System and method for gripping containers in diagnostic laboratory system

Abstract

A method of grasping a container in a diagnostic laboratory system using a robot includes capturing one or more images of the container in the diagnostic laboratory system, where each captured image includes image data. The image data is analyzed to identify one or more surface properties of the container. A gripping position of a gripper of a robot in the diagnostic laboratory system on the container is determined in response to the analysis. The container may then be gripped by the gripper at a gripping position. Other methods and systems are disclosed.

Description

[0001] Cross-application of related applications This application claims the benefit of U.S. Provisional Application No. 63 / 579,292, filed August 28, 2023, pursuant to 35 USC § 119(e). The entire contents of the patent application cited above are hereby expressly incorporated herein by reference. Technical Field

[0002] This disclosure relates to systems and methods for grasping containers in a diagnostic laboratory system. Background Technology

[0003] Diagnostic laboratory systems perform clinical chemistry tests to identify analytes or other components in biological samples or specimens (such as serum, plasma, urine, interstitial fluid, cerebrospinal fluid, etc.). Biological samples are collected in sample containers, such as sample tubes, and transported to the laboratory system, which may be located within a laboratory. Upon receipt of the sample container at the laboratory system, it is loaded into one or more sample container carriers (e.g., tube carriers or tube racks). The sample container carriers are then loaded into a sample transporter within the laboratory system. The sample transporter is an input / output module of the laboratory system that enables the system to receive and discharge sample containers. Robots can move sample containers between various locations and components within the laboratory system.

[0004] Sample containers may bear barcode labels or other markings containing patient information that should not be damaged during testing. The location of the barcode label varies depending on the sample container and its placement. If the robot used to transfer the sample container uses the same gripping position about the bottom of the container, the gripping may come into contact with and damage the barcode label. The sample container may also contain anomalous objects, which the robot's gripper or end effector should not come into contact with. Therefore, a laboratory system with a robot capable of gripping the sample container while avoiding markings and anomalous objects is sought. Summary of the Invention

[0005] According to a first aspect, a method is provided for using a robot to grasp a container in a diagnostic laboratory system. The method includes: capturing one or more images of the container in the diagnostic laboratory system, wherein each captured image includes image data; analyzing the image data to identify one or more surface properties of the container; determining, in response to the analysis, a gripping position on the container for gripping by a gripper of a robot in the diagnostic laboratory system; and using the gripper to grasp the container at the gripping position.

[0006] In another aspect, a diagnostic laboratory system is provided. The diagnostic laboratory system includes: an imaging device configured to capture images of a container; and a robot including a gripper configured to grasp the container and move it within the diagnostic laboratory system. Additionally, the diagnostic laboratory system includes a computer configured to: receive image data generated by the imaging device; analyze the image data to identify one or more surface properties of the container; determine a gripping position of the gripper on the container in response to the analysis; and generate instructions to instruct the gripper to grasp the container at the gripping position.

[0007] In another aspect, a method is provided for using a robot to grasp a sample container in a diagnostic laboratory system. The method includes: capturing one or more images of the sample container in the diagnostic laboratory system, wherein each captured image includes image data, and wherein the sample container is configured to hold a biological sample; analyzing the image data to identify one or more surface properties of the sample container, the one or more surface properties including at least one of markings, liquids, and abnormalities on the surface of the sample container; determining a gripping position on the sample container for a gripper of a robot in the diagnostic laboratory system to grasp, the gripping position avoiding the one or more surface properties; and using the gripper to grasp the sample container at the gripping position.

[0008] Further aspects, features, and advantages of this disclosure will readily become apparent from the following description and illustration of several exemplary embodiments, including the best mode conceived for carrying out this disclosure. This disclosure may also have other and different embodiments, and several details thereof may be modified in various respects without departing from the scope of this disclosure. Attached Figure Description

[0009] The accompanying drawings described below are provided for illustrative purposes and are not necessarily drawn to scale. Therefore, the drawings and descriptions are to be considered illustrative in nature and not restrictive. The drawings are not intended to limit the scope of this disclosure in any way.

[0010] Figure 1 The illustration shows a perspective view of a diagnostic laboratory system located in a laboratory according to one or more embodiments.

[0011] Figure 2 The illustration shows a sample transporter communicating with a diagnostic laboratory system according to one or more embodiments. Figure 1 A detailed view of the computer.

[0012] Figure 3 The illustration shows a robot and a sample container carrier located in a sample transporter of a diagnostic laboratory system according to one or more embodiments.

[0013] Figure 4AThe illustration shows a top perspective view of a node of a robotic gripper that undesirably grasps a sample container from a barcode label according to one or more embodiments.

[0014] Figure 4B Illustration Figure 4A A top view of the sample container, showing different locations on the surface of the sample container that are accessible to nodes according to one or more embodiments.

[0015] Figure 4C The illustration is based on one or more embodiments. Figure 4A The robot gripper is larger than Figure 4A The sample container is grasped at the high vertical position shown in the figure.

[0016] Figure 4D The illustration is based on one or more embodiments. Figure 4A Robotic gripper and gripper Figure 4A The sample containers shown are sample containers with different orientations.

[0017] Figure 4E The illustration shows a plan view of a sample container being gripped by three nodes of a robotic gripper according to one or more embodiments.

[0018] Figures 5A-5F The illustration shows a front view of sample containers with different surface properties according to one or more embodiments.

[0019] Figures 6A-6F The illustrations show examples of different sample containers and optimal and conventional gripping positions of the robotic gripper according to one or more embodiments.

[0020] Figure 7 The illustration shows a front view of a first sample container and a second sample container that move on or on a track of a diagnostic laboratory system according to one or more embodiments.

[0021] Figure 8 The diagram illustrates a flowchart of a method for grasping a sample container according to one or more embodiments.

[0022] Figure 9 The illustration shows a top view of the track of a diagnostic laboratory system according to one or more embodiments, wherein the sample container and sample mover are positioned at the transfer location.

[0023] Figure 10 The illustration is a flowchart of a method for using a robot to grasp containers in a diagnostic laboratory system according to one or more embodiments.

[0024] Figure 11 The illustration is a flowchart of a method for using a robot to grasp sample containers in a diagnostic laboratory system according to one or more embodiments. Detailed Implementation

[0025] In this patent application, nouns and pronouns referring to people generally do not specify a particular gender.

