DK202100195A1 - A relocatable robotic system for production facilities - Google Patents

Abstract

This invention relates to a relocatable robotic system and a corresponding method for the relief of manual production processes. More specifically the invention provides a robotic system designed for quick and easy conversion in a production facility with different and varying work tasks. In another aspect, the invention relates to a method for the relief of a wide range of manual work processes typically undertaken in production facilities. The method of the invention is for automating and handling a manually demanding work task using a relocatable robotic workstation. The method of the invention provides a cobot-assisted method for relief of relatively small tasks involving unilaterally repetitive work.

Description

DK 2021 00195 A1 1

A RELOCATABLE ROBOTIC SYSTEM FOR PRODUCTION FACILITIES

TECHNICAL FIELD This invention relates to a relocatable robotic system and a corresponding method for the relief of manual production processes. More specifically the invention provides a robotic system designed for quick and easy conversion in a production facility with different and varying work tasks. In another aspect, the invention relates to a method for the relief of a wide range of manual work processes typically undertaken in production facilities.

The method of the invention is for automating and handling a manually demanding work task using a relocatable robotic workstation. The method of the invention provides a cobot-assisted method for relief of relatively small tasks involving unilaterally repetitive work.

BACKGROUND ART In the slaughterhouse production, significant efforts have been made over the past many years to automate labour-intensive processes, but a considerable number of work tasks are still carried out manually, often because these tasks have so far been considered "too small" or insignificant for them to be automated.

In recent years, however, the development of small robots, intended for interaction with humans, have gained momentum, and are becoming relatively inexpensive. In contrast to traditional industrial robots, which are designed to work autonomously with safety assured by isolation from human contact, collaborative robots, also referred to as service robots, “cobots” or "co-robots”, are intended to interact with humans in a shared space, or to work safely in close proximity.

Currently, small autonomous (mobile) robots have found their way into various industrial processes, e.g., for pick and place processes, see e.g., WO2017148939 and WO2018063100, and for checking and counting stocks in warehouses, see WO2019219659. US2017269607 also describes the use of such robots in workplaces, e.g., in warehouse environments or order fulfilment environments. Also, W02019212942 discloses an autonomous robot with on demand teleoperation, i.e., the robot may be remotely operated by the operator.

The use of cobots has been suggested for e.g., the packaging industry, where they have been used to fold cardboard boxes, see e.g., WO2019137907. And WO2019158349 suggests the use of a cobot for carrying out various machining steps on a workpiece at a workstation.

DK 2021 00195 A1 2 Image capturing technologies have been suggested for navigating autonomous robots, see e.g., WO20120091804, WO2012130734, WO2018140690, WO2019191029, WO2019191592, US20190291275, and EP3567695. These systems are mobile in the sense of being self-propelled or self-navigating, in contrast to being relocatable.

However, a relocatable system as the one described according to the present invention has never been disclosed.

SUMMARY OF THE INVENTION It is an object of the present invention to provide automated solutions for use in production facilities, in particular for avoiding repetitive movements, and for increasing profitability and efficiency, by enabling automated tasks for processes that are only undertaken periodically.

The system of the invention aims at relieving unilateral repetitive work, that may lead to muscle and skeletal disorders, and may remove the bottlenecks that typically arise in consequence of the frequent job-rotations associated with the working environment in production facilities, or because of tasks that requires extraordinary staffing. Even relatively simple tasks may happen to strain the work environment.

The present invention provides a relocatable, flexible, robotic installation, that, with the help of adaptive software and a mechanically locked configuration, can solve many ordinary tasks frequently occurring in a production environment, and that can do so in a flexible manner, allowing a human operator to also use the workplace essentially without any modifications. The system of the present invention also focuses on the user interface. It is an important feature of the invention that the system can be handled by an operator, without substantial training or background knowledge. At the same time, and out of consideration of adaptability, the need for setup and programming is reduced to a minimum.

In effect, the system and related method according to the present invention improve working environment and continued efficiency with a special focus on tasks that so far have not been profitable or otherwise meaningful to automate.

Industrial processes are often characterized by a need for addressing a large variation in the number of different product types, while at the same time accompanied by a small variation within the individual product types. In contrast, biological variation plays a major role in meat processing. In fact, the slaughterhouse industry may be characterised by addressing a large individual variation within a relatively small number of product types. Not only may two pieces of meat of the same type, e.g. two tenderloins, have different appearances, but each object can also be deformed to a pronounced degree, e.g. when placed close together and possibly overlapping each other. Furthermore, the

DK 2021 00195 A1 3 occasional presence of fibrous or granulated fat, muscle, hide or bone on the meat products requires a robust vision system with a flexibility not attainable with conventional systems.

In later years, vision systems relying on deep learning techniques have emerged in many industries. This approach has clear advantages in situations with large variations because feature engineering is part of the optimization procedure of the algorithm. I.e. the features are learned by the algorithm instead of being engineered by the developer.

While deep learning techniques are successfully deployed in other industries, the meat industry is still lacking behind due to the extremely large variations within the same product types and the strict requirements on precision and capacity.

Therefore, it is an object of the present invention to provide such a system capable of handling the large product variation in the meat processing industry. Algorithm training of the imaging system enables the platform to distinguish the diverse geometric and surface-varied structures that are always present in the slaughter industry.

