Flexible cable operating method and system using double mechanical arms in confined environment

Abstract

A flexible cable operating method and system using double mechanical arms in a confined environment. The flexible cable operating method using double mechanical arms in a confined environment comprises: receiving a global planned path sent by a global planner, the global planned path being a global planned path obtained by the global planner performing global planning on the basis of a constrained bidirectional rapidly-exploring random tree method and according to an initial configuration and a target expected configuration of a flexible cable (S101); and using the global planned path as a guide path, and controlling the movement of double mechanical arms by means of real-time feedback closed-loop control until the flexible cable reaches the target expected configuration (S102).

Description

A method and system for operating flexible cables in a constrained environment using dual robotic arms

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on June 9, 2023, with application number: 202310686325.9 and titled “A method and system for operating flexible cables in a constrained environment using dual robotic arms”, the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present application relates to the technical field of flexible cable operation, and in particular to a method and system for operating flexible cables in a constrained environment using dual robotic arms. Background Art

[0004] Flexible cables are ubiquitous in our daily lives and production, and their manipulation scenarios are diverse, including cable assembly in industrial production, knotting in daily life, and suturing in medical surgeries. Manipulating flexible cables generally involves controlling their configuration, i.e., manipulating a robotic arm to manipulate the cable to the desired configuration. However, the large deformation, high degrees of freedom, and unknown model of flexible cables make manipulating them difficult. Furthermore, actual flexible cable manipulation in production often requires significant human involvement, making automation and autonomy difficult to achieve.

[0005] Existing flexible cable manipulation methods typically assume an ideal environment, where the end of the robotic arm can move freely in space, and do not consider potential collisions between the flexible cable and the environment, between the flexible cable and the robotic arm, between the robotic arm and the environment, or between the flexible cable itself. However, in actual manipulation tasks, obstacles are always present, and collisions with the robotic arm itself are also not negligible. Therefore, existing flexible cable manipulation methods are difficult to directly apply to practical tasks in constrained environments with obstacles.

[0006] Summary of the Invention

[0007] In view of the above problems, embodiments of the present application provide a method and system for operating flexible cables in a constrained environment using dual robotic arms, so as to overcome the above problems or at least partially solve the above problems.

[0008] In a first aspect of an embodiment of the present application, a method for operating a flexible cable in a constrained environment using a dual robotic arm is disclosed, which is applied to a local controller. The method includes:

[0009] receiving a global planning path sent by a global planner, wherein the global planning path is a global planning path obtained by the global planner based on a constrained bidirectional rapid exploration random tree method and performing global planning according to an initial configuration and a target desired configuration of the flexible cable;

[0010] The global planning path is used as a guide path, and the movement of the dual robotic arms is controlled by real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0011] Optionally, the globally planned path is used as a guide path, and the movement of the dual manipulators is controlled by real-time feedback closed-loop control until the flexible cable reaches a target desired configuration, including:

[0012] Taking each node in the global planning path as an intermediate desired configuration;

[0013] The real-time configuration of the dual robotic arms and the real-time configuration of the flexible cable are used as feedback to control the dual robotic arms to move iteratively along each node until the flexible cable is moved to a target desired configuration.

[0014] Optionally, controlling the dual robotic arms to iteratively move along each node until the flexible cable is moved to a target desired configuration comprises:

[0015] For each node in the global planning path, the local controller controls the dual robotic arms to move from the previous node to the current node in sequence;

[0016] When the distance between the actual configuration of the flexible cable and the expected configuration of the current node is less than a set threshold, the dual robotic arms are controlled to move to the next node.

[0017] Optionally, the local controller controls the dual robotic arms to move to the next node in the following manner:

[0018] A control cost model is constructed based on the attractive potential field of the flexible cable guiding the desired configuration of the flexible cable to the next node, the attractive potential field of the desired configuration of the dual manipulators to the next node, the spatial repulsive potential field of the flexible cable, and the spatial repulsive potential field of the dual manipulators.

[0019] The control cost model is solved to obtain a local control optimal solution, and the local optimal solution is used as a control input of the dual robotic arms to control the dual robotic arms to move to the next node.

[0020] Optionally, the method further includes:

[0021] When the flexible cable does not reach the target desired configuration but the local controller input is close to zero, a re-planning request is sent to the global planner, so that the global planner performs global planning based on the current configuration of the flexible cable and the target desired configuration to obtain a new global planning path;

[0022] The new global planning path sent by the global planner is received, and the new global planning path is used as a guide path to control the movement of the dual manipulators through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0023] A second aspect of the embodiments of the present application discloses a method for operating flexible cables in a constrained environment using dual robotic arms, which is applied to a global planner. The method includes:

[0024] Based on the constrained bidirectional rapid exploration random tree method, global planning is performed according to the initial configuration and the target desired configuration of the flexible cable to obtain a global planning path, wherein the global planning path consists of multiple nodes;

[0025] The global planning path is sent to the local controller, so that the local controller uses the global planning path as a guide path and controls the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0026] Optionally, the constrained bidirectional rapid exploration random tree method performs global planning according to the initial configuration and target desired configuration of the flexible cable to obtain a global planning path, including:

[0027] Two random trees start from a start node and a target node, respectively, and iteratively grow toward each other until the two random trees are connected, whereby the iterative growth stops, wherein each node on the two random trees includes: a configuration of the flexible cable and a configuration of the dual robotic arms;

[0028] A path from the start node to the target node is extracted from the two random trees, and the path is smoothed to obtain a global planning path.

