Method for generating a digital workspace model

Abstract

A method (M) for generating a digital workspace model (112) representing an available workspace (110) for a robot (102) in an industrial environment (100), the robot (102) having a robotic arm (106) with one or more arm segments movable in the available workspace (110) by one or more actuators, the method comprising: providing (S10) a digital arm model indicating a position of a segment representation point at the or each arm segment relative to an actuator position; providing (S12) a series of recorded actuator position sets; and generating (S18) the digital workspace model (112) by determining (S20) each position of each segment representation point based on the digital arm model and the recorded actuator position sets relative to a common coordinate system. The method generates a digital workspace model with improved model resolution and without the need for additional measurement equipment other than the robot itself.

Description

Technical Field

[0001] This invention relates to a method for generating a digital workspace model representing the available workspace of a robot in an industrial environment, the robot having a robotic arm with a single arm segment or tandemly connected arm segments, each arm segment being movable within the available workspace by one or more actuators. Furthermore, the invention also relates to a method for generating robot paths for a robot within the available workspace of a robot in an industrial environment, a method for controlling a robot within the available workspace of a robot in an industrial environment, a computer program product, a computerized device, and a robot for an industrial environment. Background Technology

[0002] The available workspace (where the robot's arm can move) is usually not known precisely and / or reliably before the robot and adjacent equipment (such as machine tools, guardrails, cables, signs, pipes, machine foundations, etc.) are installed.

[0003] One established way to determine available workspace is to deploy 3D sensors, such as laser scanners, in the area where the robot is supervised. However, this requires the installation, implementation, calibration, and maintenance of these 3D sensors.

[0004] Document US8,855,815B2 discloses a motion path search method using a six-axis vertical articulated robot as an example. The robot's posture is uniquely determined by a combination of angles from each of the six rotation axes. The operator grips the handle portion (which is part of the robot) and teaches the robot to move by directly moving the movable part of the robot. A robot shape database stores the robot shape data as closed regions using straight lines or curves. Based on the robot shape data and utilizing the combination of the six angles, the unique three-dimensional position of the movable part is calculated. Then, each voxel located inside the robot is determined as an accessible voxel.

[0005] However, determining the available workspace in this way has the following drawbacks: voxels need to be discretized before other methodological steps. That is, voxels are prone to being either too coarse to capture details or too fine to generate too much data and consume too much computation time.

[0006] Automatically deriving robot paths within a CAD model of the available workspace is known in the art. One example is employing a technique called “Rapid Exploration Random Tree,” or RRT, described in “Kuffner, James J. and LaValle, Steven M.; RRT-Connect: an efficient approach to single-querypath planning; Proceedings 2000 ICRA, Millenium conference, IEEE International Conference on Robotics and Automation; Symposia Proceedings; Cat. No.00CH37065; Vol. 2 IEEE, 2000.” Summary of the Invention

[0007] One object of this invention is to efficiently generate reliable workspace models with improved model resolution.

[0008] Therefore, a method suitable for and / or configured for generating a digital workspace model representing the available workspace of a robot in an industrial environment is proposed. The robot has a robotic arm. The robotic arm has one arm segment or two or more arm segments connected in series. Each arm segment is movable within the available workspace by one or more actuators. The method includes: providing a digital arm model indicating the position of each of one or more segment representation points located at that arm segment relative to actuator positions among the one or more actuators of that arm segment; providing a set of recorded actuator positions; and generating the digital workspace model by determining each position of each segment representation point based on the digital arm model and the recorded set of actuator positions relative to a common coordinate system.

[0009] This method generates a digital workspace model with representation points where the robotic arm has been recorded. Therefore, the proposed method's digital workspace model has improved model resolution.

[0010] In other words, the proposed approach is for an operator to control the movement of a robotic arm within an available workspace while simultaneously recording the actuator positions of that movement. Each recorded actuator position corresponds to a posture of the robotic arm within the available workspace, where no permanent obstacles such as cables, pipes, light signals, or sensors exist. Each arm segment is represented by one or more so-called (segment) representation points located in the local coordinate system of the corresponding segment. The set of recorded actuator positions is used to transform the local coordinate system to a common or global coordinate system. Thus, each representation point becomes a spatial trajectory of the representation point. All these representation points are then combined to form a three-dimensional point cloud (or, in less likely cases, a two-dimensional point cloud). This point cloud represents the known shape of the workspace, which can be used for robot path planning. For example, such a digital workspace model defined by multiple three-dimensional points can be used by methods known in toolchains that utilize data from 3D sensors, which are typically also multiple three-dimensional points. Examples of this toolchain approach are available in the "ros-planning / moveit" package, currently accessible at "https: / / github.com / ros-planning / moveit".

