Apparatus and method for predicting collision of logistics robots
The collision prediction device for logistics robots addresses the issue of collisions by learning collision criteria and outputting control signals to adjust robot and facility operations, preventing damage and ensuring stable factory processes.
Patent Information
- Application Number
- PCT/KR2023/020638
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-19
- Filing Date
- 2023-12-14
- Publication Date
- 2024-11-28
AI Technical Summary
Collisions between logistics robots and facilities within smart factories can occur due to overlapping movement paths, leading to potential damage and process disruptions, necessitating a predictive solution to prevent such incidents.
A collision prediction device and method that collects process state information, learns collision prediction criteria, and outputs control signals to prevent collisions by adjusting the movement paths of logistics robots and operating states of facilities within a predefined operation boundary.
The solution effectively predicts and prevents collisions, reducing damage to robots and equipment, and ensuring stable process operations by learning specific collision types and adjusting logistics robot and facility operations accordingly.
Smart Images

Figure KR2023020638_28112024_PF_FP_ABST
Abstract
Description
Collision prediction device and method for logistics robots
[0001] The present invention relates to a collision prediction device and method for a logistics robot for predicting and preventing collisions of logistics robots within an operation boundary including a plurality of facilities and logistics robots.
[0002]
[0003] Logistics robots are being introduced not only in general logistics warehouses and factories, but also in smart factories that manufacture products with different specifications using various parts, to ensure flexible and efficient supply and transport of parts.
[0004] Logistics robots are a general term for autonomous mobile robots (AMRs) and automated guided vehicles (AGVs), and these logistics robots can move and perform tasks under the control of a control device.
[0005] In a smart factory, logistics robots can move along optimal paths based on path planning to perform missions assigned by control devices.
[0006] Smart factories may house multiple logistics robots, and collisions between them can occur if their paths overlap. For example, if one logistics robot descends a narrow passageway in a smart factory while another ascends, the two robots could become stuck and collide. Furthermore, logistics robots may collide with equipment within the smart factory while moving.
[0007] In this way, if a collision occurs between logistics robots, not only will it cause damage to the logistics robot, but it will also have a huge impact on the progress of the entire process, which can result in significant losses. Therefore, it is necessary to propose a method that can predict and prevent the risk of collision before a collision occurs between logistics robots.
[0008]
[0009] The matters described as background technology above are only intended to enhance understanding of the background of the present invention, and should not be taken as an admission that they correspond to prior art already known to those skilled in the art.
[0010]
[0011] Accordingly, the present invention has as a technical task the provision of a collision prediction device and method for a logistics robot, which enables collision prediction criteria for a logistics robot to be learned in advance through process status information within an operation boundary, and to predict and prevent collisions of the logistics robot based on the results of the learning.
[0012]
[0013] The tasks of the present invention are not limited to the tasks mentioned above, and other tasks not mentioned will be clearly understood by those skilled in the art from the description below.
[0014]
[0015] In order to solve the above-described problem, according to one embodiment of the present invention, a collision prediction device for a logistics robot includes: a collection unit that collects process status information including facility information for a plurality of facilities arranged within a preset operation boundary and logistics robot operation information for a plurality of logistics robots provided within the operation boundary and moving through at least a part of a plurality of facility areas in which the plurality of facilities are arranged; a learning unit that learns, based on the process status information, collision prediction criteria for each collision type including a collision between any one of the plurality of facilities and any one of the plurality of logistics robots and a collision between the plurality of logistics robots; a prediction unit that predicts the occurrence of a collision for at least one of the plurality of logistics robots based on process status information collected after the learning and the learned collision prediction criteria; and an output unit that outputs a control signal corresponding to a result of the prediction.
[0016] For example, the learning unit can label the process status information by time zone and learn collision prediction criteria by collision type based on the labeled process status information.
[0017] For example, the learning unit can classify the labeled process status information based on whether a preset distance condition is satisfied for the target object including the equipment and another one of the plurality of logistics robots for the plurality of logistics robots.
[0018] For example, the learning unit can learn the collision prediction criterion based on process state information corresponding to a point in time prior to a preset time interval from the point in time at which the preset distance condition is satisfied among the labeled process state information.
[0019] For example, the learning unit can perform the learning when any one of the plurality of logistics robots satisfies the preset distance condition.
[0020] For example, the learning unit can learn the collision prediction criterion based on process state information corresponding to a point in time prior to a preset time interval from the point in time when the preset distance condition is unsatisfied among the labeled process state information.
