Flexible robot arm and method for controlling a flexible robot arm

Abstract

This application relates to the field of robotic arm technology and provides a flexible robotic arm and a control method for the flexible robotic arm. The flexible robotic arm of this application includes a drive assembly, a connecting assembly disposed on the drive assembly, and an execution assembly disposed on the connecting assembly. The connecting assembly includes multiple sequentially connected connecting units. The drive assembly drives each connecting unit to change the orientation and position of the execution assembly. The execution assembly receives the drive from the drive assembly to perform a grasping action. The flexible robotic arm described in this application improves the usability of robotic arms.

Description

Technical Field

[0001] This application relates to the field of robotic arm technology, and in particular to a flexible robotic arm and a control method for the flexible robotic arm. Background Technology

[0002] With the rapid development of intelligent manufacturing and the increasing number of specialized industrial operations, the demand for precision operations such as the detection and maintenance of confined cavities in complex industrial equipment continues to grow, placing higher demands on the flexibility and spatial capabilities of robotic arms. Currently, the industrial sector commonly uses traditional rigid robots and rigid robotic arms, which, with their fixed rigid structures, can meet the standardized operational needs of conventional open environments. However, when faced with special working conditions involving narrow interiors, curved structures, and complex passageways, the adaptability of traditional rigid equipment is insufficient, making it unable to meet the operational requirements of high-end precision in confined spaces.

[0003] Traditional rigid robots and robotic arms use fixed linkage structures, resulting in large overall size, non-adjustable shape, and limited degrees of freedom of movement due to mechanical limitations. This makes them unable to adapt to confined and complex internal spaces, flexibly adjusting their working posture, and hindering their ability to penetrate deep into equipment cavities for comprehensive detection and maintenance. Furthermore, these inherent structural defects can lead to blind spots, equipment damage from impacts, and inaccurate work. These limitations significantly restrict the operational scenarios and stability of the equipment, hindering the improvement of the robotic arm's overall performance. Summary of the Invention

[0004] In view of this, this application aims to propose a flexible robotic arm to improve the quality of use of the robotic arm.

[0005] To achieve the above objectives, the technical solution of this application is implemented as follows:

[0006] A flexible robotic arm includes a drive assembly, a connection assembly disposed on the drive assembly, and an execution assembly disposed on the connection assembly;

[0007] The connection component includes multiple sequentially connected connection units, and the driving component is used to drive each connection unit to move in order to change the orientation and position of the execution component.

[0008] The execution component is used to receive the drive from the driving component to perform the grasping action.

[0009] Furthermore, the connecting assembly also includes a flexible member passing through each of the connecting units; the driving assembly includes a first driving unit that is drively connected to the flexible member, the first driving unit retracting the flexible member to drive the connecting unit to move.

[0010] Furthermore, the connecting unit includes a connecting seat, a mounting seat disposed on the connecting seat, and a cross shaft disposed on the mounting seat; the cross shaft of one of the adjacent connecting units is connected to the connecting seat of the other connecting unit.

[0011] Furthermore, the drive assembly includes a drive base; the cross shaft of the connecting unit near the drive assembly is connected to the drive base.

[0012] Furthermore, the execution component includes an execution seat; a cross shaft of the connection unit near the execution component is connected to the execution seat.

[0013] Furthermore, the execution component includes a gripper unit, and the driving component includes a second driving unit; the second driving unit is used to drive the gripper unit to open and close to complete the gripping action.

[0014] Furthermore, a transmission unit is provided between the second drive unit and the gripper unit; the gripper unit includes a first gripper and a second gripper, and the transmission unit is used to receive the drive from the second drive unit to drive the first gripper and the second gripper to move closer or further apart.

[0015] Furthermore, the transmission unit includes a slide rail, a first sliding seat and a second sliding seat slidably disposed on the slide rail, a first rack and a second rack respectively disposed on the first sliding seat and the second sliding seat, and a gear disposed on the power output end of the second drive unit; the first gripper is disposed on the first sliding seat, the second gripper is disposed on the second sliding seat, the first rack and the second rack are both meshed with the gear, and the second drive unit drives the gear to rotate so as to drive the first rack and the second rack to drive the first sliding seat and the second sliding seat to slide on the slide rail.

[0016] Compared with related technologies, this application has the following advantages:

[0017] (1) The flexible robotic arm described in this application is easy to design and maintain independently through the modular combination of drive components, connection components and execution components. It adopts a structure in which multiple connection units are connected in series. Each connection unit contributes a certain degree of bending freedom. After the multiple units are stacked, the whole can achieve a large range of continuous bending in three-dimensional space, so that the robotic arm can bend flexibly in three-dimensional space, adapt to the operation requirements in narrow and restricted spaces, and thus improve the quality of use of the robotic arm.