[0011] Regarding terms such as “operator,” it should be noted that individuals with male, female, or other gender identities are included in the term, regardless of the grammatical usage.

[0012] Robots with robotic arms typically have sensors indicating the current arm position for precise actuation and / or control. For example, one or more servo motors may be arranged to drive the robotic arm. Since this method is preferably performed after the robot and other equipment are installed, the resulting digital workspace model is reliable because measurement errors are prevented, as it is the same robot that detects and / or records a series of said arm positions, and it will be controlled / driven according to the robot path to be planned subsequently.

[0013] A digital arm model can be understood as a model of a robotic arm. For each arm segment, the digital arm model indicates the position of each of one or more segment representation points located on that arm segment relative to the actuator position of one or more actuators in that arm segment. This can be understood as: the digital arm model indicates the position of each of one or more segment representation points virtually located in and / or on the surface of that arm segment relative to each actuator position of one or more actuators in that arm segment.

[0014] Preferably, the series has a set of recorded simultaneous actuator positions. This simplifies the steps of determining the corresponding position of each representation point at each recorded actuator position.

[0015] Because the proposed method requires no additional measuring equipment or sensors other than a robot with a driveable robotic arm itself, it is cost-effective and easy to implement in the field. This advantage applies to the entire process of generating a digital workspace model and also to the fact that the (previously unknown) workspace does not need to be discretized into voxels.

[0016] Therefore, it can be said that the proposed method is an efficient and safe measurement method that reliably measures the workspace of a robot, which can be used to plan one or more robot paths for a robotic arm.

[0017] The digital workspace model is preferably used for subsequent robot path planning. It can also be delivered as documentation / proof of a successful robot installation. Therefore, the digital workspace model is its own original digital product because it can be used in many different use cases.

[0018] The available workspace can be understood as the space in which the robotic arm can move. The available workspace can be understood as the space where the robot can move. The available workspace can be understood as a physical and / or real space. The available workspace can be understood as a space accessible to the robot and free from other objects (especially permanent objects). The available workspace can be understood as such that moving the robot and / or the robotic arm to any available location or location within the available workspace will not result in the robot and / or the robotic arm colliding with any other permanent object. The available workspace can be understood as being defined by both: a) the robot's maximum range of motion, and b) any permanent structures within that maximum range.

[0019] A robot path can be referred to as a motion path, along which the robot can be or will be motiond. A robot path can be understood as a sequence of geometric and / or spatial positions, along which a robotic arm and / or robot and / or the distal end of a robotic arm is configured to follow the sequence to complete a specific task. Preferably, a robot path includes a sequence of spatial and / or geometric positions and orientations. Preferably, a robot path is a robot trajectory. A robot trajectory can be understood as a sequence of spatial and / or geometric positions, preferably including orientation information and velocity and / or acceleration information. A robot trajectory can be understood as a sequence of motions to be executed by a robotic arm and / or robot to complete a specific task. A robot trajectory can be understood as the result of a robot control program and / or programming of the robot.

[0020] The robotic arm can be understood as having a proximal end and a distal end. The proximal end can be understood as the supported end, wherein the robotic arm is supported by a robot base or the like. The distal end can be understood as the region of the robotic arm furthest from the proximal end. The distal end can be understood as the supporting end, wherein the robotic arm is configured to support tools and / or objects.

[0021] According to one option, the tool may be attached to or be able to be attached to a robotic arm, preferably to the distal end of the robotic arm. The digital arm model may also indicate the position of each of one or more tool representation points located at the tool (within and / or on the surface of the tool) relative to the actuator position. The digital workspace model may also be generated by determining each position of each tool representation point based on the digital arm model and a set of recorded actuator positions relative to a common coordinate system. This option increases the reliability of the digital workspace model for a particular tool when created from arm positions recorded when the tool is attached.