[0021] For example, the equipment information may include at least one of the coordinates of an equipment placement area corresponding to the equipment within the operation boundary and an operation status of the equipment.
[0022] For example, the above equipment information may vary depending on the operating characteristics of the equipment.
[0023] For example, the logistics robot operation information may include at least one of the speed, movement direction, movement path, size, normal operation status, sensor detection results, braking specifications, real-time clock, characteristics of loaded items, mutual distances between some of the plurality of logistics robots, and operation priorities between the plurality of logistics robots, of each of the plurality of logistics robots.
[0024] For example, the process status information may further include detection information measured through a detection device that detects the equipment and the logistics robot within the operation boundary.
[0025] For example, the control signal may include a control signal that causes a warning notification to be displayed regarding the outcome of the prediction.
[0026] For example, the control signal may include a signal for controlling at least one of the speed and movement direction of at least one of the plurality of logistics robots when the collision type corresponds to a collision between the plurality of logistics robots.
[0027] For example, the control signal may include a signal that controls the plurality of logistics robots to operate according to the operation priority among the plurality of logistics robots.
[0028] For example, the control signal may include a signal that is transmitted to the equipment to control the operation status of the equipment when the collision type corresponds to a collision between any one of the plurality of equipment and any one of the plurality of logistics robots.
[0029] For example, the control signal may include a signal that is transmitted to the logistics robot to control at least one of the speed and movement direction of the logistics robot when the collision type corresponds to a collision between any one of the plurality of facilities and any one of the plurality of logistics robots.
[0030]
[0031] According to one embodiment of the present invention for solving the above-described problem, a method for predicting a collision of a logistics robot includes the steps of collecting process status information including facility information on a plurality of facilities arranged within a preset operation boundary and logistics robot operation information on a plurality of logistics robots provided within the operation boundary and moving through at least a part of a plurality of facility areas in which the plurality of facilities are arranged; learning collision prediction criteria for each collision type including a collision between any one of the plurality of facilities and any one of the plurality of logistics robots and a collision between the plurality of logistics robots based on the process status information; predicting the occurrence of a collision for at least one of the plurality of logistics robots based on the process status information collected after the learning and the learned collision prediction criteria; and outputting a control signal corresponding to a result of the prediction.
[0032]
[0033] According to the embodiments of the present invention as described above, collisions of logistics robots can be predicted in advance, and in particular, collision prediction performance can be improved by learning collision prediction criteria for each type of collision that may occur within the operation boundary.
[0034] In addition, collisions can be prevented by controlling logistics robots or equipment based on collision prediction results.
[0035] Through the above effects, damage to logistics robots or equipment can be prevented, and furthermore, process delays due to collisions can be prevented, thereby ensuring that the process can proceed stably.
[0036]
[0037] The effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by a person having ordinary skill in the art to which the present invention belongs from the description below.
[0038]
[0039] FIG. 1 is a block diagram showing an example of an operational boundary configuration that can be applied to embodiments of the present invention.
[0040] FIG. 2 is a drawing for explaining an example of implementation of an operational boundary that can be applied to embodiments of the present invention.
[0041] FIG. 3 is a block diagram showing an example of a control device configuration that can be applied to embodiments of the present invention.
[0042] FIG. 4 is a block diagram showing an example of a logistics robot configuration that can be applied to embodiments of the present invention.
[0043] FIG. 5 is a sequence diagram for explaining a collision prediction process of a logistics robot according to one embodiment of the present invention.
[0044]
[0045] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Regardless of the drawing numbers, identical or similar components will be given the same reference numbers and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably only for the convenience of writing the specification, and do not in themselves have distinct meanings or roles. In addition, when describing the embodiments disclosed in this specification, if it is determined that a specific description of a related known technology may obscure the gist of the embodiments disclosed in this specification, a detailed description thereof will be omitted. In addition, the attached drawings are only intended to facilitate easy understanding of the embodiments disclosed in this specification, and the technical ideas disclosed in this specification are not limited by the attached drawings, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and technical scope of the present invention.
[0046] Terms that include ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by these terms. These terms are used solely to distinguish one component from another.
[0047] When a component is referred to as being "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components intervening. Conversely, when a component is referred to as being "directly connected" or "connected" to another component, it should be understood that there are no other components intervening.
[0048] Singular expressions include plural expressions unless the context clearly indicates otherwise.
