Robotic arm end effector and positioning obstacle avoidance operation method for wind power operation and maintenance

Abstract

The present invention discloses a robotic arm end effector and a positioning and obstacle avoidance operation method for wind power operation and maintenance. It integrates the spatial perception data of the working environment and relies on the designed live working robotic arm end effector motion function structure and positioning and obstacle avoidance operation process to complete the whole process of obstacle avoidance safe arrival at the operating position of the live working point of the overlap drainage line, insulation protection shell sleeve card installation and obstacle avoidance safe exit, which helps to expand the operation function coverage of existing robot application scenarios.

Description

Technical Field

[0001] The present invention relates to a wind power operation and maintenance equipment, and more particularly to a mechanical arm end effector and a positioning and obstacle avoidance operation method for wind power operation and maintenance. Background Art

[0002] Fossil energy such as coal is non-renewable and will pollute the environment when burned to generate electricity. Therefore, it is necessary to develop clean energy such as wind power and photovoltaics for power generation. Among them, wind power, as an environmentally friendly, economical and flexible renewable energy, has developed rapidly in power generation technology, and the scale of wind power grid connection in my country is also increasing. The complexity of the operating environment of wind turbines affects the quality of their power supply and the stability of the system. Faced with the high requirements for ensuring a continuous and stable supply of electricity, the importance of the operation and maintenance of the wind turbine distribution system has become increasingly prominent. Live working is an important part of the daily operation and maintenance of the wind turbine distribution system. Faced with the harsh working conditions of high altitude, high voltage and strong electric field and the high concentration and high intensity of labor requirements, in order to solve the contradiction between power distribution reliability and the safety of manual live working, while improving labor efficiency, operational standardization and all-weather response capabilities, the current domestic and foreign technical focus is gradually focused on the assistance or replacement of humans by intelligent robots.

[0003] In recent years, China has also been catching up in the technical research and application of intelligent live-line working robots. Relevant units have developed dual-arm, 6-DOF robotic arms and multi-arm robots, and applied them in real operations. These intelligent live-line working robots with programmable, adaptable and configurable functional characteristics can realize automatic command and control of live-line working, that is, they can call corresponding response programs according to the information stored in them or some guidance information provided by the external environment, plan corresponding operation strategies, configure actuators, complete designated live-line working tasks, and improve the quality and safety of live-line working.

[0004] In the current domestic distribution network intelligent live working robot operation practice, relevant units have designed and implemented the functions of stripping the insulation shell, live connection based on different types of wire clamps, live cutting of power cables based on voice-controlled electric hydraulic wire cutters, live replacement of tension monolithic insulators, and 10kV live drainage operation. However, in general, most of the existing robot end tools have a single and limited operating function type and weak adaptability to multiple scenarios, resulting in a small number of types of workable business. Combined with factors such as cost and cost-effectiveness, it affects the further promotion and use of robots. Considering that the range of functions that the robot can complete is closely related to the end tool and its supporting response program, it is very urgent to develop execution end tools and corresponding response programs for new businesses, expand operation projects, and expand the application scenarios of existing robot operations. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a robot arm end effector and a positioning and obstacle avoidance operation method for wind power operation and maintenance. The proposed method integrates the spatial perception data of the working environment, relies on the designed live working robot arm end effector motion function structure and positioning and obstacle avoidance operation process, and completes the whole process of obstacle avoidance safe arrival at the operating position of the live working point of the lap connection drainage line, installation of the insulating protective shell sleeve and obstacle avoidance safe exit. It helps to expand the coverage of the operation functions of the live working robot for the lap connection drainage line operation scene that includes tedious processes such as connecting drainage lines, stripping, and installing insulating protective shells, and improve the utilization rate of the live working robot of the distribution network.

[0006] To achieve the above object, the present invention provides the following technical solution: a mechanical arm end effector for wind power operation and maintenance, comprising:

[0007] A connection module is used to connect with the robot arm of the live working robot and to transmit power at the end of the robot arm;

[0008] The transmission module is linked with the connection module to receive power and output it;

[0009] The clamping module is linked with the transmission module and is used to receive the power transmitted by the transmission module to generate a clamping or releasing action;

[0010] The balancing module is used to provide installation space for the transmission module, the clamping module and the connection module, and to provide a counterweight to the clamping module to adjust the center of gravity of the overall structure and balance the load;

[0011] Among them, the external host computer pre-performs kinematic modeling analysis on the robotic arm equipped with a connection module, a transmission module, a balancing module and a clamping module based on the MDH parameter method, constructs the homogeneous transformation matrix between the corresponding link coordinate systems and the posture equations between the links, and then controls the connection module, transmission module, balancing module and clamping module to perform operations according to the posture equations.