[0022] The tool may be attached to the distal end of the robotic arm and / or may be attached to the distal end of the robotic arm. The distal end of the robotic arm may be a distal arm segment of the robotic arm and / or the distal end of said distal arm segment. The distal end can be understood as the portion or segment of the robotic arm furthest from the robot base, preferably in at least one robotic arm position. The distal end may be an adapter and / or an interface configured to hold a manipulator (such as a tool). The distal end may be a manipulator permanently attached to the robotic arm. The distal end may be a manipulator detachably attached to the robotic arm.

[0023] This option offers a second benefit: the ability to select tools that occupy space during recording in order to reduce the number of arm positions (the size of the set of actuator positions being recorded) required to cover the entire available workspace, thereby reducing the time required to record these arm positions. In other words, specialized tools can be provided for recording, and / or larger tools from the existing tooling can be selectively chosen. Preferably, tools of appropriate size and / or suitable for covering a large workspace can be selected. Therefore, this option of knowing and adding tool space, combined with the option of automatically adding internal clearance described below, is particularly efficient.

[0024] According to one option, the recorded series of arm positions can be provided as a digital dataset, preferably in the form of a digital file. This option facilitates data exchange, allowing the proposed method to be implemented internally at the robot manufacturer, thereby causing less field disruption in the robot's industrial environment.

[0025] According to one option, the method may further include: recording the series of actuator positions within the available workspace using the robot's actuator position sensors while the robot arm is controlled by an operator. Preferably, one or more actuators of the robot arm are controlled. Preferably, the operator is a human operator. In this case, the proposed method can be performed on-site with direct access to the robot's workspace. For example, completing the entire process (from recording actuator / arm positions to generating a digital workspace model) at once may be more efficient. Preferably, the one or more actuator position sensors are arranged to control the one or more actuators to control the movement of the robot arm. For example, the actuators may be servo motors, each with a sensor built into the actuator. Thus, position signals are read from the same position sensors arranged to control the robot's movement, and therefore, high-precision (relative to the purpose of a digital workspace model) arm positions can be recorded.

[0026] The proposed method may further include generating a volume-based digital workspace model by adding a digital representation of a spherical space based on the representation point and a preset radius for each segment representation point (preferably also for each tool representation point). The volume-based model may be required for volume-based post-processing steps. In other words, it is proposed to assume that each sphere or ball is centered at each representation point. The preset radius can be chosen to be less than 10 centimeters, for example, 1 centimeter, so as to be less than the distance that an operator typically chooses when navigating obstacles.

[0027] According to one option, the method may further include: detecting internal gaps between representation points in the generated digital workspace model. An internal gap can be a gap larger than a preset distance value and surrounded by representation points, each of which is narrower than the preset distance value in distance from at least one other representation point. Then, preferably, one or more representation points narrower than the preset distance value are added to and / or included in and / or superimposed on the internal gaps in the digital workspace model. The internal gap can be understood as digital residual space. In other words, an internal gap can be described as a void within the digital workspace model. Such internal gaps may be caused by recorded arm positions (where arm position spaces do not overlap). An alternative approach to these internal gaps would be to only record actuator positions with overlapping robotic arm spaces, which is tedious and therefore expensive. Therefore, this option provides for the post-recording and therefore efficient removal of internal gaps.

[0028] According to one option, the method may further include: detecting peripheral gaps between representation points in the generated digital workspace model. A peripheral gap can be a gap greater than a preset distance value and only partially surrounded by representation points whose distances to each other are narrower than the preset distance value. That is, a peripheral gap can be a recess or intrusion into the point cloud from the point cloud "edge". Furthermore, the proposed method may include: visually presenting a digital representation of the space defined by the peripheral gap to one / the operator. Preferably, the space is defined by and located within the peripheral gap. Preferably, the space is visually presented to the human operator. The peripheral gap can be described as a peripheral gap of the workspace described by the digital workspace model. This peripheral gap may be caused by recorded arm positions (where arm position spaces do not overlap at the edges of the available workspace). The proposed option has the advantage of combining recording and presentation: the operator can optimize the recording of actuator position sets. Therefore, recording can be completed very quickly.