Moreover, although the system of the invention is relocatable, it is not mobile in the sense of being self-propelled or self-navigating, and unlike known robotic systems, e.g. those described above, the system according to the present invention does not use vision means for navigating the robotic system, but only for carrying out and accomplishing object analysis of the occurring objects. In fact, self-propelling or self- navigating systems are not compatible with the current infrastructure, workflow or product flow found in the typical slaughterhouse.

Therefore, in its first aspect, the invention provides a relocatable robotic system, which system is characterised as described in more details below.

In another aspect the invention provides a method for automating a manually demanding work task, which method is described in more details below.

Other objects of the invention will be apparent to the person skilled in the art from reading the following detailed description and accompanying drawings.

Any combination of two or more of the embodiments described herein is considered within the scope of the present invention.

DETAILED DISCLOSURE OF THE INVENTION The present invention provides a relocatable robotic system and a corresponding method for the relief of various manual slaughterhouse processes undertaken in production facilities.

DK 2021 00195 A1 4 The system of the invention In its first aspect, the invention relates to a relocatable robotic system (1), for use within a production facility, in particular in a slaughterhouse, for quick and effortless replacement of a human operator, without specific needs for rearranging the workplace, which system is configured for recognition, localization, determination, inspection, and/or analysis, and subsequent processing of meat items (16) introduced to the system.

The relocatable robotic system (1) of the invention may be characterised by comprising the following elements: a. a relocatable robotic workstation (2), configured to accommodate/assemble and support the necessary elements, including a means (4A) for supporting an imaging device (4), and a fixture/fitting means (7) for connecting the relocatable robotic workstation (2) to one of one or more docking means (8) located within/inside the production facility, and which relocatable robotic workstation (2) further comprises: al. a robot (3), in cooperation with an imaging device (4) and a processing means (5), which robot is configured for holding a working tool (12) suited for the intended task, and for manipulating the meat items (16) introduced to the system, and which robot is mounted on the relocatable robotic workstation (2); a2. an imaging device (4), firmly mounted on the relocatable robotic workstation (2), and configured for recording three-dimensional (3D) images of the product surface of the meat item (16) introduced to the system, and, in cooperation with a processing means (5), configured for sensing information relevant for planning, computation and safe execution of the robotic trajectories involved in a given task; a3. a processing means (5), in cooperation with the robot (3) and the imaging device (4), and configured for applying a machine learning application/algorithm on data obtained from the imaging device (4), and configured for executing one or more processing/working programs; a4. a user interface (6), configured for presenting to the operator only the available situational actions, applicable to the particular workplace and the particular circumstances/product type, and configured for receiving situational inputs from the operator; and ab. a fixture /fitting means (7), configured for firmly connecting the relocatable robotic workstation (2) to one of the one or more docking means (8) located inside the production facility; and b. one or more docking means (8), each docking means located inside the production facility and configured for quick and easy clamping of the relocatable robotic workstation (2), configured for general or task-specific supplies, and configured for acting as an information port/information provider.

DK 2021 00195 A1 Optionally, the robotic system (1) of the invention further comprises a central server/database (9) for general communication and exchange of data processes, and for feeding the system with process- or product-specific information, and may, e.g., allow for updates in respect of reporting of process status, new tasks to be accomplished, 5 automatic error messaging, and all-clear messaging after completed tasks, as well as for feeding the system with process- or product-specific information.

Moreover, the robotic system (1) of the invention may further comprise lighting means (10), i.e., one or more light sources capable of emitting light adjusted to the frequencies that the imaging device employed according to the invention is able to use.

Also, in a further embodiment, the relocatable robotic workstation (2) of the invention comprises a means (4A) for fixation/supporting/anchoring the imaging device (4).

The robotic workstation In the context of this invention, a relocatable robotic workstation (2) represents a relocatable instrument that can easily be moved (manually or powered) to a new location within the production facility.

As defined herein, the system of the invention is movable in the sense of being relocatable, i.e., capable of being moved from one workplace to another, e.g., being pushed or moved manually or power-operated, on wheels, while not being mobile in the sense of being self-propelled or self-navigating.

In one embodiment, the relocatable robotic workstation (2) of the invention may be characterised by comprising the following elements: a. a manually guided, relocatable robotic workstation (2), configured to accommodate/assemble and support the necessary elements, including a means (4A) for firmly supporting an imaging device (4), and a fixture/fitting means (7) for firmly connecting the relocatable robotic workstation (2) to one of one or more docking means (8) located within/inside the production facility, and which relocatable robotic workstation (2) further comprises: al. a robot (3), in cooperation with an imaging device (4) and a processing means (5), which robot is configured for holding a working tool (12) suited for the intended task, as calculated/proposed/selected/instructed by the processing means (5), and for manipulating the meat items (16) introduced to the system, and which robot is firmly mounted on the relocatable robotic workstation (2); a2. an imaging device (4), firmly mounted on the relocatable robotic workstation (2), and configured for recording three-dimensional (3D) images of the product surface of the product (16) introduced to the system, and, in cooperation with a processing means (5), configured for sensing information relevant for planning,

DK 2021 00195 A1 6 computation and safe execution of the robotic trajectories involved in a given, projected task; a3. a processing means (5), in cooperation with the robot (3) and the imaging device (4), and configured for applying a machine learning application/algorithm on data obtained from the imaging device (4), and configured for executing one or more processing/working programs; a4. a user interface (6), configured for presenting to the operator only the available situational actions, applicable to the particular workplace and the particular circumstances/product type, and configured for receiving situational inputs from the operator; and ab. a fixture /fitting means (7) configured for firmly connecting the relocatable robotic workstation (2) to one of the one or more docking means (8) located inside the production facility.