[0026] Automated diagnostic laboratory systems analyze (e.g., test) various biological samples, such as blood, serum, urine, and other bodily fluids. Samples are collected from the patient and placed in a sample container, such as a sample tube. The sample container, along with testing instructions, is then sent to the laboratory. The testing instructions specify which tests should be performed on the sample using instruments located in the laboratory. Technicians or software running on a computer determine which instruments in the laboratory should perform each test on each component of the sample according to the instructions.

[0027] Components in an automated diagnostic laboratory system can be broadly characterized as sample transport systems, sample movers, and instruments. Sample transport systems may include hardware such as tracks configured to move the sample mover within the laboratory system. The sample mover may house containers present on the sample transport system, such as sample containers. Instruments may be modules and / or analyzers to which the sample mover can be guided, where processing and analysis can be performed. Examples of instruments include centrifuges, chemical analyzers, cappers, storage modules, and refrigeration modules.

[0028] A typical workflow in a laboratory system may involve loading sample containers onto a sample mover and then instructing a sample transport system to transport the sample mover to one or more of these instruments. Many laboratory systems use robotic systems to move sample containers to various locations, such as within instruments and to and from transport systems. Robotic systems can use robotic grippers (e.g., end effectors) to grasp and move sample containers. In addition to sample containers, robotic systems can also move quality control kits, calibrator kits, and other items throughout the laboratory system.

[0029] Robotic systems can manipulate and move grippers using gripping rules that include multiple instructions. Given the diverse range of items a robotic system can grasp, applying the same gripping rule to every single item may not be advantageous. For example, the different geometries and surface properties of items can make the use of the same gripping rule inefficient. Furthermore, if the gripper uses the same gripping rule to grasp all containers, there is a risk of grasping short containers at unstable locations (such as near the top), potentially leading to breakage or spillage of the container's contents.

[0030] Some containers may have markings, such as barcode labels, attached to their external surfaces. These markings may include or reference patient information or testing criteria, for example. These barcode labels may need to remain intact as the containers move throughout the laboratory system. The location of the markings may vary between individual containers. Therefore, if the same gripping rules and locations are used for all containers, the grippers may damage the markings. For example, the grippers may grab the markings, potentially scratching or destroying some of them.

[0031] In some cases, the exterior of some containers may have exposed adhesives, such as from labeling tags, and / or liquids spilled onto the exterior surfaces. If a robotic gripper grasps these areas, these adhesives and / or liquids may impede the gripper's gripping action. For example, adhesives may cause the robotic gripper to adhere to the container, and liquids may cause the robotic gripper to slide relative to the container.

[0032] The methods and apparatus described herein enhance the gripping capabilities of automated diagnostic systems by identifying gripping positions on containers that allow robotic grippers to grasp the container with minimal impact on the container or its external surface. The identified position can be termed the optimal gripping position for the container. The method can take into account the degrees of freedom of the robotic gripper to provide the best possible way (e.g., position) to grip the container. The optimal gripping position enables safer and more reliable gripping while reducing potential damage to the container and system due to container breakage and spillage. Furthermore, gripping rules (such as the gripping force applied by the gripper) can take into account the surface properties of the container.

[0033] This method can capture one or more images of a container, each captured image including image data. The image data can be analyzed to identify one or more surface and / or geometric properties of the container. Geometric / surface properties may include the height of the container, the material used to make the container, markings on the exterior of the container, surface anomalies, and liquid on the exterior of the container. Based on this analysis, the gripping effect and / or position of a robotic gripper grasping the container can be determined. The gripping effect may, for example, avoid contact with markings or liquid. The gripping position of the robotic gripper on the container can be determined in response to the gripping effect. The gripping position may be a location where the gripper will not adversely affect or be adversely affected by the container. The gripping position may also be a location where the gripper can grasp the container without slipping or sticking to it. The gripping position may also be a location where the robotic gripper is unlikely to damage or otherwise harm the container. (The following is in conjunction with...) Figure 1-11 These and other systems, methods and apparatuses for determining the gripping effectiveness of robotic grippers used in laboratory systems are described in more detail.

[0034] refer to Figure 1This is a perspective view of laboratory 100. Diagnostic laboratory system 102 may be located within laboratory 100. Laboratory system 102 may be configured to perform multiple analyses or tests on a variety of different biological samples. For example, these tests may determine the amount of components or chemicals present in biological samples such as blood, urine, cerebrospinal fluid, and other biological samples. In other embodiments, laboratory system 102 may be configured to perform multiple different tests on a single biological sample type (such as serum). In other embodiments, laboratory system 102 may be configured to perform a single type of test on a single biological sample type (such as serum).

[0035] Laboratory system 102 may include multiple diagnostic instruments 104 (some labeled), which are configured to perform the same or different tests on biological samples. In some embodiments, the diagnostic instruments 104 may be transported by a transport system (e.g., track 216 - Figure 2 Interconnection. The transport system can be configured to transport biological samples between diagnostic instrument 104 and / or other devices (such as centrifuges and cap openers) in laboratory system 102. The configuration of laboratory system 102 can differ from... Figure 1 The configuration shown is illustrated. In some embodiments, the laboratory system 102 may include only a single diagnostic instrument 104.

[0036] Laboratory system 102 can be connected to computer 120, which can be located inside or outside laboratory 100. In some embodiments, some portions of computer 120 can be located inside laboratory 100, and other portions of computer 120 can be located outside laboratory 100. Computer 120 may include processor 122 and memory 124, wherein memory 124 stores program 126 configured to be executed or run on processor 122. In some embodiments, memory 124 and / or program 126 can be located outside computer 120. For example, computer 120 may be connected to the Internet to access external data, etc. Program 126 can operate diagnostic instrument 104 and process data generated by diagnostic instrument 104.

[0037] Memory 124 can be any suitable type of memory, such as, but not limited to, one or more of volatile memory and / or non-volatile memory. Memory 124 can have multiple programs 126, each consisting of instructions stored therein that, when executed by processor 122, cause processor 122 to perform various actions specified by one or more of the stored instructions. Program instructions can be provided to processor 122 to perform operations according to the current system and methods detailed in the flowcharts and / or block diagrams described herein. Processor 122 thus configured can become a special-purpose machine particularly suited to perform operations according to the current system and methods. Program instructions can be stored in a computer-readable medium, such as memory 124, which can instruct processor 122 to function in a particular manner. As used herein, the term "memory" can refer to both non-transitory memory and transient memory.