[0012] As a further improvement of the present invention, the connection module includes a docking unit and a transmission unit, the docking unit docks with the end of the robotic arm, the transmission unit receives and transmits power, the balancing module includes a fastening unit and a counterweight unit, the fastening unit is a mounting column structure, the docking unit is arranged at the bottom of the mounting column structure, the transmission unit is passed through the interior of the mounting column, the transmission module includes a driven unit, a synchronization unit, a guide shaft unit and a slider unit, a mounting plate is fixed to the upper end of the mounting column, the driven unit is mounted on the mounting plate and transmits with the transmission unit, the guide shaft unit is fixedly mounted on the mounting plate, the slider unit is slidably mounted on the guide shaft unit, the synchronization unit is mounted on the mounting plate and transmits synchronously with the driven unit, the slider unit and the synchronization unit are transmitted by a screw rod, the clamping module includes a linkage unit and a clamping claw unit, the linkage unit is mounted on the slider unit, the clamping claw unit is mounted on the linkage unit, and moves with the movement of the linkage unit.

[0013] Another aspect of the present invention provides a positioning and obstacle avoidance operation method, which mainly uses the above-mentioned wind power operation and maintenance robot end effector, and includes the following steps:

[0014] Step 1: Arrive the working robot equipped with a 6-DOF working robot arm and an end effector at the working position;

[0015] Step 2: Perform kinematic modeling analysis on the robot arm equipped with the end effector based on the MDH parameter method in advance, and construct the homogeneous transformation matrix between the corresponding link coordinate systems and the pose equations between the links;

[0016] Step 3: Based on the spatial perception data of the working environment and the position perception modeling of the working space elements, the initial position data of the robot end effector, the working target position data, and the obstacle position spatial data are obtained to model the three-dimensional motion working constraint space C of the robot arm and determine the reachable space domain C. fre and obstacle space domain set C obs , C = {C fre , C obs}, determine the initial motion point p of the end effector of the robot ini and the final operating point p ope ;

[0017] Step 4: perform positioning obstacle avoidance motion optimization calculation based on the hybrid equilibrium search mechanism, and then move to the operation point according to the calculation plan, drive the end effector to perform the operation, and the end effector performs the operation in the following way: first, the gripper unit in the clamping module grasps the insulating protective shell to be installed, and the live working robot arm connected to the docking unit in the connection module drives the motor to reverse, then drives the transmission unit, through the driven unit and the synchronization unit, and then drives the slider unit to move downward along the guide shaft, drives the gripper unit to close, and then fastens the insulating protective shell to be installed, and the motor rotates forward, then drives the slider unit to move upward along the guide shaft, and the gripper unit is separated from the protective shell; after completing the operation, according to the reverse path of the calculation plan, the robot arm and the end effector withdraw and return to the initial position.

[0018] As a further improvement of the above method, the specific steps of performing the positioning obstacle avoidance motion optimization calculation based on the hybrid balanced search mechanism in step 4 are as follows:

[0019] Step 41: Perform initialization settings;

[0020] Step 42, then perform outer loop iteration judgment, according to the judgment result of the outer loop iteration, if yes, perform inner loop traversal calculation, operation jump and dynamic update of the corresponding partial local or partial global searcher under the balanced hybrid mechanism, otherwise stop the iterative search and output the current optimal calculation plan;

[0021] Step 43: After each execution of the dynamic update of the corresponding partial local or partial global searcher under the balanced hybrid mechanism, return to continue the outer loop iteration judgment.

[0022] As a further improvement of the above method, the specific method of initialization setting in step 41 is: uniformly distributed in space C fre Inner sampling selects k points, plus the initial motion point p ini and the final operating point p ope , forming the initial point set P{p 1,0 ,…,p k+2,0}; Set k+2 initial dynamic local searchers to form an effective dynamic local searcher set LS{Ls1,…,Ls k+2}, with p i,0 As a dynamic local searcher Ls i The search starting point is p i,0 As a dynamic local searcher Ls i The search result node tree d i The root node of the node tree set D = {d1, ..., d k+2}; Set all dynamic local searchers Ls i The probability of being selected Pls i，oAnd the current search number n, initialize all Pls i，o All are 1, and n is initialized to 0; the probability threshold co1, range threshold co2, and minimum search number threshold co3 are initialized reasonably according to needs.

