Macro-micro mechanical arm task planning method

Abstract

The invention provides a macro-micro mechanical arm time-sharing task planning method which comprises the following steps: constructing macro-micro mechanical arm task characterization according to constraints among operation steps in an on-orbit task and macro-micro mechanical arm actions; constructing a constraint equation of the macro-micro mechanical arm according to the connection relation and motion characteristics of the space macro-micro mechanical arm; constructing a macro mechanical action similarity index according to the time-sharing motion mode of the macro-micro mechanical arm; according to the connection relation and the working mode of the macro mechanical arm and the micro mechanical arm, a macro mechanical action and working point matching index is constructed; and designing a macro-micro mechanical arm action generation method by considering a macro-micro mechanical arm task representation model, a macro-micro mechanical arm constraint equation, a macro mechanical action and operation point matching index and a macro-micro mechanical action similarity index. According to the technical scheme provided by the embodiment of the invention, a feasible macro-micro mechanical arm action sequence with few execution steps and short operation time is generated, so that the macro-micro mechanical arm can meet the characteristics of large span, complex process and the like of an on-orbit task in the future.

Description

[Technical field]

[0001] The invention belongs to the field of robot arm task planning and relates to a macro-micro robot arm task planning method. [Background technology]

[0002] As an important on-orbit service platform, the space station can effectively extend the on-orbit service life of satellites by replenishing fuel and replacing satellite components, thereby reducing the cost of space exploration. These on-orbit operation missions targeting satellites have the characteristics of long operating distance and high precision. Most existing space stations deploy two types of robotic arms to handle different on-orbit tasks - large robotic arms represented by Canadarm2 and dexterous robotic arms represented by Dextre. However, existing space robotic arms are unable to independently handle long-distance on-orbit delicate tasks. The emergence of macro-micro robotic arms provides a feasible technical solution for achieving the above tasks and expands the on-orbit service capabilities of space stations.

[0003] The macro-micro robotic arm is composed of two different robotic arms connected in series. The macro robotic arm has a strong carrying capacity and a large working range, and is mainly responsible for the large-scale transfer of the micro robotic arm. The micro robotic arm has the characteristics of fast response speed and high operation accuracy, and is the main executor of on-orbit operation tasks. The macro-micro robotic arm includes two motion modes: time-sharing motion and coordinated motion. The time-sharing motion control method is relatively simple, with fewer controlled joint movements, less energy required, and higher stability; while the coordinated motion mode has higher flexibility and a more complex motion control algorithm. At present, the macro-micro robotic arm mainly completes on-orbit tasks in a time-sharing motion mode.

[0004] However, each action switching between the macro- and micro-robotic arms requires additional time for starting and stopping and suppressing vibration, which greatly prolongs the task execution time of the macro- and micro-robotic arms. In addition, the various timing constraints in on-orbit tasks increase the difficulty of autonomous planning of space robots, making it difficult to directly apply existing task planning methods to macro- and micro-robotic arm systems to cope with various on-orbit tasks. [Summary of the invention]

[0005] In view of this, the present invention provides a robot arm task planning method, which is used to perform task planning for a macro-micro robot arm.

[0006] The present invention provides a robot arm task planning method, comprising:

[0007] According to the constraint information between the operation steps in the on-orbit mission and the motion data of the macro-micro robotic arm, a macro-micro robotic arm task characterization model is constructed; the macro-micro robotic arm includes a macro robotic arm and a micro robotic arm;

[0008] According to the connection relationship and motion characteristics of the macro-micro robotic arm, the constraint equation of the macro-micro robotic arm is constructed;

[0009] Based on the motion data of the macro robot arm, a similarity index of the macro robot motion is constructed;

[0010] According to the connection relationship and working mode of the macro-micro robotic arm, the matching index between the macro-mechanical action and the operation point is constructed;

[0011] Task planning is performed on the macro-micro-mechanical arm according to the task characterization model of the macro-micro-mechanical arm, the constraint equation of the macro-micro-mechanical arm, the matching index of the macro-mechanical action and the operation point, and the similarity index of the macro-micro-mechanical action.