[0029] According to one option, the method may further include: providing a first pattern for detecting peripheral gaps, wherein the first pattern defines a minimum ratio of the gap depth to the gap opening diameter of the same peripheral gap. The minimum ratio is preferably at least 1, and more preferably at least 2. The proposed method may further include: detecting peripheral gaps based on the provided second pattern. The gap opening diameter can be understood as the maximum extension tangent to the outer edge of the digital workspace model, for example, the distance between two representation points in which no representation point lies. The gap depth of the peripheral gap can be understood as the length of the peripheral gap pointing into the point cloud. The preferred minimum ratio defines a peripheral gap at least as deep as its width, making it easier (if necessary) to determine the remaining peripheral gaps. In other words, the thinner the peripheral gap, the more likely it is to be a remaining peripheral gap. Therefore, the operator is provided with means for quickly locating peripheral gaps to improve the digital workspace model.

[0030] According to one option, the method may further include: providing one or more second patterns for detecting peripheral gaps, wherein each second pattern defines a range of ratios of the gap opening area to the volume (or space) defined by the peripheral gap of the digital workspace model. Preferably, the ratio range is a maximum ratio of at most 7 (i.e., a range of "7 or less"), and more preferably at most 6. The gap opening area can be understood as the area of ​​the gap opening within the peripheral surface of the digital workspace model (point cloud). The peripheral gap is then detected based on the provided second pattern. This option is also used for rapidly locating peripheral gaps. For example, a conical peripheral gap with a base diameter of 1 and a height of 1 has a volume of approximately 0.26 and a side surface area (of the assumed cone) of approximately 1.76, so the volume-to-surface ratio is approximately 6.77. If the conical peripheral gap has a height of 2, the ratio drops to approximately 6.15. Therefore, if desired, preferred parameters include a peripheral gap at least as deep as its width.

[0031] The aforementioned options may also include: adding and / or including representation points in the digital workspace model that are narrower than the preset distance value to the peripheral gaps, according to operator instructions. That is, the operator is not only informed of the corresponding peripheral gap, but can also request the removal of the recess or the addition of space to the digital workspace model, which in many cases may be faster than moving the robotic arm and / or can be done during post-processing.

[0032] According to one aspect of the invention, a method is proposed for generating robot paths for a robot within an available workspace in an industrial environment. The proposed method includes the steps of the method described above for generating a digital workspace model representing the available workspace of a robot in an industrial environment; and generating a robot path within the space represented by the representation points of the generated digital workspace model. This option provides a complete process chain from an unexplored workspace to a completed robot path. The robot path can be provided in the form of a digital dataset (such as a file). The robot path can be understood as a trajectory and / or a motion path. This method can be referred to as a method for generating robot control data, and the robot path can be referred to as robot control data.

[0033] According to another aspect of the present invention, a method for controlling a robot within an available workspace in an industrial environment is proposed. The proposed method includes the steps of: generating a robot path for the robot within the available workspace of the robot in an industrial environment as described above; and driving the robot based on the generated robot path.

[0034] According to another aspect of the invention, a computer program product is proposed, comprising program code for executing the proposed method for generating a digital workspace model when run on at least one computerized device.

[0035] Computer program products (such as computer program devices) can be implemented as memory cards, USB sticks, CD-ROMs, DVDs, or files that can be downloaded from a server on a network. For example, such files can be provided by transmitting files including said computer program products from a wireless communication network.

[0036] According to another aspect of the invention, a computerized device is provided, preferably an industrial controller and / or an industrial personal computer, configured to execute the method for generating a digital workspace model and / or the computer program product described above. The industrial controller is preferably a programmable logic controller.

[0037] According to another aspect of the invention, a robot for industrial environments is proposed. The robot has a robotic arm. The robotic arm has a single arm segment or arm segments connected in series. Each arm segment is movable within an available workspace by one or more actuators. The robot also has the proposed computerized equipment.

[0038] Further possible embodiments or alternatives to the invention also include combinations of features described above or below with reference to the embodiments (these combinations are not explicitly mentioned herein). Those skilled in the art can also add individual or isolated aspects and features to the most basic form of the invention. Attached Figure Description

[0039] Further embodiments, features, and advantages of the invention will become apparent from the following description taken in conjunction with the accompanying drawings and dependent claims, in which: Figure 1 A schematic overlay of the available workspace of a robot in an industrial environment and a digital workspace model representing that available workspace is shown; and Figure 2 A schematic diagram illustrating the process flow of the proposed method for generating a digital workspace model according to an embodiment of the present invention is shown.

