Robot control device and robot control method
The robot control device accurately estimates the height of non-constant shaped objects by analyzing three-dimensional point cloud data and distance changes, ensuring precise robotic grasping and placement without damage.
Patent Information
- Application Number
- JP2023072401
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-04-26
- Publication Date
- 2024-11-08
AI Technical Summary
Existing technologies struggle to accurately estimate the height of objects with non-constant shapes, such as bags or pouch-packed items, when they are stacked, leading to incorrect height determination during robotic grasping and placement.
A robot control device that uses a camera to acquire three-dimensional point cloud information before and after picking an object, processes the images to recognize the object's surface, calculates the distance change, and determines the appropriate placement height using a histogram analysis of the distance data to ensure accurate robotic grasping and placement.
Enables accurate estimation of the height position of the robot's hand even when objects are loosely or angled stacked, preventing impact or crushing during robotic handling.
Smart Images

Figure 2024157810000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a robot control device and a robot control method. [Background technology]
[0002] In recent years, robots have been used to transport objects in various fields. For example, robots have been introduced for sorting specimens in medical laboratories and the like. In this case, the robot arm has a hand for grasping the specimen. The hand of the robot arm is required to grasp objects (such as test tubes) in a bulk state and set them in a designated location with high accuracy and high throughput. In particular, when many objects are piled up in a bulk state, the height of each object is different, and the robot cannot grasp the object unless it can accurately detect the height of the object to be grasped.
[0003] Patent Document 1 describes a technology for estimating the height of a rectangular parallelepiped object in a system for measuring the height of a workpiece to be transferred that can accurately measure the height of the workpiece to be transferred. That is, Patent Document 1 describes a technology having an imaging unit that acquires three-dimensional point cloud information of a plurality of loaded workpieces and acquires three-dimensional point cloud information of the plurality of loaded workpieces after the workpiece to be transferred is temporarily moved, and a box height estimation unit that selects three-dimensional point cloud information to be used as a target for plane fitting used when acquiring a plane of the workpiece required for measuring the height of the workpiece from the three-dimensional point cloud information acquired by the imaging unit. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Patent Publication No. 2021-36204 Summary of the Invention [Problem to be solved by the invention]
[0005] Existing technology for estimating the height of rectangular parallelepiped objects targets boxes as objects, estimates the bottom and side planes from the point cloud after grasping, and estimates the height of the rectangular parallelepiped shape based on this. However, when the object to be estimated is a bagged or pouch-packaged object, the shape of the object is deformed and not constant, so the bottom plane cannot be uniquely determined. Therefore, in a situation where objects that may be deformed are stacked, there is a problem that the height of the object cannot be estimated correctly.
[0006] In view of the above, an object of the present invention is to provide a robot control device and a robot control method that can accurately estimate the height at which a robot will place an object even if the shape of the object is not constant. [Means for solving the problem]
[0007] In order to solve the above problems, for example, the configurations described in the claims are adopted. The present application includes multiple means for solving the above-mentioned problems. As an example, the robot control device of the present invention is a robot control device that controls a robot that picks up a specified item at a pick position and places it at a place position, and is equipped with an image processing unit that acquires an image before the robot picks up the item and an image after the robot picks the item, recognizes a first surface of the item to be grasped from a point cloud image generated from the acquired image before picking, and acquires a partial point cloud of an area where the first surface is projected in a specified direction from the point cloud image generated from the acquired image after picking, a pick state calculation unit that calculates a first distance from the change in distance before and after the picking from the partial point cloud obtained by the image processing unit, and a height calculation unit that calculates the height at which the robot places the item based on the first distance. Effect of the Invention
[0008] According to the present invention, even if the items to be picked and placed are piled in bulk or at an angle, it is possible to correctly estimate the height position of the tip of the robot's arm. Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]
[0009] [Figure 1] 1 is a configuration diagram showing an example of an entire system according to an embodiment of the present invention; [Diagram 2] 1 is a block diagram showing an example of the configuration of a robot control device according to an embodiment of the present invention; [Diagram 3] 5 is a flowchart showing an example of processing of a robot control device according to an embodiment of the present invention. [Figure 4] A figure showing an example of an object pick-and-place process (before picking) according to one embodiment of the present invention. [Diagram 5] FIG. 2 is a diagram showing an example of an object pick-and-place process (after picking) according to an embodiment of the present invention. [Figure 6] FIG. 13 is a diagram showing an example of the place height of an object according to one embodiment of the present invention. [Figure 7] FIG. 13 is a diagram showing an example of a process for estimating an appropriate place height of an object according to an embodiment of the present invention. [Figure 8] FIG. 13 is a diagram showing an example of a histogram according to an embodiment of the present invention. [Figure 9] 10 is a flowchart illustrating an example of a histogram analysis process according to an embodiment of the present invention. [Figure 10] FIG. 11 is a diagram showing another example (part 1) of estimating the height of an object according to an embodiment of the present invention. [Figure 11] FIG. 11 is a diagram showing another example (part 2) of estimating the height of an object according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A robot control device and a robot control method according to an embodiment of the present invention (hereinafter referred to as "this embodiment") will be described below with reference to the accompanying drawings.