[0023] As a further improvement of the above method, the specific content of the outer loop iteration judgment in step 42 is: judging whether n is less than the minimum search number threshold co3 or whether there is no node tree set D that contains the initial motion point p at the same time. ini and the final operating point p ope The node tree of .

[0024] As a further improvement of the above method, the specific method of performing inner loop traversal calculation and operation jump in step 42 is: loop traversal of all currently valid dynamic local searchers, for each dynamic local searcher Ls i , calculate and update its current probability of being selected Pls i，t Equal to g (b+1) *(1-g) (m-b) ;

[0025] Among them, g is equal to the historical sampling point belonging to C fre The number of times b is divided by the total number of historical sampling times m; judge Pls i，t And determine whether the threshold co1 satisfies Pls i，t ≤co1;

[0026] After the traversal is completed, determine whether there is a probability of being selected that satisfies Pls i，t If the number of dynamic local searchers ≤co1 is greater than or equal to 1, then a partial local searcher dynamic update is performed under a balanced hybrid mechanism; otherwise, a partial global searcher dynamic update is performed under a balanced hybrid mechanism. Each time a partial local or partial global searcher dynamic update operation is performed, the current search count n=n+1 is updated.

[0027] As a further improvement of the above method, the specific method of dynamically updating the partial local searcher under the balanced hybrid mechanism in step 42 is: for each currently selected probability satisfying Pls i，t Dynamic local searcher Ls with ≤co1 i , terminate its dynamic search operation in the current local space domain, remove it from the effective dynamic local searcher set LS, and redistribute it in the space C with uniform probability fre Select a new search point p by internal sampling new ; For the new search point p that is selected new Execute node association traversal determination and path optimization operations; if point p newIf a node is added to an existing node tree, a new node is reselected and traversed, calculated, and judged accordingly; if after traversing all existing node trees in the current node tree set D, node p new If it has not been added to any existing node tree, a new dynamic local searcher Ls is created. j And add it to the effective dynamic local searcher set LS, and new As the initial point of the search, and p new Create a new node tree d for the root node new , as a new dynamic local searcher Ls j The search result node tree is added to the node tree set D;

[0028] The specific method of dynamic updating of the partial global searcher under the balanced hybrid mechanism is as follows: according to each effective dynamic local searcher Ls i The current probability distribution of being selected, sample and select one of the dynamic local searchers Ls a , a Markov chain Monte Carlo random walk mechanism based on von Mises-Fisher distribution is used to obtain a new search point p new , if Ls a The search result node tree d i The current parent node in p new The path belongs to the space C fre , p new Add to d i For the new search point p new Execute node association traversal determination and path optimization operations.

[0029] As a further improvement of the above method, the specific method of node relevance traversal determination and path optimization operation in step 42 is: traverse all existing node trees in the current node tree set D, and select the new search point p new Perform tree relevance calculations, for each existing node tree d i , if any of the nodes q l To point p new The Euclidean distance ded l,new The determination threshold co2 satisfies ded l,new ≤co2, then the node p new Add the node tree d i , with q l is its initial parent node; after the traversal of the current node tree set D is completed, if p new Only one node tree is added, and the added node tree d i The initial motion point p is included ini and the final operating point p ope At least one of them, then the node tree di Execute path reachability determination and allotropy reconnection optimization; if p new Several node trees are added, and at least one of the added node trees contains the initial motion point p ini and the final operating point p ope At least one of them is the original new Integrate other node trees into the node tree for the connection point i , followed by d i All newly added nodes are new p new Path reachability determination and allotropy reconnection optimization are performed on the node tree.

[0030] As a further improvement of the above method, the specific method of path reachability determination and allotropy reconnection optimization in step 42 is: select node tree d i Middle distance p new The node set P within the Euclidean distance co3 tem , respectively with p new and the node set P tem The connection path of each node in is used as the motion path to be tested. Based on the kinematic modeling analysis of the 6-DOF operating robot equipped with the end effector, the posture change process of the robot on the path is obtained by solving the forward and inverse kinematics according to the needs. The posture change process is compared with the feasible working space C fre and obstacle space C obs Perform collision detection and determine p new To the node set P tem The reachability of the motion paths of each node in p new The nodes that are determined to be reachable by the inter-path are added to the assumed parent node set P far ; Calculate p new The path cost to the root node, and finally select the hypothetical parent node with the lowest corresponding path cost as p new The new parent node of the node tree d is removed. i Medium new With the original parent node q l The connection relationship of P far Each node except the new parent node is represented by p new For its assumed parent node, calculate its new path cost. If the new path cost is lower than the original path cost, change the parent node of the node to p new , and remove the node tree d i The connection relationship between the node and its original parent node.