[0038] At least one of the diagnostic instrument 104 or other components may be a sample transporter 130, which is described in more detail below. Figure 1 In one embodiment, the sample carrier 130 is a component in the laboratory system 102. In other embodiments, the laboratory system 102 may include multiple sample carriers. The operations performed by the sample carrier 130 may be implemented in one or more of the diagnostic instruments 104. The sample carrier 130 may be located in various locations within the laboratory system 102, such as within individual diagnostic instruments of the diagnostic instruments 104.

[0039] Sample handler 130 is used to receive items into and dispense items from laboratory system 102. Items may include sample containers (e.g., sample container 210). Figure 2 The sample container 210, received in the sample carrier 130, may contain biological samples to be tested by one or more of the diagnostic instruments 104. The sample container 210 dispensed from the laboratory system 102 may contain residual liquid after the biological sample has been tested.

[0040] Additional References Figure 2The illustration shows a more detailed embodiment of a computer 120 communicating with a sample carrier 130. The computer 120 may include multiple programs 126 that can run on a processor 122. One of the programs 126 may be a robot controller 204, which may be configured to generate instructions that cause a robot 206 to move to a specific location as described herein. For example, these instructions may cause the robot 206 or parts thereof to move within the sample carrier 130 and perform certain operations. The instructions may also cause the robot 206 to move a sample container 210 between a sample container carrier 212 and a sample mover 214. The sample mover 214 can be transported via a transport system to a diagnostic instrument 104 ( Figure 1 Move sample container 210 between ( ) Figure 2 In one embodiment, the transport system may be a track 216 configured to move the sample mover 214.

[0041] Image processor 220 may be connected to computer 120 and may be configured to receive image data 222 generated by imaging device 224. In some embodiments, one or more portions of image processor 220 may be implemented in imaging device 224. In some embodiments, imaging device 224 may be configured to capture three-dimensional images of sample container 210, sample container carrier 212, and other items. Image processor 220 may be configured to send instructions to imaging device 224 instructing the imaging device to capture images, such as images of sample container 210 and other items in laboratory system 102. Imaging device 224 may be, for example, a digital camera or a 3D camera.

[0042] Computer 120 may include a container property recognition algorithm 226, which may be a machine learning algorithm such as a software model trained by machine learning. In some embodiments, the container property recognition algorithm may include a convolutional neural network. In other embodiments, the container property recognition algorithm 226 may include a deep neural network. The container property recognition algorithm 226 may be configured to analyze image data 222. In some embodiments, the container property recognition algorithm 226 may be configured to analyze image data 222 processed by image processor 220, which may output enhanced, merged, or optimized image data. The container property recognition algorithm 226 may identify properties such as the size of sample container 210, the container geometry of sample container 210, whether sample container 210 has a lid, barcodes and other markings, and other items. Other items may be anomalous objects, such as sample contents that have spilled from sample container 210, foreign objects or materials, residual or spilled adhesive used to attach barcode labels to sample container 210, damage or markings on barcode labels, etc.

[0043] The container property recognition algorithm 226 can also determine the height of the sample in the sample container 210 by recognizing color or brightness transitions in an image of the sample container 210. The material (e.g., glass or plastic) of the sample container 210 can be determined by analyzing surface light reflected or transmitted through the sample container 210. When imaging items other than the sample container 210 (such as reagent packages), the container property recognition algorithm 226 can analyze the properties of these other items.

[0044] The optimal gripping position algorithm 230 may be a program or machine learning software model configured to identify the optimal gripping position on the sample container 210 or other object when the robot 206 grips it. The optimal gripping position algorithm 230 may be a machine learning algorithm, such as a convolutional neural network or a rule-based network, used to implement the methods described herein. In some embodiments, the optimal gripping position algorithm 230 may include a deep neural network. Some containers may have several optimal gripping positions, and the optimal gripping position algorithm 230 may select one of these optimal gripping positions. The optimal gripping position algorithm 230 may analyze data generated by the container property recognition algorithm 226 to identify the optimal or appropriate position where the robot 206 can grip individual sample containers within the sample container 210. The optimal gripping position may avoid barcode labels, markings, abnormal objects, lids, spilled samples and other liquids, and other items that may interfere with the correct gripping of the sample container 210.

[0045] Computer 120 can be connected to workstation 240, enabling a user to communicate with computer 120. Workstation 240 may include a display 242 and a keyboard 244 and / or other input devices. Data generated by computer 120 can be displayed on display 242. A user can input data to computer 120 via keyboard 244 and / or other input devices. Display 242 can be configured to display images captured by imaging device 224 and / or other imaging devices.

[0046] Computer 120 and / or laboratory system 102 may be connected to laboratory information system (LIS) 250. In some embodiments, LIS 250 may be a program and may be executed by computer 120. LIS 250 may receive data and / or instructions from hospital information system (HIS) 252, which may be implemented at least in part in a program executed by computer 120. Healthcare professionals may enter patient-specific test requests into HIS 252. For example, a physician may request a test for a first chemical substance on blood taken from a first patient and a test for a second chemical substance on blood taken from a second patient. These test requests may be entered into HIS 252. The test requests may then be transmitted to LIS 250, which generates a test plan to be run by laboratory system 102, wherein a specific diagnostic instrument 104 ( Figure 1 ) Conduct the test.

[0047] Sample transporter 130 ( Figure 2 An embodiment of the present invention is shown to hold three sample container carriers 212, referred to as first container carrier 212A, second container carrier 212B, and third container carrier 212C, respectively. The sample container carriers 212 may be of different types, such as from different suppliers or manufacturers. Some of the container slots 232 may be occupied by sample container 210. Occupied container slots are identified by having a dark fill. Sample container 210A is shown as occupying a container slot in the third container carrier 212C and will be referenced in the examples herein. The apparatus and methods described herein can be implemented on sample containers with configurations and surface properties different from those of sample container 210A.

