CN115157286A - Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof - Google Patents
Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof Download PDFInfo
- Publication number
- CN115157286A CN115157286A CN202210980395.0A CN202210980395A CN115157286A CN 115157286 A CN115157286 A CN 115157286A CN 202210980395 A CN202210980395 A CN 202210980395A CN 115157286 A CN115157286 A CN 115157286A
- Authority
- CN
- China
- Prior art keywords
- hanging
- picking
- control box
- mechanical arm
- degree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J18/00—Arms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/08—Programme-controlled manipulators characterised by modular constructions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
Abstract
The invention discloses an automatic ground wire hanging and picking robot based on optical vision and a remote control method thereof. Automatic earth connection robot of plucking of hanging based on light vision includes: the device comprises a lifting platform, a control box, a tool support, a multi-degree-of-freedom mechanical arm, a tail end grounding wire hanging/removing tool, a first shooting mechanism and a second shooting mechanism. The control box is fixed on the lifting platform and can control the lifting platform to move longitudinally. The tool support is fixed on the lifting platform. The multi-degree-of-freedom mechanical arm is fixed on the tool support, and the control box can control the multi-degree-of-freedom mechanical arm to actuate. The tail end earth wire hanging/removing tool is fixed on a tail end joint of the multi-degree-of-freedom mechanical arm, and the control box can control the tail end earth wire hanging/removing tool to act. Therefore, the invention has simple assembly and easy disassembly, can freely move in an outdoor complex environment when the wind power is less than one level, and can carry out full-automatic hanging-picking operation without manual intervention.
Description
Technical Field
The invention relates to the technical field of high-altitude power operation, in particular to an automatic ground wire hanging and picking robot based on optical vision and a remote control method thereof.
Background
With the continuous development of robot technology and intelligent control algorithms, some robots are applied to the internal operation of a transformer substation, but mainly focus on the inspection work of equipment or lines, and the problem of high-altitude operation of high-voltage lines is not solved. Therefore, the development of robots with live working capability to replace manual work has become a necessary trend in future electric power operation and maintenance.
Although a few transformer substations adopt a mode of lifting aerial ladders to lift overhaul workers to hang and pick lines and equipment ground wires with higher voltage levels, the transformer substations still belong to a manual operation mode, physical energy consumption of the overhaul workers can be reduced, and the safety distance of the overhaul workers is greatly shortened.
The high-altitude operation robot of the existing transformer substation is easily interfered by the external environment and shakes due to insufficient rigidity of the high-altitude operation car hopper arm and the influence of high-altitude wind speed, so that the grabbing operation is difficult to position, the grabbing success rate is low, the operation efficiency is low, and the application and popularization are difficult to realize.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The invention aims to provide an automatic ground wire hanging and picking robot based on optical vision and a remote control method thereof, which are simple to assemble and easy to disassemble, can freely move in an outdoor complex environment when wind power is less than one level, and can carry out full-automatic hanging-picking operation without manual intervention.
In order to achieve the above object, an aspect of the present invention provides an automatic ground wire hanging and picking robot based on optical vision, including: the device comprises a lifting platform, a control box, a tool support, a multi-degree-of-freedom mechanical arm, a tail end grounding wire hanging/removing tool, a first shooting mechanism and a second shooting mechanism. The control box is fixed on the lifting platform and can control the lifting platform to move longitudinally. The tool support is fixed on the lifting platform. The multi-degree-of-freedom mechanical arm is fixed on the tool support, and the control box can control the multi-degree-of-freedom mechanical arm to actuate. The tail end earth wire hanging/removing tool is fixed on a tail end joint of the multi-degree-of-freedom mechanical arm, and the control box can control the tail end earth wire hanging/removing tool to move. The first shooting mechanism is fixed on the tool support and used for acquiring a terminal operation scene, and the first shooting mechanism is electrically connected with the control box. The second shooting mechanism is fixed at the tail end of the multi-degree-of-freedom mechanical arm and used for acquiring the pose of a picking and hanging point of the high-altitude wire, and the second shooting mechanism is electrically connected with the control box.
