A multi-axis joint collaboration method and system for a robot manipulator

Abstract

The present application relates to the field of agricultural fruit harvesting robot vision technology, and in particular to a multi-axis joint collaboration method and system for a robot manipulator, which collects the located joint sets of all operating postures; outputs the collaborative manipulator according to the preconfigured collaborative manipulator output template; locates the operating posture of the first manipulator as the operating posture of the second manipulator when the operating postures of the collaborative manipulator are consistently identified and matched; interrupts the second manipulator after the operating posture is located, and updates the second manipulator to the located joint set corresponding to the located operating posture; reconfigures the collaborative manipulator output template according to the valid joint set and the invalid joint set, and outputs the second manipulator that obtains a picking task from the joint set to be located, imports the collaborative manipulator into the preconfigured collaborative manipulator output template and outputs the collaborative manipulator. The present invention can effectively perform adaptive adjustment on the coordination of the manipulators in the picking process, and improve the quality level of fruit and vegetable picking.

Description

Technical Field

[0001] The present application relates to the field of agricultural fruit harvesting robot vision technology, and in particular to a multi-axis joint collaboration method and system for a robot manipulator arm. Background Art

[0002] Mechanized picking is mostly done by pulling devices to assist manual picking; during the pulling process, the picking head will squeeze the fruit, causing damage to the fruit and branches of the fruit tree, which will affect the transportation and sale of the fruit, and the secondary fruiting of the fruit tree, etc. In addition, the fruit is irregularly distributed on the branches of the fruit tree, and there are often obstacles around the fruit. At present, fruit trees are basically planted in three dimensions. Due to the wide planting area and large planting volume, there is almost no pruning, which leads to frequent obstruction of the fruit by branches and leaves. The existing picking device cannot effectively avoid obstacles such as branches and leaves, which can easily cause problems such as wrong picking, missed picking, and mistaken picking, reducing the success rate of fruit picking. It is also easy to cause irreversible damage to the fruit and branches of the fruit tree, resulting in problems such as reduced fruit tree production and unsalable fruit.

[0003] The use of robotic arms is now common in fruit and vegetable picking. However, the current use of robotic arms in picking is still relatively rigid. The picking angle and strength of a single robotic arm are determined only by the robotic arm's recognition of the picking object, ignoring operations such as fruit and vegetable collection and maintaining branch stability during the picking process. There is a lack of coordinated control of the robotic arms, and the picking quality is difficult to guarantee. Summary of the invention

[0004] To achieve the above objectives, this application provides the following technical solutions:

[0005] According to a first aspect of the present invention, the present invention claims a multi-axis joint collaboration method for a robot manipulator, which is used to position the joints used to configure a picking manipulator, comprising:

[0006] Collecting a set of positioned joints of all operation postures, wherein the set of positioned joints includes a first mechanical arm that positions the operation posture;

[0007] A second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template to output a collaborative robot arm, wherein the collaborative robot arm comprises a first robot arm and a second robot arm;

[0008] When the operation posture toward the collaborative robot arm is obtained and identified as a match, the operation posture of the first robot arm in the collaborative robot arm is positioned as the operation posture of the second robot arm;

[0009] The second mechanical arm after the positioning operation posture of the joint set to be positioned is interrupted, and the second mechanical arm after the positioning operation posture is updated to the positioned joint set corresponding to the positioned operation posture;

[0010] Analyze whether it meets the interruption location scenario;

[0011] If not, generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set;

[0012] The collaborative robot arm output template is reconfigured according to the valid joint set and the invalid joint set, and a second robot arm that obtains a picking task from the joint set to be positioned is imported into the preconfigured collaborative robot arm output template to output the collaborative robot arm.

[0013] Furthermore, before the second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template to output the collaborative robot arm, the method further includes:

[0014] The first robot arms in the located joint set are imported into a preconfigured collaborative robot arm output template, first action images of all first robot arms are output, and the first action images are saved in an action image library.

[0015] Further, a second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template to output the collaborative robot arm, including:

[0016] A second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template;

[0017] Performing video segmentation on all the imported second robotic arms in the collaborative robotic arm output template, and outputting second action images of the second robotic arms;

[0018] Calculating, in the collaborative robot arm output template, a correlation between the second action image and the first action image in a preset action image library;

[0019] A collaborative robotic arm is generated based on the first robotic arm and the second robotic arm corresponding to the first motion image and the second motion image with the greatest correlation.

[0020] Further, analyze whether it meets the interruption location scenario, including:

[0021] Calculate the number of joints of the first robot arm in the located joint set of all operation postures;

[0022] Analyze whether the number of multiple joints is not less than the preset threshold value;

[0023] If so, the diagnosis meets the interruption positioning scenario, and the positioning of the second robot arm with the joint to be positioned is terminated;

[0024] If not, the diagnosis does not meet the interruption positioning scenario, and the generation of a valid joint set and an invalid joint set from the positioned joint set and the to-be-positioned joint set is completed.

[0025] Further, generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set includes:

[0026] Select one joint from two randomly located joint sets in turn, and output multiple first-category invalid joint sets;

[0027] A plurality of second robotic arms with p and q positioning abnormalities are diagnosed from the joint set to be positioned, and a second type of invalid joint set is generated based on the second robotic arms with p positioning abnormalities, and a third type of invalid joint set is generated based on the second robotic arms with q positioning abnormalities, wherein the second type of invalid joint set and the third type of invalid joint set are collaborative robotic arms with two joints that have mismatched operation postures, wherein q is not less than p;

[0028] Diagnose two random positioned joint sets associated with the operation posture, select one joint from the two random positioned joint sets associated with the operation posture in turn, and output multiple fourth-category invalid joint sets;

[0029] Randomly obtain two joints from all located joint sets to output multiple valid joint sets.