[0031] The beneficial effects of the present invention are as follows: the present invention designs a robotic arm end effector for wind power operation and maintenance in the engineering operation scenario of installing an insulating protective shell at the live overlapping lead line operation point in wind power operation and maintenance, and proposes a positioning and obstacle avoidance operation method used in conjunction with the robotic arm end effector, which effectively fills the lack of tools and functions for interrupting lead operation in the live working robot scenario, improves the functional diversification of the live working robot, expands the coverage capability of the operation scenario, and can effectively help reduce the investment in equipment and the risk of casualties in live working operations, and has high practical value, economic value and social value. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural schematic diagram of a robot end-effector for wind power operation and maintenance according to the present invention;

[0033] Figure 2 The present invention is a flow chart of a positioning and obstacle avoidance method used in conjunction with a robot arm end effector. DETAILED DESCRIPTION

[0034] The present invention will be further described below in detail with reference to the embodiments shown in the accompanying drawings.

[0035] Reference Figure 1 As shown, this embodiment provides a robot arm end effector for wind power operation and maintenance, including:

[0036] The connection module 1 is used to connect with the robot arm of the live working robot and transmit the power of the end of the robot arm;

[0037] The transmission module 3 is linked with the connection module 1 to receive power and output it;

[0038] The clamping module 4 is linked with the transmission module 3 and is used to receive the power transmitted by the transmission module 3 to generate a clamping or releasing action;

[0039] The balancing module 2 is used to provide installation space for the transmission module 3, the clamping module 4 and the connection module 1, and to provide a counterweight to the clamping module 4 to adjust the center of gravity of the overall structure and balance the load;

[0040] Among them, the external host computer pre-carries out kinematic modeling analysis on the robot arm equipped with the connection module 1, the transmission module 3, the balance module 2 and the clamping module 4 based on the MDH parameter method, constructs the homogeneous transformation matrix between the corresponding link coordinate systems and the posture equations between the links, and then controls the connection module 1, the transmission module 3, the balance module 2 and the clamping module 4 to perform operations according to the posture equations. The following units are further provided to implement the above-mentioned functional modules. The connection module 1 includes a docking unit 11 and a transmission unit 12. The docking unit 11 docks with the end of the robot arm, and the transmission unit 12 receives and transmits power. The balance module 2 includes a fastening unit 22 and a counterweight unit 21. The fastening unit 22 is a mounting column structure. The docking unit 11 is arranged at the bottom of the mounting column structure, and the transmission unit 12 is penetrated into the interior of the mounting column. The transmission module 3 includes a driven unit The driven unit 31, the synchronization unit 32, the guide shaft unit 33 and the slider unit 34, the upper end of the mounting column is fixed with a mounting plate, the driven unit 31 is mounted on the mounting plate and transmits with the transmission unit 12, the guide shaft unit 33 is fixedly mounted on the mounting plate, the slider unit 34 is slidably mounted on the guide shaft unit 33, the synchronization unit 32 is mounted on the mounting plate and is synchronously transmitted with the driven unit 31, the slider unit 34 and the synchronization unit 32 are transmitted by a screw rod, the clamping module 4 includes a linkage unit 41 and a clamping claw unit 42, the linkage unit 41 is mounted on the slider unit 34, the clamping claw unit 42 is mounted on the linkage unit 41, and moves with the movement of the linkage unit 41, so that through the setting of the above-mentioned actuator, it can effectively realize the engineering operation scenario of installing the insulating protective shell at the operation point of the live overlap drainage line in wind power operation and maintenance.

[0041] In this embodiment, based on the above-mentioned mechanical arm end effector for wind power operation and maintenance, a method for positioning and avoiding obstacles using the actuator is further provided, including the following steps:

[0042] Step 1, determine the model of the working robot, install the end effector of the robot arm of the present invention, determine the parameter values ​​of each connecting rod based on the actual measurement of the applied six-degree-of-freedom working robot arm, limit each joint, and analyze the reachable working space of the working robot arm based on the Monte Carlo method.