[0049] In this specification, terms such as “include” or “have” are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood not to exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.
[0050] In addition, the term "Unit" or "Control Unit" included in the internal configuration names of logistics robots or control devices is merely a term widely used to name a control device (Controller) that controls a specific function, and does not mean a generic function unit. For example, each control device may include a modem / transceiver that communicates with other control devices or sensors to control the function it is responsible for, a memory that stores an operating system or logic commands and input / output information, and one or more processors that perform judgments, calculations, and decisions necessary for controlling the function it is responsible for. Depending on the implementation, one processor may be responsible for calculations for multiple control devices.
[0051] First, the configuration of the operational boundary in which the logistics robot according to the embodiment is deployed and operated is described with reference to FIGS. 1 and 2.
[0052]
[0053] FIG. 1 is a block diagram showing an example of an operational boundary configuration that can be applied to embodiments of the present invention.
[0054] Referring to FIG. 1, the operating boundary (100) may include a logistics robot (110), equipment (120), a monitoring device (130), and a control device (140).
[0055] The operational boundary (100) may be equipped with multiple logistics robots (110), multiple facilities (120), and multiple detection devices (130) depending on the production process and target production speed of the product. The operational boundary (100) may be implemented as a smart factory, and the multiple facilities may be implemented as production devices, but this is not necessarily limited to this. Each component is described below.
[0056] First, the logistics robot (110) may include an autonomous mobile robot (hereinafter, referred to as "AMR" for convenience) and an automated guided vehicle (hereinafter, referred to as "AGV" for convenience). Depending on the operation policy of the logistics robot (110) in the operation boundary (100), only one type of AGV or AMR may be operated, or both AGV and AMR may be operated together within the operation boundary (100).
[0057] AGVs generally perform required actions (movement, direction change, stop, etc.) within the operating boundary (100) by recognizing and following guidance devices placed on the floor for guidance of the AGV. Here, guidance devices may refer to optically recognizable markers (spots, 2D codes, etc.), tags that can be recognized contactlessly at close range (e.g., NFC tags, RFID tags, etc.), magnetic strips, wires, etc., but these are examples and are not necessarily limited thereto. Guidance devices may be placed continuously on the floor or may be placed discontinuously and spaced apart from each other. Since AGVs fundamentally perform operations by recognizing and following guidance devices, they require guidance devices to be installed in advance before operation. Therefore, when moving the AGV to a new path or modifying an existing path, the guidance devices must be physically installed or modified. In addition, since AGVs do not deviate from the path set by the guidance devices, if an obstacle is detected on or around the path, the AGV typically stops until the detected obstacle disappears or separate control is applied. In the operation of AGV, the control device (140) must control the AGV based on the guidance equipment, so commands such as 'drive until the third marker is recognized' or 'change the heading direction by 90 degrees when the third marker is recognized' from the current location can be transmitted to the AGV as individual command units or mission units (e.g., recovery, supply, charging, patrol, etc.) including multiple commands.
[0058] AMR can determine its current location by sensing its surroundings (i.e., positioning), and its ability to perform path planning using positioning and a map is what most distinguishes it from AGVs. Therefore, if a map with compatible coordinates is shared between the AMR and the control device (140), the control device (140) can control the AMR by instructing the AMR on a path based on the coordinates. In addition, if an obstacle is detected while driving, the AMR can set an avoidance path on its own, avoid the obstacle, and then return to the original path. The function of the control device (140) setting the path of the AMR to one or more transit coordinates can be referred to as global path planning, and the function of the AMR setting a movement path or an avoidance path between transit coordinates according to the global path planning can be referred to as local path planning.
[0059]
[0060] Next, the equipment (120) may refer to, for example, a device (robot arm, conveyor belt, etc.) that performs a production process of a product within an operation boundary (100), and in a broader sense, may refer to a device arranged to assist in the performance of a mission, such as entry and exit of a logistics robot (110), when the production process is performed by a person. The device arranged to assist in the performance of a mission may be, but is not necessarily limited to, a device that detects the status of a designated location where a pallet carried by a logistics robot (110) can be put down or collected within an area where a specific production process is performed, a device that determines the progress of the process, a means for blocking entry and exit within an area, etc.
[0061] For example, the facility (120) is controlled through a PLC (Programmable Logic Controller) and can communicate with a control device (140) in relation to the process progress.