[0030] Further, reconfiguring the collaborative robot arm output template according to the valid joint set and the invalid joint set includes:

[0031] Randomly obtain a valid joint set and an invalid joint set of a picking task and import them into the collaborative robot arm output template, and output the probability of matching the operation postures of two joint sets in all joint sets;

[0032] Calculate the picking effectiveness by the size of the possibility;

[0033] Analyze whether the picking effectiveness is less than a picking effectiveness threshold value;

[0034] If so, diagnose that the collaborative robot arm output template has completed reconfiguration, import the first robot arm in the located joint set into the reconfigured collaborative robot arm output template, output the first action images of all the first robot arms, and save the first action images in the action image library;

[0035] If not, the template index is corrected through the picking effectiveness, and a valid joint set and an invalid joint set of a randomly obtained picking task are output and imported into the collaborative robot arm output template.

[0036] Further, calculating the picking effectiveness according to the possibility size includes:

[0037] Import the matching probability of the operation postures of two joint sets in all joint sets and the label values ​​of the joint sets into the preset picking cost function, and output the picking effectiveness of all joint sets;

[0038] The importance of the diagnosis joint set is t0, t1, t2, t3, and t4, where t1 < t2 < t3 < t4, t0 is the importance of the valid joint set, and t1, t2, t3, and t4 are the importance of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set respectively;

[0039] The weighted average is calculated by the picking effectiveness and importance of all joint sets and the number of joint sets in a picking task, and the picking effectiveness is output.

[0040] According to the second invention of the present invention, the present invention claims a multi-axis joint cooperation system for a robot manipulator, which is used to position the joints used to configure the picking manipulator, comprising:

[0041] A located joint set acquisition unit, used for acquiring located joint sets of all operation postures, wherein the located joint set includes a first mechanical arm located in the operation posture;

[0042] A collaborative robot arm generating unit, used for obtaining a second robot arm for a picking task from a set of joints to be positioned, importing the second robot arm into a preconfigured collaborative robot arm output template and outputting a collaborative robot arm, wherein the collaborative robot arm comprises a first robot arm and a second robot arm;

[0043] an operation posture positioning unit, configured to position the operation posture of the first mechanical arm in the collaborative mechanical arm as the operation posture of the second mechanical arm when the operation posture facing the collaborative mechanical arm is consistent and matched;

[0044] A joint set updating unit, used for interrupting the second mechanical arm after the positioning operation posture of the joint set to be positioned, and updating the second mechanical arm after the positioning operation posture to the positioned joint set corresponding to the positioned operation posture;

[0045] A scene analysis unit is used to analyze whether the interruption positioning scene is met, and if not, complete the joint set generation unit;

[0046] A joint set generating unit, used for generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set;

[0047] A reconfiguration unit, configured to reconfigure the collaborative robot arm output template according to the valid joint set and the invalid joint set, and output a collaborative robot arm generation unit;

[0048] The multi-axis joint collaboration system for a robot manipulator arm is used to execute the multi-axis joint collaboration method for a robot manipulator arm.

[0049] The present application relates to the field of agricultural fruit harvesting robot vision technology, and in particular to a multi-axis joint collaboration method and system for a robot manipulator, which collects the located joint sets of all operating postures; outputs the collaborative manipulator according to the preconfigured collaborative manipulator output template; locates the operating posture of the first manipulator as the operating posture of the second manipulator when the operating postures of the collaborative manipulator are consistently identified and matched; interrupts the second manipulator after the operating posture is located, and updates the second manipulator to the located joint set corresponding to the located operating posture; reconfigures the collaborative manipulator output template according to the valid joint set and the invalid joint set, and outputs the second manipulator that obtains a picking task from the joint set to be located, imports the collaborative manipulator into the preconfigured collaborative manipulator output template and outputs the collaborative manipulator. The present invention can effectively perform adaptive adjustment on the coordination of the manipulators in the picking process, and improve the quality level of fruit and vegetable picking. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A workflow diagram of a multi-axis joint collaboration method for a robot manipulator arm as claimed in an embodiment of the present application;

[0051] Figure 2 A second working flow chart of a multi-axis joint collaboration method for a robot manipulator arm as claimed in an embodiment of the present application;

[0052] Figure 3 A structural module diagram of a multi-axis joint collaboration system for a robot manipulator arm as claimed in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] The terms "first", "second" and "third" in this application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. In the embodiments of this application, all directional indications (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations, actions are to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0055] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase at various locations in the specification does not necessarily refer to a matching embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0056] Embodiment 1

[0057] Figure 1 This is a flowchart of a multi-axis joint collaboration method for a robot manipulator provided in the first embodiment of the present invention. This embodiment can be applied to the joint positioning operation posture used when configuring a picking robot arm. This method can be completed by a multi-axis joint collaboration system for a robot manipulator. The multi-axis joint collaboration system for a robot manipulator can be implemented in the form of hardware and / or software and configured in an electronic device. Figure 1 As shown, the multi-axis joint collaboration method for the robot manipulator includes:

[0058] S101, collecting the positioned joint sets of all operation postures, where the positioned joint sets include the first robotic arm that has positioned the operation postures.

[0059] The joints positioned in this embodiment are used to configure a picking robot arm, which may be a robot used in a picking orchard, such as a picking robot arm embedded in a fruit and vegetable farm RPP. The picking robot arm is used to identify the robot arm control actions of the pickers and match the robot arm control actions to reply to the pickers. Therefore, the joint positioning in this embodiment refers to the operating posture to which the positioning joint belongs.

[0060] After setting multiple operating postures, this embodiment can collect and import the positioned joint sets of all operating postures, which can include a preset number of first robotic arms as example joints, and all first robotic arms have positioned operating postures.

