Robot intelligent sorting method for electric power measurement warehouse

Abstract

The invention relates to a robot intelligent sorting method for an electric power measurement warehouse, and the method comprises the following steps: carrying out the shooting of a first region through a hoisting fixed-focus camera above the first region, and collecting a first image; according to the first image, detecting metering equipment in the first area and a bar code area corresponding to the metering equipment; according to the detected metering equipment and the bar code area position, the robot mechanical arm moves at the preset height of the first area according to the optimized path track, the first area is shot multiple times through a fixed-focus camera at the tail end of the robot mechanical arm, and a second image is collected; bar code information of the bar code area is analyzed according to the second image to serve as a unique identifier of the metering equipment; and the robot mechanical arm grabs the corresponding metering devices one by one from the first area according to the needed bar code information and places the metering devices in the second area to complete sorting. On the basis of reducing the cost of the camera, the sorting efficiency of the mechanical arm of the robot is greatly improved.

Description

Technical Field

[0001] The invention relates to a robot intelligent sorting method for electric power metering equipment, belonging to the technical field of electric power metering equipment management. Background Art

[0002] The power metering warehouse, also known as the electricity metering warehouse, is a centralized storage and management place for electric energy metering equipment. It is mainly used to store meter boxes, metering instruments, calibration devices and other equipment to ensure the safe, effective, precise and reliable operation of metering equipment. It is an important part of power supply metering management. The front desk of the traditional metering asset management method is still carried out manually, with on-site inventory, handwriting, input, and summary. A large amount of manual participation has caused problems such as complex work, long time, poor real-time performance, high management difficulty, and high error rate, which ultimately reduced equipment utilization and increased operating costs.

[0003] With the development of intelligent power metering warehouses, higher requirements are put forward for the automatic storage and retrieval of metering equipment. The sorting system based on robot manipulators undertakes important intelligent storage and retrieval functions, such as Figure 1 As shown, the power metering warehouse is provided with a storage area, a sorting area and an assembly area, and the robot mechanical arm is arranged near the sorting area and the assembly area. The following is a brief description of the inbound and outbound processes of the intelligent inbound and outbound function in the prior art.

[0004] The warehousing process of metering equipment is as follows: 1) The user places the metering equipment to be stored (at least one) on the conveyor belt, and the conveyor belt transports the metering equipment to the sorting area; 2) The robot arm controls the end to determine the height in the sorting area (the working distance of the fixed-focus camera at the end of the robot arm) and moves according to the preset trajectory. The fixed-focus camera at the end of the robot arm takes multiple shots of the sorting area to collect multiple high-definition images to ensure that the image can cover the sorting area; 3) The multiple high-definition images are analyzed through the barcode detection algorithm model based on computer vision (the barcode data set is collected, which can be obtained by training based on frameworks such as Paddle Detection, Mind Yolo, and Yolo), the barcode area is identified, and the barcode information of the metering equipment in the barcode area is parsed as the unique identification of the metering equipment; 4) The metering equipment storage box is transferred to the assembly area through the conveyor belt; 5) The metering equipment is grabbed one by one from the sorting area by the robot arm and placed in the metering equipment storage box in the assembly area; 6) The storage box of the metering equipment that has been stored is transferred to the storage area through the conveyor belt; 7) The system saves the warehousing records, including the barcode information of the metering equipment, the warehousing time and other information.

[0005] The process of metering equipment outbound is as follows: 1) The user selects the type of metering equipment that needs to be outbound; 2) The system selects the metering equipment according to the type of metering equipment selected by the user, and determines the barcode information of the metering equipment; 3) The system transfers the metering equipment storage box storing the selected metering equipment to the assembly area through the conveyor belt; 4) The end of the robotic arm controls the height (working distance of the fixed-focus camera at the end of the robotic arm) in the assembly area and moves according to the preset trajectory. The fixed-focus camera at the end of the robotic arm shoots the assembly area and collects high-definition images. The system analyzes the high-definition images through the barcode detection algorithm model based on computer vision, identifies the barcode area, and parses the barcode information of the metering equipment in the barcode area. The above operations are repeated until the barcode of the metering equipment that needs to be outbound is detected; 5) The system grabs the metering equipment from the metering equipment storage box in the assembly area through the robotic arm and places it on the conveyor belt in the sorting area; 6) The system transfers the metering equipment to the user end through the conveyor belt; 7) The system saves the outbound records, including the barcode information of the metering equipment, outbound time and other information.