[0043] Step 2: The robot boom truck is lifted to the approximate location of the target work area where the insulating shell is to be installed. A 3D laser is used for global modeling to identify the main line position, the lead line position, and the positions of surrounding obstacles. Based on the modeling data, it is calculated whether the current robot parking position needs to be adjusted. If adjustment is required, the robot guides the operator to manually adjust the robot's aerial position to ensure that the work point is within the reachable work space of the 6-DOF work robot equipped with the end effector. After accurate positioning, the initial position data of the robot end effector, the work target position data, and the obstacle position space data are obtained, and the three-dimensional motion work constraint space C of the robot arm is modeled to determine the reachable space domain C. fre and obstacle space domain set C obs , C = {C fre , C obs}, determine the initial motion point p of the end effector of the robot ini and the final operating point p ope . .

[0044] Step 3: Perform kinematic modeling analysis on the robot arm equipped with the end effector in advance, and model the 6-DOF robot arm equipped with the end effector based on the MDH (Modified Denavit–Hartenberg) parameter method. Use the four parameters of link length a, link angle α, link offset d, and joint angle θ to construct the homogeneous transformation matrix between the link coordinate systems and the pose equations between the links. The homogeneous transformation matrix of coordinate i relative to coordinate system i-1 is:

[0045]

[0046] The transformation matrix of other joint coordinate systems relative to the base coordinate system is:

[0047]

[0048] by and are the posture transformation matrix and position vector respectively, then the transformation matrix of the end effector coordinate system relative to the robot base coordinate system is:

[0049]

[0050] Step 4: Perform the optimization calculation of positioning and obstacle avoidance motion based on the hybrid equilibrium search mechanism. The detailed steps are as follows:

[0051] Step 4-1, perform initialization settings, with uniform probability distribution in space C fre Inner sampling selects k points, plus the initial motion point p ini and the final operating point p ope , forming the initial point set P{p 1,0 ,…,pk+2,0}; Set k+2 initial dynamic local searchers to form an effective dynamic local searcher set LS{Ls1,…,Ls k+2}, with p i,0 As a dynamic local searcher Ls i The search starting point is p i,0 As a dynamic local searcher Ls i The search result node tree d i The root node of the node tree set D = {d1, ..., d k+2}; Set all dynamic local searchers Ls i The probability of being selected Pls i，o And the current search number n, initialize all Pls i，o are all 1, and n is initialized to 0; reasonably initialize the probability threshold co1, range threshold co2, and minimum search number threshold co3 according to needs; after initialization, execute step 4-2.

[0052] Step 4-2, execute the outer loop iteration judgment to determine whether n is less than the minimum search number threshold co3 or the node tree set D does not contain the initial motion point p ini and the final operating point p ope If yes, execute step 4-3, otherwise stop the iterative search and output the current optimal calculation solution, i.e., the initial motion point p ini To the final operating point p ope To find the minimum path cost solution, execute step 5.

[0053] Step 4-3, execute the inner loop traversal calculation, operation jump, loop traverse all currently valid dynamic local searchers, for each dynamic local searcher Ls i , calculate and update its current probability of being selected Pls i，t Equal to g (b+1) *(1-g) (m-b) , where g is equal to the historical sampling point belonging to C fre The number of times b is divided by the total number of historical sampling times m; judge Pls i，t And determine whether the threshold co1 satisfies Pls i，t ≤co1; after the traversal is completed, determine whether there is a probability of being selected that satisfies Pls i，t The number of dynamic local searchers ≤co1 is greater than or equal to 1, if so, execute step 4-4, otherwise execute step 4-5.

[0054] Step 4-4, perform dynamic update of partial local searcher under balanced hybrid mechanism, for each currently selected probability satisfying Pls i，t Dynamic local searcher Ls with ≤co1 i, terminate its dynamic search operation in the current local space domain, remove it from the effective dynamic local searcher set LS, and redistribute it in the space C with uniform probability fre Select a new search point p by internal sampling new ; For the new search point p that is selected new , execute steps 4-6 and return; if point p new If a node is added to an existing node tree, a new node is reselected and traversed, calculated, and judged accordingly; if after traversing all existing node trees in the current node tree set D, node p new If it has not been added to any existing node tree, a new dynamic local searcher Ls is created. j And add it to the effective dynamic local searcher set LS, and new As the initial point of the search, and p new Create a new node tree d for the root node new , as a new dynamic local searcher Ls j The search result node tree is added to the node tree set D; the current search times n=n+1 is updated and the execution of step 4-2 is returned.