[0061] The joint set of the robot arm includes at least a first joint, a second joint, a third joint, and a fourth joint;

[0062] The first joint is a finger joint of the robotic arm, and each robotic arm has a plurality of finger joints for contacting and picking fruits and vegetables;

[0063] The second joint is a wrist joint, and each robotic arm has a wrist joint for controlling the picking position of the picked fruits and vegetables;

[0064] The third joint is an elbow joint. Each robotic arm has an elbow joint, which is used to control the force state of picking fruits and vegetables.

[0065] The fourth joint is a shoulder root joint. Each robotic arm has a shoulder root joint for controlling the distance and height of picking fruits and vegetables.

[0066] S102, a second robotic arm for a picking task is obtained from the set of joints to be positioned, and is imported into a preconfigured collaborative robotic arm output template to output a collaborative robotic arm, wherein the collaborative robotic arm comprises a first robotic arm and a second robotic arm.

[0067] The set of joints to be positioned may be a collection of joints in unpositioned operating postures, and the collaborative robot arm output template may be various pre-configured templates for generating collaborative robot arms. In one example, the collaborative robot arm output template is a template that is configured to have the ability to generate a collaborative robot arm that matches or is associated with an operating posture. Specifically in this embodiment, the collaborative robot arm output template may generate a collaborative robot arm comprising a second robot arm and a first robot arm in the set of positioned joints.

[0068] Wherein, the first robotic arm and the second robotic arm are cooperatively operated robotic arms;

[0069] The first robotic arm completes active picking of fruits and vegetables, including positioning of fruits and vegetables and status information of fruits and vegetables; when picking, the first robotic arm controls its own picking force and picking position;

[0070] The second robotic arm completes the coordinated picking of fruits and vegetables, which at least includes collecting fruits and vegetables and stabilizing fruits and vegetables. When collecting fruits and vegetables, the second robotic arm holds a collection basket and sets the height of the collection basket from the collected fruits and vegetables according to the collection strength of the first robotic arm, so as to achieve efficient picking while preventing fruits and vegetables from falling and being damaged;

[0071] When the second robotic arm completes the coordinated picking of fruits and vegetables and is used to stabilize the fruits and vegetables, the second robotic arm supports the branches to which the fruits and vegetables are picked to ensure the stability of the fruits and vegetables when picked by the first robotic arm.

[0072] In this embodiment, the second robotic arms in the joint set to be positioned can be batched in advance, and second robotic arms for multiple picking tasks can be output. All picking tasks include at least one second robotic arm, and the second robotic arms can be imported into the collaborative robotic arm output template. Through the collaborative robotic arm output template, the first robotic arm with the most relevant operating posture can be matched for the second robotic arm from multiple first robotic arms in the joint set that have been positioned, and a collaborative robotic arm can be formed based on the first robotic arm and the imported second robotic arm.

[0073] S103: When a consistent match is obtained for the operation posture of the collaborative robot arm, the operation posture of the first robot arm in the collaborative robot arm is positioned as the operation posture of the second robot arm.

[0074] After generating multiple collaborative robot arms through the collaborative robot arm output template, the multiple collaborative robot arms can be sent to the automatic verification terminal to diagnose whether the operating posture of the second robot arm in the collaborative robot arm matches or is close to the operating posture of the first robot arm. If so, the verification personnel confirms on the automatic verification terminal. When the confirmation operation is obtained, it is diagnosed that the operating posture of the second robot arm to be positioned matches the operating posture of the positioned first robot arm, and the operating posture of the first robot arm in the collaborative robot arm can be positioned as the operating posture of the second robot arm.

[0075] S104, interrupting the second mechanical arm after positioning the operation posture in the joint set to be positioned, and updating the second mechanical arm after positioning the operation posture to the positioned joint set corresponding to the positioned operation posture.

[0076] After the second robotic arm in the joint to be positioned has positioned the operating posture, the second robotic arm in the positioned operating posture can be disconnected from the joint to be positioned and updated to the positioned joint set corresponding to the positioned operating posture. The second robotic arm in the positioned joint set is updated to become the first robotic arm, that is, from the joint to be positioned to the positioned joint, so that the number of second robotic arms to be positioned in the joint to be positioned gradually decreases, and the number of first robotic arms in the positioned joint sets of each operating posture gradually increases.

[0077] If the automatic verification terminal does not obtain confirmation operation on the collaborative robot arm, it is diagnosed that the operating postures of the first robot arm and the second robot arm in the collaborative robot arm do not match or differ too much, and the second robot arm is still set in the joint to be positioned to wait for positioning.

[0078] S105: Analyze whether the interruption positioning scenario is met.

[0079] In one embodiment, the interruption positioning scenario may be that the number of first robotic arms in the positioned joint set of each operating posture is not less than a preset threshold value, that is, the first robotic arms in each operating posture are positioned enough for configuration. In another example, the interruption positioning scenario may also be that the number of remaining second robotic arms in the joint set to be positioned is less than a preset value. When the interruption positioning scenario is met, the positioning of the joints in the joint set to be positioned is interrupted. When the interruption positioning scenario is not met, S106 is completed.

[0080] S106 , generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set.

[0081] In one embodiment, two first robotic arms may be selected from a positioned joint set of an operation posture to form a valid joint set.

[0082] In another embodiment, a second robot arm that has been imported into the collaborative robot arm output template once or more can be diagnosed from the joint set to be positioned, and an invalid joint set of collaborative robots can be generated based on the second robot arm. For example, the collaborative robot arm output when the collaborative robot arm output template is imported based on the second robot arm is used as an invalid joint set.

[0083] In another embodiment, a first mechanical arm may be selected in sequence from the positioned joint sets of two operating postures to generate an invalid joint set, or a first mechanical arm may be selected from the positioned joint sets of two associated operating postures to generate an invalid joint set.

[0084] Of course, those skilled in the art may also generate the valid joint set and the invalid joint set in other ways, and this embodiment does not limit the way to generate the valid joint set and the invalid joint set.

[0085] S107. Reconfigure the collaborative robot arm output template according to the valid joint set and the invalid joint set.