In one embodiment of the present invention, the first photographing mechanism is a panoramic camera, and the second photographing mechanism is a 3D camera.
In an embodiment of the present invention, the robot for automatically hooking and unhooking a ground wire based on optical vision further includes: moving platform, camera cloud platform and earth connection clamping jaw. The mobile platform is fixed below the lifting platform through the connecting firmware, and the control box can control the mobile platform to move. The camera cloud platform is fixed in on the end of multi freedom arm, and the 3D camera is fixed in on the camera cloud platform. And the grounding wire clamping jaw is fixed on the tool bracket.
In an embodiment of the invention, the control box can control the moving platform and the lifting platform to move to the picking and hanging positions of the target high-altitude wires according to the position information of the high-altitude wires acquired by the panoramic camera.
In an embodiment of the invention, the control box can control the multi-degree-of-freedom mechanical arm to move according to the pose information of the high-altitude wire picking and hanging point acquired by the 3D camera, and guide the tail end grounding wire hanging/picking tool to hang the grounding wire clamping jaw to the high-altitude wire picking and hanging point, so that the high-altitude wire picking and hanging point is operated.
In an embodiment of the invention, the terminal grounding wire hanging/picking tool comprises a clamp and a turnbuckle mechanism, and the control box can control the clamp and the turnbuckle to finish the operation of hanging and picking the grounding wire clamping jaw.
The invention further provides a remote control method for remotely controlling the automatic ground wire hanging and picking robot based on the optical vision. The remote control method comprises the following steps: and S1, powering up the robot system for initialization, and performing program self-inspection. And S2, identifying and acquiring the position information of the target high-altitude wire by the panoramic camera at a distance, and transmitting a signal of the position information to the control box. And S3, the control box calculates a route needing to be driven according to the position information and transmits a control signal of the route to the mobile platform. And S4, moving the moving platform according to the driving route, and moving the robot for automatically hanging and picking the grounding wire to the position of the target overhead wire. And S5, converting the final position signal transmitted by the panoramic camera into a control signal by the control box, transmitting the control signal to the lifting platform, and lifting the lifting platform according to the control signal, so that the multi-degree-of-freedom mechanical arm reaches a proper hanging and picking operation point. And S6, the control box sends a control signal to the multi-degree-of-freedom mechanical arm, so that the automatic ground wire hanging and picking robot enters an automatic hanging and picking mode. And S7, the 3D camera starts to automatically shoot hanging and picking point position information of the target wire, and accurate point cloud information of the hanging and picking points is transmitted to the control box. And S8, the control box calculates the motion trail of the multi-degree-of-freedom mechanical arm according to the accurate point cloud information and transmits a motion trail signal to the multi-degree-of-freedom mechanical arm. And S9, clamping the grounding wire clamping jaw by the grounding wire hanging/picking tool at the tail end of the multi-degree-of-freedom mechanical arm, moving according to the motion track signal, and inserting the grounding wire clamping jaw into a hanging and picking point, so that the automatic grounding wire hanging and picking robot can perform autonomous operation.
Compared with the prior art, the automatic ground wire hanging and picking robot based on the optical vision and the remote control method thereof have the following beneficial effects:
1. the invention is a modularized composition, each module is in communication connection through a middle control box, and meanwhile, the modules can complete work cooperatively;
2. the robot has a strong hardware foundation, each module has a space for redevelopment, and the robot can be quickly adjusted when dealing with complex working conditions;
3. compared with the traditional manual wire hanging/removing operation, the robot can move freely, can automatically hang/remove the grounding wire from the target wire in the whole process, and has larger working range and higher efficiency;
4. the robot has a complete system architecture, is innovative in image processing algorithm and track planning algorithm, is improved in accuracy and stability compared with the current ground wire hanging/picking robot of the transformer substation, and is suitable for wider operation working conditions;
5. the problem of current automatic string/take earth connection control system mainly instruct the maintainer operation through the picture information that the camera was shot, reference information is comparatively limited, requires higher to the operating level of maintainer is solved.