It represents an inherent challenge for these removable/relocatable workstations, that the imaging system is fixed in a position relative to the performing robot, so that any oscillations and instability are reduced in relation to the workplace. In this respect it represents a real practical advantage of the system of the invention, that the robot (3) and the imaging device (4) shares the same basis (2).

Therefore, in another embodiment, the relocatable robotic workstation (2) of the invention further comprises means (4A) for fixation/supporting/anchoring the imaging device (4). Such fixation/supporting means may be any suitable fitting, fixture, stand or structure for effectively fastening of the one or more imaging devices (4) for use according to the invention.

Also, in a further embodiment, the relocatable robotic workstation (2) of the invention further comprises means for obtaining wireless connection (WiFi) to a central server/database (9).

Connection to a central server/database (9) may be used for general communication and exchange of data processes, and for feeding the system with process- or product-specific information, and may, e.g., allow for updates in respect of reporting of process status, new tasks to be accomplished, automatic error messaging, and all-clear messaging after completed tasks, as well as for feeding the system with process- or product-specific information.

The robot The system of the invention may comprise one or more robots (3), each robot in cooperation with the imaging device (4) and the processing means (5), and each of which robots shall be configured for holding a working tool (12) suited for the intended task and for manipulating the meat items (16) introduced to the system.

DK 2021 00195 A1 7 As the robot for use according to the present invention is intended to interact directly and safely with an operator in a shared space or in close proximity, the robot may in particular be a collaborative robot, also referred to as service robots, "cobots” or "co- robots”.

The International Federation of Robotics (IFR), a global industry association of robot manufacturers and national robot associations, have defined four types of collaborative manufacturing applications: e Co-existence, where the operator and the robot work alongside each other, but with no shared workspace; e Sequential Collaboration, where the operator and the robot share all or part of a workspace but do not work on a part or machine at the same time; e Co-operation, where the operator and the robot work on the same part or machine at the same time, and both are in motion; and e Responsive Collaboration, where the robot responds in real-time to the operator's motion.

In most industrial applications of cobots today, the cobot and the operator share the same space, but complete tasks independently or sequentially (i.e. Co-existence or Sequential Collaboration). Co-operation or Responsive Collaboration are presently less common. However, the relocatable robotic workstation (2) of the invention may be employed for any of the four collaborative manufacturing applications.

Most collaborative robots comprise integrated security systems so that the robot stops at a limited collision force, and the different operations may be security approved individually.

The collaborative robot (3) for use according to the invention shall be in cooperation with the imaging device (4) and the processing means (5), and the collaborative robot shall be mounted on the relocatable robotic workstation (2).

A variety of collaborative robots are commercially available from various providers, e.g., Universal Robots, KUKA, FANUC, Denso Robotics, Yaskawa and Doosan Robotics. The particular cobot for use according to the invention should be selected based on, e.g., weight, lifting capacity, range, speed and/or safety approval. The working tool For performing the intended tasks, the system of the invention also comprises one or more working tools (12), that can be mounted on the robot (3) and is suited for the intended task.

The working tools (12) may be available from a toolbox or tool magazine (124), that may comprise one or more removable or exchangeable tools or hardware modules,

DK 2021 00195 A1 8 from which toolbox (12A) the robots (3) can pick and choose the desired tool, depending on the intended task.

Alternatively, a tool change can be performed manually, by the intervention of the operator.

In another embodiment, a multi-functional tool (3) is mounted on the robot (3) configured for operating the multi-functional tool, and able to switch between the individual tools, as need be.

Examples of working tools frequently used in abattoirs are knives, e.g., fixed knives, rotary knives, oscillating knives, and wizard knives; saws, e.g., round saws, band saws and chain saws; or combinations hereof (i.e., multi-functional tools). Tools for use according to the invention also may include scissors, e.g., scissors having multiple blades, including blades of different sizes.

In another embodiment, the working tool for use in the system of the invention is a drag head system, i.e., vacuum cleaner for aspirating undesirable material from the surface of a meat product.

The robot (3) for use according to the invention is understood to comprise manipulators, so it can be configured for holding, and capable of operating a working tool (12). In another embodiment, the robot also shall be able to choose and change tools (12), e.g., a removable or exchangeable tool or hardware module chosen from a toolbox or tool magazine (12A).

In a further embodiment, the robot (3) for use according to the invention is mounted with a working tool (12), and the imaging device (4), in communication with the processing means (5), is mounted on the tip of the robotic arm, within operating distance of the working tool (12), allowing monitoring of the process and/or guidance of the robot.

The imaging device The system of the invention shall comprise an integrated imaging device (4) firmly mounted on the relocatable robotic workstation (2) and configured for recording three- dimensional (3D) images of the product surface of the meat item/workpiece (16) introduced to the system, and, in cooperation with a processing means (5), configured for sensing information relevant for planning, computation and safe execution of the robotic trajectories involved in a given task.

The imaging devices (4) for use according to the invention shall be an adaptive imaging system, based on the use of artificial intelligence (AI). The imaging system must be robust and capable of detecting, e.g., meat products with a large biological variation. Robustness is gradually improved through self-learning.