[0086] Specifically, a collaborative robot arm can be randomly obtained from a valid joint set and an invalid joint set and imported into a collaborative robot arm output template, so as to output the possibility of the operating posture association between two joints in the collaborative robot arm through the collaborative robot arm output template, calculate the picking effectiveness through the possibility, and continue to configure the collaborative robot arm output template after correcting the template index through the picking effectiveness until the interruption configuration scenario is met, and then output S102 to continue to position the second robot arm in the joint set to be positioned, so as to improve the accuracy of joint positioning.

[0087] An embodiment of the present invention can first collect a small number of first robotic arms with all operating postures that have been positioned as example joints and update them into the positioned joint set, and then import the second robotic arm that obtains a picking task from the joint set to be positioned into the preconfigured collaborative robotic arm output template to output the collaborative robotic arm. The collaborative robotic arm has a first robotic arm and a second robotic arm. When a consistent match is obtained for the operating posture of the collaborative robotic arm, the operating posture of the first robotic arm in the collaborative robotic arm is positioned as the operating posture of the second robotic arm, and the second robotic arm after the operating posture is positioned in the joint set to be positioned is interrupted, and the second robotic arm after the operating posture is positioned is updated to the positioned joint set, and it is analyzed whether it meets the interrupt positioning scenario. If not, a valid joint set and an invalid joint set are generated from the positioned joint set and the joint set to be positioned, and the collaborative robotic arm output template is reconfigured according to the valid joint set and the invalid joint set, and the second robotic arm that obtains a picking task from the joint set to be positioned is imported into the preconfigured collaborative robotic arm output template to output the collaborative robotic arm. On the one hand, a collaborative robot arm including a second robot arm in a joint set to be positioned is generated through a collaborative robot arm output template. After manual confirmation of the collaborative robot arm, the operating posture of the first robot arm used as an example in the collaborative robot arm is positioned as the operating posture of the second robot arm. There is no need to manually position the operating posture of the joints to be positioned one by one, which improves the efficiency of joint positioning and saves time and labor costs for joint positioning. On the other hand, by manually confirming the positioned joint set and generating a positive and invalid joint set to reconfigure the collaborative robot arm output template, the accuracy of the collaborative robot arm output template can be improved, thereby improving the accuracy of the operating posture of the positioned joints.

[0088] Embodiment 2

[0089] Figure 2 A flowchart of a multi-axis joint collaboration method for a robot manipulator provided in the second embodiment of the present invention is provided. The embodiment of the present invention is optimized on the basis of the above-mentioned first embodiment. Figure 2 As shown, the multi-axis joint collaboration method for the robot manipulator includes:

[0090] S201 , collecting a set of positioned joints of all operating postures, where the set of positioned joints includes a first robotic arm that has positioned the operating posture.

[0091] After setting multiple operating postures, this embodiment can manually collect and import the located joint sets of all operating postures, and the located joint sets of all operating postures can include a small number of first robotic arms as example joints, and all first robotic arms have located operating postures.

[0092] S202, importing the first robot arm in the located joint set into a preconfigured collaborative robot arm output template, outputting first action images of all first robot arms, and saving the first action images in an action image library.

[0093] The collaborative robot arm output template of this embodiment can generate action images of the two joints after importing two joints, save the generated action images in an action image library, and calculate the correlation between the two action images. When the correlation between the action images is not less than a threshold value, the operation posture matching of the two joints corresponding to the two action images is diagnosed.

[0094] After the located joint set is collected, the first robot arm as an example joint in the located joint set can be imported into the collaborative robot arm output template, multiple first action images of the first robot arm can be output, and the first action image of the first robot arm can be saved in the action image library.

[0095] S203, obtaining a second robot arm for a picking task from the set of joints to be positioned and importing it into a preconfigured collaborative robot arm output template.

[0096] In this embodiment, the second robotic arms in the joint set to be positioned can be batched in advance, and second robotic arms for multiple picking tasks can be output. All picking tasks include at least one second robotic arm, and the second robotic arms can be imported into the collaborative robotic arm output template. The number of second robotic arms for all picking tasks can be diagnosed based on the total number of joints in the joint set to be positioned, and can also be diagnosed based on the processing performance of the collaborative robotic arm output template. This embodiment does not limit the number of second robotic arms in a picking task.

[0097] S204. Perform video segmentation on all imported second robotic arms in the collaborative robotic arm output template, and output second action images of the second robotic arms.

[0098] After the second robotic arm imports the collaborative robotic arm output template, the video of all the imported second robotic arms can be segmented to output the second action images of the second robotic arms.

[0099] S205. Calculate, in the collaborative robot arm output template, the degree of association between the second action image and the first action image in the preset action image library.

[0100] The action image library of the collaborative robot arm output template of this embodiment stores the first action image of the positioned first robot arm, and the correlation between the imported second action image of the second robot arm and the first action images of all the first robot arms in the action image library can be calculated. The correlation can be cosine distance, Euclidean distance, etc. This embodiment does not limit the method of calculating the correlation.

[0101] S206. Generate a collaborative robotic arm based on the first robotic arm and the second robotic arm corresponding to the first action image and the second action image with the greatest correlation.

[0102] After calculating the correlation between the second action image and multiple first action images, the maximum correlation is diagnosed, and a collaborative robotic arm is generated based on the first robotic arm with the largest maximum correlation and the imported second robotic arm. When there are multiple second robotic arms imported for a picking task, multiple collaborative robotic arms including the second robotic arm can be output, and the multiple collaborative robotic arms can be sent to the automatic verification terminal so that the collaborative robotic arms can be confirmed at the automatic verification terminal.