Drawings
Fig. 1 is a schematic side view of a robot for automatically hanging and picking a grounding wire based on optical vision according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a remote control method according to another embodiment of the present invention;
FIG. 3 is a logic flow diagram of a remote control method in accordance with an embodiment of the present invention;
fig. 4 is a schematic flow chart of acquiring hanging off point position information by a 3D camera according to a remote control method according to an embodiment of the present invention;
fig. 5 is a schematic view of a camera calibration board of a remote control method according to an embodiment of the present invention.
Description of the main reference numerals:
the system comprises a 1-moving platform, a 2-lifting platform, a 3-control box, a 4-multi-degree-of-freedom mechanical arm, a 5-panoramic camera, a 6-3D camera, a 7-tail end grounding wire hanging/removing tool, an 8-grounding wire clamping jaw, a 9-connecting fastener, a 10-camera holder and an 11-tool support.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
Fig. 1 is a schematic side view of an automatic ground wire hooking and detaching robot based on optical vision according to an embodiment of the present invention. As shown in fig. 1, in a first aspect, an automatic ground wire hooking and hooking robot based on optical vision according to a preferred embodiment of the present invention includes: the device comprises a lifting platform 2, a control box 3, a tool support 11, a multi-degree-of-freedom mechanical arm 4, a tail end grounding wire hanging/picking tool 7, a first shooting mechanism and a second shooting mechanism. The control box 3 is fixed on the lifting platform 2, and the control box 3 can control the lifting platform 2 to move longitudinally. The tool holder 11 is fixed to the lifting platform 2. The multi-degree-of-freedom mechanical arm 4 is fixed to the tool support 11, and the control box 3 can control the multi-degree-of-freedom mechanical arm 4 to operate. The end wire attaching/detaching tool 7 is fixed to the end joint of the multi-degree-of-freedom robot arm 4, and the control box 3 can control the end wire attaching/detaching tool 7 to be actuated. The first shooting mechanism is fixed on the tool support 11, the first shooting mechanism is used for acquiring a terminal operation scene, and the first shooting mechanism is electrically connected with the control box 3. The second shooting mechanism is fixed at the tail end of the multi-degree-of-freedom mechanical arm 4, the second shooting mechanism is used for acquiring the pose of a picking and hanging point of the high-altitude conducting wire, and the second shooting mechanism is electrically connected with the control box 3.
In an embodiment of the present invention, the first shooting mechanism is a panoramic camera 5, and the second shooting mechanism is a 3D camera 6, and the 3D camera 6 is used for collecting pose information of the picking point in a short distance.
In an embodiment of the present invention, the robot for automatically hanging and picking the grounding wire based on optical vision further includes: moving platform 1, camera cloud platform 10 and earth connection clamping jaw 8. The mobile platform 1 is fixed below the lifting platform 2 through a connecting firmware 9, and the control box 3 can control the mobile platform 1 to move. The camera pan-tilt 10 is fixed to the end of the multi-degree-of-freedom mechanical arm 4, and the 3D camera 6 is fixed to the camera pan-tilt 10. And the earth wire clamping jaw 8 is fixed to the tool holder 11.
In an embodiment of the present invention, the control box 3 can control the moving platform 1 and the lifting platform 2 to move to the picking and hanging position of the target high-altitude conductor according to the position information of the high-altitude conductor acquired by the panoramic camera 5.
In an embodiment of the invention, the control box 3 can control the multi-degree-of-freedom mechanical arm 4 to move according to the pose information of the high-altitude wire picking and hanging point acquired by the 3D camera 6, and guide the tail end grounding wire hanging/picking tool 7 to hang the grounding wire clamping jaw 8 to the high-altitude wire picking and hanging point position, so as to operate the high-altitude wire picking and hanging point.
In an embodiment of the present invention, the terminal earth wire hanging/picking tool 7 includes a clamp and a turnbuckle mechanism, and the control box 3 can control the clamp and the turnbuckle to complete the operation of hanging and picking the earth wire clamping jaw 8.