[0055] Step 4-5, perform dynamic update of partial global searcher under balanced hybrid mechanism, according to each effective dynamic local searcher Ls i The current probability distribution of being selected, sample and select one of the dynamic local searchers Ls a , a Markov Chain Monte Carlo MCMC random walk mechanism based on the von Mises-Fisher vMF distribution is used to obtain a new search point p new , if Ls a The search result node tree d i The current parent node in p new The path belongs to the space C fre , for the new search point p new , p new Add to d i , execute step 4-6 and return; update the current search times n=n+1 and return to execute step 4-2.

[0056] Step 4-6, perform node relevance traversal determination and path optimization operations, traverse all existing node trees in the current node tree set D, and select the new search point p new Perform tree relevance calculations, for each existing node tree d i , if any of the nodes q l To point p new The Euclidean distance ded l,new The determination threshold co2 satisfies ded l,new ≤co2, then the node pnew Add the node tree d i , with q l is its initial parent node; after the traversal of the current node tree set D is completed, if p new Only one node tree is added, and the added node tree d i The initial motion point p is included ini and the final operating point p ope At least one of them, then the node tree d i Execute steps 4-7 and return; if p new Several node trees are added, and at least one of the added node trees contains the initial motion point p ini and the final operating point p ope At least one of them is the original new Integrate other node trees into the node tree for the connection point i Contains the initial motion point p ini and the final operating point p ope At least one of them, in order of d i All newly added nodes are new p new Perform steps 4-7 on the node tree and return.

[0057] Step 4-7, perform path reachability determination and conformational reconnection optimization, and select node tree d i Middle distance p new The node set P within the Euclidean distance co3 tem , respectively with p new and the node set P tem The connection path of each node in is used as the motion path to be tested. Based on the kinematic modeling analysis of the 6-DOF operating robot equipped with the end effector, the posture change process of the robot on the path is obtained by solving the forward and inverse kinematics according to the needs. The posture change process is compared with the feasible working space C fre and obstacle space C obs Perform collision detection and determine p new To the node set P tem The reachability of the motion paths of each node in p new The nodes that are determined to be reachable by the inter-path are added to the assumed parent node set P far ; Calculate p new To the root node p ini 、p ope or two points to p new The path cost of the original optimal path with the smallest cost, and finally select the assumed parent node with the lowest corresponding path cost as p new The new parent node of the node tree d is removed. i Medium new With the original parent node q lThe connection relationship of P far Each node except the new parent node is represented by p new For its assumed parent node, calculate its new path cost. If the new path cost is lower than the original path cost, change the parent node of the node to p new , and remove the node tree d i The connection relationship between the node and its original parent node.

[0058] Step 5, move to the operation point according to the calculation plan, drive the end effector to perform the operation, the clamping claw unit 42 in the clamping module 4 grasps the insulating protective shell to be installed, and the live working robot arm connected to the docking unit 11 in the connection module 1 drives the motor to reverse, then drives the transmission unit 12, through the driven unit 31 and the synchronization unit 32, and then drives the slider unit 34 to move downward along the guide shaft, drives the clamping claw unit 42 to close, and then fastens the insulating protective shell to be installed, and the motor rotates forward, then drives the driving slider unit 34 to move upward along the guide shaft, and the clamping claw unit 42 is separated from the protective shell; complete the operation, according to the reverse path of the calculation plan, the robot arm and the end effector withdraw and return to the initial position.

[0059] To summarize, this embodiment provides a robotic arm end effector and a positioning and obstacle avoidance operation method for wind power operation and maintenance, integrates the spatial perception data of the working environment, relies on the designed live working robotic arm end effector motion function structure and positioning and obstacle avoidance operation process, completes the whole process of obstacle avoidance and safe arrival at the operating position of the live working point of the splicing drainage line, installation of the insulating protective shell sleeve and obstacle avoidance and safe exit, which helps to expand the operation function coverage of existing robot application scenarios.