[0103] In another embodiment, in addition to generating the collaborative robot arm, the collaborative robot arm output template also outputs the possibility of matching the operating postures of the first robot arm and the second robot arm in the collaborative robot arm. The collaborative robot arm with a possibility size less than a threshold value can be sent to the automatic verification terminal for manual confirmation. When the possibility size is not less than or equal to the threshold value, the operating posture of the first robot arm in the collaborative robot arm is directly positioned as the operating posture of the second robot arm. Thus, the collaborative robot arm with a high possibility of matching the operating posture can be directly positioned without automatic verification and confirmation, thereby reducing the number of collaborative robot arms that require automatic verification and confirmation, reducing the work pressure of the verification personnel, and improving the efficiency of joint positioning.

[0104] S207: When a consistent match is obtained for the operation posture of the collaborative robot arm, the operation posture of the first robot arm in the collaborative robot arm is positioned as the operation posture of the second robot arm.

[0105] After generating multiple collaborative robot arms through the collaborative robot arm output template, the multiple collaborative robot arms can be sent to the automatic verification terminal to diagnose whether the operating posture of the second robot arm in the collaborative robot arm matches or is close to the operating posture of the first robot arm. If so, the verification personnel confirms on the automatic verification terminal. When the confirmation operation is obtained, it is diagnosed that the operating posture of the second robot arm to be positioned matches the operating posture of the positioned first robot arm, and the operating posture of the first robot arm in the collaborative robot arm can be positioned as the operating posture of the second robot arm.

[0106] S208, interrupting the second mechanical arm after positioning the operation posture in the joint set to be positioned, and updating the second mechanical arm after positioning the operation posture to the positioned joint set corresponding to the positioned operation posture.

[0107] After the second robotic arm in the joint set to be positioned has positioned the operating posture, the second robotic arm with the positioned operating posture can be disconnected from the joint set to be positioned and updated to the positioned joint set corresponding to the positioned operating posture, so that the number of second robotic arms to be positioned in the joint set to be positioned gradually decreases, and the number of first robotic arms in the positioned joint sets of each operating posture gradually increases.

[0108] If the automatic verification terminal does not obtain the confirmation operation of the collaborative robot arm, it is diagnosed that the operating postures of the first robot arm and the second robot arm in the collaborative robot arm do not match or differ too much, and the second robot arm is still set in the joint set to be positioned.

[0109] S209: Analyze whether the interruption positioning scenario is met, diagnose whether the interruption positioning scenario is met, and terminate the positioning of the second robot arm with the joint to be positioned.

[0110] In one embodiment, the number of joints of the first robot arm in the positioned joint set of all operating postures can be calculated, and it can be analyzed whether the number of multiple joints is not less than a preset threshold value. If so, S209 is completed, and if not, S210 is completed.

[0111] When the number of positioned joints for all operating postures is sufficient, the positioning of the second robotic arm in the joint set to be positioned can be interrupted, and the positioned joint sets of each operating posture that has been positioned can be output. The picking robotic arm can be configured through the positioned joint sets of each operating posture.

[0112] S210, generating a valid joint set and an invalid joint set from the positioned joint set and the to-be-positioned joint set, and reconfiguring the collaborative robot arm output template according to the valid joint set and the invalid joint set.

[0113] In this embodiment, the valid joint set and the invalid joint set may be collaborative robotic arms. The valid joint set may be a collaborative robotic arm generated by two joints with matching operating postures, and the invalid joint set may be a collaborative robotic arm composed of two joints with unmatched operating postures.

[0114] In one embodiment, a joint can be selected in turn from two random located joint sets to output multiple first-class invalid joint sets. For example, the located joint set of operation posture R is R = (R1, R2, Ri, ..., Rp), and the located joint set of operation posture G is G = (G1, G2, Gj, ..., Gp), which can constitute the first-class invalid joint set (Ri, Gj, L1), wherein Ri is a random joint in the located joint set R of operation posture R, Gi is a random joint in the located joint set G of operation posture G, and L1 is the possibility of matching the operation posture of joint Ri with that of joint Gi.

[0115] In another embodiment, multiple second robotic arms with positioning abnormalities p times and q times are diagnosed from the joint set to be positioned, and a second type of invalid joint set is generated based on the second robotic arm with positioning abnormalities p times, and a third type of invalid joint set is generated based on the second robotic arm with positioning abnormalities q times. The second type of invalid joint set and the third type of invalid joint set are collaborative robotic arms with two joints that have mismatched operating postures. Specifically, after the collaborative robotic arm generated by S206 is sent to automatic verification, if the second robotic arm in the collaborative robotic arm diagnosed by the automatic verification does not match the operating posture of the first robotic arm, the second robotic arm is marked as having a positioning abnormality. 1 time, and so on, output the number of positioning abnormalities of all second robotic arms in the set of joints to be positioned, generate a second type of invalid joint set based on the second robotic arm with 1 positioning abnormality, for example, the collaborative robotic arm output when the second robotic arm with 1 abnormality is imported into the collaborative robotic arm output template and manually confirms the positioning abnormality as the second type of invalid joint set, and generate a third type of invalid joint set based on the second robotic arm with 3 positioning abnormalities, for example, the collaborative robotic arm output when the second robotic arm with 3 abnormalities is imported into the collaborative robotic arm output template and manually confirms the positioning abnormality as the third type of invalid joint set.

[0116] In another embodiment, two random located joint sets associated with the operation posture are diagnosed, and one joint is selected in turn from the two random located joint sets associated with the operation posture, and multiple fourth-category invalid joint sets are output. Exemplarily, the operation postures of operation posture C and operation posture D are similar, the located joint set of operation posture C is C = (C1, C2, Ci, ..., Cp), and the located joint set of operation posture D is D = (D1, D2, Dj, ..., Dp), which can constitute the fourth-category invalid joint set (Ci, Dj, L4), wherein Ci is a random joint in the located joint set C of the operation posture C, Di is a random joint in the located joint set D of the operation posture D, and L4 is the possibility of matching the operation postures of joint Ci and joint Di.

[0117] In yet another embodiment, two joints may be randomly obtained from all located joint sets to output multiple valid joint sets.