Fig. 2 is a flowchart illustrating a remote control method according to another embodiment of the present invention. Fig. 3 is a logic flow diagram of a remote control method according to an embodiment of the invention. In a second aspect, as shown in fig. 2 to 3, the present invention provides a remote control method for remotely controlling the above-mentioned automatic ground wire hooking and unhooking robot based on light vision. The remote control method comprises the following steps: and S1, powering up the robot system for initialization, and performing program self-inspection. And S2, recognizing and acquiring the position information of the target high-altitude wire at a distance by the panoramic camera 5, and transmitting a signal of the position information to the control box 3. And S3, the control box 3 calculates a route needing to be driven according to the position information and transmits a control signal of the route to the mobile platform 1. And S4, moving the moving platform 1 according to the driving route, and moving the robot for automatically hanging and picking the grounding wire to the position of the target high-altitude wire. And S5, converting the final position signal transmitted by the panoramic camera 5 into a control signal by the control box 3, transmitting the control signal to the lifting platform 2, and lifting the lifting platform 2 according to the control signal, so that the multi-degree-of-freedom mechanical arm 4 reaches a proper hanging and picking operation point. And S6, the control box 3 sends a control signal to the multi-degree-of-freedom mechanical arm 4, so that the automatic ground wire hanging and picking robot enters an automatic hanging and picking mode. And S7, the 3D camera 6 starts to automatically shoot hanging and picking point position information of the target wire, and accurate point cloud information of the hanging and picking points is transmitted to the control box 3. And S8, the control box 3 calculates the motion track of the multi-degree-of-freedom mechanical arm 4 according to the accurate point cloud information, and transmits a motion track signal to the multi-degree-of-freedom mechanical arm 4. And S9, the grounding wire hanging/picking tool 7 at the tail end of the multi-degree-of-freedom mechanical arm 4 clamps the grounding wire clamping jaw 8, moves according to the motion track signal, and inserts the grounding wire clamping jaw 8 to a hanging and picking point, so that the automatic grounding wire hanging/picking robot can perform autonomous operation.
In practical application, the automatic ground wire hanging/picking robot based on the structured light vision comprises: the device comprises a control box 3, a moving platform 1, a multi-degree-of-freedom mechanical arm 4, a lifting platform 2, a tool for hanging and picking up a ground wire rod at the tail end, a ground wire clamping jaw 8, a connecting firmware 9, a panoramic camera 5, a 3D camera 6, a tool support 11 and a camera cloud deck 10. The mobile platform 1 is electrically connected with the control box 3. The lifting platform 2 is fixed on the moving platform 1. The multi-degree-of-freedom mechanical arm 4 is fixed on the lifting platform 2, and the multi-degree-of-freedom mechanical arm 4 is electrically connected with the control box 3. The tail end hooking and detaching grounding wire tool is fixed at the tail end of the multi-degree-of-freedom mechanical arm 4 and is electrically connected with the control box 3. The lower tip of connecting firmware 9 is fixed at moving platform 1, and lift platform 2 is connected to the upper end, and panoramic camera 5 and control box 3 electric connection. One end of the tool holder 11 is fixed to the upper end of the elevating platform 2. The 3D camera 6 is fixed at the tail end of the multi-degree-of-freedom mechanical arm 4, and the 3D camera 6 is electrically connected with the control box 3. The tail end hanging and picking ground wire rod tool is fixed at one end of the multi-degree-of-freedom mechanical arm 4, and the tail end hanging and picking ground wire rod tool is electrically connected with the control box 3. The camera cloud platform 10 is arranged below the 3D camera 6, and the other end of the camera cloud platform 10 is connected with a tool with the tail end hanging and picking up the ground wire rod. The earth wire clamping jaw 8 is fixed to the tool holder 11. The 3D camera 6 is used for shooting hanging and picking points of a target lead in a short distance and obtaining pose information of the hanging and picking points; the panoramic camera 5 is used to obtain position information of the target wire from a long distance.