[0018] (2) By using a driving method in which the flexible parts are inserted through each connecting unit and independently retracted by the first driving unit, the driving source can be arranged away from the execution component, which improves the passability of the robotic arm in a confined space. Furthermore, by differentially controlling the retraction amount of each flexible part, the bending angle of the connecting component in any direction can be precisely adjusted, which is conducive to realizing the multi-degree-of-freedom spatial positioning of the execution component.

[0019] (3) By adopting a combination structure of connecting seat, mounting seat and cross shaft for the connecting unit, the structure is simple and the rotation is flexible. After multiple connecting units are connected in series, the local bending angles of each unit are superimposed, which is conducive to achieving large-angle bending and helps to improve the flexibility of the robotic arm.

[0020] (4) By connecting the cross shaft of the connecting unit near the drive assembly to the drive seat, a reliable fixed support is provided for the entire connecting assembly, effectively bearing the concentrated load generated by the tension of the flexible component at the near end of the connecting assembly, ensuring the structural stability of the robotic arm under heavy load conditions, preventing the near end of the connecting assembly from deflecting or loosening due to excessive force, and further improving the flexibility of the robotic arm.

[0021] (5) By connecting the cross axis of the connecting unit close to the execution component to the execution seat, the accurate output of the end pose can be guaranteed, and the execution seat can provide a stable mounting base, so that the execution component will not generate additional pose error due to structural loosening during the gripping operation, thereby improving gripping accuracy and reliability.

[0022] (6) By making the execution component include a gripper unit and the drive component include a second drive unit, the second drive unit independently drives the gripper unit to open and close, thus achieving decoupled control of the end gripping function and the main bending function. The two drive systems do not interfere with each other and can perform position adjustment and gripping operations simultaneously, thereby improving work efficiency.

[0023] (7) By setting up the transmission unit, the transmission unit can take over the drive of the second drive unit and distribute the driving force to the first gripper and the second gripper, driving them to move closer or further away synchronously, ensuring that the clamping force is evenly distributed on both sides of the target object, avoiding the target object from shifting or being damaged due to uneven force on one side, which is especially suitable for the stable clamping of precision parts and vulnerable parts.

[0024] (8) By making the transmission unit adopt a double-sided meshing structure of gear and first rack and second rack, the two racks move synchronously in opposite directions when the gear rotates, realizing the synchronous linear opening and closing of the first gripper and the second gripper. The gear and rack transmission is precise, the opening and closing position of the gripper can be precisely controlled by the second drive unit, and the clamping force can be precisely adjusted by the second drive unit. The structure is compact, making the structure of the execution component at the end of the robotic arm more compact, which is conducive to operation in a limited space.

[0025] This application also proposes a control method for a flexible robotic arm, applied to the aforementioned flexible robotic arm, comprising:

[0026] A dynamic model of the flexible robotic arm is established, and collision-free path planning is performed on the execution components.

[0027] The planned path is interpolated and smoothed to generate a continuous and smooth motion trajectory;

[0028] The control quantities of each of the first drive units are determined based on the desired end-effector pose.

[0029] The real-time feedback signal of the flexible robotic arm is filtered to obtain a state estimate.

[0030] The feedforward compensation amount is generated based on the tracking error of the historical cycle and superimposed on the control command of the current cycle to eliminate nonlinear interference in the transmission process of the flexible component.

[0031] Furthermore, it also includes: taking the desired end pose as input, performing nonlinear mapping through a pre-trained BP neural network, and outputting the desired rope length and tension of the flexible component corresponding to each of the first driving units; using an iterative learning control algorithm, the tracking error of the previous working cycle is processed by a feedforward compensation matrix and superimposed on the control command of the current cycle, so that the dynamic tracking error of the execution component approaches zero as the number of iterations increases, in order to overcome the nonlinear interference caused by the elastic deformation and frictional hysteresis of the flexible component.

[0032] The control method for the flexible robotic arm described in this application provides accurate physical equations for control through a dynamic model, ensures the movement safety of the robotic arm in complex obstacle environments through collision-free path planning, eliminates sudden changes in velocity and acceleration caused by discrete points on the path through interpolation smoothing, avoids motor impact, and the complete closed-loop control architecture significantly improves the end-effector positioning accuracy, which is conducive to improving the quality of use of the flexible robotic arm.