[0118] This embodiment randomly obtains two joints from all the positioned joint sets to output multiple valid joint sets, and generates a first type of invalid joint set from the positioned joint sets of two random operation postures, generates a second type of invalid joint set based on the second robotic arm with p times of positioning abnormalities, generates a third type of invalid joint set based on the second robotic arm with q times of positioning abnormalities, and generates a fourth type of invalid joint set based on the positioned joint sets associated with the two operation postures, forming a first type of invalid joint set, a second type of invalid joint set, a third type of invalid joint set, and a fourth type of invalid joint set with recognition difficulty ranging from low to high, forming multi-level invalid joint sets with different degrees of difficulty, enriching the joint set types, and the generated joint sets can fully configure the collaborative robotic arm output template, thereby improving the accuracy of the collaborative robotic arm output template in recognizing operation postures.

[0119] In one embodiment, a valid joint set and an invalid joint set for a picking task can be randomly obtained and imported into a collaborative robot arm output template, and the possibility of matching the operating postures of two joint sets in all joint sets is output. The picking effectiveness is calculated according to the possibility, and it is analyzed whether the picking effectiveness is less than the picking effectiveness threshold value. If so, the collaborative robot arm output template is diagnosed to have completed reconfiguration, and output S102 to import the first robot arm in the located joint set into the reconfigured collaborative robot arm output template, output the first action images of all the first robot arms, and save the first action images in the action image library to update the action images of the located joints saved in the action image library. If not, the template indicators are corrected according to the picking effectiveness, and the valid joint set and invalid joint set of a randomly obtained picking task are output and imported into the collaborative robot arm output template.

[0120] In one embodiment, when calculating the picking effectiveness, the possibility of matching the operation postures of two joint sets in all joint sets and the label value of the joint set can be imported into the preset picking cost function, the picking effectiveness of all joint sets can be output, and the importance of the joint sets t0, t1, t2, t3, t4 can be diagnosed, wherein t1<t2<t3<t4, t0 is the importance of the effective joint set, t1, t2, t3, t4 are the importance of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set respectively, and the weighted average is calculated by the picking effectiveness, importance and the number of joint sets of a picking task of all joint sets, and the total picking effectiveness is output. Among them, the picking cost function can be a binary cross entropy picking cost function, a mean square error picking cost function, etc., and this embodiment does not limit the picking cost function.

[0121] In this embodiment, the recognition difficulty of the operation postures of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set is from low to high, and the importance of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set is from small to large, that is, the more difficult the operation posture of the joint set is to identify, the higher the importance, and the higher the importance of its picking effectiveness in the total picking effectiveness. After the collaborative robot arm output template is configured by constraining the total picking effectiveness, the collaborative robot arm output template can identify whether the operation posture of the collaborative robot arm with high difficulty matches, so that the collaborative robot arm can be accurately generated for the joints with difficult operation posture to diagnose, and the accuracy of the joint positioning operation posture is improved. Furthermore, the collaborative robot arm with important joint sets generated by the reconfigured collaborative robot arm output template has higher accuracy, which can improve the proportion of important joint sets (joint sets with difficult operation postures) in the total joint set, avoid a certain operation posture that is difficult to identify, resulting in too few positioning joints for the operation posture, resulting in sparse joint sets and reducing the accuracy of the configured picking robot arm, thereby improving the performance of the picking robot arm.

[0122] In this embodiment, a small number of first robotic arms with all operating postures can be collected first as example joints and updated into the located joint set, and the first robotic arms in the located joint set can be imported into a preconfigured collaborative robotic arm output template, and the first action images of all first robotic arms can be output, and the first action images can be saved in the action image library, and a second robotic arm for a picking task can be obtained from the joint set to be located and imported into the preconfigured collaborative robotic arm output template, and the video of all the imported second robotic arms can be segmented in the collaborative robotic arm output template, and the second action image of the second robotic arm can be output, and the correlation between the second action image and the first action image in the preset action image library can be calculated in the collaborative robotic arm output template, and a collaborative robotic arm can be generated based on the first robotic arm and the second robotic arm corresponding to the first action image and the second action image with the largest correlation, and when a consistent match is obtained for the operating posture of the collaborative robotic arm, the operating posture of the first robotic arm in the collaborative robotic arm can be located as The operation posture of the second robotic arm is interrupted after the operation posture of the joint set to be positioned is positioned, and the second robotic arm after the operation posture is positioned is updated to the positioned joint set corresponding to the positioned operation posture, and when the interruption positioning scenario is not met, a valid joint set and an invalid joint set are generated to reconfigure the collaborative robotic arm output template. On the one hand, a collaborative robotic arm including the second robotic arm in the joint set to be positioned is generated through the collaborative robotic arm output template, and after manual confirmation of the collaborative robotic arm, the operation posture of the first robotic arm used as an example in the collaborative robotic arm is positioned as the operation posture of the second robotic arm, without the need to manually position the operation posture of the joint to be positioned one by one, thereby improving the efficiency of joint positioning and saving time and labor costs for joint positioning. On the other hand, by manually confirming the positioned joint set and generating a positive and invalid joint set to reconfigure the collaborative robotic arm output template, the accuracy of the collaborative robotic arm output template can be improved, thereby improving the accuracy of the operation posture of the positioned joint.