[0012] In the above method, a macro-micro manipulator task characterization model is constructed based on the constraint information between the operation steps in the on-orbit mission and the macro-micro manipulator motion data, including:

[0013] According to the constraint information between the operation steps in the on-orbit task and the macro-micro manipulator motion data, a macro-micro manipulator task representation model of the following form is constructed for the on-orbit task containing n operation points and the macro-micro manipulator motion:

[0014] T={t1,…,t n}

[0015] t i = {PE i C i Visited i}

[0016] A={A b A s}

[0017]

[0018]

[0019] Among them, T represents the task to be performed, t i represents the i-th operation point in task T, PE i The pose information of the task is composed of a six-dimensional vector, C i Indicates the priority constraint of the operation, describing the operation order between tasks in terms of numerical value. i Indicates whether the state has been visited. i =1 means that the operation point has been planned. Each movement of the robot arm is regarded as an action, and it is assumed that the micro-robotic arm automatically starts to perform the operation after reaching the operation point. b and A s They represent the macro-manipulator motion sequence and the micro-manipulator motion sequence respectively; represents the i-th macro robot action, represents the jth macro robot action and is executed after the i-th macro robot action.

[0020] In the above method, the constraint equations of the macro-micro robotic arm are constructed according to the connection relationship and motion characteristics of the macro-micro robotic arm, including:

[0021] According to the connection relationship and motion characteristics of the space macro-micro robotic arm, a series of constraint equations of the macro-micro robotic arm are constructed, taking into account factors such as the dynamic performance of the robotic arm, the connection relationship between the two arms, and whether the micro robotic arm can carry out tasks, as shown below:

[0022] -π≤q ma_i ≤π,i=1,2,...,7

[0023] -π≤q mi_j ≤π,j=1,2,...,7

[0024] a b ∈W ma

[0025] a s ∈W mi

[0026] PE i -a b ∈W mi

[0027] Among them, q ma_i is the joint angle of the macro robot, q ma_i is the joint angle of the micro-manipulator to prevent the joints of the macro-micro-manipulator system from exceeding the limit, a b ∈W ma and a s ∈W mi The two constraint equations represent the macro robot motion and a b and micro-manipulator motion a s In their respective workspaces ma and W mi Inside. PE i -a b ∈W mi It is the task constraint of the operation point, that is, the distance between the operation point and the macro-robotic arm motion should be within the working space of the micro-robotic arm.

[0028] In the above method, based on the motion data of the macro-micro robotic arm, a macro-mechanical motion similarity index is constructed, including:

[0029] Based on the motion data of the macro-manipulator, in order to accurately and quantitatively describe the similarity between the macro-manipulator motions, a macro-manipulator motion is defined by comprehensively considering the distance between the macro-manipulator motions and the job arrangement under each macro-manipulator motion. and Similarity index between The specific definitions are as follows:

[0030]

[0031] Among them, w i ,w j They are the target poses of the two macro manipulators, ROCP i ,ROCP j They are the closeness of the operating points under each macro robot action, and are specifically defined as follows:

[0032]

[0033] c represents the constraint function of the macro-micro manipulator, and its specific representation is as follows:

[0034]

[0035] In the above method, the matching index between the macro-mechanical action and the operation point is constructed according to the connection relationship and working mode of the macro-micro-mechanical arm, including:

[0036] According to the connection relationship and working mode of the macro-micro manipulator, in order to accurately describe the matching degree between the macro-manipulator action and the operation point, and guide the planning algorithm to search and adjust the macro-manipulator action, we design the matching index between the macro-manipulator action and the operation point based on the macro-manipulator action similarity evaluation function. The specific definition is as follows:

[0037]

[0038] Among them, B, w and ROCP are the fused action constraint satisfaction function, the macro robot action target point and the closeness of the operating point under the macro robot action, respectively, and PE is the location information of the operating point.