[0040] In the accompanying drawings, unless otherwise indicated, the same reference numerals denote the same or functionally equivalent elements. Detailed Implementation

[0041] Figure 1An industrial environment 100 is schematically illustrated. A robot 102 is arranged therein, which, for example, has a base 104 and a robotic arm 106. In the depicted example, for illustrative purposes, the robotic arm 106 has three arm segments 106a to 106c, which may also be referred to as arm sections.

[0042] Arm segments 106a to 106c are connected in series. That is, the base 104 supports a proximal arm segment 106a, which supports another arm segment 106b, which in turn supports a distal arm segment 106c.

[0043] The robotic arm 106 is configured to perform manipulation actions, which can be any interaction between the robot 102 and an object (not shown) in the industrial environment 100.

[0044] The robotic arm 106 can be configured to carry a tool 108. The tool 108 can be detachable from the robotic arm 106. The tool 108 can be one of many tools adaptable to the distal arm segment 106c of the robotic arm 106. Figure 1 In some cases, tool 108 is wider than robotic arm 106.

[0045] The robotic arm 106 and the tools 108 attached thereto can move within the available workspace 110. It can be said that one or more degrees of freedom of the robotic arm 106 and one or more possible tools 108 that can be attached thereto define the maximum range or size of the available workspace 110. Furthermore, any permanent objects in the industrial environment 100, such as protective fences or walls, may restrict the available workspace 110, even if this is within... Figure 1 Not described in the text.

[0046] For example: Robot 102 also has four actuators (not in...) Figure 1 (As shown in the diagram), each actuator is a servo motor with an integrated actuator position sensor. A fourth actuator is disposed within the distal arm segment 106c to pivot (or rotate) the tool 108 relative to the distal arm segment 106c. A third actuator is disposed within the second arm segment 106b to pivot the distal arm segment 106c together with the tool 108 relative to the second arm segment 106b. A second actuator is disposed within the first arm segment 106a to pivot the second arm segment 106b together with the distal arm segment 106c and the tool 108 relative to the first arm segment 106a. A first actuator is disposed in the base 104 to pivot the entire robotic arm 106 together with the tool 108 relative to the base 104. Each actuator position sensor reads the angular actuator position; however, the invention is not limited to rotational motion. All actuator positions together uniquely identify the attitude of the robotic arm 106.

[0047] When measurement method M (described below) is executed to generate a digital workspace model 112 of the available workspace 110, a digital workspace model 112 is created. The digital workspace model 112 represents a portion or all of the available workspace 110, either directly or by means of multiple representation points. Figure 1 In the model, the available workspace 110 has a cylindrical shape, while the digital workspace model 112 is smaller and lacks some parts of the available workspace 110.

[0048] Furthermore, the digital workspace model 112 has an internal gap 114 located within the digital workspace model 112. It also has an external gap 116 extending into the digital workspace model 112. For illustrative purposes, these two gaps 114 and 116 are depicted.

[0049] In step S10, a digital arm model of the robotic arm 106 is provided.

[0050] The position of the first arm segment 106a (also known as the proximal arm segment 106a) can be given in one or more angles in the local coordinate system of the base 104, the position of the second arm segment 106b can be given in one or more angles in the local coordinate system of the first arm segment 106a, and the position of the third arm segment 106c (also known as the distal arm segment 106c) can be given in one or more angles in the local coordinate system of the second arm segment 106b. These positions of arms 106a-c can generally be referred to as arm positions. The arm positions can be considered as a vector of all (relative) arm positions of the arm portion / segment.

[0051] The digital arm model of the embodiment indicates the relative positions of multiple segment representation points (each segment representation point represents the surface of each arm segment 106a-c) and the relative positions of tool representation points (each tool representation point represents the surface of tool 108). For example, a portion of the segment representation point describes the surface of the first arm segment 106a in the local coordinate system of the first arm segment 106a, another portion describes the surface of the second arm segment 106b in the local coordinate system of the second arm segment 106b, and yet another portion describes the surface of the third arm segment 106c in the local coordinate system of the third arm segment 106c. The tool representation point describes the surface of tool 108 in the local coordinate system of tool 108. The representation points are virtual points. The representation points can be arranged such that there are no (virtual) gaps on the surface of the robotic arm 106 wider than a preset parameter (e.g., 5 cm).