[0029] Optionally, the two random trees start from the start node and the target node respectively, and iteratively grow toward each other until the two random trees are connected and the iterative growth is stopped, comprising:

[0030] For each iteration, random sampling nodes are obtained by random sampling;

[0031] Finding a first node closest to the random sampling node from the first random tree, growing from the first node toward the random sampling node to obtain an extended node of the first random tree, and finding a second node closest to the extended node of the first random tree from the second random tree, growing from the second node toward the extended node of the first random tree to obtain an extended node of the second random tree, and determining whether the extended node of the first random tree and the extended node of the second random tree can be connected;

[0032] In the case of connection, end the iterative growth;

[0033] In the case of no connection, the next iterative growth is started from the second random tree.

[0034] Optionally, nodes in the random tree are expanded by:

[0035] interpolating the flexible cable configuration reached by the previous iterative growth in the direction of the target extended configuration to obtain a new flexible cable configuration, wherein the flexible cable configuration reached by the previous iterative growth is the flexible cable configuration corresponding to the first node or the second node, and the target extended configuration is the flexible cable configuration corresponding to the random sampling point or the extended node of the first random tree;

[0036] Projecting the new flexible cable configuration into a stable flexible cable configuration using a potential energy-based projection method;

[0037] Using inverse kinematics of the manipulator to solve the configuration of the dual manipulator corresponding to the end posture of the flexible cable in the stable flexible cable configuration, the configuration of the dual manipulator represents the joint angles of the dual manipulator;

[0038] When the inverse kinematics solution is successful and the collision detection result is no collision, an extension node of a random tree is generated based on the stable flexible cable configuration and the configuration of the dual robotic arms. The extension node of the random tree includes: a first random tree extension node and a second random tree extension node.

[0039] Optionally, the method further includes:

[0040] receiving and responding to a re-planning request sent by the local controller, performing global planning based on the current configuration and the target desired configuration of the flexible cable to obtain a new global planning path;

[0041] The new global planning path is sent to the local controller, so that the local controller uses the new global planning path as a guide path and controls the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0042] Optionally, the global planning path is planned based on a flexible cable energy model, and the flexible cable energy model is constructed in the following manner:

[0043] The position of the flexible cable is characterized by the positions of m characteristic points evenly distributed on the flexible cable. The flexible cable is modeled according to the spring mass model to obtain a flexible cable energy model. The flexible cable energy model is expressed as:

[0044] Where E is the elastic potential energy of the flexible cable energy model, λ1 is the stiffness of the type 1 spring, λ2 is the stiffness of the type 2 spring, and x k represents the kth feature point, x k+1 represents the k+1th feature point, x k+2 represents the k+2th feature point, and L is the length of the flexible cable.

[0045] In a third aspect of the embodiments of the present application, a confined environment flexible cable operation system using dual robotic arms is disclosed, which is used to perform the confined environment flexible cable operation method using dual robotic arms described in the first and second aspects of the embodiments of the present application. The system includes:

[0046] A global planner configured to generate a global planning path and send the global planning path to a local controller; receive and respond to a replanning request sent by the local controller, plan a new global planning path, and send the new global planning path to the local controller;

[0047] A local controller is configured to receive the global planning path sent by the global planner, use the global planning path as a guide path, and control the movement of the dual manipulators through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration; send a re-planning request to the global planner, receive the new global planning path sent by the global planner, and use the new global planning path as a guide path, and control the movement of the dual manipulators through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0048] The embodiments of the present application include the following advantages:

[0049] In an embodiment of the present application, in order to achieve the robust operation of the flexible cable to the desired configuration by the dual robotic arms in an obstacle-constrained environment. First, the global planner is received based on the constrained bidirectional rapid exploration random tree method, and global planning is performed according to the initial configuration and target desired configuration of the flexible cable to obtain the global planning path; then the global planning path is used as the guide path, and the movement of the dual robotic arms is controlled by real-time feedback closed-loop control until the flexible cable reaches the target desired configuration. Compared with the prior art, the embodiment of the present application takes into account the possible collision between the flexible cable and the robotic arm itself in the actual application scenario and realizes obstacle avoidance; effectively combines offline global planning with local feedback control to ensure the reliability of the overall path while greatly reducing the execution error; can be directly applied to flexible cable operation tasks of different lengths, thicknesses, and materials, and is robust to flexible cable modeling errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following is a brief introduction to the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0051] FIG1 is a flowchart of a method for operating a flexible cable in a constrained environment using dual robotic arms and applied to a local controller according to an embodiment of the present application;