[0033] Furthermore, a BP neural network is used for inverse kinematics solving. After training with a large number of samples, the network has the ability to accurately map the nonlinear inverse kinematic relationship of the flexible arm, effectively shortening the inference time and improving the real-time response capability of the system. Moreover, the iterative learning control algorithm transforms the tracking error of the previous working cycle into the feedforward compensation amount of the current cycle through the feedforward compensation matrix. As the number of iterations increases, the feedforward compensation amount gradually approaches the ideal value, and the systematic error caused by the elastic deformation and frictional hysteresis of the flexible component is gradually eliminated. The dynamic tracking error of the execution component approaches zero with the number of iterations, and the accuracy of repeated trajectory tracking is greatly improved. Attached Figure Description

[0034] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0035] Figure 1 This is a schematic diagram of the flexible robotic arm described in an embodiment of this application;

[0036] Figure 2 This is a top view structural diagram of the flexible robotic arm described in the embodiments of this application;

[0037] Figure 3 This is a partial structural schematic diagram of the flexible robotic arm described in an embodiment of this application;

[0038] Figure 4 for Figure 1 Enlarged view of section A in the image;

[0039] Figure 5 for Figure 1 Enlarged view of section B in the image;

[0040] Figure 6 This is a flowchart of the control method for the flexible robotic arm described in the embodiments of this application;

[0041] Explanation of reference numerals in the attached figures:

[0042] 1. Driver component; 2. Connection component; 3. Execution component;

[0043] 101. First drive unit; 102. Drive base; 103. Second drive unit; 201. Connecting unit; 202. Wire hole; 301. Actuator; 302. Gripper unit; 303. Transmission unit;

[0044] 2011, Connecting seat; 2012, Mounting seat; 2013, Cross shaft; 3021, First gripper; 3022, Second gripper; 3031, Slide rail; 3032, First sliding seat; 3033, Second sliding seat; 3034, First rack; 3035, Second rack; 3036, Gear. Detailed Implementation

[0045] To make the technical solution and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0047] Furthermore, it should be noted that in the description of this application, if terms such as "upper," "lower," "inner," or "outer" appear, indicating orientation or positional relationship, these are based on the orientation or positional relationship shown in the accompanying drawings and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, if terms such as "first" or "second" appear, they are also used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0048] Furthermore, in the description of this application, unless otherwise expressly defined, the terms "installation," "connection," "joining," and "connector" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application in light of the specific circumstances.

[0049] In this application, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0050] The present application will now be described in detail through exemplary embodiments. However, it should be understood that, without further description, elements, structures, and features in one embodiment may be advantageously incorporated into other embodiments.

[0051] An embodiment of the first aspect of this application provides a flexible robotic arm, which is applied in the field of robotic arm technology and is mainly used for operation in confined spaces. Furthermore, the flexible robotic arm in this embodiment, through its innovative structural design, can effectively improve the quality of use of the robotic arm.

[0052] In related technologies, with the rapid development of intelligent manufacturing and the increasing number of special industrial operation scenarios, the demand for precision operations such as the detection and maintenance of confined cavities in complex industrial equipment continues to grow, placing higher demands on the flexibility and adaptability of robotic arms and their spatial operation capabilities. Currently, the industrial sector generally uses traditional rigid robots and rigid robotic arms, which can meet the standardized operation requirements of conventional open environments due to their fixed rigid structure. However, when faced with special working conditions such as narrow interiors, curved structures, and complex passageways, the adaptability of traditional rigid equipment is insufficient, making it unable to meet the requirements of high-end precision operations in confined spaces.

[0053] Traditional rigid robots and robotic arms use fixed linkage structures, resulting in large overall size, non-adjustable shape, and limited degrees of freedom of movement due to mechanical limitations. This makes them unable to adapt to confined and complex internal spaces, flexibly adjusting their working posture, and hindering their ability to penetrate deep into equipment cavities for comprehensive detection and maintenance. Furthermore, these inherent structural defects can lead to blind spots, equipment damage from impacts, and inaccurate work. These limitations significantly restrict the operational scenarios and stability of the equipment, hindering the improvement of the robotic arm's overall performance.

[0054] In view of this, in order to overcome the shortcomings of related technologies, the flexible robotic arm in this embodiment combines... Figures 1 to 5 As shown, the overall design includes a drive component 1, a connection component 2 disposed on the drive component 1, and an execution component 3 disposed on the connection component 2.

[0055] The connecting component 2 includes multiple connecting units 201 connected in sequence. The driving component 1 is used to drive each connecting unit 201 to move, so as to change the orientation and position of the execution component 3. The execution component 3 is used to receive the drive from the driving component 1 to perform the grasping action.