[0123] Embodiment 3

[0124] Figure 3 The structure of a multi-axis joint cooperation system for a robot manipulator provided in the third embodiment of the present invention is shown in FIG. Figure 3 As shown, the multi-axis joint collaboration system for the robot manipulator is used to configure the joint positioning operation posture used by the picking manipulator, including:

[0125] A located joint set collection unit 301 is used to collect located joint sets of all operation postures, wherein the located joint set includes a first mechanical arm that has located the operation posture;

[0126] A collaborative robot arm generating unit 302 is used to obtain a second robot arm for a picking task from the set of joints to be positioned and import the second robot arm into a preconfigured collaborative robot arm output template to output a collaborative robot arm, wherein the collaborative robot arm has a first robot arm and a second robot arm;

[0127] An operation posture positioning unit 303 is used to position the operation posture of the first mechanical arm in the collaborative mechanical arm as the operation posture of the second mechanical arm when the operation posture facing the collaborative mechanical arm is consistent and matched;

[0128] A joint set updating unit 304 is used to interrupt the second mechanical arm after the positioning operation posture of the joint set to be positioned, and update the second mechanical arm after the positioning operation posture to the positioned joint set corresponding to the positioned operation posture;

[0129] A scene analysis unit 305, used to analyze whether it meets the interrupt positioning scene, and if not, complete the joint set generation unit;

[0130] A joint set generating unit 306, configured to generate a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set;

[0131] The reconfiguration unit 307 is used to reconfigure the collaborative robot arm output template according to the valid joint set and the invalid joint set, and output a collaborative robot arm generation unit.

[0132] Optionally, also include:

[0133] The action image saving unit is used to import the first robot arm in the located joint set into the preconfigured collaborative robot arm output template, output the first action images of all the first robot arms, and save the first action images in the action image library.

[0134] Optionally, the collaborative robot arm generation unit 302 includes:

[0135] A joint import unit, used to obtain a second robot arm for a picking task from the set of joints to be positioned and import it into a preconfigured collaborative robot arm output template;

[0136] An action image generating unit is used to perform video segmentation on all the imported second robotic arms in the collaborative robotic arm output template, and output a second action image of the second robotic arm;

[0137] A correlation calculation unit, used to calculate the correlation between the second action image and the first action image in the preset action image library in the collaborative robot arm output template;

[0138] The collaborative robot arm generation unit is used to generate a collaborative robot arm according to the first robot arm and the second robot arm corresponding to the first action image and the second action image with the greatest correlation.

[0139] Optionally, the scene analysis unit 305 includes:

[0140] A joint number calculation unit, used to calculate the number of joints of the first manipulator in the located joint set of all operation postures;

[0141] A joint number analysis unit is used to analyze whether the number of multiple joints is not less than a preset threshold value. If so, the interrupt positioning unit is completed. If not, the diagnosis does not meet the interrupt positioning scenario and the joint set generation unit 306 is completed;

[0142] The interrupt positioning unit is used to diagnose a scenario that meets the interrupt positioning requirement and terminate the positioning of the second robot arm with respect to the joint to be positioned.

[0143] Optionally, the joint set generating unit 306 includes:

[0144] A first manipulator set generating unit is used to select one joint from two random located joint sets in turn and output a plurality of first-category invalid joint sets;

[0145] A second robotic arm set generating unit is used to diagnose a plurality of second robotic arms with p and q positioning abnormalities from the joint set to be positioned, and generate a second type of invalid joint set based on the second robotic arms with p positioning abnormalities, and generate a third type of invalid joint set based on the second robotic arms with q positioning abnormalities, wherein the second type of invalid joint set and the third type of invalid joint set are collaborative robotic arms with two joints that have mismatched operation postures, wherein q is not less than p;

[0146] a third joint set generating unit, used for diagnosing two random positioned joint sets associated with the operation posture, selecting one joint from the two random positioned joint sets associated with the operation posture in turn, and outputting a plurality of fourth-category invalid joint sets;

[0147] The fourth joint set generating unit is used to randomly obtain two joints from all located joint sets to output multiple valid joint sets.

[0148] Optionally, the reconfiguration unit 307 includes:

[0149] A possibility size prediction unit is used to randomly obtain a valid joint set and an invalid joint set of a picking task and import them into the collaborative robot arm output template, and output the possibility size of the operation posture matching of two joint sets in all joint sets;

[0150] A picking effectiveness calculation unit, used to calculate the picking effectiveness according to the possibility size;

[0151] A picking effectiveness analysis unit, used to analyze whether the picking effectiveness is less than a picking effectiveness threshold value, if so, complete the configuration completion diagnosis unit, if not, complete the indicator correction unit;

[0152] A configuration completion diagnosis unit is used to diagnose whether the collaborative robot arm output template has completed reconfiguration and an action image storage unit has completed;

[0153] The index correction unit corrects the template index through the picking effectiveness and outputs the possibility size prediction unit.

[0154] Optionally, the picking effectiveness calculation unit includes:

[0155] The joint set picking effectiveness calculation subunit is used to import the possibility of matching the operation postures of two joint sets in all joint sets and the label values ​​of the joint sets into a preset picking cost function, and output the picking effectiveness of all joint sets;

[0156] The joint set importance diagnosis subunit is used to diagnose the importance of the joint set t0, t1, t2, t3, t4, where t1<t2<t3<t4, t0 is the importance of the valid joint set, t1, t2, t3, t4 are the importance of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set respectively;

[0157] The total picking effectiveness calculation subunit is used to calculate the weighted average through the picking effectiveness and importance of all joint sets and the number of joint sets in a picking task, and output the picking effectiveness.

[0158] The multi-axis joint collaboration system for a robot manipulator provided by an embodiment of the present invention can complete the multi-axis joint collaboration method for a robot manipulator provided by a random embodiment of the present invention, and has functional units and beneficial effects corresponding to the completion method.