[0052] FIG2 is a schematic diagram of an operation process of a flexible cable from an initial configuration to a target desired configuration provided by an embodiment of the present application;

[0053] FIG3 is a flowchart of a method for operating flexible cables in a constrained environment using dual robotic arms and applied to a global planner according to an embodiment of the present application;

[0054] FIG4 is a schematic diagram of a flexible cable energy model provided in an embodiment of the present application;

[0055] FIG5 is a schematic diagram of a path planning method based on a constrained bidirectional rapid exploration random tree method provided in an embodiment of the present application;

[0056] FIG6 is a comparison diagram of an original randomly sampled configuration and a projected configuration of a flexible cable provided in an embodiment of the present application;

[0057] FIG7 is a schematic diagram of an overall solution of a method for operating a flexible cable in a constrained environment using dual robotic arms provided in an embodiment of the present application;

[0058] FIG8 is a structural block diagram of a flexible cable operation system for a constrained environment using dual robotic arms provided in an embodiment of the present application. Specific embodiments

[0059] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0060] Referring to FIG1 , FIG1 shows a flowchart of the steps of a method for operating a flexible cable in a confined environment using a dual robotic arm and applied to a local controller according to an embodiment of the present application. As shown in FIG1 , the method for operating a flexible cable in a confined environment using a dual robotic arm and applied to a local controller according to an embodiment of the present application may include steps S101 to S102:

[0061] Step S101: receiving a global planning path sent by a global planner, wherein the global planning path is a global planning path obtained by the global planner based on a constrained bidirectional rapid exploration random tree method and performing global planning according to an initial configuration and a target desired configuration of the flexible cable.

[0062] In this embodiment, during the operation of the flexible cable, the dual robotic arms grasp both ends of the flexible cable and operate the flexible cable from the initial configuration to the target desired configuration. During this process, neither the flexible cable nor the dual robotic arms can collide with obstacles. The configuration of the flexible cable refers to the shape of the flexible cable, that is, the initial configuration of the flexible cable refers to the initial shape of the flexible cable, and the target desired configuration of the flexible cable refers to the desired shape of the flexible cable. The configuration x of the flexible cable is represented by the positions of m (m is a positive integer) feature points evenly distributed on the flexible cable, wherein the position of the kth (k≤m) feature point is recorded as Then the configuration vector of the flexible cable is expressed as

[0063] In this embodiment, the global planning path refers to the path that the dual robotic arms take to control the flexible cable from its initial configuration to its desired target configuration. This global planning path is a relatively rough, collision-free global path planned offline by the global planner to ensure the solvability of the flexible cable's operation from its initial configuration to its desired target configuration. The global planning path consists of multiple nodes, each of which includes the configuration of the flexible cable and the configuration of the dual robotic arms. When the local controller controls the dual robotic arms to operate the flexible cable, it needs to receive the global planning path sent by the global planner so that it can perform local control based on this global planning path in subsequent steps.

[0064] Step S102: Using the global planning path as a guide path, the movement of the dual robotic arms is controlled by real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0065] In this embodiment, since the global planning path is an offline planning path, error compensation cannot be performed during the planning process like closed-loop control. Therefore, due to the influence of planning errors, the globally planned path cannot enable the dual robotic arms to operate the flexible cable to reach the target desired position completely without collision.

[0066] Therefore, in step S102, the local controller tracks the global planning path planned offline by the global planner as the guidance path, and simultaneously uses real-time feedback to adjust the dual-manipulator motion in a closed loop, thereby ensuring that the actual motion of the flexible cable and the dual-manipulator motion is collision-free and that the target desired configuration of the flexible cable can be accurately reached.

[0067] In an optional embodiment, the globally planned path is used as a guide path, and the movement of the dual robotic arms is controlled by real-time feedback closed-loop control until the flexible cable reaches the target desired configuration, including:

[0068] Taking each node in the global planning path as an intermediate desired configuration;

[0069] The real-time configuration of the dual robotic arms and the real-time configuration of the flexible cable are used as feedback to control the dual robotic arms to move iteratively along each node until the flexible cable is moved to a target desired configuration.

[0070] In this embodiment, the local controller uses a single-step control approach, iteratively moving to each node using each node in the globally planned path as an intermediate desired configuration. Specifically, for each node in the globally planned path, the local controller controls the dual robotic arms to move sequentially from the previous node to the current node. If the distance between the actual configuration of the flexible cable and the desired configuration of the current node is less than a set threshold, the dual robotic arms are controlled to move to the next node.

[0071] For example, as shown in Figure 2, the flexible cable needs to be operated from the upper semicircular configuration (i.e., the initial configuration) to the lower semicircular configuration (i.e., the target desired configuration). The local controller controls the dual robotic arms through three iterations, moving the first node to the second node, then from the second node to the third node, and finally from the third node to the fourth node, so that the flexible cable reaches the target desired configuration.

[0072] In an optional embodiment, the local controller controls the dual robotic arms to move to the next node in the following manner:

[0073] A control cost model is constructed based on the attractive potential field of the flexible cable guiding the desired configuration of the flexible cable to the next node, the attractive potential field of the desired configuration of the dual manipulators to the next node, the spatial repulsive potential field of the flexible cable, and the spatial repulsive potential field of the dual manipulators.