[0011] [System Configuration] FIG. 1 shows an example of the overall configuration of a robot 100 controlled by a robot control device 200 of this embodiment. 1, the robot 100 of this example is configured as a gripping device and includes an arm 101 as a manipulator that picks up and places an object 103. A hand 102 that adsorbs the object 103 is installed at the tip of the arm 101, and a camera 104 is installed near the arm 101 as a 3D point cloud information acquisition device. 1, an object 103 is picked up from a pick container 105, and the picked object 103 is placed at an appropriate height in a place container 106. A camera 104 is set to capture an image of the inside of the pick container 105 from directly above.
[0012] The camera (3D point cloud information acquisition device) 104 acquires object information indicating the state of the target object from the entire or partial area of the working space. The camera 104 used is one that can acquire the distance to each point in the captured image. The object information acquired by the camera 104 includes data on the size, shape, attitude, position, etc. of the target object. The robot 100 may have a plurality of object information acquisition devices that acquire the same type of object information, or may have a plurality of object information acquisition devices that acquire different types of object information.
[0013] The arm 101 is composed of a link and a joint, and has a drive device (not shown) for moving the link in a predetermined direction with the joint as a fulcrum.
[0014] The hand 102 has a suction rod or the like for suctioning the object 107. Note that, although an example of holding an object by suction has been given here, the object may be grasped by a hand with fingers or the like. In this case, the hand may have a force sensor, a tactile sensor, or the like, and may hold the object 107 with an appropriate force. The hand 102 and the arm 101 may include a plurality of devices of the same type.
[0015] The robot controller 110 communicates with the robot control device 200 and receives instructions from the robot control device 200 to cause the hand 102 to pick and place the object 103 .
[0016] [Robot control device configuration] FIG. 2 shows an example of the configuration of the robot control device 200. The robot control device 200 is configured as, for example, a computer, and includes a processor 201, a main memory device 202, a secondary memory device 203, a network interface 204, and an input / output interface 205. These elements 201 to 205 are connected to each other via an internal bus.
[0017] The processor 201 executes a program stored in the main memory device 202. When the processor 201 executes processing according to the program, functional units (modules) that realize specific functions are configured in the main memory device 202. For example, as shown in Fig. 2, an image processing unit 211, a pick state calculation unit 212, a height calculation unit 213, etc. are configured. The processes performed by these functional units (modules) 211, 212, and 213 will be described later.
[0018] The main memory device 202 is a storage device such as a dynamic random access memory (DRAM) that stores programs executed by the processor and information used by the programs. The main memory device 202 is also used as a work area used by the programs.
[0019] The programs and information may be stored in a device or apparatus other than the main memory device 202, such as the secondary memory device 203. In this case, the processor 201 reads the programs and information from the device or apparatus and loads them into the main memory device 202. The secondary storage device 203 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores data permanently.
[0020] The network interface 204 is an interface for communicating with external devices such as the robot 100 via a network connected to the robot control device 200. This network interface 204 transmits commands to the robot 100, receives commands from the camera 104, and the like.
[0021] The input / output interface 205 is an interface for connecting the robot control device 200 to an input device, an output device, etc. The input device is a keyboard, a mouse, a touch panel, etc., and the output device is a display, a printer, etc.