DK 2021 00195 A1 9 The imaging device shall be firmly mounted on the relocatable robotic workstation (2) in order to ensure a sharp and accurate image capturing.

Therefore, in another embodiment, and the imaging device (4) for use according to the invention is firmly attached the robotic workstation/carriage (2) by use of means (4A) for fixation of the imaging device (4).

However, although firmly mounted on the workstation/carriage (2), or on the means for fixation (4A), the one or more cameras included in the imaging device (4) may be individually set, in different positions or angles, depending on the intended task.

The imaging device (4) for use according to the invention may be considered as part of a machine imaging system. Machine imaging systems typically comprise a 3D- camera with a lens, an image sensor, an imaging processing means, and communication means.

In further embodiments, the system of the invention comprises two or more cameras, e.g., three cameras, one directed forward and two directed to the sides.

Moreover, the imaging device (4) for use according to the invention shall be a three-dimensional (3D) imaging system. 3D-imaging systems typically comprise multiple cameras or one or more laser displacement sensors. Commercially available 3D scanners or range cameras include the time-of-flight camera, the structure light camera, stereo camera or 3D camera, e.g., the Sick 3D ruler.

In another embodiment, the imaging device for use according to the invention is a LIDAR-based 3D-camera, that illuminates the target with laser light, and measures the reflection with a sensor.

In further embodiments, the system of the invention may be supplemented with additional sensors configured for measuring, determining and/or detecting various properties or events, e.g., distance sensors, positions sensors, X-ray sensors, ultra- sound sensors, infrared sensors, radar sensors, etc.

The single components of the imaging device (4) for use according to the invention are commercially available. Optionally the imaging system for use according to the invention also include lighting means. The processing means The relocatable robotic workstation (2) of the invention shall comprise a processing means/CPU (5), in cooperation with the robot (3) and the imaging device (4), capable of receiving and processing data obtained from the central database (9) and of guiding the robot (3) and configured for applying artificial intelligence applications/machine learning applications/algorithms on data obtained from the imaging device (4) and configured for executing one or more processing/working programs.

DK 2021 00195 A1 10 Typical processing/working programs include algorithms for moving, processing, and/or orienting workpieces, e.g., for cutting, trimming, deboning, cleaning and inspection.

In a particular embodiment of the invention, the processing means/CPU (5), in cooperation with the robot (3), may also be configured to process robot information including posture information of the robot main body, and load information of a load acting on the robot main body.

The processing means/CPU (5) for use according to the invention may be any commercially available processor/PC.

The central database Usually, at modern slaughterhouses, information about the carcasses supplied, and the products derived therefrom, may be stored on a central server/database (9), from which database this information may again be retrieved.

It is no prerequisite, that the system of the invention has access to such centrally stored information, and in one embodiment the system may be operated without any exchange of any information with a central server/database (9).

However, for optimal performance of various work tasks, the system according to the invention may, however, benefit from having access to centrally stored information relating to the subject in question and/or to the actual workstation.

Therefore, in another embodiment, the system of the invention comprises a processing means (5), in communication with the robot (3), and in communication with a central server/database (9) containing data about the slaughtered animal/carcass, and/or the meat item/workpiece, to be processed, and/or information about the specific workstation.

Such information, stored on the central server/database (9), may include, e.qg., process- or product-specific information, and may, e.g., allow for updates in respect of reporting of process status, new tasks to be accomplished, automatic error messaging, and all-clear messaging after completed tasks.

Artificial Intelligence (AI) applications/algorithms In the context of this invention, and in accordance with established terminology, an Artificial Intelligence (AI) application/algorithm covers machine learning and deep learning algorithms.

Machine learning is an application of artificial intelligence (AI) which use statistical techniques to perform automated decision-making, and optionally improve performance on specific tasks based on experience without being explicitly programmed.

DK 2021 00195 A1 11 Several approaches to machine leaning exists, e.g., supervised learning, reinforced learning, imitation learning and un-supervised learning.

In supervised learning, the computer may be presented with example inputs and their desired outputs, as given by the supervisor, and the goal is to learn a general rule that maps inputs to outputs. Using large sets of reference data, covering different product types, along with the desired output for each element in the data sets, a supervised learning algorithm is trained to learn a general rule that maps inputs to outputs.

As data is building up, the precision and robustness of the system increases, but occasionally, e.g., caused by quality errors or recognition errors, manual assistance may be needed, for additional training of the system.

Meat products are characterized by large variation in physical appearance. This variation comes from biological variation as well as machine and/or operator induced variations from previous stages in the processing. The variation is characterized not only by differences between different types of products, but in particular by differences between products of the same type, e.g., in deformation, flexibility, colouring, shape, size, surface texture (e.g., wet, greasy, icy).

Since the products typically are very difficult to handle and require custom tools, additional variation is introduced regarding position and orientation of the tool which handles the product. In many cases, the products are also closely spaced, stacked or partly covering each other, which amplifies the variations described above. As the products need to be handled in a production setup, these variations introduce strong requirements on precision, capacity, and robustness of the imaging system.

To meet these demands, a combination of state-of-the-art and custom deep learning algorithms are designed using, e.g., image segmentation and regression, trained specifically to identify meat products and related information on how to handle the products. The architecture and parameters of the algorithms are optimized for speed and precision in a production setup.

In one embodiment, the system of the invention applies a new algorithm for each workplace, and/or for each working tool, based on or as determined by information provided i.a. via the docking means/information port (8), or obtained by use of the imaging device (4).

