| F represents the Frobenius norm of the Jacobian matrix J, J + represents the pseudo-inverse of J, λ max (J) is the maximum singular value of J, λ min (J) is the minimum singular value of J, σ(x) is the stress value on the shape, σ avg (x) is the mean value of the stress on the shape, σ(x) and σ avg (x) are both obtained through finite element simulation, ||σ(x) - σ avg (x)||2 represents the two-norm of the difference between the stress value σ(x) and the stress mean value σ avg (x).

Images