[0056] At this point, with the above configuration, the modular combination of drive component 1, connection component 2 and execution component 3 facilitates independent design and maintenance. Furthermore, the structure of multiple connection units 201 connected in series allows each connection unit 201 to contribute a certain degree of bending freedom. After the multiple units are stacked, the whole can achieve a wide range of continuous bending in three-dimensional space, enabling the robotic arm to bend flexibly in three-dimensional space, adapting to the operational needs in narrow and confined spaces, and thus improving the quality of use of the robotic arm.

[0057] Based on the above overview, specifically, the flexible robotic arm in this embodiment can be applied to, for example, the detection and maintenance of narrow cavities in complex industrial equipment. Through the setting of the connecting component 2, it receives the drive of the driving component 1 and, under the coordinated action of multiple connecting units 201, can change the orientation of the execution component 3 and ultimately perform the grasping action. Furthermore, the aforementioned execution component 3 can also be selected according to the usage requirements to complete other types of tasks.

[0058] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, include a flexible member passing through each connecting unit 201, and a driving component 1 including a first driving unit 101 that is driveably connected to the flexible member. The first driving unit 101 retracts and extends the flexible member to drive the connecting unit 201 to move.

[0059] It is understandable that by driving the flexible parts through each connecting unit 201 and independently retracting them by the first driving unit 101, the driving source can be arranged away from the execution component 3, which improves the manipulator's passability in confined spaces. Furthermore, by differentially controlling the retraction and extension of each flexible part, the bending angle of the connecting component 2 in any direction can be precisely adjusted, which is beneficial for achieving multi-degree-of-freedom spatial positioning of the execution component 3.

[0060] In specific implementation, the aforementioned flexible component can be, for example, a flexible tendon rope, and each connecting unit 201 can be provided with a wire hole 202 for the flexible component to pass through. The wire hole 202 can be covered with a protective layer to reduce the wear of the wire hole 202 on the flexible tendon rope. The number of the aforementioned flexible components can be adjusted according to the degree of freedom required when the connecting component 2 moves.

[0061] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, include a connecting unit 201 including a connecting seat 2011, a mounting seat 2012 disposed on the connecting seat 2011, and a cross shaft 2013 disposed on the mounting seat 2012, wherein the cross shaft 2013 of one of the adjacent connecting units 201 is connected to the connecting seat 2011 of the other connecting unit 201.

[0062] It is understandable that by adopting a combination structure of connecting seat 2011, mounting seat 2012 and cross shaft 2013 for connecting unit 201, the structure is simple and the rotation is flexible. After multiple connecting units 201 are connected in series, the local bending angles of each unit are superimposed, which is conducive to achieving large-angle bending and helps to improve the flexibility of the robotic arm.

[0063] In specific implementation, the aforementioned connector 2011 may be circular, for example. As mentioned above, the wire hole 202 is provided on the connector 2011 and is located in the area near the edge of the connector 2011. The aforementioned cross shaft 2013 is mounted on the mounting base 2012. The other two free ends of the cross shaft 2013 are connected to the connector 2011 of the adjacent connecting unit 201. For example, a cross shaft 2013 known to those skilled in the art can be used, which will not be described in detail here.

[0064] Continue to combine Figures 1 to 5As shown, in some exemplary embodiments, this embodiment may, for example, include a drive assembly 1 including a drive base 102, with a cross shaft 2013 near the connection unit 201 of the drive assembly 1 connected to the drive base 102.

[0065] It is understandable that by connecting the cross shaft 2013 of the connecting unit 201 near the drive assembly 1 to the drive seat 102, a reliable fixed support is provided for the entire connecting assembly 2, effectively bearing the concentrated load generated by the tension of the flexible component at the near end of the connecting assembly 2, ensuring the structural stability of the robotic arm under heavy load conditions, preventing the near end of the connecting assembly 2 from deflecting or loosening due to excessive force, and further improving the flexibility of the robotic arm.

[0066] In a specific implementation, the drive base 102 is used to assemble the first drive unit 101. The first drive unit 101 can be, for example, multiple units arranged on the drive base 102 corresponding to the number of flexible parts. Each first drive unit 101 can be, for example, a motor. The power output end of the first drive unit 101 can rotate and wind the flexible part. By adjusting the length of the flexible part wound by each first drive unit 101, the direction of the connecting unit 201 can be achieved.

[0067] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, include an execution component 3 including an execution seat 301, with a cross shaft 2013 connected to the connection unit 201 near the execution component 3.