[0052] For any given set of simultaneous actuator positions, the relative position and orientation of the local coordinate system are known. Therefore, for any given set of simultaneous actuator positions, the location of the surface of the robot arm 106 is known.

[0053] In step S12, a series of recorded simultaneous actuator position groups are provided. That is, for each timing (e.g., every 100ms), the angular position of each of the four actuators is provided, for example, within the dataset.

[0054] Step S12 may include step S14, in which the set of actuator positions is recorded. That is, during method M, the operator controls one or more actuators (in this case, four actuators) that control the arm posture of the robotic arm 102.

[0055] Step S12 may include step S16, wherein the recorded set of actuator positions is provided in the form of a digital file, for example from a data storage device, network interface, etc. That is, method M is provided with a pre-recorded set of actuator positions.

[0056] Then, in step S18, a digital workspace model 112 is generated. The digital workspace model 112 describes / indicates the space occupied by the robotic arm 102 (to date) during the recording of actuator position sets.

[0057] Step S18 comprises step S20, wherein at each actuator position group, each position of each segment representation point is determined in a common coordinate system. That is, for each group of (simultaneous) actuator positions, the coordinates of each segment representation point in a single global coordinate system are calculated based on the provided digital arm model and a series of recorded actuator position groups. Thus, a three-dimensional point cloud is generated, which is the digital workspace model 112. The digital workspace model 112 can therefore be understood as a digital description of all the space occupied by the main body / multiple bodies of the robotic arm 102 under any arm posture given in S12 to S16.

[0058] In optional step S22, at each actuator position group, each position of each tool representation point is also determined in a common coordinate system. Step S22 determines the tool representation points in the same way that step S20 determines the segment representation points. Therefore, step S22 generates a more detailed point cloud compared to the point cloud after step S20. Furthermore, reflecting the tool in the digital arm model, in the actuator position groups, and in the digital workspace model 112 makes it easier to determine the robot path for the (realistic) case where the tool 108 is attached to the distal arm segment 108.

[0059] The digital workspace model 112 is preferably generated in the form of a digital dataset.

[0060] Since the digital workspace model 112 indicates a space in which the robotic arm 106 has moved during recording, this space can have a complex and / or jagged shape. However, in most applications, a simple and / or smooth shape is desired. In step S24, any internal gaps 114 in the digital workspace model 112 are detected. An internal gap 114 is a part of the digital workspace model 112, specifically representing gaps 114 between points that are larger than a preset distance value and are surrounded by points whose distances from each other are narrower than the preset distance value.

[0061] If one or more internal gaps 114 are detected, steps S26 to S30 can be performed for each of these internal gaps 114. It should be noted that there are two preferred methods for handling the internal gaps 114.

[0062] The first approach is to ask the operator (in steps S26 and S28) whether they want to add additional representation points (in step S30) to the corresponding internal gap 114 in the digital workspace model 112. In this case, in step S26, before the user request (e.g., a request to add or not add representation points to the internal gap 114) is read in step S28, the space or volume representing the internal gap 114 is visually presented to the operator. Then, in step S30, the internal gap 114 can be filled by adding the added representation points to the digital workspace model 112.

[0063] In most cases, the internal gap 114 is not caused by the presence of a real object at that location. Therefore, the second approach is to skip steps S26 and S28, and then automatically add representation points to any internal gap 114 in the digital workspace model 112 in step S30, without performing steps S26 and S28.

[0064] The steps S24 (automatic detection of internal gaps) and S30 (automatic filling of internal gaps) enable the recording of a series of actuator positions in a very rapid manner. Specifically, a longer tool 108 extending in at least one direction can be selected for recording, and then this longer tool 108 can be moved (e.g., rotated) to create an internal gap 114. In other words, the robotic arm does not move in a manner that causes the covered spaces to overlap, but rather in a manner that causes only the space covered by the tip of the tool 108 to overlap. Since the space does not need to be covered by the robotic arm 106, the recording steps can be much faster. Therefore, including steps S24 and S30 in method M is advantageous.