[0022] [Control processing by the robot control device] FIG. 3 is a flow chart showing an example of processing when the hand 102 at the tip of the arm 101 picks up and places an object under the control of the robot control device 200 of this example. First, the image processor 211 of the robot controller 200 acquires point cloud information of an object existing in the working space before picking from the camera 104, which is a three-dimensional point cloud information acquisition device. Then, the image processor 211 estimates the upper side surface of the object to be picked from the acquired point cloud (step S11). For example, as shown in Fig. 4, when an object 411 among a plurality of objects 411, 412, and 413 arranged in a pick container 105 is the object to be picked, the image processor 211 of the robot controller 200 estimates the upper side surface (first surface) 400 from the point cloud information.
[0023] Here, the range in which the robot control device 200 acquires point cloud information from the camera 104 may be determined by a user who operates the robot control device 200. Alternatively, the range in which the robot controller 110 of the robot 100 acquires point cloud information may be determined.
[0024] Next, the robot controller 200 calculates a position for suction (pick position) and a movement position for placing based on the estimated object information of the pick target. The robot controller 200 also generates a trajectory plan to the movement position and moves the target object after picking it up (step S12). Here, the trajectory plan generated in step S12 is a plan formulated regarding the movement trajectories of the arm 101 and the hand 102 for moving the hand from the gripping position to the movement position.
[0025] Then, the image processing unit 211 of the robot control device 200 acquires point cloud information after the item is picked from the camera 104, projects the upper surface of the item to be picked that was estimated before the pick onto the point cloud after the pick, and acquires a partial point cloud within that area (step S13). For example, as shown in FIG. 5, when the hand 102 picks up an item 411, the image processing unit 211 projects the upper side of the item to be picked from the upper side of the location (FIG. 4) where the item to be picked 411 was estimated to be located before the pick, onto a point cloud after the pick in a direction perpendicular to the upper side, and then obtains a partial point cloud inside the area 401.
[0026] When the image processing unit 211 acquires a partial point cloud inside this region 401, the points closest to the upper surface 400 are extracted as neighboring points from the point clouds included in each divided grid. In the example of Figure 5, the state in which the item 411 has been moved to the place container 106 is shown, but if a partial point cloud can be obtained after picking, point cloud information within the area 401 may be obtained before picking.
[0027] Thereafter, the pick state calculation unit 212 of the robot control device 200 calculates a first distance 402 between the upper side estimated by a predetermined procedure and the items 412, 413 in the pick container 105 from the partial point cloud (step S14). Here, the pick state calculation unit 212 performs a process of calculating the first distance from the partial point cloud based on a change in distance before and after picking, and a specific example of the procedure for calculating the first distance 402 in step S14 will be described later.
[0028] Furthermore, the height calculation unit 213 of the robot control device 200 calculates an appropriate placing height based on the distance calculated in step S14, and generates control information for the corresponding placing height (step S15). For example, as shown in Fig. 5, the height calculation unit 213 calculates the height 403 at which the target item 411 is placed in the placing container 106, and generates control information for the hand 102 to place the item 411 at that height. Then, the hand 102 places the target item based on the control information thus generated by the robot control device 200 (step S16).
[0029] FIG. 6 is a diagram showing cases where the place height is appropriate and where it is not appropriate. For example, as shown on the left side of FIG. 6, if an article 501 picked up by the hand 102 is released at an inappropriate height, that is, at a position higher than the height of the article 501, it will be subjected to a drop impact, which is undesirable. Also, as shown in the center of FIG. 6, if an item 501 picked up by the hand 102 is placed at a position lower than its height, there is a possibility that the item 501 will be crushed.
[0030] On the other hand, as shown on the right side of Figure 6, by correctly calculating the placement height by the height calculation unit 213, the item 503 picked up by the hand 102 can be placed in the place container 106 without being dropped or crushed.
[0031] [Process to estimate the height of the place] Fig. 7 is a diagram for explaining the process of estimating an appropriate place height for an object by the robot control device 200. In the example of Fig. 7, as shown on the left side of Fig. 7, multiple items 611-614 are stacked in the pick container 105, and it is assumed here that the slightly tilted topmost item 611 is to be picked. At this time, the robot control device 200 estimates the upper side surface 400 of the item 611 to be picked from the point cloud before picking, and further captures an image of the same location as the upper side surface 400 of the point cloud after picking with the camera 104.