[0068] It is understandable that by connecting the cross axis 2013 of the connection unit 201 near the execution component 3 to the execution seat 301, the accurate output of the end position posture can be guaranteed, and the execution seat 301 can provide a stable mounting base, so that the execution component 3 will not generate additional posture errors due to structural loosening during the gripping operation, thereby improving gripping accuracy and reliability.

[0069] In specific implementation, the aforementioned actuator 301 is used to assemble the actuator 3, and one side of the actuator 301 is connected to the cross shaft 2013 of the connecting unit 201 near the actuator 301. The end of the flexible part can be connected to the actuator 301, for example, so as to better control the orientation of the actuator and facilitate the execution of actions at more angles.

[0070] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, include an execution component 3 including a gripper unit 302 and a drive component 1 including a second drive unit 103, the second drive unit 103 being used to drive the gripper unit 302 to open and close in order to complete the gripping action.

[0071] It is understandable that by making the execution component 3 include the gripper unit 302 and the drive component 1 include the second drive unit 103, the gripper unit 302 is opened and closed independently by the second drive unit 103, thus achieving decoupled control of the end gripping function and the main bending function. The two drive systems do not interfere with each other and can perform posture adjustment and gripping operations simultaneously, thereby improving work efficiency.

[0072] In practical implementation, the aforementioned execution component 3 may include not only the gripper unit 302, but may also be replaced with other execution units as needed, such as a cutting unit, welding unit, or camera unit, so as to select different execution units for different usage scenarios. The aforementioned second drive unit 103 may, for example, be a motor in actual implementation.

[0073] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, provide a transmission unit 303 between the second drive unit 103 and the gripper unit 302. The gripper unit 302 includes a first gripper 3021 and a second gripper 3022. The transmission unit 303 is used to receive the drive of the second drive unit 103 to drive the first gripper 3021 and the second gripper 3022 to move closer or further apart from each other.

[0074] It is understandable that, through the setting of the transmission unit 303, the transmission unit 303 can take over the drive of the second drive unit 103, and distribute the driving force to the first gripper 3021 and the second gripper 3022, driving them to move closer or further away synchronously, ensuring that the clamping force is evenly distributed on both sides of the target object, avoiding the target object from shifting or being damaged due to uneven force on one side, which is especially suitable for the stable clamping of precision parts and vulnerable parts.

[0075] In a specific implementation, the first gripper 3021 and the second gripper 3022 may be arranged symmetrically, and the second driving unit 103 may be correspondingly located at the middle position of the line connecting the first gripper 3021 and the second gripper 3022, so as to synchronously drive the first gripper 3021 and the second gripper 3022 to move.

[0076] Continue to combine Figures 1 to 5 As shown, in some exemplary embodiments, this embodiment may, for example, include a transmission unit 303 comprising a slide rail 3031, a first sliding seat 3032 and a second sliding seat 3033 slidably disposed on the slide rail 3031, a first rack 3034 and a second rack 3035 respectively disposed on the first sliding seat 3032 and the second sliding seat 3033, and a gear 3036 disposed on the power output end of the second drive unit 103.

[0077] The first gripper 3021 is disposed on the first sliding seat 3032, the second gripper 3022 is disposed on the second sliding seat 3033, the first rack 3034 and the second rack 3035 are both meshed with the gear 3036, and the second drive unit 103 drives the gear 3036 to rotate so as to drive the first rack 3034 and the second rack 3035 to drive the first sliding seat 3032 and the second sliding seat 3033 to slide on the slide rail 3031.

[0078] It is understandable that by making the transmission unit 303 adopt a double-sided meshing structure of gear 3036 with first rack 3034 and second rack 3035, the two racks move synchronously in opposite directions when gear 3036 rotates, realizing the synchronous linear opening and closing of first gripper 3021 and second gripper 3022. Moreover, the gear 3036 rack transmission is precise, the opening and closing position of the gripper can be precisely controlled by the second drive unit 103, and the clamping force can be precisely adjusted by the second drive unit 103. The structure is compact, making the structure of the execution component 3 at the end of the robotic arm more compact, which is conducive to operation in a limited space.

[0079] In specific implementation, the power output end of the second drive unit 103 is fixedly connected to the gear 3036 and can drive the gear 3036 to rotate. As the gear 3036 rotates, the first rack 3034 and the second rack 3035 meshing on both sides of the gear 3036 also move together, causing the first gripper 3021 and the second gripper 3022 to move closer or further away from each other.

[0080] It is worth noting that, regarding the flexible robotic arm of this embodiment, based on the above exemplary implementations, in specific implementation, as a preferred embodiment, it is still composed of... Figures 1 to 5 As shown, it may include, for example, a driver component 1, a connection component 2, and an execution component 3.