[0062] The monitoring device (130) can perform a function of acquiring information for determining the situation within the operating boundary (100) and transmitting the information to the control device (140). For example, the monitoring device (130) may include a camera, a proximity sensor, etc., but is not necessarily limited thereto.
[0063] The control device (140) can communicate with the aforementioned components (110, 120, 130) to obtain information necessary for the operation of the operation boundary (100) or control each component. For example, the control device (140) can perform dispatching of the logistics robot (110), route setting, mission assignment, process management by product, material management, etc.
[0064] In implementation, the control device (140) may include a local control device (ACS: AMR / AGV Control System) that controls surrounding process facilities based on the location of the AGV / AMR and performs mission-based control of the AGV / AMR, and an integrated control device (MoRIMS: Mobile Robot Integrated Monitoring System) that integrates and controls two or more local control devices. The integrated control device may perform status and route, logistics flow setting, and traffic control of all logistics robots (110) within the operation boundary (100) from each of a plurality of local control devices. For example, when the local control device (ACS) is equipped in units of logistics robots of the same manufacturer or the same model, the integrated control device may perform integrated control for collision prevention, such as bottleneck level analysis of intersection / overlapping areas, driving acceleration / deceleration control, and regeneration of avoidance paths, through traffic distribution control between heterogeneous types based on information acquired through a plurality of local control devices (ACS).
[0065] In addition, the integrated control device can have a manufacturing execution system (MES) as its upper control subject, and the manufacturing execution system (MES) can be linked to an automated scheduler (APS: Advanced Planning & Scheduling).
[0066] In addition to the configuration (110, 120, 130, 140) of the operation boundary (100) described above, it goes without saying that devices for mutual communication between components such as beacons, repeaters, APs (Access Points), chargers for charging logistics robots (110), loading spaces for storing or loading parts, spaces for storing finished or intermediate products, traffic lights, circuit breakers, waiting spaces for idle logistics robots (110), etc. can be appropriately placed within the operation boundary (100).
[0067] Below, an implementation example of an operating boundary (100) that can be applied to embodiments of the present invention is described with reference to FIG. 2.
[0068] FIG. 2 is a diagram illustrating an implementation example of an operational boundary applicable to embodiments of the present invention. The implementation example illustrated in FIG. 2 primarily illustrates components related to the description of the present invention, and actual implementations of the operational boundary (100) may include more or fewer components.
[0069] Referring to FIG. 2, a plurality of equipment areas (Z) may be provided within an operation boundary (100), and equipment (120) of the same or different types may be placed within each equipment area (Z). A logistics robot (110) moves through at least some of the multiple equipment areas (Z) within the operation boundary (100), and a plurality of logistics robots (110) may be placed within the operation boundary (100).
[0070] When a logistics robot (110) moves within an operation boundary (100), collisions may occur between logistics robots (110) or between logistics robots (110) and equipment (120). If a collision occurs between logistics robots (110), not only may the logistics robot (110) be damaged and cause damage, but the entire process may also be halted, thereby hindering process efficiency. Therefore, it is proposed to predict and prevent such collisions between logistics robots (110) in advance through a collision prediction device (150) according to one embodiment of the present invention.
[0071] Below, the configuration of a control device (140) that can be applied to embodiments of the present invention is described with reference to FIG. 3.
[0072]
[0073] FIG. 3 is a block diagram illustrating an example of a control device configuration applicable to embodiments of the present invention. Each component illustrated in FIG. 3 primarily represents components related to embodiments of the present invention, and in the actual implementation of the control device (140), more or fewer components may be included.
[0074] Referring to FIG. 3, the control device (140) may include a firmware management unit (141), a traffic control unit (142), a process management unit (143), a production / logistics management unit (144), an inventory management unit (145), a communication unit (146), a monitoring unit (147), a map management unit (148), and a work schedule management unit (149).
[0075] The firmware management unit (141) can obtain the latest firmware of the logistics robot (110) through the communication unit (146) and transmit it to the logistics robot (110) to perform a firmware update, thereby keeping the firmware of the logistics robot (110) up to date.
[0076] The traffic control unit (142) controls traffic lights and barriers based on the route of the logistics robot (110), and can also recalculate the route of the logistics robot (110) according to traffic.
[0077] The process management department (143) can define the process for each product and manage missions such as process progress and progress location.
[0078] The production / logistics management department (144) can dispatch logistics robots (110) based on missions.
[0079] The inventory management unit (145) manages the location and quantity of each material, and this information can be useful for more efficient process operation, such as sending the logistics robot (110) to the destination earlier than the time when actual assembly / consumption of materials is detected for pallet pickup or retrieval.