[0006] However, the above-mentioned intelligent in-and-out process relies on the preset covered moving track of the end of the robot arm. For example, when entering the warehouse, the barcode of the metering equipment may not be within the range of the moving track covered by the robot arm sorting area, resulting in the barcode recognition efficiency of the incoming metering equipment cannot be guaranteed; and when leaving the warehouse, it relies on the preset robot arm assembly area to search the moving track, resulting in the low efficiency of searching and locating the outgoing metering equipment based on the metering equipment barcode. Even if the barcode of the metering equipment to be out of the warehouse is known, it is difficult to accurately and quickly locate the metering equipment from the metering equipment storage box. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a method for automatically sorting metering equipment in an electric power metering warehouse accurately and quickly, thereby improving the sorting efficiency of a robot mechanical arm.

[0008] In order to solve the above technical problems, the technical solution proposed by the present invention is: a robot intelligent sorting method for an electric power metering warehouse, which is used to sort electric power metering equipment from a first area to a second area of ​​the electric power metering warehouse, comprising the following steps: The first area is photographed by a fixed-focus camera mounted above the first area to collect a first image, and the first image is ensured to cover the first area. Based on a computer vision detection algorithm, a measuring device located in the first area and a barcode area corresponding to the measuring device are detected according to the first image; According to the position of the detected measuring device and the barcode area, the robot arm moves at a preset height in the first area according to the optimized path trajectory, takes multiple shots of the first area through a fixed-focus camera at the end of the robot arm, collects a second image, and analyzes the barcode information of the barcode area according to the second image as a unique identifier of the measuring device; The robot arm grabs the corresponding metering equipment from the first area one by one according to the required barcode information and places it in the second area to complete the sorting; The method for determining the optimized path trajectory of the robot manipulator in the first area is as follows: The movement principle of the robot arm is as follows: when the fixed-focus camera at the end of the robot arm covers a single metering device, the target moving position of the robot arm end is the center of the metering device; when the fixed-focus camera at the end of the robot arm covers two metering devices, the target moving position of the robot arm end is the midpoint of the line connecting the centers of the two metering devices; when the fixed-focus camera at the end of the robot arm covers more than two metering devices, the target moving position of the robot arm end is the position with the shortest sum of distances to the centers of multiple metering devices; The robot arm starts from the current position and moves towards the uncovered measuring equipment; each time it moves, it records the measuring equipment covered and the moving time, and also records the moving path of the robot arm, the measuring equipment that has been covered, and the time spent scanning the covered measuring equipment; when the robot arm covers the same measuring equipment through different paths, it retains the minimum time spent scanning the covered measuring equipment and the corresponding moving path; The above operation is repeated until all the measuring devices are covered. At this time, the moving path corresponding to the minimum time spent on scanning all the measuring devices is the optimized path trajectory.

[0009] The present invention sets a hanging fixed-focus camera just above the first area to ensure that the field of view of the hanging fixed-focus camera can cover the first area, because the cost of a camera with a large coverage range and sufficient image clarity to parse barcode information is very high. The hanging fixed-focus camera of the present invention only requires the field of view to cover the first area, and does not need to parse the barcode information, thus greatly reducing the cost. The present invention finds the position of the metering equipment and the position of the barcode area through a hanging fixed-focus camera, and then parses the barcode information of the metering equipment through the fixed-focus camera at the end of the robot mechanical arm, and searches for the barcode information along the optimized path trajectory. On the basis of reducing the cost of the camera, the required metering equipment can be accurately and quickly identified, thereby quickly completing the sorting work, and greatly improving the sorting efficiency of the robot mechanical arm. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The present invention will be further described below in conjunction with the accompanying drawings.

[0011] Figure 1 This is a schematic diagram of the regional distribution of electricity metering warehouses. DETAILED DESCRIPTION

[0012] This embodiment shows a robot intelligent sorting method for an electric power metering warehouse, which is used to sort electric power metering equipment from a first area to a second area of ​​the electric power metering warehouse, and includes the following steps: The first area is photographed by a fixed-focus camera mounted above the first area to collect a first image, ensuring that the first image covers the first area. Since the cost of a fixed-focus camera with a large coverage area and sufficient image clarity to resolve barcode information is very high, this embodiment only requires that the field of view can cover the first area, and does not require sufficient image clarity to resolve barcode information, thereby reducing the cost of the camera.