[0048] Additional References Figure 3 The illustration shows a front perspective view of an embodiment of a robot 206 that grasps a sample container 210A. The robot 206 may include a gripper 304 (e.g., an end effector) configured to grasp the sample container 210 and move the sample container 210 within the sample carrier 130, including moving the sample container carrier 212 in and out of the sample container carrier 212. Figure 2 ).exist Figure 3 In one embodiment, robot 206 is illustrated as moving sample container 210A into and out of a third container carrier 212C. Robot 206 can be configured to move sample container 210A into and out of all sample container carriers 212C. Figure 2 ).

[0049] Robot 206 may include multiple gantries enabling gripper 304 to move along the x-, y-, and z-directions. A first gantry 310 may be configured to move gripper 304 along the y-direction. A second gantry 312 may be configured to move gripper 304 along the x-direction. A third gantry 314 may be configured to move gripper 304 along the z-direction. The gantries may be controlled by motors (not shown), which may receive commands from robot controller 204. Figure 2 The instructions generated.

[0050] Robot 206 may include an arm 320 to which gripper 304 may be attached. In some embodiments, arm 320 may be attached to a third gantry 314. Imaging device 224 may also be attached to arm 320. Thus, robot 206 may be configured to move imaging device 224 within the entire sample handler 130 to capture images such as sample container 210 from various viewpoints as described herein. Figure 2 (Image of an item.)

[0051] Additional References Figure 4A Its illustration Figure 3 Enlarged view of the 304 gripper. Figure 4A The gripper 304 illustrated in the diagram includes two fingers 400, referred to as the first finger 400A and the second finger 400B, respectively. The ends of the fingers 400 may have nodes 404 configured to contact the sample container 210A. Friction between the nodes 404 and the sample container 210A enables the gripper 304 to grasp and move the sample container 210A, as described herein. Figure 4A In this embodiment, each of the two fingers 400 includes two nodes, referred to as nodes 404A, 404B, 404C, and 404D. The gripper 304 is moved by the robot 206, and the movement of the gripper 304 includes the movement of the fingers 400 and nodes 404.

[0052] Figure 4B The diagram illustrates a top view of sample container 210A, showing the position where node 404 can contact the outer surface of sample container 210A. Grip 304 can be configured to rotate or pivot relative to sample container 210A along arc R41. Node 404, shown as a solid line, indicates a first gripping position where node 404 can contact sample container 210A. Node 404, shown as a dashed line, indicates a second gripping position where node 404 can contact sample container 210A. For example, gripper 304, and therefore finger 400, can rotate as shown by arc R41, such that node 404 can contact sample container 210A at various arcuate positions on the surface of sample container 210A.

[0053] The gripper 304, and therefore the pointer 400 and node 404, can be configured to move along the z-direction to grip the sample container 210A at various vertical positions or heights. Figure 4C Node 404 in the diagram is compared to Figure 4A The node 404 shown grasps the sample container 210A at a vertical position higher than its vertical position. Figure 4C In this configuration, the gripper 304 and the finger 400 may have positioned the node 404 such that the node 404 does not contact the barcode label 410 located on the surface of the sample container 210A. Figure 4A In the configuration, the vertical position of node 404 is... Figure 4C The medium-low temperature and contact with barcode label 410 are present. The methods and apparatus described herein can prevent node 404 from... Figure 4A The node 404 contacts the barcode label as shown in the diagram. In some embodiments, node 404 may contact the barcode label, but direct contact with the barcode itself may be avoided.

[0054] Additional References Figure 4D The diagram shows a gripper 304 configured to have a large number of degrees of freedom, such as six. These degrees of freedom enable the gripper 304 to have many different postures to grasp and move the sample container 210A. These degrees of freedom may also be referred to as posture degrees of freedom. The gripper 304 may have other degrees of freedom. Figure 4D The gripper 304 can be configured to move or rotate along the arc R42. Therefore, the gripper 304, fingers 400, and nodes 404 can grip the sample container 210A when it is tilted or in various different orientations. The ability of the gripper 304 to move along the arc R42 can be as follows: Figure 4B and Figure 4C The additional capability of movement along the z-direction and along arc R41 as described above. In some embodiments, the gripper 304 may also be configured to move along an arc perpendicular to arc R42.

[0055] The gripper 304 has been described as having two fingers 400 and four nodes 404. The gripper 304 can have different configurations of fingers and nodes. (Reference) Figure 4E The diagram illustrates a top view of sample container 210A, with three nodes 412 contacting the exterior of sample container 210A. The three nodes 412 may be equidistantly spaced around the circumference of the outer surface of sample container 210A. Grip 304 may have other numbers of nodes configured to contact the outer surface of sample container 210A.

[0056] Different types of sample containers 210 can be used in the sample transporter 130, and the sample containers 210 can have different surface and geometric properties. Additional References Figures 5A-5FThe illustration shows front views of sample containers 500A-500F with different surface and geometric properties. Sample containers 500A-F can be similar to sample container 210A and are manufactured using procedure 126 ( Figure 2 The determined method is used to capture and move the data within the entire laboratory system 102. Figure 2 ).

[0057] Figure 5A The diagram shows a front view of sample container 500A, with barcode label 502 approximately located in the center of sample container 500A. Sample container 500A has a height H51, a width W51, and a lid 503, which are geometric and / or surface properties of sample container 500A. If gripper 304 grips the center section of sample container 500A, gripper 304 may come into contact with barcode label 502 and / or the barcode itself, and may therefore damage barcode label 502 or the barcode. Figure 5B As shown, due to gripper 304 or node 404 ( Figure 4A Defect 504, caused by unfavorable contact with barcode label 502, may damage barcode label 502. Defect 504 may prevent barcode label 502 from being read correctly by a barcode reader (not shown) (unless otherwise stated, barcodes and barcode labels may be used interchangeably herein).

[0058] Additional References Figure 5C The illustration shows sample container 500C, where barcode label 508 is positioned higher on the sample container than barcode label 502. Sample container 500C has a height H52, a width W52, and excludes the lid; these are geometric and / or surface properties of sample container 500C. Height H52 may be shorter than height H51, and width W52 may be wider than width W51. The optimal gripping position of gripper 304 on sample container 500C may differ from the optimal gripping position on sample container 500A to avoid contact with barcode label 508. For example, optimal gripping position algorithm 230 may determine to grip sample container 500C above barcode label 508, but not too close to the opening (which could cause sample container 500C to break). In other cases, optimal gripping position algorithm 230 may determine that sample container 500C should be gripped below barcode label 508.