[0080] The communication unit (146) can communicate with internal components of the operation boundary (100), such as a logistics robot (110), equipment (120), and a monitoring device (130), as well as external entities, such as a firmware update server.
[0081] The monitoring unit (147) can monitor the location, route, battery status, communication status, power train status, etc. of individual logistics robots (110). Here, the route is a concept that includes a waypoint-based global route and a real-time local route. In addition, the battery status may include voltage, current, temperature, peak voltage and current, state of charge (SOC), state of health (SOH), etc. The communication status may include information on the currently activated communication protocol (such as Wi-Fi), connected AP, distance to the AP, channel in use, etc. In addition, the power train status may include load, temperature, RPM, etc. of the drive system.
[0082] In addition, the monitoring unit (147) can also check the mission, operation mode, firmware version, etc. currently assigned to each logistics robot (110).
[0083]
[0084] The map management unit (148) may obtain map data in the form of a grid map obtained when an AMR among logistics robots (110) drives within the operation boundary (100), and may provide a tool that allows a factory manager to edit the obtained map data. By editing the map data, a zone, a virtual lane, an intersection, a no-entry zone, etc., in which the logistics robot (110) performs one or more preset actions upon entry may be set, but this is merely an example and is not necessarily limited thereto. In addition, the map management unit (148) may distribute the corresponding map to the remaining logistics robots (110) other than the logistics robot (110) that initially obtained the grid map through actual driving, through the communication unit (146).
[0085] The work schedule management unit (149) can manage and monitor the mission of the logistics robot (110) based on the process information of the operation boundary (100) received from the equipment (120) and the monitoring device (130) through the communication unit (146). In addition, the work schedule management unit (149) can select a specific logistics robot (110) and assign a mission, and set the global route of the logistics robot (110) according to the assigned mission.
[0086] Next, a logistics robot is described with reference to Fig. 4.
[0087]
[0088] FIG. 4 is a block diagram showing an example of a logistics robot configuration that can be applied to embodiments of the present invention.
[0089] Referring to FIG. 4, the logistics robot (110) may include a driving unit (111), a sensing unit (112), a loading unit (113), a communication unit (114), and a control unit (115). Each component is described below.
[0090] The driving unit (111) may include a driving source, wheels, suspension, etc. involved in the movement, steering, and stopping of the logistics robot (110). The driving source may be an electric motor supplied with power from a built-in battery (not shown). The wheels may include one or more driving wheels that receive driving force from the driving source, and non-driving wheels that rotate by the movement of the vehicle body without receiving driving force. Depending on the implementation, when multiple driving wheels are provided, the driving source may be matched to each driving wheel so that the rotation of each driving wheel can be independently controlled. In this case, by making the rotation directions of different driving wheels different, the vehicle body can be rotated and steering can be performed without a separate steering means. At least some of the non-driving wheels may be configured as caster-type wheels, but this is exemplary and is not necessarily limited thereto.
[0091] The sensing unit (112) is for detecting the surrounding environment of the logistics robot (110) or its own operating status, and may include at least one of a 2D laser scanner (e.g., LiDAR), a 3D vision (stereo) camera, a multi-axis gyro sensor, an acceleration sensor, a wheel encoder, and a proximity sensor.
[0092] An encoder can output information that can determine how much the wheel has rotated by using light emitted from a light-emitting element (e.g., a photodiode). For example, the encoder can count the number of slits arranged along the circumference of the wheel or a disk rotating with the wheel per unit time. The control unit (115) can perform odometry, which estimates displacement by analyzing the amount of position change over time using data acquired through the encoder and gyro sensor. However, the displacement estimated based on the encoder data may have an error from the actual displacement due to wheel slip or wear (wheel-related diameter change). Therefore, when performing odometry, the control unit (115) can perform noise and error correction on the information collected from the wheel and gyro sensor using a predetermined algorithm (e.g., EKF: Extended Kalman Filter) to output a result that tends to be close to the actual value. This odometry can be particularly useful when localization using a 2D laser scanner, as described later, is not possible.
[0093] 2D laser scanners scan their surroundings by projecting laser light onto a rotating reflector and detecting the reflected signal. By analyzing the intensity of the reflected signal and the time difference between the projection and reception, they can output detection results in the form of a point cloud.