[0065] Next, steps S32 to S46 are presented, which are used to process the peripheral gap 116. The peripheral gap 116 is a gap between points that is greater than a preset distance value, said gap being only partially surrounded by points whose distances to each other are narrower than said preset distance value. That is, the peripheral gap 116 differs from the internal gap 114 in that the peripheral gap 116 has an opening toward the outside of the digital workspace model 112. This further means that the peripheral gap 116 can also be caused by real objects protruding into the usable workspace 110 (e.g., signal lights hanging from the ceiling), or simply by inaccuracies during the recording of the series of actuator position groups.

[0066] In step S32, a first mode for detecting the peripheral gap 116 is provided. The first mode defines a minimum ratio of the gap depth of the peripheral gap 116 to the gap opening diameter of the same peripheral gap 116, wherein the minimum ratio is preferably, for example, 1.

[0067] In step S34, a second mode for detecting the peripheral gap 116 is provided. The second mode defines a range of ratios between the gap opening area of ​​the recess in the digital workspace model and the space defined by the recess. For example, the second mode may provide that any peripheral gap 116 having a (side) surface area to volume ratio (the ratio of the side surface area of ​​the peripheral gap to the volume enclosed by the peripheral gap) equal to or less than 7 is identified as the peripheral gap 116 to be presented to the operator.

[0068] Then, in step S36, any peripheral gaps 116 in the digital workspace model 112 are detected. In sub-step S38, detection is performed based on the first mode. Figure 1 The exemplary peripheral gap 116. In addition, in step S40, the exemplary peripheral gap 116 is also detected based on the second mode.

[0069] In step S42, for any peripheral gap 116 detected in S36, the space represented by that peripheral gap 116 is visually presented to the operator. Then, in step S44, a user request can be read. The user request can be a user's choice of a prompt that provides to "fill" the space of the peripheral gap, i.e., add representation points to the detected peripheral gap 116 in the digital workspace model 112. The user request can also be a user's choice of a prompt that provides to ignore using the peripheral gap 116 for creation in the digital workspace model 112. Then, in step S46, based on the read user request, additional representation points can be added to the peripheral gap in the digital workspace model, each of the added representation points being narrower than the preset distance value from at least one other representation point.

[0070] Therefore, the proposed method M efficiently obtains a digital workspace model 112 that reliably represents the available workspace 110 of the robot 102 in the industrial environment 100.

[0071] However, in step S48, method M can continue by generating a robot path within the space represented by the representation points of the digital workspace model 112. Then, in step S50, the robot 102 can be controlled based on the generated robot path. That is, preferably, one or more servo motors of the robot 102 can be controlled to drive the robot 102, causing the robotic arm 106 to move along the robot path.

[0072] In other words, method M can be extended to method M for controlling robot 102 in available workspace 110 of industrial environment 100 through step S48 or steps S48, S50.

[0073] Furthermore, step S48 or these steps S48, S50 can be performed separately after the separately generated digital workspace model 112 has been provided.

[0074] The above method M can be executed by the computerized control device of robot 102.

[0075] Although the invention has been described with reference to preferred embodiments, it will be apparent to those skilled in the art that modifications can be made in all embodiments.

[0076] List of reference numerals 100 Industrial Environment 102 robots 104 bases 106 robotic arms 106a Proximal / First Arm Segment 106b Second Arm Segment 106c distal / third arm segment 108 tools 110 available workspaces 112 Digital Workspace Model 114 Internal clearance 116 Peripheral gap M is a measurement method used to generate digital workspace models. S10 provides a digital arm model of the robotic arm. S12 provides a series of recorded actuator position groups. S14 Record the series of actuator position groups S16 provides a series of recorded actuator position sets as a digital file. S18 Generates a Digital Workspace Model S20 determines the position of each segment representation point based on the digital arm model and the recorded actuator position set. S22 determines each position of each tool representation point based on the digital arm model and the recorded actuator position set. S24 Detects internal gaps in a digital workspace model S26 visual presentation of space defined by internal gaps S28 Read user request S30 adds representation points to the internal gaps in the digital workspace model. S32 provides a first mode for detecting peripheral clearances. S34 provides a second mode for detecting peripheral gaps. S36 Detects peripheral gaps in a digital workspace model S38 detects peripheral gaps based on the first mode. S40 detects peripheral gaps based on the second mode. S42 visual presentation of space defined by peripheral gaps S44 Read User Request S46 Adds the representation point to the outer gap in the digital workspace model. S48 Generates robot paths in the digital workspace model. S50 driven robot.