[0013] Based on the computer vision detection algorithm, the metering device located in the first area and the barcode area corresponding to the metering device are detected according to the first image. Identifying the metering device and the barcode area is an existing technology, for example, the metering device and the barcode area can be detected based on PaddleDetection, Mind Yolo or Yolo framework, etc.

[0014] According to the position of the metering device and the barcode area to be detected, the robot arm moves at a preset height in the first area according to the optimized path trajectory, and the first area is photographed multiple times by a fixed-focus camera at the end of the robot arm to collect a second image, and the barcode information of the barcode area is parsed according to the second image as the unique identifier of the metering device. Preferably, the barcode information in the second image can be detected based on Paddle Detection, Mind Yolo or Yolo framework.

[0015] When determining the optimized path trajectory movement, the movement principle of the robot arm is: when the fixed-focus camera at the end of the robot arm covers a single measuring device, the target moving position of the end of the robot arm is the center of the measuring device; when the fixed-focus camera at the end of the robot arm covers two measuring devices, the target moving position of the end of the robot arm is the midpoint of the line connecting the centers of the two measuring devices; when the fixed-focus camera at the end of the robot arm covers more than two measuring devices, the target moving position of the end of the robot arm is the position with the shortest sum of distances to the centers of multiple measuring devices.

[0016] Determine the optimized path trajectory movement, that is, determine the position where the end of the robot arm moves. The specific method is: the robot arm starts from the current position and moves towards the uncovered metering equipment; each time it moves, it records the metering equipment covered and the moving time, and also records the moving path of the robot arm, the metering equipment that has been covered, and the time spent scanning the covered metering equipment; when the robot arm covers the same metering equipment through different paths, it retains the minimum time spent scanning the covered metering equipment and the corresponding moving path. Repeat the above operation until all metering equipment is covered. At this time, the moving path corresponding to the minimum time spent scanning all metering equipment is the optimized path trajectory.

[0017] During implementation, the base of the robot arm is taken as the origin, the height data is ignored, a global two-dimensional coordinate system is established, and the coordinates of the center position of the measuring device in the first area, the position of the barcode area, and the end position of the robot arm are calculated; Assume that there are n measuring devices in the first area, and their center points are d1, d2, d3, ..., dn respectively; Assume that the end of the robot arm needs to move m times, and its positions are r1, r2, r3, ..., rm; Assume that T({ri}|{dj}) represents the time taken for the end of the robot arm to move to position {ri} in sequence and complete the scan of the measuring device {dj}, where 1≤i≤m, 1≤j≤n, {ri} represents a list of i positions, whose order represents the path trajectory, and {dj} represents a list of j measuring devices; Take the robot arm moving from position r1 to position r2 as an example. According to the motion logic of the robot arm accelerating to the maximum speed, maintaining a constant speed, and then decelerating to stop, let s(r1,r2) be the distance from position r1 to position r2, t(r1,r2) be the time it takes for the robot arm to move from position r1 to position r2, a be the acceleration of the robot arm, and vmax be the maximum moving speed of the robot arm. Then: When s(r1,r2)<=(vmax 2 ) / a, t(r1,r2)=sqrt(4s(r1,r2) / a); When s(r1,r2)>(vmax 2 ) / a, t(r1,r2)=vmax / a+s(r1,r2) / vmax, Assume that when the fixed-focus camera of the robot arm is at position r1, all the second images collected cover the measuring devices d1 and d2, and the time spent on scanning is calculated as T((r1, r2)|(d1, d2, d3)). When it is at position r2, all the second images collected cover the measuring devices d1, d2 and d3, and the time spent on scanning is calculated as T((r1, r2)|(d1, d2, d3)). Then: T((r1,r2)|(d1,d2,d3))= T((r1,r2)|(d1,d2,d3))+t(r1,r2).