[0039] In the above method, task planning is performed on the macro-micro robotic arm according to the macro-micro robotic arm task characterization model, the constraint equation of the macro-micro robotic arm, the matching index between the macro-mechanical action and the operation point, and the macro-mechanical action similarity index, including:

[0040] Initialize all the operation points in the macro-micro robotic arm task T and the state of the macro-micro robotic arm;

[0041] Determine the set of all currently executable work points t_set, that is, all work points with the highest priority constraints;

[0042] According to the operation point set t_set, a hierarchical clustering algorithm from bottom to top is used to generate a macro robot action set {a B} and the corresponding cluster {cluster};

[0043] According to the matching index between the macro mechanical action and the operation point, determine the macro mechanical arm action to be optimized a B And the corresponding operation point cluster;

[0044] Select the operation point with the highest matching index between the macro mechanical action and the operation point of the current macro mechanical arm;

[0045] If the matching index s=0, skip the current step. Otherwise, add the new operation point to the operation cluster for the action of the current macro-manipulator, and return to the step of selecting the operation point with the highest matching index between the macro-mechanical action and the operation point of the current macro-manipulator;

[0046] If there are still unexecuted work points in the macro-micro robotic arm task T, return to the step of determining the set of all currently executable work points t_set. Otherwise, according to the joint travel of the macro-micro robotic arm, the cluster traveling salesman problem (CTSP) and configuration optimization algorithm are used to generate the action sequence A of each arm in the macro-micro robotic arm. b and A s .

[0047] The macro-micro robotic arm task planning method proposed in the present invention reduces the difficulty of planning and solving by first determining the macro robotic arm action and then generating the micro robotic arm action; various constraints in on-orbit operation are handled through a clustering algorithm and an iterative mechanism, the action generation of the macro-micro robotic arm is realized and the number of motion switching of the macro-micro robotic arm is reduced, thereby avoiding extra time for starting and stopping the macro-micro robotic arm and suppressing vibration; finally, a cluster traveling salesman solution algorithm and a configuration optimization algorithm are used for optimization and solving, thereby obtaining a macro-micro robotic arm action sequence with the shortest task execution time.

Brief Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creativity and labor.

[0049] Figure 1 It is a flowchart of a macro-micro robotic arm task planning method provided by an embodiment of the present invention;

[0050] Figure 2 It is the macro-micro mechanical arm model used in the simulation experiment in the embodiment of the present invention.

[0051] Figure 3 This is the simulation task scenario in this experimental embodiment.

[0052] Figure 4 This is a visual display of the macro-micro robotic arm planning results in this experimental embodiment.

[0053] Figure 5 This is a visual display of the macro-micro robotic arm planning results solved by the conventional robotic arm [Specific embodiment]

[0054] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0055] It should be clear that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] The embodiment of the present invention provides a macro-micro robotic arm motion planning method based on motion iteration, please refer to Figure 1 , which is a flow chart of a macro-micro robotic arm motion planning method based on motion iteration provided by an example of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0057] Step 101, constructing a macro-micro robotic arm task characterization model based on the constraint information between the operation steps in the on-orbit mission and the macro-micro robotic arm motion data.

[0058] Specifically, according to the on-orbit mission constraints, the tasks to be performed are defined as T = {t1,…,t n}, n is the number of work points in the task, where t i Represents the i-th operation point in the task. It is defined as follows:

[0059] t i = {PE i C i Visited i}

[0060] Among them, PE i The pose information representing the task is composed of a six-dimensional vector, C i Indicates the priority constraint of the operation, describing the operation order between tasks in terms of numerical value. i Indicates whether the state has been visited. i =1 means that the operation point has been planned.

[0061] According to the motion data of the macro-micro robotic arm, the motion sequence of the macro-micro robotic arm is constructed as follows:

[0062] A={Ab A s}

[0063] Among them, A b and A s They represent the macro-manipulator motion sequence and the micro-manipulator motion sequence respectively. represents the i-th macro robot action, Represents the jth macro robot action and is executed after the ith macro robot action.

[0064] Step 102, constructing the constraint equation of the macro-micro robotic arm according to the connection relationship and motion characteristics of the macro-micro robotic arm.