The system of the invention may also be addressing random meat items/workpieces (16), or may be confronted with predetermined items, e.g., as determined by information provided from the central server/database (9), i.a. via the docking means/information port (8), or information obtained by use of the imaging device (4).

DK 2021 00195 A1 12 The vision system for use according to this invention may comprise both state- of-the-art and custom designed deep learning algorithms for detecting objects and positions in the images, combined in a way which meets the capacity requirements.

A key component to reaching high precision is to carefully optimize every component that goes into training the algorithm. The physical setup of the camera position and the designated workspace for the robot is designed to minimize the variation in physical appearance of the objects to be processed. At the same time, a special data augmentation technique is developed which artificially increases the data volume during training of the algorithm thereby enabling the system to deal with the remaining variation.

This applies both to variations in the objects themselves, but also to variations of the surface on which the objects are located (e.g., colours, patterns, or structures), as well as lighting conditions and other external factors.

Finally, a post-processing step may be applied using space and time averaging to provide stability and robustness.

The user-interface For operating the system, the relocatable robotic workstation (2) of the invention also comprises a user interface (6). This user interface may include, e.g., an interactive display, a virtual reality display, an augmented reality display, and imitation learning interfaces.

This user interface shall be configured for presenting to the operator only the available situational actions, applicable to the particular workplace, i.e., depending on the actual location of the system within the production facility, and depending on the particular circumstances, e.g., on the particular product type in question, e.g., as established by the imaging device (5).

The user interface shall also be configured for receiving situational inputs from the operator, i.e. only those inputs that are required and relevant in the given location and in the given situation, and which removes operator user-interface choices that are not compatible with the work site for reducing risks of operator faults, and which is compact enough that a human can operate the site with no further modifications than relocation of the robotic work station.

This interface shall allow the operator to interact quickly and easily with the system, at start or during operation, e.g., by offering the operator to select a particular processing/working program to be executed by the system.

The user interface/interactive display (6) for use according to the invention shall require little or moderate training. It shall be intuitive, with limited flexibility, and shall not offer extensive setting for the operator.

DK 2021 00195 A1 13 Also, the user interface (6) shall be sturdy and resistant to water and cleaning agents. The connection means: Fixtures and docking means The system of the invention shall comprise one or more docking means (8), each docking means located inside the production facility and configured for quick and easy clamping of the relocatable robotic workstation (2), configured for general or task-specific supplies, and configured for acting as an information port/information provider.

More specifically, the system of the invention shall comprise one or more docking means (8), each docking means located inside the production facility and configured for quick and easy clamping of the relocatable robotic workstation (2), and configured for feeding the workstation with general or task-specific supplies such as electrical voltage or pressurized air, and shall be configured for communicating/providing/supplying positional information and/or other task-specific digital or analogue input to the processing means (5), and shall be configured for providing data or internet connection to facilitate real-time sporadic, periodic or continuous access to sensor data and system statistics, and shall provide the workstation with a geometrical/spatial reference to other parts relevant to the task, e.g. a delivery box or obstacles.

The docking means (8) of the invention shall act as workstation lock and be capable of supporting and providing physical stability and positional repeatability to the workstation (2).

Moreover, the docking means (8) of the invention shall be configured for feeding the workstation with general or task-specific supplies such as electrical voltage or pressurized air, and it shall be configured for communicating/providing/supplying positional information and/or other task-specific digital or analogue input to the processing means (5).

Finally, the docking means (8) of the invention shall act as information port/information provider, i.e., be configured for feeding the workstation with process- or product-specific information, e.g., for providing data or internet connection to facilitate real-time sporadic, periodic or continuous access to sensor data and system statistics, and it shall provide the workstation with a geometrical/spatial reference to other parts relevant to the task, e.g., a delivery box or obstacles. Additional information provided via the docking means/information port (8) may include, e.g., product ID specifications, production receipts, etc.

For a stationary docking station to work properly, the connecting machine part must hold a matching fixture, like a key into a lock.

DK 2021 00195 A1 14 Therefore, the relocatable robotic workstation (2) of the invention must comprise one or more fixtures/fitting keys (7) for plugging in the relocatable robotic workstation (2) to one of the one or more docking means/workstation locks (8).

As defined herein, a fixture/fitting means (7) for connecting the relocatable robotic workstation (2) to one of the one or more docking means/workstation locks (8) located inside the production facility represents the "key-like” component that matches the docking station.

The system of the invention comprises at least one docking means/workstation locks (8). However, for the system to serve more workstations, the system may comprise two or more docking means/workstation locks (8). In this case, the two or more docking means/locks may be located at different places inside the production facility for securing a quick and easy plugging-in/coupling/clamping of the relocatable robotic workstation (2) to one of one or more on-site docking means.

First, the connection means for use according to the invention shall ensure mechanical stability of the relocatable robotic workstation (2) of the invention.

Next, the connecting means for use according to this invention shall comprise means for providing power supply to the relocatable robotic workstation (2), and/or network (cable) access to a central server/database (9). The connecting means may also provide for compressed air and/or vacuum.

Finally, the connecting means may also offer a water or saline connection.

The tool holder and the (end-of-arm) working tools For performing the intended tasks, the system of the invention also comprises one or more working tools (12), that can be attached to the robot (3) and is suited for the intended task.