Before the automatic grounding wire hanging/picking robot carries out operation, the robot is powered on the system firstly, so that the robot can communicate normally. Then, the automatic ground wire hanging/picking robot based on the structured light vision recognizes and acquires the position information of the target lead at a distance through the panoramic camera 5, transmits a signal of the position information to the control box 3, calculates a route required to be driven by the device according to the position information through the control box 3, and transmits a control signal of the route to the mobile platform 1. And finally, moving the moving platform 1 to enable the earth wire hanging and picking robot to move to the position below the target guide wire. In the process, the panoramic camera 5 transmits the position information to the control box 3 in real time, and the control box 3 can make real-time adjustment according to the relative position, so that the hook-picking robot can accurately move to the target wire position.
After the device has been moved to the target position, the mobile platform 1 stops moving. Next, the control box 3 converts the final position signal transmitted by the panoramic camera 5 into a control signal, transmits the control signal to the multi-degree-of-freedom mechanical arm 4, and then the multi-degree-of-freedom mechanical arm 4 moves according to the control signal and rises to a proper hanging and picking operation point.
When the multi-degree-of-freedom mechanical arm 4 reaches a proper hanging and picking operation point, the control box 3 then sends a control signal to the multi-degree-of-freedom mechanical arm 4, and the hanging and picking robot enters an automatic hanging and picking mode. The 3D camera 6 starts to shoot hanging and picking point position information of the target wire actively, and transmits accurate point cloud information of hanging and picking points to the control box 3. The control box 3 calculates the motion trail of the multi-degree-of-freedom mechanical arm 4 according to the point cloud information, and transmits the motion trail signal to the multi-degree-of-freedom mechanical arm 4. The tool for hanging and picking the ground wire rod at the tail end of the multi-degree-of-freedom mechanical arm 4 clamps the ground wire clamping jaw 8, and after a motion track control signal transmitted by the control box 3 is obtained, the ground wire clamping jaw 8 at the tail end of the multi-degree-of-freedom mechanical arm 4 is moved according to the control signal, and is gradually inserted into a hanging and picking point.
When the multi-degree-of-freedom mechanical arm 4 is inserted into a target wire hanging and picking point, the 3D camera 6 obtains hanging and picking point position information in real time and transmits the position information to the control box 3. The control box 3 performs point cloud segmentation and attitude calculation according to the information, and calculates the motion plan of the multi-degree-of-freedom mechanical arm 4.
After the multi-degree-of-freedom mechanical arm 4 is successfully inserted into the hanging and picking point, the automatic ground wire hanging/picking robot starts to perform autonomous operation on the hanging and picking point based on structured light vision. After the operation is finished, the operation is stopped when the earth wire rod hanging and picking tool is hung and picked at the tail end of the multi-degree-of-freedom mechanical arm 4, the multi-degree-of-freedom mechanical arm 4 is gradually moved away, and the initial pose is recovered.
Fig. 4 is a schematic flow chart of acquiring hanging off point position information by a 3D camera according to a remote control method according to an embodiment of the present invention. As shown in fig. 4, when the 3D camera reaches the proper photographing area, the 3D camera starts scanning, resulting in an RGB image and a depth image. And identifying the RGB image through YoloV5 to obtain a region of interest (ROI), and fusing the depth image and the ROI image to obtain an ROI point cloud. And obtaining a target pose through attitude calculation by using the ROI point cloud after amplitude limiting, jitter eliminating and filtering, and finishing the task of hanging and picking the grounding wire when the mechanical arm starts to move and goes to the target pose.
In order to obtain a transformation matrix of a camera coordinate system relative to a mechanical arm tail end coordinate system
And a transition matrix of tool end center (TCP) versus robot end
The hand-eye calibration and the TCP calibration operations are required. For eye-in-hand calibration, a special calibration disc composed of a white dot color bottom surface as shown in fig. 5 is adopted, so that the re-projection error of calibration can be reduced. The hand-eye calibration algorithm is a self-calibration algorithm based on a 3D camera and a mechanical arm system, and is a method proposed by Tsai et al.