[0081] The connecting component 2 is located on the driving component 1, and the execution component 3 is located on the connecting component 2. The connecting component 2 includes multiple connecting units 201 connected in sequence. The driving component 1 is used to drive each connecting unit 201 to change the orientation and position of the execution component 3. The execution component 3 is used to receive the drive from the driving component 1 to perform the grasping action.

[0082] The connecting assembly 2 further includes flexible members passing through each connecting unit 201. The driving assembly 1 includes a first driving unit 101 that is drively connected to the flexible members. The first driving unit 101 retracts and expands the flexible members to drive the connecting unit 201 to move. The connecting unit 201 includes a connecting seat 2011, a mounting seat 2012 disposed on the connecting seat 2011, and a cross shaft 2013 disposed on the mounting seat 2012. The cross shaft 2013 of one of the adjacent connecting units 201 is connected to the connecting seat 2011 of the other connecting unit 201.

[0083] The drive assembly 1 includes a drive base 102, and a cross shaft 2013 near the connecting unit 201 of the drive assembly 1 is connected to the drive base 102. The execution assembly 3 includes an execution base 301, and a cross shaft 2013 near the connecting unit 201 of the execution assembly 3 is connected to the execution base 301. The execution assembly 3 includes a gripper unit 302, and the drive assembly 1 includes a second drive unit 103. The second drive unit 103 is used to drive the gripper unit 302 to open and close to complete the gripping action. A transmission unit 303 is provided between the second drive unit 103 and the gripper unit 302. The gripper unit 302 includes a first gripper 3021 and a second gripper 3022. The transmission unit 303 is used to receive the drive from the second drive unit 103 to drive the first gripper 3021 and the second gripper 3022 to move closer or further apart.

[0084] The transmission unit 303 includes a slide rail 3031, a first sliding seat 3032 and a second sliding seat 3033 slidably disposed on the slide rail 3031, a first rack 3034 and a second rack 3035 respectively disposed on the first sliding seat 3032 and the second sliding seat 3033, and a gear 3036 disposed on the power output end of the second drive unit 103. A first gripper 3021 is disposed on the first sliding seat 3032, and a second gripper 3022 is disposed on the second sliding seat 3033. The first rack 3034 and the second rack 3035 are both meshed with the gear 3036. The second drive unit 103 drives the gear 3036 to rotate so as to drive the first rack 3034 and the second rack 3035 to drive the first sliding seat 3032 and the second sliding seat 3033 to slide on the slide rail 3031.

[0085] In the preferred embodiment of the above flexible robotic arm, the specific settings and arrangements of the drive component 1, the connection component 2, the execution component 3, etc. can still be referred to the descriptions in the above exemplary embodiments. Furthermore, in this preferred embodiment, the beneficial effects brought about by the design of the drive component 1, the connection component 2, and the execution component 3, etc., can also be referred to the descriptions in the above exemplary embodiments.

[0086] The flexible robotic arm in this embodiment adopts the above design, which facilitates independent design and maintenance. Each connecting unit 201 contributes a certain degree of bending freedom. After multiple units are stacked, the whole can achieve a large range of continuous bending in three-dimensional space, enabling the robotic arm to bend flexibly in three-dimensional space, adapting to the operation requirements in narrow and restricted spaces, and thus improving the quality of use of the robotic arm.

[0087] The second aspect of this application provides a control method for a flexible robotic arm, which is applied to the flexible robotic arm described in the first aspect of this application, in conjunction with... Figure 6 As shown, the control method includes the following steps:

[0088] Step S1: Establish the dynamic model of the flexible robotic arm and perform collision-free path planning for the execution component 3.

[0089] In step S1, for example, based on the Lagrange method and the assumption of constant curvature, the Cosserat elastic rod theory can be introduced to establish a dynamic model that includes gravity, elastic potential energy, and friction dissipation terms. The assumption of constant curvature simplifies each flexible arm segment into a uniformly curved arc segment. The Cosserat elastic rod theory introduces elastic deformation correction on this basis, so that the model can accurately describe the deformation behavior of the flexible arm under actual stress, providing physical equation support for subsequent inverse kinematics solution and trajectory tracking control.

[0090] Subsequently, an improved RRT* algorithm is used to search for a collision-free path from the starting position to the target position of execution component 3 within the 3D workspace. In the workspace, the passable area is defined as free space, and the obstacle area is defined as prohibited area. The algorithm continuously generates random sampling points in the free space and connects them to the search tree, while reconnecting and updating the cost of the generated paths, so that the planned path continuously optimizes the path length and quality while satisfying the collision-free constraint.