The relocatable robotic workstation (2) of the invention shall be configured for holding, and capable of operating a working tool (12). For fixture of a particular working tool, the robot (3) for use according to the invention may comprise a robotic tool holder/tool changer (11).

This tool holder/changer (11) may be for a dedicated working tool (12), or it may be a holder for a universal (i.e., multi-functional) working tool (12).

In another embodiment, a multi-functional tool (12) is mounted on the robot (3) that is configured for operating the multi-functional tool, and able to switch between the individual tools, as need be.

In a further embodiment, the various working tools (12) may be available from a toolbox or tool magazine (12A), that may comprise one or more removable or exchangeable tools or hardware modules, from which toolbox (12A) the robots can pick and choose the desired tool, depending on the intended task.

DK 2021 00195 A1 15 Examples of working tools (12) frequently used in abattoirs include different models of grippers with range sizes and styles fitting for the repertoire of tasks. These can be specialized to the particular task, process or program, or they can be more general tools, and will typically be hard or soft or flexible finger grippers, suction cups or other manipulation devices.

The working tools (12) may be available from a toolbox or tool magazine (12A), that may comprise one or more removable or exchangeable tools or hardware modules, from which toolbox (12A) one or more robots can pick and choose the desired tool, depending on the intended task.

Lighting means In order for the imaging device (4) used according to the invention to work properly, the workpieces to be handled by the system of the invention shall be sufficiently illuminated. To this end, the system of the invention may optionally comprise lighting means (10), i.e., one or more light sources capable of emitting light adjusted to the frequencies that the imaging device used according to the invention is able to sense.

This usually involves light in the ranges of from about 350 nm to about 1700 nm to (i.e. from UV to shortwave-infrared regions), and it may be accomplished by use of diode light (LED light).

Also, the lighting means (10) for used according to the invention may be located on and fixed to the relocatable robotic workstation (2), or it may be placed off the workstation (2), in the vicinity of the particular workplace.

In a particular embodiment, the lighting means (10) for use according to the invention may be mounted on the relocatable robotic workstation (2), and it may additionally comprise a diode light (LED light) or a laser light capable of emitting a strip of light adjusted to indicate the exact extent of the working space, e.g., as a warning to the operator or any other human passing by the system.

Additional features The relocatable robotic workstation (2) represents a movable instrument that can easily be moved to a new location within the production facility. This may be accomplished by manually moving the workstation, or by means of an assistance of a powered engine.

In case the robotic workstation (2) shall be moved manually, a pair of handles (13) for guiding the unit will be useful.

Also, the relocatable robotic workstation (2) should comprise one or more wheels (14) for easy transport of the unit, whether the unit shall be moved manually or by power. The wheel may be rotatable and/or may be in a fixed position.

DK 2021 00195 A1 16 The relocatable robotic workstation (2) also may comprise means for communicating the status, i.e., a system status indicator (15) of the ongoing process/program to the operator, e.g., a light-, sound- or voice-controlled feed-back mechanism.

Finally, the system of the invention may additionally comprise a centrally located monitoring unit for supervision and status updating of one or more relocatable robotic systems (1) operating in the production facility. Such a centrally placed monitoring unit may, e.g., provide status feed-back in the form of light-, sound- or voice-generated warnings or messages.

The method of the invention In another aspect, the invention relates to a method for the relief of a wide range of manual work processes typically undertaken in various production facilities.

The method of the invention may, e.g., be for automating and handling a manually demanding work task using a relocatable robotic workstation, that is quickly positioned at the relevant workplace, and is easy and intuitive to operate, thanks to the system being able to gather the relevant information stored, and continually updated, on a central server/database (9).

Therefore, the method of the invention provides a robot-assisted method for relief of relatively small tasks involving unilaterally repetitive work.

The method of the invention is for automating a manually demanding work task in a production facility, using a relocatable robotic system (1), configured for recognition, localization, determination, inspection, and/or analysis, and subsequent processing of meat items introduced to the system, and the method of the invention may be characterised by comprising the subsequent steps of: i. bringing the relocatable robotic workstation (2) to a docking means (8) selected from the one or more docking means (8) located within the production facility; ii. connecting the relocatable robotic workstation (2) to the chosen docking means (8), ensuring mechanical stability and repeatability of the relocatable robotic workstation (2), and allowing the relocatable robotic workstation (2) access to a power supply and/or network (cable or wireless) access to a central server/database (9), for communication (i.e., for receiving and/or exchange information) with a central server/database (9); iii. based on input from the central server (9) and/or the imaging device (4), and in communication with the processing means (5), the user interface (6) of the relocatable robotic workstation (2) presents one or more processing options to the operator, and,

DK 2021 00195 A1 17 using the user interface (6), the operator selects the desired processing/working program, or, alternatively, define a new process to be accomplished; and iv. allowing the system to accomplish the selected process/working program.

The relocatable robotic workstation (2) contemplated used according to the method of the invention is the relocatable robotic workstation (2) of the invention described above.

The method of the invention includes the step of connecting the relocatable robotic workstation (2) to the one of the one or more docking means (8), located within the production facility. Therefore, the method of the invention includes the use of a system (1) that comprises at least one docking means/workstation locks (8).

However, for the system (1) to serve more workstations, the system may comprise two or more docking means/workstation locks (8). In this case, the two or more docking means/locks may be located at different places inside the production facility for securing a quick and easy plugging-in/coupling/clamping of the relocatable robotic workstation (2) to one of one or more on-site docking means.