The TCP calibration algorithm adopts a five-point calibration method, the algorithm needs to move the TCP teaching of the mechanical arm to a fixed point, convert five different postures and calculate the displacement of the TCP relative to the center of the tail end of the mechanical arm.
Wherein the content of the first and second substances,
representing the rotation matrix of the coordinate system of the tail end of the mechanical arm relative to the base coordinate system in the ith posture, B P iE a displacement matrix of the origin of the coordinate system at the tail end of the mechanical arm relative to the origin of the base coordinate system in the ith posture is shown, E P tcp and the displacement matrix of the TCP relative to the origin of the coordinate system at the tail end of the mechanical arm is represented. After the displacement matrix is obtained, the two postures of 4 and 5 are utilized to calibrate the X-axis direction and the Z-axis direction of the tool end coordinate system relative to the tail end coordinate system of the mechanical arm, and the posture calibration can be completed to obtain
To complete the identification of the region of interest, image identification is performed using YOLOv 5. The YOLOv5 network currently has a total of 4 versions: yolov5s, yolov5m, yolov5l, yolov5x. The invention uses YOLOv5s which is the minimum network in the YOLOv5 series, and the network model has the advantages of small parameter number, small size, high flexibility, high detection precision and high speed, can effectively identify the detected target, and simultaneously meets the real-time property of target detection. The network structure of YOLOv5s is mainly divided into the following three parts: backbone network Dacknet-53, neck network PANet and predictive network. In the model training stage, YOLOv5s uses the technologies of Mosaic data enhancement, adaptive anchor frame calculation, adaptive picture scaling and the like, thereby greatly improving the network training speed and reducing the model memory. The backbone network extracts features from the input image based on the Focus module, CSP1_ X module, and SPP (spatial pyramid pool) module, and transmits them to the neck network. The neck network uses an FPN-PAN structure, and enhances the detection capability of the targets with different scales through the bidirectional fusion of the bottom-layer spatial features and the high-layer semantic features. Finally, the prediction network performs regression analysis on the features with different sizes, and the classification result, the coordinate position and the confidence coefficient are obtained after NMS.
YOLO v5 can directly and effectively reason single images, batch processed images, videos and even input of a port of a network camera, and the identification speed is high.
After image recognition based on YOLO v5 is completed, an interesting image area is obtained, interesting point clouds are obtained through fusion, the point clouds are subjected to condition filtering, only points with coordinate values larger than zero are reserved, and target point clouds are obtained through removing outliers. The steps of obtaining the target 6D pose after obtaining the target point cloud are shown in the formulas (2) to (4): the center P (x) of the target point cloud is obtained by the following formula c ,y c ,z c ) Wherein x is c ，y c ，z c The central values of the three coordinate axes x are respectively, and the calculation formula is as follows:
respectively intercepting x according to conditional filtering>x c And x<x c Partially, and finding out the centers P1 and P2 and vectors of corresponding point clouds
Knowing the normal vector of the ground from the camera coordinate system
By vectors
Sum vector
Can obtain the passing vector
And normal to the plane perpendicular to the ground
Wherein
X is a vector cross product.
By vectors
Sum vector
Available vector
Namely, it is
Position of the target object
Posture of the vehicle
And simultaneously obtaining the 6D pose of the target point under the camera coordinate system.
Finally obtaining the attitude calibration
And the wire grabbing gesture, the final grabbing pose is obtained through gesture transformation, and the mechanical arm carries out path planning to complete grabbing.