[0074] The control cost model is solved to obtain a local control optimal solution, and the local optimal solution is used as a control input of the dual robotic arms to control the dual robotic arms to move to the next node.

[0075] Specifically, the attractive potential field that guides the flexible cable to move to the desired configuration of the flexible cable at the next node is Expressed as:

[0076] Among them, x represents the current configuration of the flexible cable, x d Indicates the desired configuration of the flex cable at the next node.

[0077] The attractive potential field of the desired configuration of the dual manipulators towards the next node Expressed as:

[0078] Among them, q represents the current configuration of the dual manipulator (i.e., the joint angle of the dual manipulator), q d Represents the desired configuration of the dual manipulator at the next node.

[0079] In order to achieve local obstacle avoidance for flexible cables and dual manipulators, a repulsive potential field is designed. Define the repulsive potential field U where a single point p collides with an obstacle CO in Cartesian space. r Expressed as:

[0080] Among them, c(p) is the distance between the single point p and the nearest obstacle, ε c is the maximum distance affected by the repulsive potential field.

[0081] Then the spatial repulsive potential field of the flexible cable Expressed as:

[0082] Among them, x k represents the position of the kth feature point, and m represents the number of feature points.

[0083] The positions of the y control points on the dual manipulator in Cartesian space ξ(q)=[ξ1(q); ...; ξ y (q)], where ξ i (·) represents the forward kinematics of the i-th point. A control point is selected from each manipulator, and the spatial repulsive potential field of the dual manipulator is Expressed as:

[0084] Then, a control cost model is constructed, which includes a control cost function and constraints, wherein the control cost function aims to minimize the control loss. Specifically, the control cost model is expressed as:

[0085] in, They are the attractive potential field that guides the flexible cable to move to the desired configuration of the flexible cable to the next node, the attractive potential field of the desired configuration of the dual manipulator to the next node, the spatial repulsive potential field of the flexible cable, and the weight coefficient of the spatial repulsive potential field of the dual manipulator. δt is the time step, K u is the weight coefficient of the control input, J arm is the Jacobian matrix of the manipulator, u is the given control input, and the given control input refers to the joint velocity of the dual manipulator, C dof represents the constraint matrix of the degree of freedom of the end of the manipulator, u max Indicates the maximum joint velocity of the dual manipulator.

[0086] In this embodiment, the local control optimal solution of the control cost model is used as the single-step control input of the dual robotic arms, that is, as the control instruction of the dual robotic arms, so that the dual robotic arms can continuously move the flexible cable to the desired configuration in a constrained environment.

[0087] In an optional embodiment, considering the high degree of freedom of the flexible cable and the fact that the closed-loop system is underactuated, during operation over a large range and with large deformation, the local controller may fall into a local optimum, causing the flexible cable operation to stagnate. Therefore, to prevent the local controller from falling into a local optimum, the method further includes steps A1 and A2:

[0088] Step A1: When the flexible cable has not reached the target desired configuration, but the local controller input is close to zero, a replanning request is sent to the global planner, so that the global planner performs global planning based on the current configuration of the flexible cable and the target desired configuration to obtain a new global planning path;

[0089] Step A2: Receive the new global planning path sent by the global planner, and use the new global planning path as a guide path to control the movement of the dual manipulators through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0090] In this embodiment, when the flexible cable has not reached the target desired configuration, but the local controller input is close to zero, it indicates that the local controller is stuck in a local optimum. Therefore, in order to continue operating the flexible cable, a replanning request must be sent to the global planner. Specifically, the replanning request includes: the current and target desired configurations of the flexible cable, and the current and target configurations of the dual robotic arms. Upon receiving the replanning request, the global planner performs global planning based on the current and target desired configurations of the flexible cable, and the current and target configurations of the dual robotic arms, to obtain a new global planned path and send it to the local controller. The local controller then operates the flexible cable based on the new global planned path to achieve the target desired configuration.

[0091] Based on the above method, by detecting local optimality and replanning the global path, the problem of the local controller being limited to the local optimality is overcome, ensuring the smooth completion of the flexible cable operation task.

[0092] In this embodiment, in order to enable the dual robotic arms to robustly operate the flexible cable to the desired configuration in an obstacle-constrained environment, offline global planning and local feedback control are effectively combined to ensure the reliability of the overall path while greatly reducing the execution error; it can be directly applied to flexible cable operation tasks of different lengths, thicknesses, and materials, and is robust to flexible cable modeling errors.

[0093] Referring to FIG3 , FIG3 shows a flowchart of the steps of a method for operating a flexible cable in a constrained environment using a dual robotic arm and applied to a global planner according to an embodiment of the present application. As shown in FIG3 , the method for operating a flexible cable in a constrained environment using a dual robotic arm and applied to a global planner according to an embodiment of the present application may include steps S301 and S302:

[0094] Step S301: Based on the constrained bidirectional rapid exploration random tree method, global planning is performed according to the initial configuration and the target expected configuration of the flexible cable to obtain a global planning path, where the global planning path consists of multiple nodes.