[0094] A 3D vision camera can calculate the distance to an object based on the parallax between two cameras spaced a certain distance apart, i.e., the pixel distance between the images captured by each camera. A texture projector that projects infrared light in a predetermined pattern may also be included to enable detection of objects of the same color, such as flat surfaces (e.g., white walls).
[0095] Typically, 2D laser scanners are used for mapping, navigation, object recognition, etc., and 3D cameras can be used for navigation, especially for obstacle avoidance, but these are examples and are not necessarily limited to this.
[0096] The loading section (113) is a means for loading items to be transported, and may be a top plate on the upper part of the vehicle body itself, a table placed on the top plate, a lift, a turntable rotating along a vertical axis, a forklift, a conveyor, or a combination thereof. Similar to a forklift, a forklift may also support telescopic and tilting functions.
[0097] The communication unit (114) can communicate with other components within the operation boundary (100), such as equipment (120) and control devices (140), and can also support communication between logistics robots (110), and can also communicate with a charger when performing a charging mission.
[0098] The control unit (115) is a subject that performs overall control of each of the aforementioned components (111, 112, 113, 114), and can perform current mission, current location, destination determination, route planning, load control, etc. based on information obtained from the control device (140) through the communication unit (114).
[0099] Hereinafter, a collision prediction process of a logistics robot according to one embodiment of the invention will be described in detail with reference to FIG. 5.
[0100] FIG. 5 is a sequence diagram for explaining a collision prediction process of a logistics robot according to one embodiment of the present invention.
[0101] Referring to FIG. 5, a collision prediction device (150) according to an embodiment of the present invention includes a collection unit (151), a learning unit (152), a prediction unit (153), and an output unit (154), and may be connected to a control device (140). However, FIG. 5 mainly illustrates components related to the description of the present invention, and an actual collision prediction device (150) may be implemented by including more or fewer components than this. In addition, the collision prediction device (150) may be implemented as a component of the control device (140) instead of being connected to the control device (140). Hereinafter, a collision prediction process according to an embodiment of the present invention will be described in detail.
[0102] First, the collection unit (151) can collect process status information including facility information on a plurality of facilities (120) arranged within a preset operation boundary (100) and logistics robot operation information on the operations of a plurality of logistics robots (110) that move through at least a part of a plurality of facility areas (Z) provided within the operation boundary (100) and in which a plurality of facilities (120) are arranged (S510). In this case, the process status information can be obtained from the control device (140), and the control device (140) can obtain logistics robot operation information through communication with the logistics robot (110) (S501'). In addition, the control device (140) can obtain facility information and detection information through communication with the facility (120) and the monitoring device (130) and transmit the same to the collection unit (151).
[0103] More specifically, the facility information may include at least one of the coordinates of the facility area (Z) corresponding to the facility within the operation boundary (100) and the operation status of the facility, and in particular, the facility information may vary depending on the operation characteristics of the facility (120). For example, if the facility (120) is not fixed to a specific location within the facility area (Z) but has an operation characteristic in which the range it occupies within the facility area (Z) varies depending on the operation status, such variations may be reflected in the facility information, thereby improving the learning and judgment performance of collision prediction.
[0104] Meanwhile, the logistics robot operation information may include at least one of the speed, movement direction, movement path, size, normal operation status, sensor detection results, braking specifications, real-time clock, characteristics of loaded items, mutual distances between some of the plurality of logistics robots, and operation priorities between the plurality of logistics robots, of each of the plurality of logistics robots (110).
[0105]
[0106] The collection unit (151) inputs the collected process status information into the learning unit (512) (S502), and the learning unit (512) can learn collision prediction criteria for each collision type, including collisions between any one of the plurality of facilities (120) and any one of the plurality of logistics robots (110) and collisions between some of the plurality of logistics robots, based on the process status information (S503). Thereafter, the learning unit (512) can input the learned collision prediction criteria into the prediction unit (153) (S504).
[0107] In particular, the learning unit (152) can label process status information by time zone and learn collision prediction criteria by collision type based on the labeled process status information.
[0108] More specifically, the learning unit (152) can classify labeled process status information based on whether a preset distance condition is satisfied for a target object including another one of the plurality of logistics robots (110) and the equipment (120), depending on the collision type. In this case, the preset distance condition may be satisfied, for example, when the logistics robot (110) and the target object come within a certain distance.