Claims

1. A method (M) for generating a digital workspace model (112), the digital workspace model representing an available workspace (110) of a robot (102) in an industrial environment (100), the robot (102) having a robotic arm (106) having a single arm segment (106a-c) or cascaded arm segments (106a-c), each arm segment (106a-c) being movable within the available workspace (110) by one or more actuators, the method comprising: Provide (S10) a digital arm model, the digital arm model indicating the position of each of one or more segment representation points located at the arm segment (106a-c) relative to the actuator position of the one or more actuators of the arm segment for each arm segment (106a-c); Provide (S12) a series of recorded actuator positions; as well as The digital workspace model (112) is generated (S18) by determining (S20) each position of each segment representation point based on the digital arm model and the recorded actuator position set relative to the common coordinate system.

2. The method according to claim 1, wherein, The tool (108) is attached to or can be attached to the robotic arm (106). The digital arm model further indicates the position of each of one or more tool representation points located at the tool (108) relative to the actuator position, and The digital workspace model is generated (112) by determining (S22) each position of each tool representation point based on the digital arm model and the recorded actuator position set relative to the common coordinate system.

3. The method according to claim 1 or 2, wherein, A series of recorded actuator position groups are provided as a digital file (S16).

4. The method according to claim 1 or 2, further comprising: While the robotic arm (106) is controlled by the operator, the actuator position sensors of the robot (102) record (S14) the series of actuator positions within the available workspace (110).

5. The method according to any one of the preceding claims further comprises: A volume-based digital workspace model is generated by adding a digital representation of a spherical space based on the representation point and a preset radius for each segment representation point, and preferably also for each tool representation point.

6. The method according to any one of the preceding claims further comprises: In the generated digital workspace model (112), an internal gap (114) between representation points is detected (S24), the internal gap (114) being greater than a preset distance value and the internal gap (114) being surrounded by representation points, each of the representation points being narrower than the preset distance value in distance from at least one other representation point, and representation points are added (S30) to the internal gap (114) in the digital workspace model (112), the distance between the representation points being narrower than the preset distance value.

7. The method according to any one of the preceding claims further comprises: In the generated digital workspace model (112), the peripheral gap (116) between the representation points is detected (S36), the peripheral gap (116) is greater than a preset distance value, and the peripheral gap (116) is only partially surrounded by representation points whose distance to each other is narrower than the preset distance value, and a digital representation of the space defined by the peripheral gap (116) is visually presented to the operator (S42).

8. The method according to claim 7, further comprising: A first pattern for detecting the peripheral gap (116) is provided (S32), wherein the first pattern defines a minimum ratio of gap depth to gap opening diameter, and wherein the peripheral gap (116) is detected based on the provided first pattern.

9. The method according to claim 7 or 8, further comprising: Provide (S34) a second mode for detecting the peripheral gap (116), wherein the second mode defines a range of ratios of the gap opening area of ​​the peripheral gap (116) to the volume defined by the peripheral gap (116), and wherein the peripheral gap (116) is detected based on the provided second mode.

10. The method according to any one of claims 7 to 9, further comprising: According to the operator's instructions (S44), add (S46) representation points to the peripheral gap (116) in the digital workspace model (112), wherein the distance between the added representation points is narrower than the preset distance value.

11. A method for generating a robot path for a robot (102) within an available workspace (110) of a robot (102) in an industrial environment (100), the method comprising: The steps of the method (M) according to any one of claims 1 to 10; as well as A robot path (S48) is generated within the space represented by the points of the generated digital workspace model (112).

12. A method for controlling a robot (102) within an available workspace (110) of the robot (102) in an industrial environment (100), the method comprising: The steps of the method according to claim 11; as well as The robot (102) is controlled based on the generated robot path (S50).

13. A computer program product comprising program code for performing the method according to any one of claims 1 to 12 when run on at least one computerized device.

14. A computerized device, preferably an industrial controller and / or an industrial personal computer, configured to perform the method according to any one of claims 1 to 12 and / or the computer program product according to claim 13.

15. A robot (102) for an industrial environment (100), the robot (102) having a robotic arm (106) having a single arm segment (106a-c) or a series of connected arm segments (106a-c), each arm segment (106a-c) being movable in an available workspace (110) by one or more actuators, and the robot (102) having a computerized device according to claim 14.

Images