[0032] The right side of Fig. 7 shows the point cloud of the upper side surface 400. The upper right corner of Fig. 7 is a diagram of the point cloud 601 viewed from the side, and the lower right is a diagram of the point cloud 601 viewed from the top surface. A point cloud 601 shown in FIG. 7 is a set of points within each of the divided regions obtained by dividing a region 603 on the upper surface 400 into an arbitrary number (N×N: N is an arbitrary integer).
[0033] Here, a point group (contact points) 601 that is closest to the upper surface 400 within each divided region is selected, and the pick state calculation unit 212 calculates the distance 602 between that point and the upper surface 400. Then, the pick state calculation unit 212 estimates the distance (height) of the picked item 611 from each calculated distance, and the height calculation unit 213 estimates the optimal place height of the item 611 based on the estimated height. As a processing method for estimating the distance (height) of the item 611 from each calculated distance, for example, a histogram is used.
[0034] [Processing to estimate place height using histogram] Fig. 8 shows a histogram of distances 602 between each point of the point cloud 601 and the upper surface 400. In the histogram in Fig. 8, the horizontal axis 700 shows 10 bins (distance values) ranging from value 1 to value 10, and the vertical axis 701 lists the number of points that fall into each bin. Values 1 to 10 are provisionally set examples for ease of explanation.
[0035] First, the pick state calculation unit 212 uses the minimum and maximum distance values found in each region to generate distance values (Bins) for multiple items and a histogram of the distances. That is, the distance from the camera 104 to the upper surface 400 before picking is the minimum distance, and the distance from the camera 104 to the same point after picking is the maximum distance. Then, the pick state calculation unit 212 obtains the height of each area of the picked object by calculating (maximum distance - minimum distance) (pick state calculation process). The pick state calculation unit 212 divides this calculated distance by the number of sections for each area, averages it, and sets it as the Bin size.
[0036] That is, the bin size is "(maximum distance - minimum distance) / number of sections". For example, when the number of sections is 10, the bin size is (maximum distance - minimum distance) / 10. Note that the number of sections (number of bins) here may be changed according to the situation. Next, the pick state calculation unit 212 creates a histogram of the height of each area, and estimates the place height from the histogram created by the height calculation unit 213 (height calculation process).
[0037] As shown in Fig. 8, the pick state calculation unit 212 puts the distance information found for the points in each area into each bin by distance and aggregates it. However, bins close to the maximum distance (e.g., bins 702 with values 9 and 10 in Fig. 8) are ignored because they may have recognized a gap between items below the item to be picked. In other words, the pick state calculation unit 212 determines whether the bin value of the section close to the maximum value in the histogram 700 is relatively large compared to other sections. If this determination is that the bin value is relatively large compared to other sections, the pick state calculation unit 212 excludes it from the candidates for estimation described below because it may have recognized a gap between items below the item to be picked.
[0038] Then, the pick state calculation unit 212 searches for the largest bin (for example, bin 704 with value 5 in FIG. 8) and the second largest bin (for example, bin 703 with value 2 in FIG. 8). Then, the pick state calculation unit 212 determines the distance between bin 704 and bin 703 that has the largest distance value (the distance of bin 704 in the example of FIG. 8) as the distance (height) of the item to be picked. Based on this distance of the item to be picked, the height calculation unit 213 sets an appropriate height for the robot's hand place.
[0039] FIG. 9 is a flow chart showing the flow of a process for estimating the optimum place height of an article using the histogram of FIG. First, the pick state calculation unit 212 acquires distance information Bin of the point cloud within the region, and generates a histogram shown in FIG. 8 (step S21). Next, when the value of the section close to the maximum value of the histogram is relatively large compared to other sections, the pick state calculation unit 212 ignores information Bin of the section close to the relatively large maximum value in the generated histogram (step S22).