[0109] Additionally, the learning unit (152) can learn the collision prediction criteria based on the process status information corresponding to a point in time prior to a preset time interval from the point in time when the preset distance condition is satisfied among the labeled process status information. In this case, learning of the collision prediction criteria can be performed when any one of the plurality of logistics robots (110) satisfies the preset distance condition.
[0110] Alternatively, the learning unit (152) may learn the collision prediction criterion based on the process state information corresponding to a point in time prior to a preset time interval from the point in time when the preset distance condition is unsatisfied among the labeled process state information.
[0111]
[0112] In addition to learning information and learning results, process status information can be continuously input into the prediction unit (513) as judgment information for predicting collisions (S505), and the prediction unit (513) can predict the occurrence of a collision for at least one of a plurality of logistics robots (110) based on the process status information collected after learning and the learned collision prediction criteria.
[0113] Thereafter, the prediction unit (153) inputs the prediction result to the output unit (154) (S506), and the output unit (154) can output a control signal corresponding to the input prediction result (S507). In this case, the output control signal can be transmitted to the control device (140) and transmitted to the logistics robot (110) or equipment (120), or can be directly transmitted to the logistics robot (110) and equipment (120).
[0114] More specifically, the control signal may include a control signal that causes a warning notification to be displayed regarding the result of the prediction, and the warning notification may be transmitted to a control device (140), equipment (120), logistics robot (110), etc.
[0115] In addition, the control signal may include a signal for controlling at least one of the speed and movement direction of at least one of the logistics robots (110) when the collision type corresponds to a collision of some of the plurality of logistics robots (110). Through this, when a collision is predicted, the logistics robot (110) may be controlled to decelerate, stop, or perform an evasive maneuver. For example, the control signal may be a control signal for causing the logistics robot to wait after a preliminary stop, and may be transmitted to the logistics robot (110) via the control device (140) (S508).
[0116] Additionally, the control signal may control some logistics robots (110) to operate according to the operational priorities among the logistics robots (110). For example, a logistics robot (110) with a high priority may be controlled to maintain an operational state, while a logistics robot (110) with a low priority may be controlled to stop, decelerate, or perform an evasive maneuver.
[0117] Additionally, the control signal may be transmitted to the facility (120) when the collision type corresponds to a collision between any one of the plurality of facilities (120) and any one of the plurality of logistics robots (110), and may include a signal for controlling the operation status of the facility (120). For example, the operation of the facility (120) may be controlled to be stopped through such a control signal.
[0118] Additionally, the control signal may be transmitted to the logistics robot (110) when the collision type corresponds to a collision between any one of the plurality of facilities (120) and any one of the plurality of logistics robots (110), and may include a signal for controlling at least one of the speed and movement direction of the logistics robot (110).
[0119] That is, the collision prediction device (150) can prevent a collision by controlling the operation of at least one of the logistics robot (110) and the equipment (120).
[0120] Meanwhile, the control device (140) can request the logistics robot (110) or equipment (120) to restart work after the collision problem for the logistics robot (110) is resolved (S509), and the logistics robot (110) that receives this can supplement its driving algorithm according to the collision prediction result (S510). After supplementing the algorithm, the logistics robot (110) reconfirms the mission with the control device (140) (S511), and after confirming the mission, the control device (140) requests the logistics robot (110) to proceed with the mission operation (S512), so that the operation of the logistics robot can finally be resumed.
[0121]
[0122] According to the embodiments of the present invention as described above, collisions of logistics robots can be predicted in advance, and in particular, collision prediction performance can be improved by learning collision prediction criteria for each type of collision that may occur within the operation boundary.
[0123] In addition, collisions can be prevented by controlling logistics robots or equipment based on collision prediction results.
[0124] Through the above effects, damage to logistics robots or equipment can be prevented, and furthermore, process delays due to collisions can be prevented, thereby ensuring that the process can proceed stably.
[0125]
[0126] Although the present invention has been illustrated and described with respect to specific embodiments thereof as described above, it will be apparent to those skilled in the art that the present invention may be variously improved and modified without departing from the technical spirit of the present invention as defined by the following claims.
Claims
1. A collection unit that collects process status information including facility information on a plurality of facilities arranged within a preset operation boundary and logistics robot operation information on a plurality of logistics robots that move through at least a part of a plurality of facility areas provided within the operation boundary and in which the plurality of facilities are arranged; A learning unit that learns collision prediction criteria for each collision type, including a collision between any one of the plurality of facilities and any one of the plurality of logistics robots and a collision between the plurality of logistics robots, based on the above process status information; A prediction unit that predicts the occurrence of a collision for at least one of the plurality of logistics robots based on the process status information collected after the above learning and the learned collision prediction criteria; and A collision prediction device for a logistics robot including an output unit that outputs a control signal corresponding to the result of the above prediction.