[0065] Specifically, based on the connection relationship and motion characteristics of the space macro-micro robotic arm, the dynamic performance of the robotic arm, the connection relationship between the two arms, and whether the micro robotic arm can carry out tasks are considered, and a series of constraint equations of the macro-micro robotic arm are constructed, as follows:

[0066]

[0067] Among them, q ma_i is the joint angle of the macro robot, q ma_i is the joint angle of the micro-manipulator to prevent the joint of the macro-micro-manipulator from exceeding the limit, a b ∈W ma and a s ∈W mi The two constraint equations represent the macro robot motion and a b and micro-manipulator motion a s In their respective workspaces ma and W mi Inside. PE i -a b ∈W mi It is the task constraint of the operation point, that is, the distance between the operation point and the macro-robotic arm motion should be within the working space of the micro-robotic arm.

[0068] Step 103, constructing a macro-mechanical action similarity index based on the macro-mechanical arm action data;

[0069] Specifically, based on the action data of the macro robot arm, the macro robot arm action is constructed and Similarity index between

[0070]

[0071] Among them, w i ,w j They are the target poses of the two macro manipulators, ROCP i ,ROCP jThey are the closeness of the operating points under each macro robot action, and are specifically defined as follows:

[0072]

[0073] c represents the constraint function of the macro-micromanipulator, and its specific representation is as follows:

[0074]

[0075] Step 104, constructing a matching index between a macro-mechanical action and an operation point according to the connection relationship and working mode of the macro-micro-mechanical arm;

[0076] Specifically, according to the connection relationship and working mode of the macro-micro robotic arm, the macro robotic arm action a is constructed. b The matching index s(a b ,t):

[0077]

[0078] Among them, B, w and ROCP+ are the fused action constraint satisfaction function, the macro robot action target point and the closeness of the operation point under the macro robot action, and PE is the location information of the operation point.

[0079] Step 105, designing a macro-micro robotic arm motion generation method based on the macro-micro robotic arm task characterization model, the macro-micro robotic arm constraint equation, the macro-mechanical action and operation point matching index, and the macro-mechanical action similarity index:

[0080] Specifically, the macro-micro robot arm motion sequence is solved according to the following process:

[0081] Step 1: Initialize all the operation points in the macro-micro robotic arm task T and the macro-micro robotic arm status;

[0082] Step 2: Determine the set of all currently executable job points t_set, that is, all the job points with the highest priority;

[0083] Step 3: Based on the operation point set t_set, a low-to-up hierarchical clustering algorithm is used to generate a macro robot action set {a B} and the corresponding cluster {cluster};

[0084] Step 4: Determine the macro robot arm action to be optimized based on the matching index between the macro robot action and the operation point B And the corresponding operation point cluster;

[0085] Step 5: According to the macro robot action a B, the distribution cluster of the operation points to be iterated and the information of the operation points to be iterated t, select the matching index s (a b ,t) The highest job point to be iterated;

[0086] Step 6: If the matching index s=0, skip the current step. Otherwise, for the action of the current macro-manipulator, add the new operation point to the operation cluster, and return to the step of selecting the operation point with the highest matching index between the macro-mechanical action and the operation point of the current macro-manipulator;

[0087] Step 7: If there are still unexecuted work points in the macro-micro robotic arm task T, return to the step of determining the set of all currently executable work points t_set. Otherwise, based on the joint travel of the macro-micro robotic arm, the Clustering Travelling Salesman Problem (CTSP) algorithm and the configuration optimization algorithm are used to generate the action sequence A of each arm in the macro-micro robotic arm. b and A s .

[0088] According to the above method provided by the embodiment of the present invention, a simulation experiment was conducted on a macro-micro arm composed of two redundant mechanical arms. Figure 2 , where the dotted box represents the macro robot arm, the solid box represents the micro robot arm, and the macro and micro robot arms are connected by a transfer piece. The DH parameters of the macro and micro robot arm are shown in Table 1, where the parameters of the macro robot arm are d1 = d7 = 0.6m, d2 = d3 = d4 = d5 = 0.35m, a3 = a4 = 5m, and the parameters of the micro robot arm are d1 = d7 = 0.6m, d2 = d3 = d4 = d5 = 0.25m, a3 = a4 = 2.5m:

[0089] Table 1 DH parameters of macro-micro robot arm

[0090] <![CDATA[α i ]]> <![CDATA[a i ]]> <![CDATA[d i ]]> <![CDATA[θ i ]]> 1 0 0 <![CDATA[d1]]> -180 2 0 0 <![CDATA[d2]]> -90 3 -90 <![CDATA[a3]]> <![CDATA[d3]]> 0 4 0 <![CDATA[a4]]> <![CDATA[d4]]> 0 5 0 0 <![CDATA[d5]]> 0 6 90 0 0 90 7 90 0 <![CDATA[d7]]> 180

[0091] Please refer to Figure 3 , where each point represents an operation point. The specific parameters and constraint relationships of the operation points are shown in Table 2. The first column is the number of the operation point, the second column is the six-dimensional pose information of the operation point, and the third column is the constraint relationship of the operation point, which indicates the execution order of the operation point. The macro-micro manipulator needs to execute in order from small to large according to the constraint value. There is no order under the same value.

[0092] Table 2 Task information

[0093]

[0094]

[0095] The simulation experiment was carried out on a computer with a CPU of 7-7700k and a memory of 16G. For the planning results, please refer to Figure 4 and Figure 5 , where the motion trajectory of the macro robot arm is a continuous solid line segment, and the motion trajectory of the micro robot arm is a dotted line segment. Since the macro-micro robot arm adopts time-sharing motion, the state of the micro robot arm will change during the motion of the macro robot arm, making the motion trajectory of the micro robot arm discontinuous. The ball in the figure represents the approximate workspace of the micro robot arm under the action of the macro robot arm. It can be seen that the algorithm in this paper can effectively generate the corresponding robot arm action covering all task points, indicating the effectiveness of the method of the present invention.

[0096] The comparison of the two methods in terms of action sequence length, joint travel, running time, etc. is shown in Table 3.

[0097] Table 3 Comparison of planning results of two methods

[0098]

[0099] The macro-micro robotic arm motion sequence generated by the motion generation method based on motion iteration is significantly shorter than that of the conventional robotic arm motion sequence generation method. The sequence length is reduced by 18.75%. The reduction in the number of macro-robotic arm movements can improve the stability of the macro-micro robotic arm during task execution. The joint stroke and task execution time are reduced by an average of 16.9% and 20% respectively. If the vibration suppression process during the macro-micro robotic arm motion switching is considered, the algorithm proposed in this paper will further reduce the macro-micro robotic arm task execution time, which proves the rationality of the method of the present invention.

[0100] The technical solution of the embodiment of the present invention has the following beneficial effects:

[0101] According to the constraints between the operation steps and the macro-micro manipulator movements in the on-orbit mission, the macro-micro manipulator task representation and the macro-micro manipulator constraint equations are constructed. Combined with the special time-sharing motion mode of the macro-micro machine, the macro-mechanical movement similarity index and the macro-mechanical movement and operation point matching index are constructed. The clustering method and the iterative mechanism are integrated to design the macro-micro manipulator task planning method, decouple the coupling relationship of the macro-micro manipulator, and generate a feasible macro-micro manipulator action sequence with fewer execution steps and shorter running time, so as to improve the execution efficiency of the macro-micro manipulator's large-span on-orbit fine tasks.

[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0103] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A robot arm task planning method, characterized in that: The method comprises: According to the constraint information between the operation steps in the on-orbit mission and the motion data of the macro-micro robotic arm, a macro-micro robotic arm task characterization model is constructed; the macro-micro robotic arm includes a macro robotic arm and a micro robotic arm; According to the connection relationship and motion characteristics of the macro-micro robotic arm, the constraint equation of the macro-micro robotic arm is constructed; Based on the motion data of the macro robot arm, a similarity index of the macro robot motion is constructed; According to the connection relationship and working mode of the macro-micro robotic arm, the matching index between the macro-mechanical action and the operation point is constructed; Task planning is performed on the macro-micro-mechanical arm according to the task characterization model of the macro-micro-mechanical arm, the constraint equation of the macro-micro-mechanical arm, the matching index of the macro-mechanical action and the operation point, and the similarity index of the macro-micro-mechanical action.