[0095] In this embodiment, before the local controller operates the flexible cable, an offline global planner is first used to obtain a relatively rough collision-free global planning path to ensure that the flexible cable is feasible from the initial configuration to the target desired configuration.

[0096] In order to globally plan the flexible cable operation process, existing methods require precise flexible cables and incur huge computational time costs to ensure the reliability of the planned path. However, since offline planning cannot use real-time feedback for compensation like closed-loop control, the impact of model errors is greater, which can easily lead to the flexible cable not moving along the expected path.

[0097] Therefore, in this embodiment, the global planning path is planned based on the flexible cable energy model, and the flexible cable energy model is constructed in the following way:

[0098] The position of the flexible cable is characterized by the positions of m characteristic points evenly distributed on the flexible cable. The flexible cable is modeled according to the spring mass model to obtain a flexible cable energy model. The flexible cable energy model is expressed as:

[0099] Where E is the elastic potential energy of the flexible cable energy model, λ1 is the stiffness of the type 1 spring, λ2 is the stiffness of the type 2 spring, and x k represents the kth feature point, x k+1 represents the k+1th feature point, x k+2 represents the k+2th feature point, and L is the length of the flexible cable.

[0100] In practical applications, the corresponding spring stiffness value is selected according to the type of flexible cable being operated to characterize different flexible cables. As shown in Figure 4, the flexible cable is ultimately modeled as a series of masses connected by springs, with the i-th mass connected to the (i±1)-th mass via the first-type spring and to the (i±2)-th mass via the second-type spring.

[0101] In this embodiment, global planning is performed based on the flexible cable energy model. Compared with the existing technology, there is no need to calculate the precise flexible cables, which greatly reduces the calculation cost, achieves efficient planning of the global planning path, and ensures the feasibility of flexible cable operation.

[0102] Step S302: Send the global planned path to the local controller, so that the local controller uses the global planned path as a guide path and controls the movement of the dual manipulators through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0103] In this embodiment, after completing global planning, the global planner sends the global planning path to the local controller so that the local controller tracks the global planning path as a guide path. At the same time, based on the real-time configuration of the dual robotic arms and the real-time configuration of the flexible cable as feedback, the global planning path error caused by the modeling error of the flexible cable energy model during global path planning in step 301 is compensated in a closed loop, and local obstacle avoidance is achieved to ensure the accuracy of flexible cable operation.

[0104] In an optional embodiment, the constrained bidirectional rapid exploration random tree method in step S301 performs global planning according to the initial configuration and the target desired configuration of the flexible cable to obtain a global planning path, including steps B1 and B2:

[0105] Step B1: Two random trees start from a start node and a target node respectively, and iteratively grow toward each other until the two random trees are connected, and the iterative growth stops, wherein each node on the two random trees includes: a configuration of the flexible cable and a configuration of the dual robotic arms.

[0106] In this embodiment, the starting node is generated from the initial configuration of the flexible cable and the initial configuration of the dual manipulators, while the target node is generated from the target desired configuration of the flexible cable and the target configuration of the dual manipulators. For the starting node, both the initial configuration of the flexible cable and the initial configuration of the dual manipulators are known, but for the target node, only the target desired configuration of the flexible cable is known. Therefore, to determine the target configuration of the dual manipulators, the joint angles of the dual manipulators corresponding to the inverse kinematic solution of the end-point pose of the target desired configuration of the flexible cable are used as the target configuration of the dual manipulators.

[0107] In this embodiment, a constrained bidirectional rapid exploration random tree method is used as the path planning framework, and iterative growth is performed until two random trees are connected. Furthermore, iterative growth is also stopped if it is found that the generation of a new node fails or the distance between the new node and the target node is greater than the distance between the initial node and the target node. Since bidirectional growth is performed from the starting node and the target node, the efficiency of path planning is improved.

[0108] Specifically, the two random trees start from the start node and the target node respectively, and iteratively grow toward each other until the two random trees are connected and the iterative growth stops, including:

[0109] For each iteration, random sampling nodes are obtained by random sampling;

[0110] Finding a first node closest to the random sampling node from the first random tree, growing from the first node toward the random sampling node to obtain an extended node of the first random tree, and finding a second node closest to the extended node of the first random tree from the second random tree, growing from the second node toward the extended node of the first random tree to obtain an extended node of the second random tree, and determining whether the extended node of the first random tree and the extended node of the second random tree can be connected;

[0111] In the case of connection, end the iterative growth;

[0112] In the case of no connection, the next iterative growth is started from the second random tree.