[0059] Figure 5DThe sample container 500D shown has an anomaly 510 located on the sample container 500D above the barcode label 512 and below the lid 514. Some of the sample 516 located in the sample container 500D (e.g., liquid) may have spilled out of the sample container 500D, resulting in the anomaly 510. If the gripper 304 comes into contact with the anomaly 510, the gripper 304 may be contaminated, or the gripper 304 may fail to grip the sample container 500D correctly. The optimal gripping position algorithm 230 can determine that the sample container 500D should be gripped below the anomaly 510 and above the barcode label 512.

[0060] refer to Figure 5E A barcode label 522 located on sample container 500E has an anomaly 524. The anomaly 524 may be a torn portion of the barcode label 522 or a portion with exposed adhesive. If the gripper 304 comes into contact with the anomaly 524, the gripper 304 may become contaminated, or may further tear or damage the barcode label 522. Therefore, the optimal gripping position algorithm 230 can determine that the gripper 304 should grip sample container 500E while avoiding both the barcode label 522 and the anomaly 524. For example, the optimal gripping position algorithm 230 can determine that the gripper 304 should grip the sample container above the anomaly 524 and below the cap 526.

[0061] exist Figure 5F In this case, barcode label 528 is not properly attached to sample container 500F. If gripper 304 grasps barcode label 528, barcode label 528 may become completely detached from sample container 500F. In other cases, some parts of barcode label 528 may adhere to gripper 304, which may interfere with the operation of gripper 304. Therefore, optimal gripping position algorithm 230 can determine that gripper 304 should grip sample container 500F above barcode label 528 to avoid further detachment of barcode label 528.

[0062] Barcode labels attached to sample containers may make it difficult to identify geometric properties. (Reference) Figure 5F A partially detached barcode label 528 may obscure the true geometric properties of the sample container 500F. A container property recognition algorithm 226 can be trained to recognize the sample container even when the corresponding barcode label is not correctly attached. For example, the container property recognition algorithm 226 could be a CNN that uses the correct sample container geometry as a basic fact.

[0063] Although the robotic sample-grabbing container 210 described herein is shown ( Figure 2 However, the robot can also be configured to grasp laboratory system 102 ( Figure 1Other items used in the process. For example, the robot can grasp quality control packages and calibration packages. The robot can also be configured to grasp different types of sample containers, such as capillaries, sample cups with tube tops, and other containers configured to hold biological samples. Container property recognition algorithm 226 can identify these other containers and objects, and optimal grasping position algorithm 230 can generate instructions for gripper 304 to grasp these other containers and objects.

[0064] Additional References Figure 2 The laboratory system 102 described in this paper ( Figure 1 Images of sample container 210 are captured by a method and can be analyzed by computer 120. For example, image data 222 can be analyzed by image processor 220 and / or container property recognition algorithm 226. Analysis such as that performed by container property recognition algorithm 226 can identify surface properties of sample container 210 that may interfere with the operation of gripper 304 or may damage barcode labels (which would hinder processing of sample container and / or samples therein). Optimal gripping position algorithm 230 can then determine the best or appropriate position on sample container 210 for gripper 304 to grip.

[0065] Computer 120 can generate gripping parameters or rules instructing gripper 304 to grip sample container 210. For example, robot controller 204 can generate instructions instructing gripper 304 to move to a specific position and grip sample container 210 at the optimal gripping position. Gripping parameters may include, but are not limited to: the three-dimensional (3D) position of the gripping point or gripping area on the surface of sample container 210; a (3D) vector indicating the direction of gripper 304 approaching sample container 210; the joint angles of gripper 304 to achieve the optimal gripping position within robot constraints; the joint trajectory of gripper 304 to characterize approaching sample container 210; and the force applied by gripper 304 to reliably hold sample container 210. In some embodiments, program 126 may determine that specific surface properties may require gripper 304 to apply specific forces to sample container 210.

[0066] Laboratory system 102 identifies the optimal gripping position of sample container 210 for gripper 304. In some cases, the gripping properties may enable gripper 304 or another robot to grasp and move other items within laboratory system 102, such as reagent packages. The laboratory system 102 and methods described herein can take into account the degrees of freedom of robot 206 to provide the optimal manner for handling sample container 210. The optimal manner of gripping sample container 210 provides safe and reliable gripping while reducing potential damage to sample container 210 caused by gripper 304 (e.g., breakage and spillage of sample container 210).

[0067] Additional References Figures 6A-6F The illustration shows examples of images of different sample containers 600A-600G and optimal and conventional gripping positions 602. The optimal gripping position refers to the gripping position determined by the laboratory system 102 and / or methods described herein, such as by the optimal gripping position algorithm 230. The optimal gripping position may be node 404 on the sample containers 600A-600G (… Figure 4C The optimal gripping position 602 is shown as a horizontal solid line uniquely positioned over each of the sample containers 600A-600G. As illustrated, the optimal gripping position is unique based on the different surface and / or geometry of each individual sample container 600A-600G. (Note: The text also mentions the position where the robot gripper would normally grasp sample containers 600A-600G, indicated by a dashed line.)

[0068] Sample container 600A is relatively short, with a barcode label 604 positioned in the middle of the test tube section, and has no lid. Anomaly 605 is located on the upper portion of barcode label 604. Sample container 600A is also almost full of sample 606, indicated by the shading. The regular gripping position 602 on sample container 600A is positioned very close to barcode label 604 and on anomaly 605. Therefore, during regular gripping, gripper 304 ( Figure 4A It may come into contact with barcode label 604 and / or anomaly 605, and may cause barcode label 604 to resemble Figure 5B The barcode label 502 is unreadable. The optimal gripping position 608 has been determined to be above the barcode label 604, so that the gripper 304 does not contact the barcode label 604 or the abnormal object 605. The optimal gripping position 608 is also below the top of the open sample container 600A to prevent the gripper 304 from damaging the sample container 600A.