[0091] Step S2: Perform interpolation smoothing on the planned path to generate a continuous and smooth motion trajectory.

[0092] In step S2, when generating the motion trajectory, for example, a fifth-order B-spline can be used to interpolate the generated discrete path points to generate a continuous and smooth motion trajectory of the position, velocity and acceleration of the execution component 3.

[0093] Step S3: Solve the control quantities of each first drive unit 101 according to the desired end pose.

[0094] In step S3, the desired end-effector pose at each moment of the motion trajectory is used as input. A pre-trained backpropagation (BP) neural network is used for nonlinear mapping to directly output the desired rope length and tension of the flexible component corresponding to each first drive unit 101. The training process of the BP neural network uses a large number of pose-rope length sample pairs generated by the dynamic model as the training dataset. The network weights are iteratively optimized through the backpropagation algorithm, enabling the network to accurately map the nonlinear relationship of the inverse kinematics of the flexible arm. Compared with the traditional Jacobian matrix inversion method, the BP neural network can complete the inverse kinematics solution in milliseconds, significantly improving the real-time response capability of the system.

[0095] Step S4: Filter the real-time feedback signal of the flexible robotic arm to obtain the state estimate.

[0096] In step S4, when obtaining the state estimate, a Kalman filter can be used to filter the real-time feedback signals from the encoder and tension sensor to obtain an accurate state estimate. The Kalman filter sequentially performs four steps: state prediction, prediction covariance calculation, Kalman gain calculation, and state update. It weights and fuses the model predictions with the sensor measurements to obtain the optimal state estimate under the minimum mean square error condition. After Kalman filtering, the least squares method is used to linearly fit the relationship between the filtered state estimate and the true value, establishing a linear correction model that includes a proportionality coefficient and a fixed deviation. This model performs a secondary correction on the state estimate, eliminating the fixed-distance offset introduced by sensor installation errors.

[0097] Step S5: Generate a feedforward compensation amount based on the tracking error of the historical cycle, and superimpose it on the control command of the current cycle to eliminate nonlinear interference in the transmission process of the flexible component.

[0098] In step S5, the aforementioned tracking error, based on experience, includes structural errors generated during the assembly process, which cannot disappear over time or through filtering. Therefore, it is necessary to eliminate these errors by generating a feedforward compensation amount through the tracking error.

[0099] The control method of the flexible robotic arm in this embodiment, by adopting the flexible robotic arm in the embodiment of the first aspect of this application, can provide accurate physical equation support for control through dynamic model, ensure the motion safety of the robotic arm in complex obstacle environment through collision-free path planning, eliminate the sudden changes in speed and acceleration caused by path discrete points through interpolation smoothing processing, avoid motor impact, and the complete closed-loop control architecture significantly improves the end-effector positioning accuracy, which is conducive to improving the quality of use of flexible robotic arm.

[0100] Furthermore, in some exemplary embodiments, the control method in this embodiment may further include the following steps:

[0101] Step S6: Using the desired end pose as input, perform nonlinear mapping through a pre-trained BP neural network to output the desired rope length and tension of the flexible component corresponding to each first driving unit 101.

[0102] Step S7: Using an iterative learning control algorithm, the tracking error of the previous working cycle is processed by the feedforward compensation matrix and superimposed on the control command of the current cycle. As the number of iterations increases, the dynamic tracking error of the execution component 3 approaches zero, so as to overcome the nonlinear interference caused by the elastic deformation and frictional hysteresis of the flexible component.

[0103] Specifically, in steps S6 and S7, when performing nonlinear mapping using a pre-trained BP neural network, a desired trajectory value is set as the starting point for single-point iteration. The desired trajectory value of the current iteration is input into the continuum manipulator model to obtain the actual output response. The error between the desired trajectory and the actual output is calculated, and the desired trajectory is corrected based on the error. The convergence condition is checked, i.e., the absolute value of the error is less than a preset threshold, or the number of iterations exceeds the set maximum number of iterations. If the condition is met, the iteration of the current single point is stopped; otherwise, the iteration loop continues.

[0104] The control method for the flexible robotic arm in this embodiment uses a BP neural network for inverse kinematics solving. After training with a large number of samples, the network has the ability to accurately map the nonlinear inverse kinematic relationship of the flexible arm, effectively shortening the inference time and improving the real-time response capability of the system. Furthermore, the iterative learning control algorithm uses a feedforward compensation matrix to convert the tracking error of the previous working cycle into the feedforward compensation amount of the current cycle. As the number of iterations increases, the feedforward compensation amount gradually approaches the ideal value, and the systematic error caused by the elastic deformation and frictional hysteresis of the flexible component is gradually eliminated. The dynamic tracking error of the execution component 3 approaches zero with the number of iterations, and the accuracy of repeated trajectory tracking is greatly improved.