[0159] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not completed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0160] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the content of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0161] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A multi-axis joint collaboration method for a robot manipulator, characterized in that: Used to configure the joint positioning operation posture used by the picking robot arm, including: Collecting a set of positioned joints of all operation postures, wherein the set of positioned joints includes a first mechanical arm that positions the operation posture; A second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template to output a collaborative robot arm, wherein the collaborative robot arm comprises a first robot arm and a second robot arm; When the operation posture toward the collaborative robot arm is obtained and identified as a match, the operation posture of the first robot arm in the collaborative robot arm is positioned as the operation posture of the second robot arm; The second mechanical arm after the positioning operation posture of the joint set to be positioned is interrupted, and the second mechanical arm after the positioning operation posture is updated to the positioned joint set corresponding to the positioned operation posture; Analyze whether it meets the interruption location scenario; If not, generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set; Reconfigure the collaborative robot arm output template according to the valid joint set and the invalid joint set, and output the collaborative robot arm by importing the second robot arm that obtains a picking task from the joint set to be located into the preconfigured collaborative robot arm output template; Generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set includes: Select one joint from two randomly located joint sets in turn, and output multiple first-category invalid joint sets; A plurality of second robotic arms with p and q positioning abnormalities are diagnosed from the joint set to be positioned, and a second type of invalid joint set is generated based on the second robotic arms with p positioning abnormalities, and a third type of invalid joint set is generated based on the second robotic arms with q positioning abnormalities, wherein the second type of invalid joint set and the third type of invalid joint set are collaborative robotic arms with two joints that have mismatched operation postures, wherein q is not less than p; Diagnose two random positioned joint sets associated with the operation posture, select one joint from the two random positioned joint sets associated with the operation posture in turn, and output multiple fourth-category invalid joint sets; Randomly obtain two joints from all located joint sets to output multiple valid joint sets; Reconfiguring the collaborative robot arm output template according to the valid joint set and the invalid joint set, comprising: Randomly obtain a valid joint set and an invalid joint set of a picking task and import them into the collaborative robot arm output template, and output the probability of matching the operation postures of two joint sets in all joint sets; Calculate the picking effectiveness by the size of the possibility; Analyze whether the picking effectiveness is less than a picking effectiveness threshold value; If so, diagnose that the collaborative robot arm output template has completed reconfiguration, import the first robot arm in the located joint set into the reconfigured collaborative robot arm output template, output the first action images of all the first robot arms, and save the first action images in the action image library; If not, the template index is corrected through the picking effectiveness, and a valid joint set and an invalid joint set of a randomly obtained picking task are output and imported into the collaborative robot arm output template.

2. The method according to claim 1, characterized in that Before the second robot arm for a picking task is obtained from the joint set to be positioned and imported into the preconfigured collaborative robot arm output template to output the collaborative robot arm, the method further includes: The first robot arms in the located joint set are imported into a preconfigured collaborative robot arm output template, first action images of all first robot arms are output, and the first action images are saved in an action image library.

3. The method according to claim 1, characterized in that The second robot arm for a picking task is obtained from the set of joints to be positioned and imported into the preconfigured collaborative robot arm output template to output the collaborative robot arm, including: A second robot arm for a picking task is obtained from the set of joints to be positioned and imported into a preconfigured collaborative robot arm output template; Performing video segmentation on all the imported second robotic arms in the collaborative robotic arm output template, and outputting second action images of the second robotic arms; Calculating, in the collaborative robot arm output template, a correlation between the second action image and the first action image in a preset action image library; A collaborative robotic arm is generated based on the first robotic arm and the second robotic arm corresponding to the first motion image and the second motion image with the greatest correlation.

4. The method according to claim 1, characterized in that Analyze whether it meets the interruption location scenario, including: Calculate the number of joints of the first robot arm in the located joint set of all operation postures; Analyze whether the number of multiple joints is not less than the preset threshold value; If so, the diagnosis meets the interruption positioning scenario, and the positioning of the second robot arm with the joint to be positioned is terminated; If not, the diagnosis does not meet the interruption positioning scenario, and the generation of a valid joint set and an invalid joint set from the positioned joint set and the to-be-positioned joint set is completed.

5. The method according to claim 4, characterized in that The picking effectiveness is calculated by the possibility size, including: Import the matching probability of the operation postures of two joint sets in all joint sets and the label values ​​of the joint sets into the preset picking cost function, and output the picking effectiveness of all joint sets; The importance of the diagnosis joint set is t0, t1, t2, t3, and t4, where t1 < t2 < t3 < t4, t0 is the importance of the valid joint set, and t1, t2, t3, and t4 are the importance of the first type of invalid joint set, the second type of invalid joint set, the third type of invalid joint set, and the fourth type of invalid joint set respectively; The weighted average is calculated by the picking effectiveness and importance of all joint sets and the number of joint sets in a picking task, and the picking effectiveness is output.

6. A multi-axis joint collaboration system for a robot manipulator, characterized in that: Used to configure the joint positioning operation posture used by the picking robot arm, including: A located joint set acquisition unit, used for acquiring located joint sets of all operation postures, wherein the located joint set includes a first mechanical arm located in the operation posture; A collaborative robot arm generating unit, used for obtaining a second robot arm for a picking task from a set of joints to be positioned, importing the second robot arm into a preconfigured collaborative robot arm output template and outputting a collaborative robot arm, wherein the collaborative robot arm comprises a first robot arm and a second robot arm; an operation posture positioning unit, configured to position the operation posture of the first mechanical arm in the collaborative mechanical arm as the operation posture of the second mechanical arm when the operation posture facing the collaborative mechanical arm is consistent and matched; A joint set updating unit, used for interrupting the second mechanical arm after the positioning operation posture of the joint set to be positioned, and updating the second mechanical arm after the positioning operation posture to the positioned joint set corresponding to the positioned operation posture; A scene analysis unit is used to analyze whether the interruption positioning scene is met, and if not, complete the joint set generation unit; A joint set generating unit, used for generating a valid joint set and an invalid joint set from the located joint set and the to-be-located joint set; A reconfiguration unit, configured to reconfigure the collaborative robot arm output template according to the valid joint set and the invalid joint set, and output a collaborative robot arm generation unit; The multi-axis joint collaboration system for a robot manipulator arm is used to execute the multi-axis joint collaboration method for a robot manipulator arm as described in any one of claims 1-5.

Images