Finally, the system (1) of the invention may comprise one or more relocatable robotic systems (1) operating in the production facility, and the system may additionally comprise a centrally located monitoring unit for supervision and status updating of one or more relocatable robotic systems (1) operating in the production facility.

The use of docking stations (8) ensures that the equipment is placed stably and in exact positions within the production facility. The fact that fixed positions are in place, allows for information to be deduced from these positions. Such information may be stored in the processing means (5), or it may be communicated from the central server/database (9).

Fixed operating positions also provides for increased repeatability, and obviated automatic or manual compensation for positional variations between relocations to the same workstation. This in turn increases ease of use and the overall robustness of the system.

Finally, the system of the invention may additionally comprise a centrally located monitoring unit for supervision and status updating of one or more relocatable robotic systems (1) operating in the production facility. Such a centrally placed monitoring unit may, e.g., provide status feed-back in the form of light-, sound- or voice-generated warnings or messages.

In step ii according to the method of the invention, connection of the relocatable robotic workstation (2) of the invention to the chosen docking means (8) may additionally provide for access to compressed air and/or vacuum, and/or may also offer a water or saline connection.

DK 2021 00195 A1 18 The working tools (12) may be available from a toolbox or tool magazine (12A), that may comprise one or more removable or exchangeable tools or hardware modules, from which toolbox (12A) one or more robots can pick and choose the desired tool, depending on the intended task.

In a further embodiment, the method of the invention may further comprise a step (iii-a) for manual or automatic shift of working tool (12).

For a manual shift of working tool (12) the operator selects and mounts the decide tool on the robotic arm.

For automatic shift of working tool (12) the robot (3), in cooperation with an imaging device (4) and a processing means (5), choses a tool (12) suited for the intended process from a toolbox or tool magazine (12A), for performing the intended (programmed or self-taught) task, as calculated/proposed/selected/instructed by the processing means (5).

For communication with the system (1) of the invention, the method of the invention also includes the use of a user interface (6), as described above, for the operator to select the desired processing/working program.

When a program has been selected, the system will, as far as possible, carry out the specified operation. However, the operator may wish to interfere with the program during the process, or, alternatively, may want to define a new process to be performed.

This may be accomplished by the operator, either supervised by the operator or in collaboration with the operator.

Typical tasks for the system to accomplish include manual processes such as lifting, moving, tipping, flipping, positioning, inspecting, tactile inspection, etc.

An example of the process to be accomplished by the method of the invention may be a matter of moving a pile of items, e.g., contained in a box, individually onto a conveyor belt.

Another example of the process accomplished by the method of the invention could be to perform a value-adding cutting operation on meat products, e.g., by leading a product to a stationary cutting tool.

In a third example of the process accomplished by the method of the invention, the robot may assist the operator in the performance of wear-and-tear manoeuvres, such as repetitive pulling/pushing/lifting/twisting of the workpiece.

In a fourth example, the robot may perform a selection task where products of insufficient quality are removed from a running conveyor belt or otherwise split from prime products. The imaging system's machine learning enables many detection tasks, in particular tasks that have not hitherto been possible with conventional imaging systems.

DK 2021 00195 A1 19 In a fifth example, the robot may refill certain operated production devices during operation so as to better utilize the operator's time for other value-adding tasks.

Workpieces suited for the method of the invention include, e.g., backs, belly meat, tenderloins, shoulder, toes, or similar meat products, frequently handles in abattoirs.

The objects contemplated according to the method of the invention also may be production consumables or equipment, like packing tools or plastics, labels, fixation devices, lids, etc.

The production facility The invention provides a cobot-assisted method for relief of relatively small tasks involving unilaterally repetitive work, and for quick and easy conversion of work processes taking place in a production facility facing different and varying work tasks. Such work processes are undertaken in various industries.

However, processes which can advantageously be relieved by the method according to the invention typically are undertaken in slaughterhouses or meat processing industries, e.g., for handling fresh or packaged meat products at all stages, from cold storage to unloading, etc.

DETAILED DESCRIPTION OF THE FIGURES The present invention is further illustrated by reference to the accompanying drawing, in which: Fig. 1 shows a graphical representation of the essential elements of the relocatable, robotic system of the invention (1): Robot-carrying carriage/robotic working station (2); Robot (3); Imaging device (4); Processing means/CPU (5); User interface display (6); Fixture/fitting means (7) for connection to a docking means/workstation lock (8); Docking station for connecting means (8); Lighting means/light source (10); Figs. 2A-2C show an example of the relocatable robotic workstation (2) of the invention: Robot (3); Imaging device/imaging device (4); Means for fixation/supporting/anchoring the imaging device (4A); User interface (6); Fixture/fitting means for connection to docking means (7); Tool holder/tool changer (11); Handles for guiding the robotic workstation (13); Wheels (rotatable and/or fixed) for transporting the robotic workstation (14); Fig. 3 shows another example of the relocatable robotic workstation of the invention: Robot-carrying carriage/robotic working station (2); Inspection hatch (2A); Robot (3); Imaging device (4); Fixture/fitting means for connection to docking means

DK 2021 00195 A1 20 (7); Tool holder/tool changer (11); Working tool (12); System status indicator (15); Workpiece (16); Working bench (17); and Figs. 4A-4B show an example of a fixture/fitting means for connection to docking means (7) and a docking means/docking station/workstation lock/information port (8) for use according to the invention: 4A in an open position, and 4B in a closed position. List of reference signs This is a listing of various elements relating to the present invention. Alternative/synonymous designations are separated by slashes.