[0113] For example, as shown in Figure 5, random tree A and random tree B represent the starting node and target node, respectively. The current iteration starts to grow from random tree A. First, a random sampling point 1 is randomly generated. From point 1 closest to random sampling point 1 in random tree A, growth is extended to random sampling point 1 to obtain node 3. Then, point 2 closest to node 3 is found from random tree B. Growth is extended from point 2 to point 3 to obtain point 4. At this time, it is found that there is an obstacle between points 3 and 4 and they cannot be connected, so the next iteration starts; the next iterative growth starts to grow from random tree B, a random sampling point 2 is randomly generated, and growth is extended from point 4 closest to random tree B to random sampling point 2 to obtain point 5. Then, point 3 closest to point 5 is found from random tree A. Growth is extended from point 3 to point 5 to obtain point 6. At this time, it is found that points 5 and 6 can be connected, so the iterative growth ends.

[0114] Step B2: extracting a path from the start node to the target node from the two random trees, and performing smoothing on the path to obtain a global planning path.

[0115] In this embodiment, when two random trees are connected, a path from the start node to the target node is planned. All nodes on the connected path are then extracted as the path from the start node to the target node. As shown in Figure 5 , "Start Node," "Point 1," "Point 3," "Point 6," "Point 5," "Point 4," "Point 2," and "Target Node" are sequentially extracted as the path. Because the nodes in the path are randomly grown, there are a large number of broken lines in the path, resulting in poor path smoothness. Therefore, the extracted path needs to be smoothed and used as the global planned path. Path smoothing can be performed using a path smoothing algorithm (e.g., a short-cutting algorithm).

[0116] In an optional embodiment, during each expansion and growth process, the nodes in the random tree are expanded and grown by following steps C1 to C4:

[0117] Step C1: interpolate the flexible cable configuration reached by the previous iterative growth in the direction of the target extended configuration to obtain a new flexible cable configuration, wherein the flexible cable configuration reached by the previous iterative growth refers to the flexible cable configuration corresponding to the first node or the second node, and the target extended configuration refers to the flexible cable configuration corresponding to the random sampling point or the first random tree extended node.

[0118] In this embodiment, a preset step size is used based on actual needs. The flexible cable configuration reached in the previous iterative growth process is interpolated toward the target configuration at the preset step size to obtain a new flexible cable configuration. Linear interpolation is used for the centroid position of the flexible cable, while spherical interpolation is used for relative deformation of the flexible cable that is not related to translation. This maintains the shape of the linear flexible body and prevents excessive compression of the flexible cable.

[0119] Step C2: Projecting the new flexible cable configuration into a stable flexible cable configuration based on a potential energy projection method.

[0120] In this embodiment, the configuration vector of the flexible cable is represented as As can be seen, the configuration space of the flexible cable is 3m-dimensional; however, only one subspace contains a stable configuration, which is called a manifold. Therefore, when the global planner performs global planning, the configuration of the flexible cable must be constrained to this manifold, that is, the configuration of the flexible cable at each node must remain stable. The new flexible cable configuration obtained by interpolation in step C1 is an unstable configuration, and thus needs to be projected into a stable flexible cable configuration.

[0121] The potential energy projection method specifically refers to: assuming that the elastic potential energy E of the flexible cable energy model is completely determined by the configuration x of the flexible cable, a stable configuration of the flexible cable with fixed postures at both ends needs to satisfy the local minimum of E, that is:

[0122] The two leftmost and rightmost feature points of the flexible cable are fixed, indicating that the terminal position of the flexible cable is fixed. The new flexible cable position x obtained by interpolation in step C1 is 0 As the initial value of the variable, the local minimization potential E is projected onto the manifold of the stable configuration Then x stable As a stable configuration.

[0123] Therefore, in step C2, the new flexible cable configuration obtained in step C1 is projected into a stable flexible cable configuration according to the above-mentioned potential energy projection method.

[0124] Step C3: using the inverse kinematics of the manipulator to solve the configuration of the dual manipulator corresponding to the end position of the flexible cable at the stable flexible cable configuration, the configuration of the dual manipulator represents the joint angles of the dual manipulator.

[0125] Step C4: When the inverse kinematics solution is successful and the collision detection result is no collision, an extension node of a random tree is generated based on the stable flexible cable configuration and the configuration of the dual robotic arms. The extension node of the random tree includes: a first random tree extension node and a second random tree extension node.

[0126] In this embodiment, the inverse kinematics of the manipulator is used to determine the dual manipulator configuration corresponding to the flexible cable's end position in the stable flexible cable configuration. If multiple solutions exist, the solution closest to the manipulator configuration from the previous iteration is selected as the dual manipulator configuration. Then, in step C4, a collision check is performed based on the stable flexible cable configuration and the dual manipulator configuration to determine whether the resulting stable flexible cable configuration and dual manipulator configuration are feasible.

[0127] In an optional embodiment, obtaining a random sampling node through random sampling includes randomly sampling the flexible cable in the original space to obtain an original random sampling configuration of the flexible cable. The original random sampling configuration of the flexible cable is not a stable configuration, so it is projected onto a manifold of stable configurations based on a local minimization potential to obtain a stable random sampling configuration. As shown in FIG6 , FIG6 illustrates a comparison of the original random sampling configuration of the flexible cable and the projected configuration.