2. The method according to claim 1, characterized in that The macro-micro robotic arm task characterization model is constructed based on the constraint information between the operation steps in the on-orbit task and the macro-micro robotic arm motion data, including: According to the constraint information between the operation steps in the on-orbit task and the macro-micro manipulator motion data, a macro-micro manipulator task representation model of the following form is constructed for the on-orbit task containing n operation points and the macro-micro manipulator motion: T={t1,…,t n } t i ={PE i C i Visited i } A={A b A s } Where T represents the on-orbit mission to be performed, t i represents the i-th operation point in the on-orbit task T, PE i represents the position information of the on-orbit mission, which is composed of a six-dimensional vector. i Indicates the priority constraint of the operation, which is used to describe the operation order between tasks in terms of numerical value; Visited i Indicates whether the state has been visited. i =1 means that the on-orbit mission has been planned and completed; A b and A s They represent the macro-manipulator motion sequence and the micro-manipulator motion sequence respectively; represents the i-th macro robot action, Represents the jth macro robot action and is executed after the ith macro robot action.

3. The method according to claim 1, characterized in that: The constraint equations of the macro-micro robotic arm are constructed based on the connection relationship and motion characteristics of the macro-micro robotic arm, including: Based on the connection relationship and motion characteristics of the macro-micro robotic arm, as well as the dynamic performance of the macro-micro robotic arm, the connection relationship between the two arms, and whether the micro robotic arm can carry out tasks, the constraint equations of the macro-micro robotic arm are constructed, including: -π≤q ma_i ≤π,i=1,2,...,7 -π≤q mi_j ≤π,j=1,2,...,7 a b ∈W ma a s ∈W mi ON i -a b ∈W mi Among them, q ma_i is the joint angle of the macro robot, q ma_i is the joint angle of the micro-manipulator, a b ∈W ma and a s ∈W mi The two constraint equations represent the macro robot arm motion a b and micro-manipulator motion a s In their respective workspaces ma and W mi Inside; PE i -a b ∈W mi is the task constraint of the operation point, which means that the distance between the operation point and the macro-robotic arm action should be within the working space of the micro-robotic arm.

4. The method according to claim 1, characterized in that The method of constructing a similarity index of a macro-mechanical action based on the action data of the macro-mechanical arm includes: Construct macro robot movements based on the macro robot movement data and Similarity index between include: Among them, w i ,w j They are the target poses of the two macro manipulators; ROCP i ,ROCP j The closeness of the operating points under each macro robot action, including: Among them, c represents the constraint function of the macro-micromanipulator, including:

5. The method according to claim 1, characterized in that The macro-mechanical action and operation point matching index is constructed based on the connection relationship and working mode of the macro-micro-mechanical arm, including: According to the connection relationship and working mode of the macro-micro robotic arm, the matching index between the macro-mechanical action and the operation point is constructed, including: Among them, B, w and ROCP are the fused action constraint satisfaction function, the macro robot action target point and the closeness of the operation point under the macro robot action, and PE is the location information of the operation point.

6. The method according to claim 1, characterized in that According to the task characterization model of the macro-micro mechanical arm, the constraint equation of the macro-micro mechanical arm, the matching index between the macro mechanical action and the operation point, and the similarity index of the macro-micro mechanical action, the task planning of the macro-micro mechanical arm is performed, including: Initialize all operation points in the macro-micro robotic arm task T and the macro-micro robotic arm status; Determine the set of all currently executable job points t_set; According to the operation point set t_set, a hierarchical clustering algorithm is used to generate a macro robot action set {a B } and the corresponding cluster {cluster}; According to the matching index between the macro mechanical action and the operation point, the macro mechanical arm action aB to be optimized and the corresponding operation point cluster cluster are determined; Select the operation point with the highest matching index between the macro mechanical action and the operation point of the current macro mechanical arm; If the matching index s=0, skip the current step. Otherwise, add the new operation point to the operation cluster for the action of the current macro-manipulator, and return to the step of selecting the operation point with the highest matching index between the macro-mechanical action and the operation point of the current macro-manipulator; If there are still unexecuted work points in the macro-micro robotic arm task T, return to the step of determining the set of all currently executable work points t_set. Otherwise, according to the joint travel of the macro-micro robotic arm, the cluster traveling salesman algorithm CTSP algorithm and the configuration optimization algorithm are used to generate the action sequence A of each arm in the macro-micro robotic arm. b and A s .

Images