[0040] Then, the pick state calculation unit 212 searches for the bin with the highest occurring distance in the histogram and the bin with the second highest occurring distance (step S23). Furthermore, the pick state calculation unit 212 determines the bin with the larger distance value of the two distances found in step S23 as the distance of the target item (step S24). Here, obtaining the distance information that occurs first and second most frequently in the histogram is just an example, and the distance that occurs most frequently may be determined as the distance of the target item, for example. Alternatively, the distance information that occurs most frequently from first to a predetermined number (e.g., third) may be obtained, and the distance value among them may be determined as the largest.
[0041] As described above, the robot control device 200 of this example can correctly estimate the appropriate robot hand placement height even when the items to be picked are in various forms (boxes, bags, pouches, etc.) and are piled loose or at an angle. Therefore, the article can be transported properly without being subjected to a drop shock or crushed due to an inappropriate place height.
[0042] [Other processes to estimate place height from histograms] The process of estimating the place height using the histogram described with reference to FIGS. 8 and 9 is a preferred example, and the place height may be estimated from the histogram using other processes. For example, in the histogram shown in Fig. 8, if the value of the section close to the maximum value of Bin is relatively large compared to other sections, the height calculation unit 213 determines that the part supporting the picked item is not dense and has a cavity, etc. If it determines that there is a cavity, the height calculation unit 213 excludes the corresponding section and calculates the place height.
[0043] On the other hand, if the value of the section close to the maximum value of Bin is not relatively large compared to other sections, the height calculation unit 213 determines that it is a dense case, includes that section, and estimates the place height using the procedure described in the flowchart of Figure 9. This makes it possible to properly estimate the place height even when there is a cavity between the loaded items and the items below.
[0044] Furthermore, when the height calculation unit 213 calculates the distance to the target article, if the value of the section close to the maximum value of the histogram is relatively large compared to other sections, the number of divisions is reduced when dividing the partial point cloud shown in Fig. 7 into a grid of a predetermined size. Furthermore, if the value of the section close to the maximum value of the histogram is not relatively large compared to other sections, the height calculation unit 213 increases the number of divisions when dividing the partial point cloud shown in Fig. 7 into a grid of a predetermined size when calculating the distance to the target article. By changing the number of divisions in this manner, the height calculation unit 213 can estimate an appropriate place height in accordance with the actual state of the object, thereby improving the accuracy of height estimation.
[0045] Furthermore, when the item to be picked is an item that does not deform, such as a box, one point closest to the upper side surface of the item 811 to be picked estimated from the partial point cloud shown in Fig. 7 may be extracted as a nearby point 800, as shown in Fig. 10. In the case of Fig. 10, the item 811 is placed at an angle due to contact with another item 812. This allows the height calculation unit 213 to estimate the height position of the hand 102 at the tip of the arm 101 at which the robot places the article, based on the distance 801 between the extracted nearby point 800 and the estimated surface. In this way, in the case of a non-deformable article, the height calculation unit 213 can correctly estimate the placing height with simpler calculations.
[0046] Furthermore, when the item to be picked is a non-deformable item such as a box and is placed flat, as shown in Fig. 11, the height calculation unit 213 detects an average value 900 of the estimated distance to the upper side of the item 911 to be picked. Then, based on the detected average distance, the height calculation unit 213 estimates the height position of the hand 102 at the tip of the arm 101 at which the robot places the item. In this way, when the item is non-deformable and placed flat, the height calculation unit 213 can correctly estimate the placing height with a simpler calculation.
[0047] [Variations] The embodiment described above has been described in detail to make the present invention easier to understand, and is not necessarily limited to those having all of the configurations described. Also, the configurations and processes described in the above embodiment can be modified or changed in various ways. For example, in the above-described embodiment, the robot is configured to pick up an object by suction with the hand 102 at the tip of the arm 101. However, the present invention may be applied to a robot that grasps an object with a hand having fingers.
[0048] Furthermore, the shapes of the arm 101 and the hand 102 are also merely examples, and the present invention may be applied to a robot that picks and places objects using arms and hands with shapes other than those mentioned above. Moreover, the number of divided regions and the number of point groups shown in FIG. 7 are merely examples, and the present invention is not limited to those shown in the drawings. Furthermore, in the above-described embodiment, the distance is acquired from an image captured by the camera 104, but a measuring device other than a camera may be used as the three-dimensional point cloud information acquisition device.