[0018] Each time coverage is performed, the time consumption of different choices (i.e., when covering different metering devices through different paths) is recorded. For the scanning time of covering the same metering device through different paths, the smallest value is retained. The above operation is repeated until all metering device positions d1, d2, d3, ..., dn are covered, and the smallest T({rm}|{dn}) is retained, where the order of the m positions in {rm} is the optimized path trajectory.

[0019] This embodiment plans the path trajectory of the robot arm and selects the path trajectory with the least time consumption from multiple possible path trajectories as the optimized path trajectory. When the fixed-focus camera at the end of the robot arm shoots the first area, it moves along the optimized path trajectory, so that all the metering equipment that needs to be photographed can be accurately and quickly covered, thereby improving the sorting efficiency.

[0020] When the robot intelligent sorting method of the power metering warehouse of this embodiment is used to perform the storage operation on the power metering equipment, the first area is the sorting area of ​​the power metering warehouse, and the second area is the metering equipment storage box located in the assembly area of ​​the power metering warehouse. Figure 1 The specific storage process is as follows: 1. The user places the metering equipment (at least one) to be stored on the conveyor belt, and the conveyor belt transports the metering equipment to the sorting area; 2. Use a fixed-focus camera mounted above the sorting area to capture a first image and ensure that the first image covers the sorting area; 3. Based on computer vision detection algorithm, detect the measuring equipment and barcode area.

[0021] 4. According to the position of the detected measuring equipment and barcode area, based on the current position data of the robot arm, plan the motion trajectory of the end of the robot arm and determine the optimized path trajectory; 5. The robot arm controls the end to move along the optimized path trajectory at a determined height in the sorting area (the working distance of the fixed-focus camera at the end of the robot arm). The fixed-focus camera at the end of the robot arm takes multiple photos of the sorting area and collects multiple high-definition second images to ensure that all second images can cover all barcode areas. 6. Associate the detected barcode area with the high-definition second image, and parse the barcode information of the measuring equipment in the barcode area as the unique identification of the measuring equipment; 7. Transport the metering equipment storage box to the assembly area via a conveyor belt; 8. The robot arm grabs the metering equipment from the sorting area one by one and places them in the metering equipment storage box in the assembly area; 9. The stored metering equipment storage box is transferred to the storage area via a conveyor belt; 10. Save the warehousing records, including the barcode information of the measuring equipment, warehousing time and other information.

[0022] When the robot intelligent sorting method of the power metering warehouse of this embodiment is used to carry out the outbound operation of the power metering equipment, the first area is the metering equipment storage box in the assembly area of ​​the power metering warehouse, and the second area is the sorting area. The specific outbound process is as follows: 1. The user selects the type of metering equipment that needs to be shipped out; 2. Select the measuring device according to the type of measuring device selected by the user, and determine the barcode information of the measuring device; 3. The metering equipment storage box storing the selected metering equipment is transferred to the assembly area via a conveyor belt; 4. The fixed-focus camera mounted above the assembly area takes pictures and collects the first image; 5. Detect the measuring equipment and barcode area based on computer vision detection algorithm; 6. According to the detected measuring equipment and barcode area position, based on the robot arm and current position data, plan the motion trajectory of the robot arm end to determine the optimized path trajectory; 7. The robot arm controls the end to move, determines the height in the assembly area (working distance of the fixed-focus camera at the end of the robot arm), and moves according to the optimized path trajectory. The fixed-focus camera at the end of the robot arm shoots the assembly area, collects a high-definition second image, and analyzes the second image through a barcode detection algorithm based on computer vision to identify the barcode area and parse the barcode information of the metering equipment in the barcode area. If the target metering equipment barcode exists in the current second image, the end of the robot arm stops moving. Otherwise, it plans the optimized path trajectory based on the current position data of the robot arm, and repeats the above steps until the barcode information of the metering equipment that needs to be shipped out is detected; 8. The robot arm grabs the metering equipment that needs to be shipped out from the metering equipment storage box in the assembly area and places it on the conveyor belt in the sorting area; 9. Transport the metering equipment to the user end via a conveyor belt; 10. Save the outbound records, including the barcode information of the measuring equipment, outbound time and other information.

[0023] Because the hanging fixed-focus camera is only used to cover the first area, and the first area is different when the goods are shipped out and when they are put into the warehouse, in order to simplify the operation, preferably, the field of view of the hanging fixed-focus camera just covers the assembly area and sorting area of ​​the power metering warehouse. In this way, only one hanging fixed-focus camera can meet the needs of shipping out and putting into the warehouse, further saving costs.