[0069] Images of the sample container 600A can be obtained from imaging device 224 ( Figure 2The image data 222 generated by the imaging device 224 is captured and analyzed by the image processor 220, which can process the image data 222. The container property recognition algorithm 226 can identify surface and / or geometric properties in the image data 222 or the processed image data 222. Surface properties may include the condition and / or size of the barcode label 604, and the position of the barcode label 604 and the anomaly 605. Geometric properties may include the height and width of the sample container 600A. Surface and / or geometric properties may also include the absence of a lid and the height of the sample 606 within the sample container 600A. Based on the surface and / or geometric properties, the optimal gripping position algorithm 230 can identify an optimal gripping position 608. The optimal gripping position 608 can be determined to be higher than the normal gripping position 602, which can prevent the gripper 304 from damaging the barcode label 604 and contacting the anomaly 605.

[0070] The sample container 600B is relatively tall, including the lid 610, and a barcode label 614 is positioned in the upper part of the test tube. The standard gripping position 602 is on the barcode label 614, which may cause the gripper 304 ( Figure 4C As described above, the barcode label 614 is damaged. An image of the sample container 600B can be obtained from the imaging device 224. Figure 2 The image data 222 is captured and analyzed to identify the surface and / or geometric properties of the sample container 600B. The container property identification algorithm 226 can identify surface and / or geometric properties in the image data 222, including the height and width of the sample container 600B, lid status, barcode label size and position, and any anomalies on the surface of the sample container 600B. Based on the surface and / or geometric properties, the optimal gripping position algorithm 230 can identify an optimal gripping position 616, which can be located below the barcode label 614 to prevent the gripper 304 from being caught. Figure 4C 614. Damaged barcode label.

[0071] Images of sample container 600C can be captured and analyzed to identify its surface and / or geometric properties. Surface properties may include the condition and location of barcode label 624, and the location and size of anomaly 625. Geometric properties may include the presence of lid 628 and the height of sample 629 contained within sample container 600C. Sample container 600C has the same height as sample container 600A, but barcode label 624 on sample container 600C is lower than barcode label 604. Optimal gripping position 630 can be determined by optimal gripping position algorithm 230 to be slightly lower than the normal gripping position 602, which prevents gripper 304 from interfering with lid 628 and anomaly 625.

[0072] The sample container 600D is relatively tall and does not include a lid. The container property recognition algorithm 226 can identify the height, width, and / or lid status. Based on these properties, the optimal gripping position algorithm 230 can determine an optimal gripping position 634, which is below the top of the sample container 600D and above the normal gripping position 602, to prevent the gripper 304 from damaging the sample container 600D.

[0073] The sample container 600E includes a barcode label 640 that is skewed and not properly attached to the sample container 600E. The image data 222 of the sample container 600E can be analyzed by the container property recognition algorithm 226. Figure 2 The algorithm can identify skewed and partially detached barcode labels 640. The optimal grasping position algorithm 230 can identify the optimal grasping position 642 as being much higher than the barcode label 640, in order to prevent node 404 ( Figure 4C This could further damage the barcode label 640 or cause it to detach. In some embodiments, the optimal gripping position algorithm 230 may instruct the robot controller 204 ( Figure 2 Generate instructions to prevent node 404 from touching barcode label 640 while approaching and leaving sample container 600E.

[0074] Sample container 600F is skewed. For example, sample container 600F may already be in the sample mover (e.g., one of sample movers 214). Figure 2 The image data of sample container 600F can be analyzed by container property recognition algorithm 226. Figure 2 The algorithm can identify skewed sample containers 600F. The optimal grasping position algorithm 230 can identify the optimal grasping position 648 as being much higher than the barcode label 646, in order to prevent node 404 ( Figure 4C Damaged barcode label 646. Optimal gripping position 648 can be tilted to be perpendicular to the side of sample container 600F. In some embodiments, optimal gripping position algorithm 230 can instruct robot controller 204 ( Figure 2 ) generates instructions for gripper 304 to grip sample container 600E at an angle or orientation similar to or the same as that of sample container 600F. Therefore, node 404 ( Figure 4C It can fully contact or approach the optimal gripping position 648.

[0075] refer to Figure 7The illustration shows a first sample container 700 and a second sample container 702 located on or moving on track 216. The first sample container 700 is transported by a first sample mover 706, and the second sample container 702 is transported by a second sample mover 708. Sample movers 706 and 708 may experience wear or sudden movement, which may cause sample container 702 to become skewed, as shown by sample container 600F. The first sample container 700 has a vertical orientation shown by dashed line 710, which may be perpendicular to track 216. The second sample container 702 has a skewed orientation shown by dashed line 712, which may be caused by movement of the second sample mover 708 during startup and shutdown, or by wear and tear on the second sample mover 708.

[0076] Additional References Figure 4D The gripper 304 shown can be moved to grasp the second sample container 702. For example, the image data 222 of the second sample container 702 can be analyzed by the container property recognition algorithm 226. Figure 2 The optimal gripping position algorithm 230 can instruct the robot controller 204 to move the gripper 304 based on the orientation of the second sample container 702 to grip the skewed second sample container 702.

[0077] Laboratory System 102 ( Figure 1 ) can be configured to move sample mover 214 ( Figure 2 The device moves to a transfer position, where the imaging device 224 can generate image data 222 and can grasp sample containers 700 and 702. Figure 7 In one embodiment, track 216 has a transfer position 716 at which the first sample mover 706 moves, such that the first sample container 700 can be imaged and / or moved by robot 206. Figure 3 ) Grasping. In some cases, due to problems with the track 216 or other objects in the laboratory system 102, the first sample mover 706 may move to an offset transfer position 718 instead of the original transfer position 716. When the first sample container 700 is at the offset transfer position 718, image data 222 of the first sample container 700 can still be generated. Container property recognition algorithm 226 ( Figure 2 The image data 222 can be analyzed to identify the position of the first sample container 700 at the offset transfer position 718. Then, the optimal gripping position algorithm 230 can instruct the robot controller 204 to grip the clamp 304 ( Figure 4C Move to the offset transfer position 718, at which the gripper 304 can correctly grip the first sample container 700 as described herein.

[0078] Additional References Figure 8 It is illustrated as a sample container (such as...) Figure 7 The flowchart of method 800 (for the first sample container 700) is shown below. Method 800 includes waiting in block 802 for a sample container (e.g., the first sample container 700) to reach a transition position 716 (and, in some embodiments, a transition position 716 within a predetermined offset, such as an offset transition position 718). Processing proceeds to decision block 804, where it is determined whether the sample container is located at transition position 716. If no sample container is located at transition position 716, the process continues to wait as described in block 802.