[0128] In this embodiment, based on the constrained bidirectional rapid exploration random tree method, a path from the initial configuration to the target desired configuration is obtained, and obstacle avoidance is achieved on the path by discarding collision random sampling points during the path planning process, thereby ensuring the feasibility of flexible cable operation.

[0129] In an optional embodiment, when the flexible cable operation of the local controller falls into a local optimum, the method further includes:

[0130] Step D1: receiving and responding to a re-planning request sent by the local controller, performing global planning according to the current configuration and target desired configuration of the flexible cable, and obtaining a new global planning path.

[0131] Step D2: Send the new global planning path to the local controller, so that the local controller uses the new global planning path as a guide path and controls the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0132] In this embodiment, a re-planning request sent by a local controller is received, wherein the re-planning request includes: the current position and target expected position of the flexible cable, the current position and target position of the dual robotic arms, and then the global planner performs global planning based on the current position and target expected position of the flexible cable, and the current position and target position of the dual robotic arms to obtain a new global planning path.

[0133] Figure 7 illustrates a schematic diagram of the overall scheme of a method for operating a flexible cable in a constrained environment using dual robotic arms in this embodiment. First, based on the flexible cable energy model, the global planner efficiently plans a relatively rough global planning path according to the initial and target configurations of the flexible cable, thereby ensuring the feasibility of flexible cable operation. Subsequently, the local controller tracks the global planning path as a guide path, using the real-time configurations of the dual robotic arms and the flexible cable as feedback. This closed-loop compensation for global planning path errors caused by modeling errors in the flexible cable energy model is implemented to control the movement of the dual robotic arms until the flexible cable reaches the target configuration.

[0134] In the embodiment of the present invention, in order to achieve the robust operation of the flexible cable to the desired configuration by the dual robotic arms in an obstacle-constrained environment. First, the global planner is received based on the constrained bidirectional rapid exploration random tree method, and global planning is performed according to the initial configuration and target desired configuration of the flexible cable to obtain the global planning path; then the global planning path is used as the guide path, and the movement of the dual robotic arms is controlled by real-time feedback closed-loop control until the flexible cable reaches the target desired configuration. Compared with the prior art, the embodiment of the present application takes into account the possible collision between the flexible cable and the robotic arm itself in the actual application scenario and realizes obstacle avoidance; effectively combines offline global planning with local feedback control to ensure the reliability of the overall path while greatly reducing the execution error; can be directly applied to flexible cable operation tasks of different lengths, thicknesses, and materials, and is robust to flexible cable modeling errors.

[0135] 8 , a structural block diagram of a constrained environment flexible cable operation system using dual robotic arms according to an embodiment of the present application is shown. The constrained environment flexible cable operation system using dual robotic arms performs the constrained environment flexible cable operation method using dual robotic arms described in the above embodiment. As shown in FIG8 , the system may include: a global planner and a local controller. Specifically:

[0136] A global planner configured to generate a global planning path and send the global planning path to a local controller; receive and respond to a replanning request sent by the local controller, plan a new global planning path, and send the new global planning path to the local controller;

[0137] A local controller is configured to receive the global planning path sent by the global planner, use the global planning path as a guide path, and control the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration; send a re-planning request to the global planner, receive the new global planning path sent by the global planner, and use the new global planning path as a guide path, and control the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

[0138] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0139] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods and systems according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0140] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce computer-implemented processing, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0142] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic inventive concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0143] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0144] The above is a detailed introduction to the flexible cable operation method and system for a constrained environment using dual robotic arms provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for operating flexible cables in a constrained environment using dual robotic arms, characterized in that: Applied to a local controller, the method comprises: Receiving a global planning path sent by a global planner, wherein the global planning path is a global planning path obtained by the global planner based on a constrained bidirectional rapid exploration random tree method and performing global planning according to an initial configuration and a target expected configuration of the flexible cable; The global planning path is used as a guide path, and the movement of the dual robotic arms is controlled by a real-time feedback closed-loop control method until the flexible cable reaches the target desired configuration.

2. The method according to claim 1, characterized in that The method uses the global planning path as a guide path and controls the movement of the dual robotic arms by means of real-time feedback closed-loop control until the flexible cable reaches the target desired configuration, including: Taking each node in the global planning path as an intermediate desired configuration; The real-time configuration of the dual robotic arms and the real-time configuration of the flexible cable are used as feedback to control the dual robotic arms to move iteratively along each node until the flexible cable is moved to a target desired configuration.

3. The method according to claim 2, characterized in that The step of controlling the dual robotic arms to iteratively move along each node until the flexible cable is moved to a target desired configuration comprises: For each node in the global planning path, the local controller controls the dual robotic arms to move from the previous node to the current node in sequence; When the distance between the actual configuration of the flexible cable and the expected configuration of the current node is less than a set threshold, the dual robotic arm is controlled to move to the next node.

4. The method according to claim 3, characterized in that The local controller controls the dual robotic arms to move to the next node in the following manner: A control cost model is constructed based on the attractive potential field of the flexible cable guiding the flexible cable to move to the desired configuration of the next node, the attractive potential field of the dual manipulators to the desired configuration of the dual manipulators to the next node, the spatial repulsive potential field of the flexible cable, and the spatial repulsive potential field of the dual manipulators; The control cost model is solved to obtain a local control optimal solution, and the local optimal solution is used as a control input of the dual robotic arm to control the dual robotic arm to move to a next node.