1. The relocatable robotic system of the invention

2. Relocatable robotic workstation/robot-carrying working station/robot carriage 2A. Inspection hatch

3. Robot/cobot/co-robot 4, Imaging device/vision device 4A. Means for fixation/supporting/anchoring the imaging device

5. Processing means/processor/ CPU

6. User interface/interactive display

7. Fixture/fitting means for connection to docking means

8. Docking means/docking station/workstation lock/information port

9. Central server/database

10. Lighting means/light source

11. Tool holder/tool changer

12. Working tool 12A. Toolbox/tool magazine

13. Handles for guiding the robotic workstation

14. Wheels (rotatable and/or fixed) for transporting the robotic workstation

15. System status indicator/communication means

16. Meat item/workpiece

17. Working bench

Claims (6)

DK 2021 00195 A1 1 CLAIMS

1. A relocatable robotic system (1), for use within a production facility, and configured for recognition, localization, determination, inspection, and/or analysis, and subsequent processing of meat items introduced to the system, which system comprises the following elements: a. a relocatable robotic workstation (2), configured to accommodate/assemble and support the necessary elements, including a means (4A) for supporting an imaging device (4), and a fixture/fitting means (7) for connecting the relocatable robotic workstation (2) to one of one or more docking means (8) located within/inside the production facility, and which relocatable robotic workstation (2) further comprises: al. a robot (3), in cooperation with an imaging device (4) and a processing means (5), which robot is configured for holding a working tool (12) suited for the intended task, and for manipulating the meat items (16) introduced to the system, and which robot is mounted on the relocatable robotic workstation (2); a2. an imaging device (4), firmly mounted on the relocatable robotic workstation (2), and configured for recording three-dimensional (3D) images of the product surface of the meat item (16) introduced to the system, and, in cooperation with a processing means (5), configured for sensing information relevant for planning, computation and safe execution of the robotic trajectories involved in a given task; a3. a processing means (5), in cooperation with the robot (3) and the imaging device (4), and configured for applying a machine learning application/algorithm on data obtained from the imaging device (4), and configured for executing one or more processing/working programs; a4. a user interface (6), configured for presenting to the operator only the available situational actions, applicable to the particular workplace and the particular circumstances/product type, and configured for receiving situational inputs from the operator; and ab. a fixture/fitting means (7), configured for firmly connecting the relocatable robotic workstation (2) to one of the one or more docking means (8) located inside the production facility; and b. one or more docking means (8), each docking means located inside the production facility and configured for quick and easy clamping of the relocatable robotic workstation (2), configured for general or task-specific supplies, and configured for acting as an information port/information provider.

DK 2021 00195 A1 2

2. The robotic system of claim 1, which system further comprises lighting means (10).

3. The robotic system of claim 1, wherein the relocatable robotic workstation (2) further comprises a means (4A) for fixation/supporting/anchoring the imaging device (4).

4. A method for automating a manually demanding work task in a production facility, which method comprises the subsequent steps of: i. bringing a relocatable robotic workstation (2) to a docking means (8) selected from the one or more docking means (8) located within the production facility, which relocatable robotic workstation (2) comprises: al. a robot (3), in cooperation with an imaging device (4) and a processing means (5), which robot is configured for holding a working tool (12) suited for the intended task, and for manipulating the meat items (16) introduced to the system, and which robot is mounted on the relocatable robotic workstation (2); a2. an imaging device (4), firmly mounted on the relocatable robotic workstation (2), and configured for recording three-dimensional (3D) images of the product surface of the meat item (16) introduced to the system, and, in cooperation with a processing means (5), configured for sensing information relevant for planning, computation and safe execution of the robotic trajectories involved in a given task; a3. a processing means (5), in cooperation with the robot (3) and the imaging device (4), and configured for applying a machine learning application/algorithm on data obtained from the imaging device (4), and configured for executing one or more processing/working programs; a4. a user interface (6), configured for presenting to the operator only the available situational actions, applicable to the particular workplace and the particular circumstances/product type, and configured for receiving situational inputs from the operator; and ab. a fixture/fitting means (7), configured for firmly connecting the relocatable robotic workstation (2) to one of the one or more docking means (8) located inside the production facility; and b. one or more docking means (8), each docking means located inside the production facility and configured for quick and easy clamping of the relocatable robotic workstation (2), configured for general or task-specific supplies, and configured for acting as an information port/information provider; ii. connecting the relocatable robotic workstation (2) to the chosen docking means (8), ensuring mechanical stability and repeatability of the relocatable robotic workstation (2), and allowing the relocatable robotic workstation (2) access to a

DK 2021 00195 A1 3 power supply and/or network (cable or wireless) access to a central server/database (9), for communication (i.e., for receiving and/or exchange information) with a central server/database (9); iii. based on input from the central server (9) and/or the imaging device (4), and in communication with the processing means (5), the user interface (6) of the relocatable robotic workstation (2) presents one or more processing options to the operator, and, using the user interface (6), the operator selects the desired processing /working program, or, alternatively, define a new process to be accomplished; and iv. allowing the system to accomplish the selected process/working program.

5. The method of claim 4, wherein two or more docking means (8) are located within the production facility.

6. The method of claim 5, comprising two or more relocatable robotic workstations (2).