2. In claim 1, The above learning department, A collision prediction device for a logistics robot, characterized in that the process status information is labeled by time zone and a collision prediction criterion is learned by collision type based on the labeled process status information.
3. In claim 2, The above learning department, A collision prediction device for a logistics robot, characterized in that it classifies the labeled process status information according to whether a preset distance condition is satisfied for a target object including the equipment and another one of the plurality of logistics robots with respect to the plurality of logistics robots.
4. In claim 3, The above learning department, A collision prediction device for a logistics robot, characterized in that the collision prediction criterion is learned based on process status information corresponding to a point in time prior to a preset time interval from a point in time at which the preset distance condition is satisfied among the labeled process status information.
5. In claim 4, The above learning department, A collision prediction device for a logistics robot, characterized in that the learning is performed when any one of the plurality of logistics robots satisfies the preset distance condition.
6. In claim 3, The above learning department, A collision prediction device for a logistics robot, characterized in that the collision prediction criterion is learned based on process state information corresponding to a point in time prior to a preset time interval from a point in time when the preset distance condition is unsatisfied among the labeled process state information.
7. In claim 1, The above equipment information is: A collision prediction device for a logistics robot, characterized in that it includes at least one of the coordinates of an equipment placement area corresponding to the equipment within the above operating boundary and the working status of the equipment.
8. In claim 1, The above equipment information is: A collision prediction device for a logistics robot, characterized in that it changes according to the operating characteristics of the above equipment.
9. In claim 1, The above logistics robot operation information is: A collision prediction device for a logistics robot, characterized in that it includes at least one of the speed, movement direction, movement path, size, normal operation status, sensor detection results, braking specifications, real-time clock, characteristics of loaded items, mutual distances between some of the plurality of logistics robots, and operation priorities between the plurality of logistics robots.
10. In claim 1, The above process status information is: A collision prediction device for a logistics robot, characterized in that it further includes sensor information measured through a detection device that detects the equipment and the logistics robot within the above operation boundary.
11. In claim 1, The above control signal is, A collision prediction device for a logistics robot, characterized in that it includes a control signal for displaying a warning notification regarding the result of the above prediction.
12. In claim 1, The above control signal is, A collision prediction device for a logistics robot, characterized in that it includes a signal for controlling at least one of the speed and movement direction of at least one of the plurality of logistics robots when the above collision type corresponds to a collision between the plurality of logistics robots.
13. In claim 1, The above control signal is, A collision prediction device for a logistics robot, characterized in that it includes a signal for controlling the plurality of logistics robots to operate according to the operation priorities among the plurality of logistics robots.
14. In claim 1, The above control signal is, A collision prediction device for a logistics robot, characterized in that it includes a signal that is transmitted to the facility and controls the operation status of the facility when the above collision type corresponds to a collision between any one of the plurality of facilities and any one of the plurality of logistics robots.
15. In claim 1, The above control signal is, A collision prediction device for a logistics robot, characterized in that it includes a signal that is transmitted to the logistics robot and controls at least one of the speed and movement direction of the logistics robot when the above collision type corresponds to a collision between any one of the plurality of facilities and any one of the plurality of logistics robots.
16. A step of collecting process status information including facility information on a plurality of facilities arranged within a preset operation boundary and logistics robot operation information on a plurality of logistics robots moving through at least a part of a plurality of facility areas provided within the operation boundary and in which the plurality of facilities are arranged; A step of learning collision prediction criteria for each collision type, including a collision between any one of the plurality of facilities and any one of the plurality of logistics robots and a collision between the plurality of logistics robots, based on the above process status information; A step of predicting the occurrence of a collision for at least one of the plurality of logistics robots based on the process status information collected after the above learning and the learned collision prediction criteria; and A method for predicting a collision of a logistics robot, comprising the step of outputting a control signal corresponding to the result of the above prediction.
Citation Information
Patent Citations
Machine tool and machine learning apparatus
JP2018092428A
Method for planning path for avoiding collision between multi-mobile robot
KR1020150137166A
Method for planning path for avoiding collision between multi-mobile robot
KR1020160021161A
Method and control device for avoiding collisions between cooperating robots
US20030225479A1
KR20210116004A