[0060] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A positioning and obstacle avoidance method for a manipulator end effector used in wind power operation and maintenance, characterized in that: The end effector of the robot arm includes: A connection module (1) is used to connect with the robot arm of the live working robot and to transmit power from the end of the robot arm; The transmission module (3) is linked with the connection module (1) and is used to receive power and output it; The clamping module (4) is linked with the transmission module (3) and is used to receive power transmitted by the transmission module (3) to generate a clamping or releasing action; The balancing module (2) is used to provide installation space for the transmission module (3), the clamping module (4) and the connection module (1), and to provide a counterweight to the clamping module (4) so ​​as to adjust the center of gravity of the overall structure and balance the load; The external host computer preliminarily performs kinematic modeling analysis on a robot arm equipped with a connection module (1), a transmission module (3), a balancing module (2) and a clamping module (4) based on the MDH parameter method, constructs a homogeneous transformation matrix between corresponding link coordinate systems and a posture equation between each link, and then controls the connection module (1), the transmission module (3), the balancing module (2) and the clamping module (4) to perform operations according to the posture equation; the connection module (1) comprises a docking unit (11) and a transmission unit (12), the docking unit (11) docks with the end of the robot arm, the transmission unit (12) receives power and transmits it out, the balancing module (2) comprises a fastening unit (22) and a counterweight unit (21), the fastening unit (22) is a mounting column structure, the docking unit (11) is arranged at the bottom of the mounting column structure, the transmission unit (12) is penetrated inside the mounting column, the transmission module (3) comprises a driven unit (31), a synchronization unit ( The invention relates to a method for controlling the connection module (1), the transmission module (3), the balancing module (2) and the clamping module (4). The method comprises: controlling the connection module (1), the transmission module (3), the balancing module (2) and the clamping module (4) to move with the movement of the connection module (41); and controlling the connection module (1), the transmission module (3), the balancing module (2) and the clamping module (4) to move with the movement of the connection module (41). The method comprises: Step 1: The working robot equipped with a 6-DOF working manipulator and an end effector arrives at the working position; Step 2: Perform kinematic modeling analysis on the robot arm equipped with the end effector based on the MDH parameter method in advance, and construct the homogeneous transformation matrix between the corresponding link coordinate systems and the pose equations between the links; Step 3: Based on the spatial perception data of the working environment and the position perception modeling of the working space elements, the initial position data of the robot end effector, the working target position data, and the obstacle position spatial data are obtained to model the three-dimensional motion working constraint space C of the robot arm and determine the reachable space domain C. fre and obstacle space domain set C obs , C={C fre , C obs }, determine the initial motion point p of the end effector of the robot ini and the final operating point p ope ; Step 4, performing a positioning obstacle avoidance motion optimization calculation based on a hybrid equilibrium search mechanism, then moving to the operation point according to the calculation scheme, driving the end effector to perform the operation, and the end effector performs the operation in the following manner: first, the gripper unit (42) in the gripper module (4) grasps the insulating protective shell to be installed, and the live working robot arm connected to the docking unit (11) in the connection module (1) drives the motor in reverse, then drives the transmission unit (12), through the driven unit (31) and the synchronization unit (32), and then drives the slider unit (34) to move downward along the guide shaft, driving the gripper unit (42) to close, and then fastening the insulating protective shell to be installed, and the motor rotates forward, then drives the slider unit (34) to move upward along the guide shaft, and the gripper unit (42) is separated from the protective shell; after completing the operation, according to the reverse path of the calculation scheme, the robot arm and the end effector exit and return to the initial position; the specific steps of performing the positioning obstacle avoidance motion optimization calculation based on the hybrid equilibrium search mechanism in step 4 are as follows: Step 41: Perform initialization settings; Step 42, then perform outer loop iteration judgment, according to the judgment result of the outer loop iteration, if yes, perform inner loop traversal calculation, operation jump and dynamic update of the corresponding partial local or partial global searcher under the balanced hybrid mechanism, otherwise stop the iterative search and output the current optimal calculation plan; Step 43: After each execution of the dynamic update of the corresponding partial local or partial global searcher under the balanced hybrid mechanism, return to continue the outer loop iteration judgment; the specific method of initialization setting in step 41 is: uniformly distribute the space C with probability fre Inner sampling selects k points, plus the initial motion point p ini and the final operating point p ope , forming the initial point set P{p 1,0 ,…,p k+2,0 }; Set k+2 initial dynamic local searchers to form an effective dynamic local searcher set LS{Ls1,…,Ls k+2 }, with p i,0 As a dynamic local searcher Ls i The search starting point is p i,0 As a dynamic local searcher Ls i The search result node tree d i The root node of the node tree set D = {d1, ..., d k+2 }; Set all dynamic local searchers Ls i The probability of being selected Pls i，o And the current search number n, initialize all Pls i，o are all 1, and n is initialized to 0; the probability threshold co1, range threshold co2, and minimum search number threshold co3 are initialized reasonably according to the needs; the specific content of the outer loop iteration judgment in step 42 is: determine whether n is less than the minimum search number threshold co3 or the node tree set D does not contain the initial motion point p at the same time ini and the final operating point p ope The specific method of performing the inner loop traversal calculation and operation jump in step 42 is: loop traversing all currently valid dynamic local searchers, for each dynamic local searcher Ls i , calculate and update its current probability of being selected Pls i，t Equal to g (b+1) *(1-g) (m-b) ; Among them, g is equal to the historical sampling point belonging to C fre The number of times b is divided by the total number of historical sampling times m; judge Pls i，t And determine whether the threshold co1 satisfies Pls i，t ≤co1; After the traversal is completed, determine whether there is a probability of being selected that satisfies Pls i，t If the number of dynamic local searchers ≤co1 is greater than or equal to 1, then a partial local searcher dynamic update is performed under a balanced hybrid mechanism; otherwise, a partial global searcher dynamic update is performed under a balanced hybrid mechanism. Each time a partial local or partial global searcher dynamic update operation is performed, the current search count n=n+1 is updated.