Claims

1. A robot intelligent sorting method for an electric power metering warehouse, for sorting electric power metering equipment from a first area to a second area of ​​the electric power metering warehouse, comprising the following steps: The first area is photographed by a fixed-focus camera mounted above the first area to collect a first image, and the first image is ensured to cover the first area. Based on a computer vision detection algorithm, a measuring device located in the first area and a barcode area corresponding to the measuring device are detected according to the first image; According to the position of the metering device and the barcode area to be detected, the robot arm moves at a preset height in the first area according to the optimized path trajectory, and the first area is photographed multiple times by a fixed-focus camera at the end of the robot arm to collect a second image, and the barcode information of the barcode area is parsed according to the second image as the unique identification of the metering device; The robot arm grabs the corresponding metering equipment from the first area one by one according to the required barcode information and places it in the second area to complete the sorting; The method for determining the optimized path trajectory of the robot manipulator in the first area is as follows: The movement principle of the robot arm is as follows: when the fixed-focus camera at the end of the robot arm covers a single metering device, the target moving position of the robot arm end is the center of the metering device; when the fixed-focus camera at the end of the robot arm covers two metering devices, the target moving position of the robot arm end is the midpoint of the line connecting the centers of the two metering devices; when the fixed-focus camera at the end of the robot arm covers more than two metering devices, the target moving position of the robot arm end is the position with the shortest sum of distances to the centers of multiple metering devices; The robot arm starts from the current position and moves towards the uncovered measuring equipment; each time it moves, it records the measuring equipment covered and the moving time, and also records the moving path of the robot arm, the measuring equipment that has been covered, and the time spent scanning the covered measuring equipment; when the robot arm covers the same measuring equipment through different paths, it retains the minimum time spent scanning the covered measuring equipment and the corresponding moving path; The above operation is repeated until all the measuring devices are covered. At this time, the moving path corresponding to the minimum time spent on scanning all the measuring devices is the optimized path trajectory.

2. The robot intelligent sorting method for the power metering warehouse according to claim 1 is characterized in that: When the electric power metering equipment is put into storage, the first area is the sorting area of ​​the electric power metering warehouse, and the second area is the metering equipment storage box located in the assembly area of ​​the electric power metering warehouse.

3. The robot intelligent sorting method for the power metering warehouse according to claim 2 is characterized in that: When the fixed-focus camera at the end of the robot arm collects the second image, it is necessary to ensure that the collected second image can cover all the measuring devices in the first area.

4. The robot intelligent sorting method for the power metering warehouse according to claim 1 is characterized in that: When the electric power metering equipment is shipped out of the warehouse, the first area is the metering equipment storage box in the assembly area of ​​the electric power metering warehouse, and the second area is the sorting area.

5. The robot intelligent sorting method for the power metering warehouse according to claim 4 is characterized in that: When the electric power metering equipment is shipped out of the warehouse, when determining the optimized path trajectory of the robot manipulator in the first area, when parsing the barcode information of the barcode area of ​​the metering equipment according to the second image, if all the second images currently taken already have the required barcode information, then no other metering equipment will be photographed; otherwise, based on the current position of the robot manipulator, continue to determine the optimized path trajectory of the first area until the barcode information of all the metering equipment that needs to be shipped out is found.

6. The robot intelligent sorting method for the power metering warehouse according to any one of claims 1 to 5, characterized in that: Based on Paddle Detection, Mind Yolo or Yolo framework, the measuring device and the barcode area in the first image are detected.

7. The robot intelligent sorting method for the power metering warehouse according to any one of claims 1 to 5, characterized in that: Detect the barcode information in the second image based on Paddle Detection, Mind Yolo, or Yolo framework.

8. The robot intelligent sorting method for the power metering warehouse according to any one of claims 1 to 5, characterized in that: The field of view of the suspended fixed-focus camera just covers the assembly area and sorting area of ​​the electricity metering warehouse.

9. The robot intelligent sorting method for the power metering warehouse according to any one of claims 1 to 5, characterized in that: After the robot arm completes the sorting of the measuring equipment, it records the barcode information and the grabbing time of the grabbed measuring equipment.

Images