[0079] If a sample container is present at transfer position 716, the process proceeds to block 806, where an image of the first sample container 700 is captured (e.g., by imaging device 224) and analyzed to detect surface and / or geometric properties. For example, container property recognition algorithm 226 can analyze image data 222 to identify the surface and / or geometric properties of the first sample container 700. The process then proceeds to block 808, where one or more optimal gripping positions are determined, and gripping parameters of robot 206, such as the orientation and gripping position of gripper 304, are also determined.

[0080] When the optimal gripping position is determined, processing proceeds to frame 810, where gripper 304 grips the first sample container 700 using gripping parameters. For example, robot controller 204 ( Figure 2 It can generate instructions to move the gripper 304 as determined by the optimal gripping position algorithm 230.

[0081] refer to Figure 9 The diagram shows a top view of track 216, where the first sample container 700 and sample mover 706 are positioned at transfer position 716. Figure 9 In this embodiment, there are three imaging devices 900 positioned close to orbit 216, referred to as the first imaging device 902, the second imaging device 904, and the third imaging device 906, respectively. Each of the imaging devices 900 can generate image data 222 from a different field of view of the first sample container 700. Figure 2 These different fields of view are shown as dashed lines extending from the imaging device 900. The image processor 220 can merge or stitch together the images generated by the imaging device 900 to form a single image for analyzing the 360-degree field of view of the first sample container 700. The above-described processing can be applied using the 360-degree field of view of the first sample container 700. For example, the determinations made by the container property recognition algorithm 226 and the optimal grasping position algorithm 230 can be based on the 360-degree field of view.

[0082] Determining the optimal gripping position can include the material type of the item being gripped (e.g., a sample container) (such as glass or plastic). For example, if a particular material makes the item prone to breakage, the optimal gripping position might be located away from a top opening, which may be the weakest point in the material where gripping could potentially cause the item to break. Grip properties may also take into account other factors such as the mechanical constraints of robot 206, including joint limitations, force / torque considerations, and the available workspace of robot 206. Surface friction of sample container 210 at the optimal gripping position may be a factor in determining the force exerted by gripper 304 on sample container 210.

[0083] Now for reference Figure 10 This is a flowchart of method 1000 for using a robot (e.g., robot 206) to grasp a container (e.g., sample container 210) in a diagnostic laboratory system (e.g., laboratory system 102). The method includes, in block 1002, capturing one or more images of the container in the diagnostic laboratory system, wherein each captured image includes image data (e.g., image data 222). Method 1000 includes, in block 1004, analyzing the image data to identify one or more surface properties of the container. Surface properties may be markings (such as barcodes, barcode labels), anomalous objects (such as adhesive from barcode labels and spilled liquid). Surface properties may also be a lid on a sample container. Method 1000 includes, in block 1006, determining, in response to the analysis, a grasping position (e.g., grasping position 608, 616, 630) on the container for a gripper (e.g., gripper 304) of the robot (e.g., robot 206) in the diagnostic laboratory system to grasp the container. The grasping position may be an optimal grasping position that avoids surface properties that may obstruct grasping. Therefore, the gripper 304 can avoid contact with barcodes, barcode labels, liquids, adhesives, and abnormalities on the surface of the container. Method 1000 includes, in frame 1008, using the gripper to grasp the container at the gripping position.

[0084] Now for reference Figure 11The illustration shows a flowchart of method 1100, which uses a robot (e.g., robot 26) to grasp a sample container (e.g., sample container 210) in a diagnostic laboratory system (e.g., laboratory system 102). Method 1100 includes, in block 1102, capturing one or more images of the sample container in the diagnostic laboratory system, wherein each captured image includes image data (e.g., image data 222), and wherein the sample container is configured to hold a biological sample. Method 1100 includes, in block 1104, analyzing the image data to identify one or more surface properties of the sample container, including at least one of markings (e.g., barcode label 410), liquids, and abnormalities (e.g., abnormality 510) on the surface of the sample container. Method 1100 includes, in block 1106, determining gripping positions (e.g., gripping positions 608, 616, 630) on the sample container for a gripper (e.g., gripper 304) of a robot (e.g., robot 206) in a diagnostic laboratory system to grip the sample container, these gripping positions avoiding the one or more surface properties. Method 1100 includes, in block 1108, using the gripper to grip the sample container at the gripping positions.

[0085] Non-limiting illustrative examples Illustrative Example 1. A method for using a robot to grasp a container in a diagnostic laboratory system, the method comprising: capturing one or more images of the container in the diagnostic laboratory system, wherein each captured image includes image data; analyzing the image data to identify one or more surface properties of the container; determining, in response to the analysis, a gripping position on the container for gripping by a gripper of a robot in the diagnostic laboratory system; and using the gripper to grasp the container at the gripping position.

[0086] Illustrative Example 2. The method according to Illustrative Example 1 further includes generating instructions to instruct the gripper to grip the container at the gripping position.

[0087] Illustrative Example 3. The method according to one embodiment of the foregoing embodiments, wherein the container is a sample container configured to hold a biological sample.

[0088] Illustrative Example 4. The method according to one embodiment of the foregoing examples, wherein the container is configured to hold chemicals consumed in a diagnostic laboratory system.

[0089] Illustrative Example 5. The method according to one embodiment of the foregoing embodiments, wherein: the gripper has at least one degree of freedom; and determining the gripping position includes determining the gripping position of the gripper in response to the at least one degree of freedom.

[0090] Illustrative Example 6. The method according to one embodiment of the foregoing embodiments, wherein the container has one or more geometric properties, and wherein determining the gripping position includes determining the gripping position of the gripper in response to the one or more geometric properties.

[0091] Illustrative Example 7. The method according to one embodiment of the foregoing embodiments further includes determining the material of the container, wherein determining the gripping position includes determining the gripping position of the gripper in response to determining the material of the container.

[0092] Illustrative Example 8. The method according to one embodiment of the foregoing embodiments, wherein the analysis includes identifying the location of a mark on the container, wherein determining the gripping location includes determining the gripping location of the gripper in response to determining the location of the mark.