2. The positioning and obstacle avoidance operation method of the end effector of a manipulator arm for wind power operation and maintenance according to claim 1 is characterized in that: The specific method of dynamically updating the partial local searcher under the balanced hybrid mechanism in step 42 is: for each currently selected probability satisfying Pls i，t Dynamic local searcher Ls with ≤co1 i , terminate its dynamic search operation in the current local space domain, remove it from the effective dynamic local searcher set LS, and redistribute it in the space C with uniform probability fre Select a new search point p by internal sampling new ; For the new search point p that is selected new Execute node relevance traversal determination and path optimization operations; If point p new If a node is added to an existing node tree, a new node is reselected and traversed, calculated, and judged accordingly; if after traversing all existing node trees in the current node tree set D, node p new If it has not been added to any existing node tree, a new dynamic local searcher Ls is created. j And add it to the effective dynamic local searcher set LS, and new As the initial point of the search, and p new Create a new node tree d for the root node new , as a new dynamic local searcher Ls j The search result node tree is added to the node tree set D; The specific method of dynamic updating of the partial global searcher under the balanced hybrid mechanism is as follows: according to each effective dynamic local searcher Ls i The current probability distribution of being selected, sample and select one of the dynamic local searchers Ls a , a Markov chain Monte Carlo random walk mechanism based on von Mises-Fisher distribution is used to obtain a new search point p new , if Ls a The search result node tree d i The current parent node in p new The path belongs to the space C fre , p new Add to d i For the new search point p new Execute node association traversal determination and path optimization operations.

3. The positioning and obstacle avoidance method of the end effector of a manipulator arm for wind power operation and maintenance according to claim 2 is characterized in that: The specific method of node relevance traversal determination and path optimization operation in step 42 is: traverse all existing node trees in the current node tree set D, and select the new search point p new Perform tree relevance calculations, for each existing node tree d i , if any of the nodes q l To point p new The Euclidean distance ded l,new The determination threshold co2 satisfies ded l,new ≤co2, then the node p new Add the node tree d i , with q l is its initial parent node; After the traversal of the current node tree set D is completed, if p new Only one node tree is added, and the added node tree d i The initial motion point p is included ini and the final operating point p ope At least one of them, then the node tree d i Execute path reachability determination and allotropy reconnection optimization; if p new Several node trees are added, and at least one of the added node trees contains the initial motion point p ini and the final operating point p ope At least one of them is the original new Integrate other node trees into the node tree for the connection point i , followed by d i All newly added nodes are new p new Path reachability determination and allotropy reconnection optimization are performed on the node tree.

4. The method for positioning and avoiding obstacles of a manipulator end effector for wind power operation and maintenance according to claim 3, characterized in that: The specific method of path reachability determination and allotropy reconnection optimization in step 42 is: select node tree d i Middle distance p new The node set P within the Euclidean distance co3 tem , respectively with p new and the node set P tem The connection path of each node in is used as the motion path to be tested. Based on the kinematic modeling analysis of the 6-DOF operating robot equipped with the end effector, the posture change process of the robot on the path is obtained by solving the forward and inverse kinematics according to the needs. The posture change process is compared with the feasible working space C fre and obstacle space C obs Perform collision detection and determine p new To the node set P tem The reachability of the motion paths of each node in p new The nodes that are determined to be reachable by the inter-path are added to the assumed parent node set P far ; Calculate p new The path cost to the root node, and finally select the hypothetical parent node with the lowest corresponding path cost as p new The new parent node of the node tree d is removed. i Medium new With the original parent node q l The connection relationship of P far Each node except the new parent node is represented by p new For its assumed parent node, calculate its new path cost. If the new path cost is lower than the original path cost, change the parent node of the node to p new , and remove the node tree d i The connection relationship between the node and its original parent node.

Images