In a word, the automatic ground wire hanging and picking robot based on the optical vision and the remote control method thereof have the following beneficial effects:
1. the invention is a modularized composition, each module is in communication connection through a middle control box, and meanwhile, the modules can complete work cooperatively;
2. the robot has a strong hardware foundation, each module has a space for redevelopment, and the robot can be quickly adjusted when dealing with complex working conditions;
3. compared with the traditional manual wire hanging/removing operation, the robot can move freely, can automatically hang/remove the grounding wire from the target wire in the whole process, and has a larger working range and higher efficiency;
4. the robot provided by the invention has a complete system architecture, is innovative in image processing algorithm and track planning algorithm, is improved in accuracy and stability compared with the current ground wire hanging/picking robot of a transformer substation, and is suitable for wider operation working conditions;
5. the problem of current automatic string take off earth control system mainly guide the maintainer operation through the picture information that the camera was shot, reference information is comparatively limited, requires higher to the operating level of maintainer is solved.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (7)
1. The utility model provides an automatic earth connection robot of plucking of hanging based on light vision which characterized in that includes:
a lifting platform;
the control box is fixed on the lifting platform and can control the lifting platform to move longitudinally;
the tool bracket is fixed on the lifting platform;
the multi-degree-of-freedom mechanical arm is fixed on the tool support, and the control box can control the multi-degree-of-freedom mechanical arm to actuate;
the tail end grounding wire hanging/removing tool is fixed on a tail end joint of the multi-degree-of-freedom mechanical arm, and the control box can control the tail end grounding wire hanging/removing tool to actuate;
the first shooting mechanism is fixed on the tool support and used for acquiring a tail end operation scene and is electrically connected with the control box; and
the second shooting mechanism is fixed at the tail end of the multi-degree-of-freedom mechanical arm and used for acquiring the pose of a picking and hanging point of the high-altitude lead, and the second shooting mechanism is electrically connected with the control box.
2. The light vision based automatic ground wire hooking and unhooking robot of claim 1, wherein the first photographing mechanism is a panoramic camera and the second photographing mechanism is a 3D camera.
3. The light vision based automatic ground wire hanging and picking robot of claim 2, further comprising:
the mobile platform is fixed below the lifting platform through a connecting firmware, and the control box can control the mobile platform to move;
the camera cloud deck is fixed at the tail end of the multi-degree-of-freedom mechanical arm, and the 3D camera is fixed on the camera cloud deck; and
and the grounding wire clamping jaw is fixed on the tool bracket.
4. The light vision based automatic ground wire hanging and picking robot is characterized in that the control box can control the moving platform and the lifting platform to move to the hanging and picking positions of target high-altitude wires according to the position information of the high-altitude wires acquired by the panoramic camera.
5. The light vision-based automatic ground wire hanging and picking robot as claimed in claim 4, wherein the control box is capable of controlling the multi-degree-of-freedom mechanical arm to move according to the pose information of the high-altitude wire hanging and picking point acquired by the 3D camera, and guiding the tail end ground wire hanging/picking tool to hang the ground wire clamping jaw to the position of the high-altitude wire hanging and picking point, so as to operate the high-altitude wire hanging and picking point.
6. The light vision based automatic ground wire hanging and picking robot as claimed in claim 5, wherein the terminal ground wire hanging/picking tool comprises a clamp and a turnbuckle mechanism, and the control box can control the clamp and the turnbuckle to complete the operation of hanging and picking the ground wire clamping jaw.