[0049] In the above-described embodiment, the robot control device 200 is configured such that a computer having the configuration shown in Fig. 2 performs arithmetic processing based on a program implemented therein, but this is merely an example, and other configurations may be used. For example, a part or the whole of the robot control device 200 may be realized by dedicated hardware such as a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC).
[0050] Furthermore, when the robot control device 200 is configured to perform arithmetic processing based on an implemented program, the program may be prepared in the main memory device 202 or the secondary memory device 203, or may be stored in an external memory, an IC card, an SD card, an optical disk, or other recording medium and transferred to the robot control device 200.
[0051] In addition, in the configuration diagram of Figure 2, only control lines and information lines that are considered necessary for explanation are shown, and not all control lines and information lines in the product are necessarily shown. In reality, it may be considered that almost all components are connected to each other. In addition, as for the flowcharts shown in Figures 3 and 9, the processing order may be changed or multiple processes may be executed simultaneously as long as the processing results are the same. [Explanation of symbols]
[0052] 100...robot, 101...arm, 102...hand, 103...target object, 104...camera (3D point cloud information acquisition device), 105...pick container, 106...place container, 110...robot controller, 200...robot control device, 201...processor, 202...main memory device, 203...sub-memory device, 204...network interface, 205...input / output interface, 211...image processing unit, 212...pick state calculation unit, 213...height calculation unit, 400...upper surface, 411...target item
Claims
1. A robot control device that controls a robot that picks up a predetermined item at a pick position and places it at a place position, an image processing unit that acquires an image of the article before the robot picks up the article and an image of the article after the robot picks up the article, recognizes a first surface of the article to be grasped from a point cloud image generated from the acquired image before picking, and acquires a partial point cloud of an area where the first surface is projected in a predetermined direction from a point cloud image generated from the acquired image after picking; a pick state calculation unit that calculates a first distance from a change in distance before and after picking from the partial point cloud obtained by the image processing unit; a height calculation unit that calculates a height at which the robot places the item based on the first distance. Robot control device.
2. The pick state calculation unit divides the partial point cloud into a grid of a predetermined size, extracts a point closest to the first surface from the point cloud included in each divided grid as a neighboring point, and calculates the first distance. The robot control device according to claim 1 .
3. the height calculation unit creates a histogram of a predetermined division from maximum and minimum values of distances between each neighboring point and the first surface; A class value of a predetermined rank of the histogram is calculated as the first distance. The robot control device according to claim 2 .
4. Output whether the value in the section close to the maximum value of the histogram is relatively large compared to other sections The robot control device according to claim 3.
5. when a value in a section close to a maximum value of the histogram is not relatively large compared to other sections, the height calculation unit detects a distance between the first surface and a point closest to the first surface from a group of points included in each lattice obtained by dividing the partial point cloud into a lattice shape of a predetermined size; Calculate a position where the robot will place the item based on an average value of the distances detected for each grid. The robot control device according to claim 4.
6. The height calculation unit reduces the number of divisions when dividing the partial point cloud into a grid of a predetermined size when a value in a section close to the maximum value of the histogram is relatively large compared to other sections, and reduces the number of divisions when dividing the partial point cloud into a grid of a predetermined size when a value in a section close to the maximum value of the histogram is not relatively large compared to other sections. The robot control device according to claim 4.
7. the height calculation unit detects a distance between the first surface and a neighboring point in the partial point cloud that is closest to the first surface, or an average distance between the partial point cloud and the first surface; Calculating the height at which the robot places the item based on the distance or the average distance between the neighboring point and the first surface The robot control device according to claim 1 .
8. A robot control method for controlling a robot that picks up a predetermined item at a pick position and places it at a place position, comprising: image processing for acquiring an image of the article before the robot picks up the article and an image of the article after the robot picks up the article, recognizing a first surface of the article to be grasped from a point cloud image generated from the acquired image before picking, and acquiring a partial point cloud of an area where the first surface is projected in a predetermined direction from a point cloud image generated from the acquired image after picking; a pick state calculation process for calculating a first distance from a change in distance before and after picking from the partial point cloud obtained by the image processing; a height calculation process for calculating a height at which the robot places the item based on the first distance. A method for controlling a robot.
Citation Information
Patent Citations
Transfer object work-piece height measuring system
JP2021036204A