5. The method according to claim 1, characterized in that The method further comprises: When the flexible cable does not reach the target desired configuration but the local controller input is close to zero, a re-planning request is sent to the global planner so that the global planner performs global planning according to the current configuration of the flexible cable and the target desired configuration to obtain a new global planning path; A new global planning path sent by the global planner is received, and the new global planning path is used as a guide path, and the movement of the dual manipulators is controlled by a real-time feedback closed-loop control method until the flexible cable reaches the target desired configuration.

6. A method for operating flexible cables in a constrained environment using dual robotic arms, characterized in that: Applied to the global planner, the method comprises: Based on the constrained bidirectional rapid exploration random tree method, global planning is performed according to the initial configuration and the target expected configuration of the flexible cable to obtain a global planning path, wherein the global planning path is composed of multiple nodes; The global planning path is sent to the local controller, so that the local controller uses the global planning path as a guide path and controls the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

7. The method according to claim 6, characterized in that The constraint-based bidirectional rapid exploration random tree method performs global planning according to the initial configuration and target expected configuration of the flexible cable to obtain a global planning path, including: Two random trees start from a start node and a target node respectively, and iteratively grow toward each other until the two random trees are connected and then the iterative growth stops, wherein each node on the two random trees includes: a configuration of a flexible cable and a configuration of a dual robotic arm; A path from the start node to the target node is extracted from the two random trees, and the path is smoothed to obtain a global planning path.

8. The method according to claim 7, characterized in that The two random trees start from a start node and a target node respectively, and iteratively grow toward each other until the two random trees are connected and the iterative growth is stopped, including: For each iteration, a random sampling node is obtained by random sampling; Find the first node closest to the random sampling node from the first random tree, expand and grow from the first node to the random sampling node to obtain the first random tree extension node, and find the second node closest to the first random tree extension node from the second random tree, expand and grow from the second node to the first random tree extension node to obtain the second random tree. random tree expansion nodes, and determining whether the first random tree expansion node and the second random tree expansion node can be connected; In the case of connection, end the iterative growth; In the case of no connection, the next iterative growth starts from the second random tree.

9. The method according to claim 8, characterized in that The nodes in the random tree grow in the following way: Interpolate the flexible cable configuration reached by the last iterative growth in the direction of the target extended configuration to obtain a new flexible cable configuration, wherein the flexible cable configuration reached by the last iterative growth refers to the flexible cable configuration corresponding to the first node or the second node, and the target extended configuration refers to the flexible cable configuration corresponding to the random sampling point or the first random tree extended node; Projecting the new flexible cable configuration into a stable flexible cable configuration based on a potential energy projection method; Using inverse kinematics of the robot arm to solve the configuration of the dual robot arm corresponding to the terminal posture of the flexible cable in the stable flexible cable configuration, the configuration of the dual robot arm represents the joint angles of the dual robot arm; When the inverse kinematics solution is successful and the collision detection result is no collision, an extension node of a random tree is generated based on the stable flexible cable configuration and the configuration of the dual robotic arms, and the extension node of the random tree includes: a first random tree extension node and a second random tree extension node.

10. The method according to claim 6, characterized in that The method further comprises: Receiving and responding to a re-planning request sent by the local controller, performing global planning according to a current configuration and a target desired configuration of the flexible cable, and obtaining a new global planning path; The new global planning path is sent to the local controller, so that the local controller uses the new global planning path as a guide path and controls the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

11. The method according to any one of claims 6 to 10, characterized in that: The global planning path is planned based on the flexible cable energy model, and the flexible cable energy model is constructed in the following way: The position of the flexible cable is characterized by the positions of m feature points evenly distributed on the flexible cable, and the flexible cable is modeled according to the spring mass model to obtain a flexible cable energy model, which is expressed as: Where E is the elastic potential energy of the flexible cable energy model, λ1 is the stiffness of the type 1 spring, λ2 is the stiffness of the type 2 spring, and x k represents the kth feature point, x k+1 represents the k+1th feature point, x k+2 represents the k+2th feature point, and L is the length of the flexible cable.

12. A flexible cable operating system in a constrained environment using dual robotic arms, characterized in that: The system is used to perform the method for operating flexible cables in a constrained environment using dual robotic arms as described in any one of claims 1 to 11, the system comprising: A global planner, the global planner is used to generate a global planning path and send the global planning path to the local controller; receive and respond to the re-planning request sent by the local controller, plan a new global planning path, and send the new global planning path to the local controller; A local controller is used to receive the global planning path sent by the global planner, use the global planning path as a guide path, and control the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration; send a re-planning request to the global planner, receive a new global planning path sent by the global planner, and use the new global planning path as a guide path, and control the movement of the dual robotic arms through real-time feedback closed-loop control until the flexible cable reaches the target desired configuration.

Images