[0093] Illustrative Example 9. The method according to one embodiment of the foregoing examples, wherein the grab position is not on the marker.

[0094] Illustrative Example 10. The method according to one embodiment of the foregoing embodiments, wherein the analysis includes identifying anomalies on the container, and wherein determining the gripping position includes determining the gripping position of the gripper in response to identifying the anomaly.

[0095] Illustrative Example 11. The method according to one embodiment of the foregoing examples, wherein the grasping location is not on the abnormal object.

[0096] Illustrative Example 12. The method according to one embodiment of the foregoing examples, wherein the abnormal substance is a liquid.

[0097] Illustrative Example 13. The method according to one embodiment of the foregoing embodiments, wherein the abnormal object is part of a label attached to the container.

[0098] Illustrative Example 14. The method according to one embodiment of the foregoing embodiments, wherein the analysis includes identifying the orientation of the container and further includes adjusting the orientation of the gripper in response to identifying the orientation of the container.

[0099] Illustrative Example 15. The method according to one embodiment of the foregoing embodiments, wherein the analysis determines whether the container has a lid.

[0100] Illustrative Example 16. The method according to one embodiment of the foregoing embodiments, wherein analyzing image data or determining at least one of the grasping locations includes using a deep neural network.

[0101] Illustrative Example 17. According to an embodiment of the method described in the foregoing embodiments, determining the gripping position of the robot's gripper on the container includes determining the height at which the gripper grips the container.

[0102] Illustrative Example 18. A diagnostic laboratory system includes: an imaging device configured to capture images of a container; a robot including grippers configured to grasp the container and move it within the diagnostic laboratory system; and a computer configured to: receive image data generated by the imaging device; analyze the image data to identify one or more surface properties of the container; determine a gripping position of the grippers on the container in response to the analysis; and generate instructions to instruct the grippers to grasp the container at the gripping position.

[0103] Illustrative Example 19. A diagnostic laboratory system according to Illustrative Example 18, wherein the container is a sample container configured to hold biological samples.

[0104] Illustrative Example 20. A method for using a robot to grasp a sample container in a diagnostic laboratory system, the method comprising: capturing one or more images of the sample container in the diagnostic laboratory system, wherein each captured image includes image data, and wherein the sample container is configured to hold a biological sample; analyzing the image data to identify one or more surface properties of the sample container, the one or more surface properties including at least one of markings, liquids, and abnormalities on the surface of the sample container; determining a gripping position on the sample container for a gripper of a robot in the diagnostic laboratory system to grasp, the gripping position avoiding the one or more surface properties; and grasping the sample container at the gripping position using the gripper.

[0105] While this disclosure is open to various modifications and alternatives, specific methods and apparatus embodiments have been illustrated by way of example in the accompanying drawings and are described in detail herein. However, it should be understood that the specific methods and apparatus disclosed herein are not intended to limit this disclosure.

Claims

1. A method for using a robot to grasp containers in a diagnostic laboratory system, the method comprising: Capture one or more images of containers in a diagnostic laboratory system, wherein each captured image includes image data; Analyze the image data to identify one or more surface properties of the container; In response to the analysis, a gripping position on the container is determined for the gripper of the robot in the diagnostic laboratory system to grasp; and The gripper is used to grasp the container at the grasping position.

2. The method of claim 1, further comprising generating instructions to instruct the gripper to grip the container at the gripping position.

3. The method according to claim 1, wherein, The container is a sample container configured to hold biological samples.

4. The method according to claim 1, wherein, The container is configured to hold the chemicals consumed in the diagnostic laboratory system.

5. The method according to claim 1, wherein: The gripper has at least one degree of freedom; and Determining the gripping position includes determining the gripping position of the gripper in response to the at least one degree of freedom.

6. The method according to claim 1, wherein, The container has one or more geometric properties, and wherein determining the gripping position includes determining the gripping position of the gripper in response to the one or more geometric properties.

7. The method of claim 1, further comprising determining the material of the container, wherein, Determining the gripping position includes determining the gripping position of the gripper in response to determining the material of the container.

8. The method according to claim 1, wherein, The analysis includes identifying the location of a mark on the container, wherein determining the gripping location includes determining the gripping location of the gripper in response to determining the location of the mark.

9. The method according to claim 8, wherein, The grab position is not on the marker.

10. The method according to claim 1, wherein, The analysis includes identifying anomalies on the container, and wherein determining the gripping position includes determining the gripping position of the gripper in response to identifying the anomalies.

11. The method according to claim 10, wherein, The grabbing location is not on the abnormal object.

12. The method according to claim 10, wherein, The abnormal substance is a liquid.

13. The method according to claim 10, wherein, The anomaly is part of a label attached to the container.

14. The method according to claim 1, wherein, The analysis includes identifying the orientation of the container and adjusting the orientation of the gripper in response to identifying the orientation of the container.

15. The method according to claim 1, wherein, The analysis determines whether the container has a lid.

16. The method according to claim 1, wherein, The analysis of the image data or the determination of the capture location includes the use of a deep neural network.

17. The method according to claim 1, wherein, Determining the gripping position of the robot's gripper on the container includes determining the height at which the gripper grips the container.

18. A diagnostic laboratory system comprising: An imaging device configured to capture an image of a container; The robot includes a gripper configured to grasp the container and move the container within the diagnostic laboratory system; as well as Computer, the computer is configured to: Receive image data generated by the imaging device; Analyze the image data to identify one or more surface properties of the container; In response to the analysis, the gripping position of the gripper on the container is determined; and Generate instructions to instruct the gripper to grasp the container at the gripping position.

19. The diagnostic laboratory system of claim 18, wherein, The container is a sample container configured to hold biological samples.

20. A method for using a robot to grasp a sample container in a diagnostic laboratory system, the method comprising: Capture one or more images of a sample container in a diagnostic laboratory system, wherein each captured image includes image data, and wherein the sample container is configured to hold a biological sample; The image data is analyzed to identify one or more surface properties of the sample container, the one or more surface properties including at least one of markings, liquids, and abnormalities on the surface of the sample container; Determine a gripping position on the sample container for the gripper of a robot in the diagnostic laboratory system to grasp, the gripping position avoiding the one or more surface properties; and The sample container is grasped at the grasping position using the gripper.

Images