[0105] The above descriptions are merely some embodiments of this application and are not intended to limit this application. The technical features or structures in the foregoing different embodiments can be arbitrarily combined to form other specific technical solutions as needed. For those skilled in the art, this application can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of the claims of this application.

Claims

1. A flexible robotic arm, characterized in that: It includes a driving component (1), a connection component (2) disposed on the driving component (1), and an execution component (3) disposed on the connection component (2). The connection component (2) includes a plurality of sequentially connected connection units (201), and the drive component (1) is used to drive each connection unit (201) to move in order to change the orientation and position of the execution component (3); The execution component (3) is used to receive the drive component (1) to perform the grasping action.

2. The flexible robotic arm according to claim 1, characterized in that: The connecting component (2) further includes a flexible element that passes through each of the connecting units (201); The drive assembly (1) includes a first drive unit (101) that is connected to the flexible member for transmission. The first drive unit (101) retracts and extends the flexible member to drive the connecting unit (201) to move.

3. The flexible robotic arm according to claim 2, characterized in that: The connecting unit (201) includes a connecting seat (2011), a mounting seat (2012) disposed on the connecting seat (2011), and a cross shaft (2013) disposed on the mounting seat (2012). The cross shaft (2013) of one of the adjacent connecting units (201) is connected to the connecting seat (2011) of the other connecting unit (201).

4. The flexible robotic arm according to claim 3, characterized in that: The drive assembly (1) includes a drive housing (102); The cross shaft (2013) of the connecting unit (201) near the drive assembly (1) is connected to the drive seat (102).

5. The flexible robotic arm according to claim 3, characterized in that: The execution component (3) includes an execution seat (301); The cross shaft (2013) of the connection unit (201) near the execution component (3) is connected to the execution seat (301).

6. The flexible robotic arm according to any one of claims 1-5, characterized in that: The execution component (3) includes a gripper unit (302), and the drive component (1) includes a second drive unit (103). The second drive unit (103) is used to drive the gripper unit (302) to open and close in order to complete the gripping action.

7. The flexible robotic arm according to claim 6, characterized in that: A transmission unit (303) is provided between the second drive unit (103) and the gripper unit (302). The gripper unit (302) includes a first gripper (3021) and a second gripper (3022). The transmission unit (303) is used to receive the drive of the second drive unit (103) to drive the first gripper (3021) and the second gripper (3022) to move closer or further apart from each other.

8. The flexible robotic arm according to claim 7, characterized in that: The transmission unit (303) includes a slide rail (3031), a first sliding seat (3032) and a second sliding seat (3033) slidably disposed on the slide rail (3031), a first rack (3034) and a second rack (3035) respectively disposed on the first sliding seat (3032) and the second sliding seat (3033), and a gear (3036) disposed on the power output end of the second drive unit (103). The first gripper (3021) is disposed on the first sliding seat (3032), the second gripper (3022) is disposed on the second sliding seat (3033), the first rack (3034) and the second rack (3035) are both meshed with the gear (3036), and the second drive unit (103) drives the gear (3036) to rotate so as to drive the first rack (3034) and the second rack (3035) to drive the first sliding seat (3032) and the second sliding seat (3033) to slide on the slide rail (3031).

9. A control method for a flexible robotic arm, applied to the flexible robotic arm according to any one of claims 1 to 8, characterized in that, include: Establish a dynamic model of the flexible robotic arm and perform collision-free path planning for the execution component (3); The planned path is interpolated and smoothed to generate a continuous and smooth motion trajectory; The control quantities of each of the first driving units (101) are calculated based on the desired end pose; The real-time feedback signal of the flexible robotic arm is filtered to obtain a state estimate. The feedforward compensation amount is generated based on the tracking error of the historical cycle and superimposed on the control command of the current cycle to eliminate nonlinear interference in the transmission process of the flexible component.

10. The control method for the flexible robotic arm according to claim 9, characterized in that, Also includes: With the desired end pose as input, a pre-trained BP neural network is used for nonlinear mapping to output the desired rope length and tension of the flexible component corresponding to each first driving unit (101). An iterative learning control algorithm is adopted, and the tracking error of the previous working cycle is processed by the feedforward compensation matrix and superimposed on the control command of the current cycle. As the number of iterations increases, the dynamic tracking error of the execution component (3) approaches zero, so as to overcome the nonlinear interference caused by the elastic deformation and frictional hysteresis of the flexible component.

Images