7. A remote control method for remotely controlling the light vision based automatic ground wire hooking and hooking robot as claimed in claims 1 to 6, comprising:
step S1, a robot system is powered on to be initialized, and program self-checking is carried out;
s2, recognizing and acquiring position information of a target high-altitude wire at a distance by the panoramic camera, and transmitting a signal of the position information to the control box;
s3, the control box calculates a route needing to be driven according to the position information and transmits a control signal of the route to the mobile platform;
s4, moving the moving platform according to a driving route to enable the robot for automatically hanging and picking the grounding wire to move to a position of a target high-altitude wire;
s5, converting the final position signal transmitted by the panoramic camera into a control signal by the control box, transmitting the control signal to the lifting platform, and ascending the lifting platform according to the control signal so as to enable the multi-degree-of-freedom mechanical arm to reach a proper hanging and picking operation point;
s6, the control box sends a control signal to the multi-degree-of-freedom mechanical arm to enable the automatic ground wire hanging and picking robot to enter an automatic hanging and picking mode;
s7, the 3D camera starts to automatically shoot hanging and picking point position information of the target wire, and accurate point cloud information of the hanging and picking points is transmitted to the control box;
s8, the control box calculates the motion trail of the multi-degree-of-freedom mechanical arm according to the accurate point cloud information and transmits a motion trail signal to the multi-degree-of-freedom mechanical arm;
and S9, clamping the ground wire clamping jaw by a tail end ground wire hanging/picking tool of the multi-degree-of-freedom mechanical arm, moving according to the motion track signal, and inserting the ground wire clamping jaw to a hanging and picking point, so that the automatic ground wire hanging and picking robot can perform autonomous operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210980395.0A CN115157286A (en) | 2022-08-16 | 2022-08-16 | Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210980395.0A CN115157286A (en) | 2022-08-16 | 2022-08-16 | Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115157286A true CN115157286A (en) | 2022-10-11 |
Family
ID=83479748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210980395.0A Pending CN115157286A (en) | 2022-08-16 | 2022-08-16 | Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115157286A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116652951A (en) * | 2023-06-08 | 2023-08-29 | 广东技术师范大学 | Method and device for positioning robot vision in unstructured large working space |
-
2022
- 2022-08-16 CN CN202210980395.0A patent/CN115157286A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116652951A (en) * | 2023-06-08 | 2023-08-29 | 广东技术师范大学 | Method and device for positioning robot vision in unstructured large working space |
| CN116652951B (en) * | 2023-06-08 | 2024-04-05 | 广州鑫帅机电设备有限公司 | Method and device for positioning robot vision in unstructured large working space |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107767423B (en) | mechanical arm target positioning and grabbing method based on binocular vision | |
| CN111421539A (en) | Industrial part intelligent identification and sorting system based on computer vision | |
| CN111267095B (en) | Mechanical arm grabbing control method based on binocular vision | |
| CN110614638A (en) | Transformer substation inspection robot autonomous acquisition method and system | |
| CN111243017A (en) | Intelligent robot grabbing method based on 3D vision | |
| CN110120074B (en) | Cable positioning method for live working robot in complex environment | |
| CN112819943B (en) | Active vision SLAM system based on panoramic camera | |
| CN108229587A (en) | A kind of autonomous scan method of transmission tower based on aircraft floating state | |
| CN111300405B (en) | Visual identification positioning device and method for mobile platform | |
| CN111702755B (en) | Intelligent mechanical arm control system based on multi-view stereoscopic vision | |
| CN113276106A (en) | Climbing robot space positioning method and space positioning system | |
| Momeni-k et al. | Height estimation from a single camera view | |
| CN112102403B (en) | High-precision positioning method and system for autonomous inspection unmanned aerial vehicle in power transmission tower scene | |
| CN115157286A (en) | Automatic ground wire hanging and picking robot based on optical vision and remote control method thereof | |
| CN115816460B (en) | Mechanical arm grabbing method based on deep learning target detection and image segmentation | |
| CN114155301A (en) | Robot target positioning and grabbing method based on Mask R-CNN and binocular camera | |
| CN115629066A (en) | Method and device for automatic wiring based on visual guidance | |
| CN111880522A (en) | Novel autonomous assembly robot path planning autonomous navigation system and method | |
| CN114241269A (en) | A collection card vision fuses positioning system for bank bridge automatic control | |
| CN113525631A (en) | Underwater terminal docking system and method based on optical visual guidance | |
| CN111452045A (en) | Reinforcing steel bar identification marking system and method based on stereoscopic vision | |
| CN114750154A (en) | Dynamic target identification, positioning and grabbing method for distribution network live working robot | |
| CN114434456A (en) | Machine room inspection robot and inspection method thereof | |
| CN115861780B (en) | Robot arm detection grabbing method based on YOLO-GGCNN | |
| CN114187312A (en) | Target object grabbing method, device, system, storage medium and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |