Mobile robot unit, hybrid path moving mobile robot system and method for controlling the same
The hybrid path-driving mobile robot system efficiently connects discontinuous paths by generating virtual nodes and links, optimizing path generation and ensuring rapid, accurate connectivity to destination points.
Patent Information
- Application Number
- KR1020250129322
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2024-11-07
- Filing Date
- 2025-09-10
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2045-09-10
AI Technical Summary
Conventional robot path generation methods face challenges in responding to constraints and newly occurring destination points or environmental conditions, leading to inefficient path generation and unnecessary detours when the destination point is outside the constrained path.
A hybrid path-driving mobile robot system that generates a virtual connection path using current and destination information, forming virtual nodes and links to connect discontinuous constraint paths, optimizing path generation by combining segmented paths into a single continuous path.
The system achieves rapid and accurate path search, preventing unnecessary routes and deriving optimized paths by ensuring connectivity to points outside the limited constraint path.
Smart Images

Figure 112025104221191-PAT00003_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a control method for a mobile robot unit that performs path generation, and more specifically, to a mobile robot unit that drives by performing a new path search by executing an efficient connection process with a point outside a limited provided path, a hybrid path driving mobile robot system equipped with the same, and a control method thereof. Background Technology
[0003] Robots are expanding their range of applications in various industrial fields and daily life. From robot vacuum cleaners used in homes to manufacturing robots used in welding or press processes in automobile factories, as well as various robots used in logistics systems such as AGVs and AMRs, a wide variety of robots are being researched, produced, and used.
[0004] In particular, the scope of utilization and application of transport and logistics robots that are guided or autonomously driven, such as AGVs (Automatic Guided Vehicles) or AMRs (Autonomous Mobile Robots), is increasing dramatically. That is, due to the development of drive technology, beyond the need for operation in clean facilities such as semiconductor factories, there is a trend toward expanded application in transportation service industries such as shopping malls and freight transport, as well as various logistics delivery industries operating fulfillment systems, and even in heavy industries that require the transport of large loads, such as heavy industry.
[0005] In general, for logistics robots, users generate a path for the robot to travel by considering a constrained link-node path on a map. However, conventional robot path generation has been accompanied by problems that make it difficult to respond to constraints and newly occurring destination points or environmental conditions. That is, when driving based on a conventional constrained path, when a destination point is outside the constrained path, there are significant limitations in path search in situations where there is no prior planning when setting a path to reach that point, leading to problems such as unnecessary path detours or inefficient path generation due to collisions with obstacles. The problem to be solved
[0007] Accordingly, the present invention aims to provide a mobile robot unit that achieves a single efficient path search by executing a virtual path connection connecting existing paths and target points, etc., based on a constrained path, a hybrid path-driving mobile robot system equipped with the same, and a control method thereof. means of solving the problem
[0009] The present invention for achieving the above-mentioned technical challenges may have the following configuration.
[0010] According to one aspect of the present invention, the present invention provides a hybrid path driving mobile robot system comprising a current position information and destination information of a robot that drives in a work environment including preset node and edge information, a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and a mobile robot path generation unit (21) that generates a virtual connection path using the constraint path and the current position information and the destination information when the current position information and the destination information and the constraint path are separated or the constraint path is discontinuous, thereby generating and determining a hybrid path for driving the mobile robot.
[0011] In the above hybrid path driving mobile robot system, the mobile robot path generation unit (21) may comprise: a position information verification unit (211) that verifies current position information and destination information of a robot performing driving in a work environment including preset node and edge information; a path attribute verification unit (212) that verifies a constraint path composed of node information and edge information required to move from the current position information to the destination information; a path connectivity verification unit (213) that verifies connectivity between the current position information, the destination information, and the constraint path; a virtual connection path information generation unit (214) that generates a virtual node and a virtual link to connect the current position information, the destination information, and the constraint path when a non-connection is confirmed in the path connectivity verification unit (213); and a hybrid path information determination unit (218) that generates a hybrid path from the constraint path and the virtual connection path.
[0012] In the above-described hybrid path driving mobile robot system, the virtual connection path information generation unit (214) may include: a virtual circle forming unit (2141) that forms a virtual circle from at least one of the current location information and destination information that is not connected from the constraint path and the end node of the constraint path when a non-connection is confirmed in the path connectivity verification unit (213); a virtual node generating unit (2143) that forms a virtual node at a contact point between the virtual circle and the constraint path; a virtual link generating unit (2145) that generates a virtual link connecting the virtual node and at least one of the current location information and destination information or connecting a non-continuous constraint path; and a virtual connection path generating unit (2147) that generates a virtual connection path using the virtual node and the virtual link.
[0013] In the above-described hybrid path driving mobile robot system, the virtual circle formed by the virtual circle forming unit (2141) may be formed around the current location information or the destination information.
[0014] In the above-described hybrid path driving mobile robot system, the virtual circle formed by the virtual circle forming unit (2141) may be formed around any end node of the constraint path.
[0015] In the above-described hybrid path driving mobile robot system, the mobile robot path generation unit (21) may include a hybrid path candidate generation unit (215) that generates a hybrid path candidate using the virtual connection path and the constraint path generated by the virtual connection path information generation unit (214).
[0016] In the above-described hybrid path driving mobile robot system, the mobile robot path generation unit (21) may further include an obstacle path verification unit (216) that checks obstacle information among the hybrid path candidates and generates an obstacle bypass path.
[0017] In the above-described hybrid path driving mobile robot system, the obstacle path verification unit (216) may include: an obstacle information verification unit (2161) that verifies obstacle information among the hybrid path candidates, and an obstacle bypass path generation unit (2163) that generates an obstacle bypass path based on the obstacle information.
[0018] In the above-described hybrid path driving mobile robot system, the mobile robot path generation unit (21) may include a hybrid path candidate cost calculation unit (217) that calculates a hybrid path candidate cost formed according to a preset cost standard for the hybrid path candidate.
[0019] In the above-described hybrid path driving mobile robot system, the hybrid path candidate cost calculation unit (217) may include: a constraint cost calculation unit (2171) that calculates a constraint link node cost for link driving and node rotation driving from the link information and node information of the constraint path of the hybrid path candidate.
[0020] In the above-described hybrid path driving mobile robot system, the constraint link node cost may include: a link cost (CCL) for link driving calculated from the link information of the constraint path of the hybrid path candidate, and a node cost (CCR) for turning driving calculated from the node information of the hybrid path candidate.
[0021] In the above-described hybrid path driving mobile robot system, the hybrid path candidate cost calculation unit (217) may include: a virtual cost calculation unit (2173) that calculates a virtual link node cost for link driving and node rotation driving from the virtual link information and virtual node information of the virtual connection path of the hybrid path candidate.
[0022] In the above-described hybrid path driving mobile robot system, the virtual link node cost may include: a link cost (CVL) for link driving calculated from the link information of the virtual connection path of the hybrid path candidate, and a node cost (CVR) for turning driving calculated from the node information of the hybrid path candidate.
[0023] In the above-described hybrid path-driving mobile robot system, the virtual link node cost may further include an obstacle bypass extension cost (CDE) for an obstacle bypass path.
[0024] In the above hybrid path driving mobile robot system, the obstacle bypass expansion cost (CDE) may include: a gradient bypass expansion cost (CDGE) for a multi-angle bypass path formed to bypass an obstacle, a heuristic bypass expansion cost (CDHE) for a straight non-bypass path that does not bypass an obstacle, a congestion bypass expansion cost (CDCE) due to obstacle detection for a curve bypass path calculated from the multi-angle bypass path, and a slope bypass expansion cost (CDSE) due to the curve slope for a curve bypass path calculated from the multi-angle bypass path.
[0025] According to another aspect of the present invention, the present invention provides a control method for a hybrid path driving mobile robot system, wherein a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, and a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, wherein when the current position information and destination information and the constraint path are discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information, and a hybrid path for driving the mobile robot is generated and determined from the constraint path and the virtual connection path.
[0026] The above method for controlling a hybrid path driving mobile robot system may include a position information verification step (S10) in which current position information and destination information of a robot executing driving in a work environment including preset node and edge information are verified; a constraint path verification step (S20) in which a constraint path composed of node information and edge information required to move from the current position information to the destination information is verified; a path connectivity verification step (S30) in which connectivity between the current position information, the destination information, and the constraint path is verified; a virtual connection path information generation step (S40) in which virtual nodes and virtual links are generated to connect the current position information, the destination information, and the constraint path when a non-connection is confirmed in the path connectivity verification unit (213); and a hybrid path information determination step (S80) in which a hybrid path is generated from the constraint path and the virtual connection path.
[0027] In the above method for controlling a hybrid path driving mobile robot system, the virtual connection path information generation step (S40) may include: a virtual circle formation step (S41) in which a virtual circle is formed from at least one of the current location information, the destination information, and the end of the constraint path when a non-connection is confirmed in the path connectivity verification step (S30); a virtual node generation step (S43) in which a contact point between the virtual circle and the constraint path is formed as a virtual node; a virtual link generation step (S45) in which a virtual link is generated connecting the virtual node and at least one of the current location information, the destination information, and the end of the constraint path; and a virtual connection path generation step (S47) in which a virtual connection path is generated using the virtual node and the virtual link.
[0028] In the above method for controlling a hybrid path driving mobile robot system, the virtual circle formed in the virtual circle forming step (S41) may be formed around the current location information or the destination information.
[0029] In the above method for controlling a hybrid path driving mobile robot system, the virtual circle formed in the virtual circle forming step (S41) may be formed around any end node of the constraint path.
[0030] In the above method for controlling a hybrid path driving mobile robot system, between the virtual connection path information generation step (S40) and the hybrid path information determination step (S80), a hybrid path candidate generation step (S50) may be included in which a hybrid path candidate is generated using the virtual connection path and the constraint path.
[0031] In the above method for controlling a hybrid path driving mobile robot system, between the hybrid path candidate generation step (S50) and the hybrid path information determination step (S80), an obstacle path verification step (S60) may be further included, in which obstacle information among the hybrid path candidates generated in the hybrid path candidate generation step (S50) is verified and an obstacle bypass path is generated.
[0032] In the above method for controlling a hybrid path driving mobile robot system, the obstacle path verification step (S60) may include: an obstacle information verification step (S61) in which obstacle information among the hybrid path candidates is verified, and an obstacle bypass path generation step (S63) in which an obstacle bypass path is generated using the obstacle information.
[0033] The above method for controlling a hybrid path driving mobile robot system may include a hybrid path candidate cost calculation step (S70) in which a hybrid path candidate cost formed according to a preset cost standard is calculated for the hybrid path candidate.
[0034] In the above method for controlling a hybrid path driving mobile robot system, the hybrid path candidate cost calculation step (S70) may include: a constraint cost calculation step (S71) in which a constraint link node cost for link driving and node rotation driving is calculated from the link information and node information of the constraint path of the hybrid path candidate.
[0035] In the above method for controlling a hybrid path driving mobile robot system, the constraint cost calculation step (S71) may include: a link cost calculation step (S711) for calculating a link cost (CCL) for link driving calculated from the link information of the constraint path of the hybrid path candidate, and a node cost calculation step (S713) for calculating a node cost (CCR) for turning driving calculated from the node information of the hybrid path candidate.
[0036] In the above method for controlling a hybrid path driving mobile robot system, the hybrid path candidate cost calculation step (S70) may include: a virtual cost calculation step (S73) in which a virtual link node cost for link driving and node rotation driving is calculated from the virtual link information and virtual node information of the virtual connection path of the hybrid path candidate.
[0037] In the above method for controlling a hybrid path driving mobile robot system, the virtual cost calculation step (S73) may include: a virtual link cost calculation step (S731) in which a link cost (CVL) for link driving calculated from link information of the virtual connection path of the hybrid path candidate is calculated, and a virtual node cost calculation step (S733) in which a node cost (CVR) for turning driving calculated from node information of the virtual connection path of the hybrid path candidate is calculated.
[0038] In the above method for controlling a hybrid path driving mobile robot system, the virtual cost calculation step (S73) may further include an obstacle bypass expansion cost calculation step (S75) for calculating an obstacle bypass expansion cost (CDE) for an obstacle bypass path.
[0039] In the above method for controlling a hybrid path driving mobile robot system, the obstacle bypass expansion cost calculation step (S75) may include: a gradient bypass expansion cost calculation step (S751) in which a gradient bypass expansion cost (CDGE) for a multi-angle bypass path formed to bypass an obstacle is calculated; a heuristic bypass expansion cost calculation step (S753) in which a heuristic bypass expansion cost (CDHE) for a straight non-bypass path that does not bypass an obstacle is calculated; a congestion bypass expansion cost step (S755) in which a congestion bypass expansion cost (CDCE) due to obstacle detection for a curve bypass path calculated from the multi-angle bypass path is calculated; and a slope bypass expansion cost calculation step (S757) in which a slope bypass expansion cost (CDSE) due to the curve slope for a curve bypass path calculated from the multi-angle bypass path is calculated.
[0040] According to another aspect of the present invention, the present invention provides a hybrid path driving mobile robot unit comprising a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and a mobile robot path generation unit that generates a virtual connection path using the constraint path and the current position information and the destination information when the current position information and the destination information and the constraint path are separated or the constraint path is discontinuous, thereby generating and determining a hybrid path for driving the mobile robot.
[0041] According to another aspect of the present invention, the present invention provides a hybrid path driving mobile robot unit control method comprising a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, and a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, wherein when the current position information and destination information and the constraint path are discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information, and a hybrid path for driving the mobile robot is generated and determined from the constraint path and the virtual connection path. Effects of the invention
[0043] The effects of the mobile robot unit, hybrid path-driving mobile robot system, and control method thereof of the present invention configured as described above are as follows.
[0044] First, the mobile robot unit, the hybrid path-driving mobile robot system, and the control method thereof according to the present invention may achieve rapid and fast path search by ensuring faster and more accurate connectivity to points other than the limited constraint path, thereby preventing the generation of paths with unnecessary routes.
[0045] Second, the mobile robot unit, hybrid path driving mobile robot system, and control method thereof of the present invention may derive an optimized path generation by generating multiple path candidates that combine discontinuous, segmented, and differentiated paths of a limited constraint path into a single continuous path. Brief explanation of the drawing
[0047] FIG. 1 is a schematic diagram of a mobile robot unit (10) according to an embodiment of the present invention and a hybrid path-driving mobile robot system (1) equipped with the same. FIG. 2 is a schematic diagram of a mobile robot unit (10) and a mobile robot server (2) according to an embodiment of the present invention. FIG. 3 is a detailed configuration diagram of a mobile robot unit (10) according to an embodiment of the present invention. FIG. 4 is a configuration diagram of a mobile robot server (2) according to an embodiment of the present invention. FIG. 5 is a detailed configuration diagram of a mobile path generation unit (21) of a mobile robot server (2) according to an embodiment of the present invention. FIG. 6 is a detailed configuration diagram of a virtual connection path information generation unit (214) of a mobile path generation unit (21) according to an embodiment of the present invention. FIG. 7 is a detailed configuration diagram of an obstacle path verification unit (216) of a mobile path generation unit (21) according to an embodiment of the present invention. FIG. 8 is a detailed configuration diagram of a hybrid path candidate cost calculation unit (217) of a mobile path generation unit (21) according to an embodiment of the present invention. FIG. 9 is a flowchart of a method for controlling a hybrid path-driving mobile robot system according to an embodiment of the present invention. FIG. 10 is a state diagram of a schematic path when destination information exists outside the constraint path in a virtual connection path information generation process according to an embodiment of the present invention. FIGS. 11 and FIGS. 12 are state diagrams of the connection status of virtual links to target information separated from the constraint path (Pc). FIG. 13 is a diagram showing the incorrect type of connection status method of a virtual link to target information separated from the constraint path (Pc) of FIG. 11 and FIG. 12. Figure 14 is a diagram showing the connection status of a virtual link to a modified form of a constraint path (Pc) and a separated destination information. FIGS. 15 and 16 are state diagrams of the process of forming virtual nodes (Nv) and virtual links through multiple forms of virtual circles. Figure 17 is a state diagram showing the virtual connection path creation status for destination information. FIGS. 18 to 20 are state diagrams illustrating the process of generating an obstacle bypass path, including an obstacle detection process. FIGS. 21 to 23 are state diagrams illustrating the process of generating a hybrid candidate path. Figures 24 to 28 are state diagrams illustrating the process of calculating hybrid candidate path costs. FIGS. 29 to 34 are example diagrams of the hybrid path calculation process. FIGS. 35 to 41 are flowcharts of a control method for a hybrid path-driving mobile robot system according to the present invention. Specific details for implementing the invention
[0048] Hereinafter, specific details for implementing the configuration of the mobile robot unit, mobile robot system, and control method thereof according to the present invention are described based on examples with reference to the drawings. These examples are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that various embodiments of the present invention are different but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the present invention in relation to one embodiment. Furthermore, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the following detailed description is not intended to be limiting, and the scope of the present invention is limited only by the appended claims, including all equivalents to those claimed by the claims, provided they are properly described. Similar reference numerals in the drawings refer to the same or similar functions across various aspects.
[0049] Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.
[0050] In the embodiments described below, the server may be implemented as a standalone device or as part of a general-purpose processing unit in the form of a collection of hardware including a central processing unit, a user interface, an operating system (not shown), a memory for storing various necessary data, an external communication port, a printing device interface, etc., which executes various software programs and / or instruction sets stored in a memory unit to perform various functions required for the present invention described below, and software that drives the same.
[0052] A mobile robot unit (10) according to one embodiment of the present invention can communicate with a mobile robot server (not shown), and in some cases, a manager terminal (not shown), such as a computer of a work manager, may also be included in the overall system to form a hybrid path driving mobile robot system. Each component constituting these mobile robot systems (1) can communicate with each other through a communication network and transmit and receive data.
[0053] FIG. 1 schematically illustrates the positional state within a work environment in which a mobile robot unit (10) detects the work environment and acquires detection information through a unit detection unit (11), such as a LiDAR detection unit (111, described later). The mobile robot unit (10) moves from the current position information to the destination information (Io), and a virtual connection path (Pv) is shown that includes a constraint path (Pc) composed of node information (Nc) and link information (Lc) described later, and link information (Lv) connecting them, and a driving state diagram is shown in which the mobile robot unit (10) bypasses an obstacle (Obst) on the constraint path (Pc).
[0054] Here, communication, communication network, or communication network may include, for example, at least one of a cellular communication protocol such as LTE, LTE-A, 5G, WCDMA, CDMA, UMTS, Wibro, or GSM. Additionally, the communication network may be configured regardless of the type of communication, such as wired or wireless, and may be implemented as various communication networks such as a Personal Area Network (PAN), a Local Area Network (LAN), a Metropolitan Area Network (MAN), or a Wide Area Network (WAN). Furthermore, the communication, communication network, or communication network may be the known World Wide Web (WWW) and may include various communication methods such as Infrared Data Association (IrDA), Bluetooth, or Bluetooth Low Energy and RF wireless communication.
[0055] First, the mobile robot unit (10) of the hybrid path driving mobile robot system may be provided with one or more units. The mobile robot unit (10) drives a reference path, for example, a path pre-set to be mainly used, to execute a predetermined task, and enables stable task execution. The mobile robot unit (10) of the present invention enables a series of driving and task executions by extracting accurate information about the surrounding environment. At this time, the present invention performs path search generation to enhance adaptability to areas other than the pre-set path and to achieve accurate driving under such a work environment.
[0056] The mobile robot unit (10) of the present invention can be implemented in various types, such as a differential type, a left-right 2WS-2WD type, a front-rear 2WS-2WD type, a front-wheel WS-WD tricycle type, and is not limited to a specific type.
[0057] More specifically, FIGS. 2 and 3 illustrate a hybrid path-driving mobile robot system, wherein the mobile robot unit (10) of the hybrid path-driving mobile robot system includes a unit detection unit (11), a unit control unit (12), a unit storage unit (13), a unit calculation unit (14), a unit input unit (15), a unit communication unit (16), a unit output unit (17), and a unit driving unit (18).
[0058] The unit detection unit (11) is a detection means for acquiring information on the driving state and surrounding environment of the mobile robot unit (10). As shown in FIG. 3, the unit detection unit (11) includes a laser sensor. The laser sensor is implemented as a LiDAR sensor (111) and is a component that performs a scan detection function. It detects the presence of other mobile robot units driving in opposite directions requiring driving or passing, or objects that may collide, i.e., obstacles that affect driving motion, and extracts point cloud-shaped data of surrounding objects such as fixed pillars and walls, thereby enabling map creation and position estimation of the driving environment.
[0059] Additionally, the unit detection unit (11) may further include an encoder detection unit (115) as a motor detection unit or wheel detection unit that detects the driving operation of the unit driving unit. Additionally, depending on the case, an image sensor (not shown) may capture an image of the surroundings while the mobile robot unit (10) is driving and convert and provide image information.
[0060] The unit drive unit (18) includes a unit drive motor (not shown), a unit drive transmission (not shown), and a unit drive wheel (not shown). Caster wheels may be positioned at the front and rear of the center of the mobile robot unit (10) in FIG. 1 as auxiliary wheels to enable stable driving. Here, the unit drive motor (not shown) includes a driving drive motor that causes rotation of the unit drive wheel (not shown), as well as a steering drive motor for steering the unit drive wheel (not shown).
[0061] Here, the unit detection unit (11) may accumulate errors or experience a decrease in precision due to wear and tear from use or load from usage frequency caused by exposure to various environments, such as deformation due to impact during use, and thus a predetermined calibration may be performed. For example, the target path may be a rectangular path, but the actual path may take a deformed path in the form of movement and rotation. Therefore, regarding the sensor output value of the unit detection unit (11) of the present invention, the unit control unit (12) may implement a correction execution through a self-correction calibration operation.
[0062] The unit control unit (12) may apply a detection operation control signal to the unit detection unit (11) of the mobile robot unit (10), receive the detected detection information to generate a map or enable location estimation, and transmit it to the mobile robot server (not shown) to generate a map. The mobile robot path generation unit described below is explained mainly in a structure placed on the server side, but various configurations are possible depending on the usage environment, such as being directly mounted on the mobile robot unit.
[0063] The unit control unit (12) includes a map generation control module (121), a alignment control module (126), a position estimation control module (127), a calibration control module (128), and a drive control module (129). Here, the drive control module (129) can control the drive operation of a motor that generates driving power as well as a motor that generates steering power.
[0064] The map generation control module (121) generates a map through the occupancy status of cells within the map corresponding to the driving environment from the detection information, and the map generation control module (121) may be individually equipped with a mobile robot path generation unit described later.
[0065] The alignment control module (126) enables the position estimation of the mobile robot unit (10) through an alignment process using sensing information. In this embodiment, it may be executed using the ICP (Iterative Closest Point) method, but the alignment control method used in the alignment control module (126) is not limited to this and various methods may be used.
[0066] The position estimation control module (127) can perform position estimation using the transformation matrix calculated using the reference position and other information and the LiDAR detection information at the current position, using the matching result from the matching control module (126).
[0067] The unit storage unit (13) is operated according to the unit storage control signal of the unit control unit (12) and may perform the function of storing unit prior data for driving operation or storing update records, etc. through image information or detection data in the form of a point cloud detected by the unit detection unit (11), calibration, path information, obstacle information, etc.
[0068] The unit calculation unit (14) may execute a predetermined calculation process according to the unit calculation control signal of the unit control unit (12) by utilizing unit dictionary data stored in the unit storage unit, and may also perform calculation functions in the process of calculating transformation matrices such as movement and rotation during the alignment process. In addition, it may be possible to directly input data to the mobile robot unit (10) through the unit input unit (not shown), in addition to input through a terminal (not shown) of a user or management operator.
[0069] The unit communication unit (16) communicates with the mobile robot server (not shown) via wired or wireless communication, and can receive motion control signals transmitted from the mobile robot server (not shown) or transmit detection data detected by the unit detection unit (11) of the mobile robot unit (10).
[0070] The unit output section (17) is attached to the mobile robot unit (10) to provide an external output of a signal, and the unit output section (17) can be implemented with various output elements such as a speaker or a display.
[0071] As described above, the unit drive unit (18) includes a unit drive motor (181), a unit drive transmission (183), and a unit drive wheel (185), and includes various power generation and power transmission means such as wheels, drive motors, and reduction gears of the mobile robot unit. As described above, the mobile robot unit (10) of this embodiment is not limited to a specific drive type and can be varied in various ways, such as a single drive type, a differential type, a left-right 2WS-2WD type, a front-rear 2WS-2WD type, or a front-wheel WS-WD tricycle type.
[0072] The mobile robot server (20, FIG. 4) may create a map of the driving environment detected by one or more mobile robot units (10) and perform position estimation.
[0073] The server control unit (22) receives detection information detected from the mobile robot unit (10) to enable the creation of a map or location estimation, and may also transmit it to the mobile robot unit (10) to move and create a map.
[0074] The server storage unit (23) is operated according to the server storage control signal of the server control unit (22) and may perform the function of storing server prior data for operation or storing image information or detection data in the form of a point cloud detected by the unit detection unit (11) of the map-based mobile robot unit. The server computation unit (24) may execute a predetermined computation process according to the server computation control signal of the server control unit (22) by utilizing the server prior data, etc. stored in the server storage unit.
[0075] As described below, the mobile robot path generation unit (21) generates a driving path for the mobile robot unit (10). Although the mobile robot path generation unit (21) is described as being provided as a separate component in this embodiment, it may be included within the server control unit (22), and various configurations are possible within the scope of performing its inherent function.
[0076] This series of mobile robot servers (not shown) may create a map of the driving environment detected by one or more mobile robot units (10) and perform position estimation. It may also receive information detected from the mobile robot units and perform position estimation of the mobile robot units using the overall detected information, and may be configured to perform path search and generation for driving from a given location or current location to a destination and transmit it to the robot units in real time.
[0078] Hereinafter, with reference to the drawings of the present invention, a hybrid path-driving mobile robot system (1) and a control method thereof, comprising a mobile robot server (20) and / or a mobile robot unit (10) including a mobile robot path generation unit (21) according to the present invention, will be described. As described above, the mobile robot path generation unit (21) is configured to be mounted on the mobile robot server (20), but in some cases, it may be configured to be mounted on the mobile robot unit (10), and the description will focus on the configuration in which the mobile robot path generation unit (21) is mounted on the mobile robot server (20).
[0080] First, the mobile robot unit (10) according to the present invention performs driving in a work environment including pre-set node information and edge information and executes assigned or assigned tasks. In this embodiment, the mobile robot server (2) includes a server control unit (20), and the server control unit (200) is equipped with a mobile robot path generation unit (21).
[0081] The mobile robot path generation unit (21) in the hybrid path driving mobile robot system (1) of the present invention generates and determines a hybrid path for driving the mobile robot.
[0082] The constraint path consists of pre-set node information and edge information connecting the nodes, which are constrained or restricted in order for the mobile robot unit (10) to perform major movements while driving in a work environment. That is, typically, the mobile robot unit (10) of the present invention drives or performs a predetermined task in the work environment, and the current location information (Ip) refers to the occupied node information regarding the current occupied location of the mobile robot unit (10), and the destination information (Io) refers to the destination information to which the mobile robot unit (10) must move.
[0083] A constrained path refers to a pre-configured path in a working environment that must be used to move from current location information (Ip) to destination information (Io), and the constrained path is formed through node information and edge information. A constrained path may refer to a typical driving path primarily used in a working environment, in a form that excludes constraints such as driving being prohibited in the working environment.
[0084] Accordingly, if the current location information (Ip) and destination information (Io) exist on a constraint path and no separate adjustment is required, the constraint path between the current location information (Ip) and destination information (Io) can be set as the driving path used for driving. For example, if the current location information (Ip) and destination information (Io) exist on a constraint path and the constraint path is continuous, the driving path from the starting location information to the destination information can be identified using the knob information and edge information.
[0085] However, conversely, when a discontinuous state is formed between the current location information (Ip), the destination information (Io), and the constraint path, that is, when at least one of the current location information (Ip) and the destination information (Io) does not exist on the constraint path or when the constraint path is fragmented rather than a single path, the mobile robot path generation unit (21) generates a hybrid path through a combination of the constraint path and the virtual connection path by creating a virtual connection path as a connection path with the previous constraint path, rather than pre-configured node information and edge information, in order to ensure continuity between the current location information (Ip), the destination information (Io), and the constraint path.
[0086] In particular, as illustrated in FIG. 5, the mobile robot path generation unit (21) of the present invention comprises a location information verification unit (211), a path attribute verification unit (212), a path connectivity verification unit (213), a virtual connection path information generation unit (214), and a hybrid path information determination unit (218).
[0087] First, the location information verification unit (211) verifies the current location information and destination information of the mobile robot unit (10) that is driving in a work environment that includes pre-set node and edge information. That is, it verifies the current location status by verifying the node or edge information currently occupied by the mobile robot unit (10), and verifies the destination information as node information for the destination to which the mobile robot unit (10) must move.
[0088] The path attribute verification unit (212) verifies path information using the current location information and destination information of the mobile robot unit (10) verified by the location information verification unit (211). More specifically, the path attribute verification unit (212) verifies a constraint path composed of node information and edge information required to move from at least the current location information to the destination information. That is, the path attribute verification unit (212) verifies the attributes of the path, i.e., the constraint path, by using the current location information and destination information to select and verify constraint information suitable for movement among the constraint information as numerous path information for the work environment stored in the server storage unit (23), etc., the path attribute verification unit (212) can verify the constraint path, which is composed of the node information used and the edge information connecting them.
[0089] The path connectivity verification unit (213) verifies the connectivity between the current location information (Ip), destination information (Io), and constraint path (Pc). That is, the path connectivity verification unit (213) verifies whether the current location information (Ip), destination information (Io), and constraint path (Pc) form a continuous path configuration without separation.
[0090] Here, checking connectivity between current location information (Ip), destination information (Io), and constraint path (Pc) refers to checking whether the current location information (Ip) and destination information (Io) are on or off the constraint path (Pc), or whether the constraint path (Pc) is not a single path but is fragmented into multiple separate paths. The path connectivity checking unit (213) checks for such separated location states or fragmented paths. Through this, the necessity of securing a path that forms continuity between separated location states or fragmented paths can be confirmed.
[0091] Accordingly, the connectivity between the current location information (Ip), destination information (Io), and constraint path (Pc) is verified through the path connectivity verification unit (213), and if connectivity is not secured, a predetermined operation to secure connectivity is executed through the virtual connection path information generation unit (214).
[0092] As illustrated in FIG. 6, the virtual connection path information generation unit (214) generates virtual nodes and virtual links when a non-connection is confirmed in the path connectivity verification unit (213). Through the generated virtual nodes and virtual links, the current location information, destination information, and constraint path can all be connected, and the generated virtual nodes and virtual links form a virtual connection path. That is, when the current location information (Ip) and destination information (Io) are not on the constraint path (Pc), the current location information (Ip) or destination information (Io) is connected to the constraint path (Pc) through the virtual connection path, and the constraint path (Pc), which is fragmented or in a discontinuous state into multiple parts, can form a single path driving path through the virtual connection path.
[0093] Through this, the hybrid path information determination unit (218) generates a hybrid path using the virtual connection path generated by the constraint path and the virtual connection path information generation unit (214).
[0095] Here, more specifically, the virtual connection path information generation unit (214) includes a virtual circle formation unit (2141), a virtual node generation unit (2143), a virtual link generation unit (2145), and a virtual connection path generation unit (2147).
[0096] If it is confirmed in the path connectivity verification unit (213) that a single path has not been secured by connecting the current location information (Ip), destination information (Io), and constraint path (Pc), the virtual circle forming unit (2141) forms a virtual circle from at least one of the terminal nodes of the current location information (Ip), destination information (Io), and constraint path (Pc) that are not connected from the constraint path.
[0097] At this time, the virtual circle formed by the virtual circle forming unit (2141) may be formed around current location information or destination information. Additionally, the virtual circle formed by the virtual circle forming unit (2141) may be formed around any terminal node of the constraint path.
[0098] That is, the virtual circle forming unit (2141) sets the starting point for forming a virtual virtual connection path as at least one of the current location information (Ip) and destination information (Io) that is not on the constraint path, or even if both the current location information (Ip) and destination information (Io) exist on the constraint path (Pc), if the constraint path (Pc) is formed into multiple segmented or fragmented paths and cannot form a single path, the terminal node information of the segmented or fragmented multiple paths is formed as the center point of the virtual circle (Cv), and the radius (r; r1, r2) is expanded.
[0099] The virtual node generation unit (2143) forms a virtual node (Nv) at the point of contact between the virtual circle (Cv) and the strain path (Pc). At this time, even if the first virtual node (Nv) is identified, it is possible to check whether there are additional virtual nodes by increasing the radius within a preset radius change range, or to derive virtual nodes by increasing the preset radius change value until a preset number of contact points occur, and various change options are possible depending on the design specifications.
[0100] FIG. 10 illustrates a case where destination information (Io) exists outside the constraint path (Pc), and a virtual circle (CV; CV1, CV2) is formed around the destination information (Io). Through the virtual node generation unit (2143), a virtual node (Nv) is formed at the contact point between the virtual circle (Cv) and the constraint path (Pc). When the virtual node (Nv) is formed at the contact point between the virtual circle (Cv) and the constraint path (Pc) through the virtual node generation unit (2143), the virtual link generation unit (2145) generates a virtual link (Lv), and the virtual link (Lv) connects the virtual node (Nv) with at least one of the current location information (Ip) and destination information (Io), or connects a discontinuous constraint path (Pc).
[0101] In FIGS. 11 and 12, when a point is located outside the constraint path (Pc), a virtual node (Nv) that is an intersection with the virtual circle and a virtual link (Lv) are formed by connecting the virtual node and the point, and the constraint path (Pc) and the virtual link (Lv) can take an orthogonal structure. When forming a virtual link structure through such an orthogonal structure, the creation of a virtual link with an inclined structure as shown in FIGS. 13 (a) and (b) is not appropriate.
[0102] That is, these virtual nodes (Nv) and virtual links (Lv) are node information and edge information newly formed under those conditions through virtual circles (Cv), rather than node information and edge information pre-set in the working environment.
[0103] The virtual connection path generation unit (2147) generates a virtual connection path (Pv) using a virtual node (Nv) and a virtual link (Lv).
[0104] Meanwhile, in the preceding embodiment, when setting the constraint path and the virtual path, instead of setting the orthogonal contact point as a virtual node, a virtual connection path may be formed having an inclined virtual path (Lv1, Lv2, Lv3) that connects the point forming the angle between the constraint path and the virtual path to the target point (Io) using a virtual node (Nv; Nv1, Nv2, Nv3), wherein the point forming the preset angle (θ; θ1, θ2, θ3) in the backward direction from the orthogonal point is set as a virtual node, and a virtual path is generated through a virtual link. When such an inclined virtual path is adopted, a method for generating virtual nodes for selecting a driving path can be adopted to prevent abrupt angle changes when entering destination information (Io) that is separated from the constraint path, thereby increasing the cost in the link information described later but reducing the turning driving cost in the node information to minimize risk costs. At this time, the preset angles (θ; θ1, θ2, θ3) of the inclined virtual path may be varied in magnitude according to comprehensive information, such as the condition of the road surface in the driving environment and the load condition of the object being transported (not shown) in the driving environment. In selecting such virtual nodes, a more appropriate path search configuration can be achieved by considering the conditions of the working environment and the driving environment.
[0105] As illustrated in FIGS. 15 and 16, the interval of the expanding radius can be increased by setting a preset radius change value, the radius of the virtual circle (Cv1, Cv2) changes to R1, R2, the point of contact between the constraint path (Pc) and the virtual circle is set as a virtual node (Nv1, Nv2), and a virtual path is established consisting of destination information (Io), virtual nodes (Nv1, Nv2), and virtual links (Lv1, Lv2) connecting them. Through this process, multiple virtual paths may be formed.
[0107] That is, the mobile robot path generation unit (21) includes a hybrid path candidate generation unit (215). The hybrid path candidate generation unit (215) generates a hybrid path candidate (Pcnd) using a virtual connection path and a constraint path generated by the virtual connection path information generation unit (214). In cases where only one single driving path is formed through the generated virtual connection path and the pre-set constraint path, a separate selection process may be unnecessary as a single candidate path. However, in the case of the present invention, multiple single paths may be formed by the virtual connection path and the constraint path. In this case, the hybrid path candidate generation unit (215) may generate multiple hybrid path candidates using the virtual connection path and the constraint path generated by the virtual connection path information generation unit (214).
[0108] As illustrated in FIG. 17, a virtual connection path formed by virtual nodes and virtual links through multiple virtual circles is created, and multiple hybrid path candidates (Pcand1, Pcand2) described later are generated through combination with a constraint path.
[0110] (Obstacle detection)
[0111] In addition, a mobile robot path generation unit (21) according to an embodiment of the present invention may further include an obstacle path verification unit (216), wherein the obstacle path verification unit (216) verifies obstacle information by checking whether there is an obstacle on the path among hybrid path candidates, and if there is obstacle information, generates an obstacle bypass path.
[0112] More specifically, as illustrated in FIG. 7, the obstacle path verification unit (216) includes an obstacle information verification unit (2161) and an obstacle bypass path generation unit (2163). The obstacle information verification unit (2161) verifies the existence of obstacle information indicating that there is an obstacle on the path among the hybrid path candidates.
[0113] The obstacle bypass path generation unit (2163) generates an obstacle bypass path using obstacle information when the existence of obstacle information is confirmed.
[0114] At this time, the obstacle bypass path generation unit (2163) generates a path that bypasses an obstacle. When an obstacle within a hybrid path candidate formed by a constraint path and a virtual connection path is on the constraint path, the obstacle bypass path branches off from the constraint path, bypasses the obstacle, and returns to the constraint path. When an obstacle is on the virtual connection path, the obstacle bypass path branches off from the virtual connection path, bypasses the obstacle, and returns to the virtual connection path. This is merely a configuration described in one embodiment of the present invention, and various variations are possible, such as connecting from the constraint path to the virtual connection path or forming a connection in the opposite case.
[0115] In some cases, the obstacle bypass path generation unit (2163) may include a bypass path generation unit (21631) and a path smoothing unit (21633). In this embodiment, when generating an obstacle bypass path, the bypass path generation unit (21631) generates a path that bypasses the obstacles found using an A* path search algorithm. Various options are possible, such as using the Dijkstra algorithm when searching for an obstacle bypass path. Then, the A* path, which is the found obstacle bypass path, is smoothed through the smoothing unit (21633) to derive an optimized curved path using interpolation methods such as spline curve interpolation, methods such as Bézier curves, or connecting curved paths of key points using the least squares method or gradient descent method. By curving the obstacle bypass path in this way, the smoothness of the path is maximized to prevent dangerous conditions such as rollover during driving due to sudden changes, and the original purpose of obstacle avoidance is achieved without compromising the intent of the original hybrid path candidate, while maintaining similarity to the original path.
[0116] FIGS. 18 to 20 illustrate an example of an obstacle detection process. In the process of moving from current location information (Ip) to destination information (Io), obstacles (Obst), such as static obstacles like fixed obstacles or dynamic obstacles like other mobile robot units, are identified through the LiDAR detection unit (111) of the unit detection unit (11) while moving along a hybrid path candidate (Pcand). FIG. 18 illustrates the case where no separate obstacle is detected during the obstacle detection process before the hybrid path candidate is generated, FIG. 19 illustrates the case where no obstacle is detected during driving after the hybrid path candidate is generated, and FIG. 20 illustrates the case where an obstacle (Obst) is detected while moving along a hybrid path formed by a virtual connection path.
[0117] At this time, such detection is performed by searching for fixed static obstacles along the hybrid path discovered from the initial starting point, and the expected location of another mobile robot unit located on the hybrid path according to the expected schedule may be considered as a dynamic obstacle for search.
[0118] In addition, since this obstacle detection has the characteristic of a hybrid path that is updated in real time even if it is a confirmed hybrid path, when moving from the point shown in FIG. 18 and FIG. 19 to occupy the point shown in FIG. 20 as current location information, the obstacle detection process is executed again at the point moved on the hybrid path to generate new path information. That is, in the case of the present invention, a continuous obstacle detection process is executed while driving along the hybrid path, and the update of the hybrid path through this may be executed.
[0119] In addition, the cost for the path candidate (Pcnd) generated by the hybrid path candidate generation unit (215) of the mobile robot path generation unit (21) is calculated. In this embodiment, the cost is calculated after the path candidate is derived, but depending on the case, various options are possible, such as executing it simultaneously during path generation.
[0120] The mobile robot path generation unit (21) of the present invention includes a hybrid path candidate cost calculation unit (217). That is, the hybrid path candidate cost calculation unit (217) calculates and generates a cost for a hybrid path candidate (Pcnd) including a constraint path (Pc), a virtual connection path (Pv), and an obstacle bypass path (Pd), according to a preset cost standard.
[0121] And, the hybrid path information determination unit (218) generates a hybrid path from at least a constraint path and a virtual connection path.
[0122] In the case where a single hybrid path candidate (Pcnd) is generated by the mobile robot path generation unit (21) of the present invention, or a plurality of such candidates are formed, the hybrid path candidate cost calculation unit (217) calculates a cost for each hybrid path candidate (Pcnd) according to a preset cost standard, and the hybrid path information determination unit (218) selects a path having the minimum cost according to the preset standard, and the final hybrid path candidate according to the preset standard is selected as the hybrid path (Phyb) as the driving target path.
[0124] As illustrated in FIG. 8, the hybrid path candidate cost calculation unit (217) includes a constraint cost calculation unit (2171). The constraint cost calculation unit (2171) calculates the constraint link node cost for link driving and node turning driving from the link information and node information of the constraint path of the hybrid path candidate.
[0125] In this embodiment, the constraint link node cost (CoC) includes the link cost (CCL) for link travel and the node cost (CCR) for rotation travel.
[0126] The link cost (CCL) for link travel is calculated from the link information of the constraint path of the hybrid path candidate, and refers to a pre-set cost or the sum thereof for the movement and passing or occupying node information on the edge link edge information used to move from one node to another node.
[0127] The node cost for turning (CCR) is calculated from the node information of the hybrid path candidate, and refers to the pre-allocated cost for turning, such as a 90-degree turn, in the node information of the corresponding node.
[0128] The constraint cost calculation unit (2171) calculates a constraint link node cost (CoC) including a link cost (CCL) for link driving and a node cost (CCR) for turning driving.
[0129] Additionally, the hybrid path candidate cost calculation unit (217) includes a virtual cost calculation unit (2173), and the virtual cost calculation unit (2173) calculates virtual link node costs for link driving and node rotation driving from virtual link information and virtual node information of the virtual connection path of the hybrid path candidate.
[0130] Here, the virtual link node cost (CoV) includes the link cost (CVL) for link travel and the node cost (CVR) for turn travel in a pre-configured manner, similar to the case of the preceding constraint cost; the link cost (CVL) for link travel is calculated from the link information of the virtual connection path of the hybrid path candidate, and the node cost (CVR) for turn travel is calculated from the node information of the hybrid path candidate.
[0131] That is, the sum of the occupied nodes, the link edges connecting them, and the rotation costs allocated to rotation operations on the nodes are calculated for a virtual connection path formed to form a single continuous path from multiple fragmented constraint paths where current location information or destination information does not exist on the constraint path.
[0132] Additionally, the hybrid path candidate cost calculation unit (217) includes an obstacle bypass expansion cost calculation unit (2175), the obstacle bypass expansion cost calculation unit (2175) calculates an obstacle bypass expansion cost (CDE) for an obstacle bypass path, and the obstacle bypass expansion cost (CDE) is calculated for a path generated by the bypass path generation unit (21631) and the path smoothing unit (21633) of the obstacle bypass path generation unit (2163), with a preset obstacle bypass cost calculated.
[0133] In this case, the Constraint Link Node Cost (CoC) refers to the cost for a constraint path confirmed to be formed through the node and link information used among the actual candidate paths among the pre-configured constraint paths, and the costs for overlapping node and link information among the Constraint Link Node Cost (CoC) and Virtual Link Node Cost (CoV) are calculated such that overlapping calculations are excluded.
[0134] FIGS. 21 and 22 illustrate the relationship of cost details for each path element for hybrid candidate path 1 (Pcand1) and hybrid candidate path 2 (Pcand2) with respect to a constraint link node cost (CoC) including a link cost (CCL) for link driving and a node cost (CCR) for turning driving according to an embodiment of the present invention, and a virtual link node cost (CoV) including a link cost (CVL) for link driving and a node cost (CVR) for turning driving. Here, hybrid candidate path 1 (Pcand1) refers to a candidate path passing through the first virtual circle (Cv1) and the virtual node (Nv1) which is the point of contact of the constraint path, and hybrid candidate path 2 (Pcand2) refers to a candidate path passing through the second virtual circle (Cv2) and the virtual node (Nv2) which is the point of contact of the constraint path, and the hybrid candidate path cost (CoCND) is calculated for the hybrid candidate path (Pcnd) considering the constraint link node cost (CoC) and virtual link node cost (CoV) for each of these hybrid candidate paths.
[0135] At this time, the hybrid path candidate cost (CoCND) may include the obstacle bypass expansion cost (CDE). In this embodiment, the obstacle bypass expansion cost (CDE) includes the gradient bypass expansion cost (CDGE), the heuristic bypass expansion cost (CDHE), the congestion bypass expansion cost (CDCE), and the slope bypass expansion cost (CDSE).
[0136] The gradient bypass expansion cost (CDGE) represents the cost for a multi-directional bypass path formed to bypass an obstacle, and the heuristic bypass expansion cost (CDHE) represents the cost for a straight non-bypass path that does not bypass an obstacle. In this case, as shown in FIG. 24, the multi-directional bypass path is a path that bypasses an obstacle (Obst) from current location information (Ip) to destination information (Io) and is composed of a multi-directional bypass link (Lg) and a multi-directional bypass node (Ng). As illustrated in the drawing, when an obstacle exists on the path, the gradient bypass expansion cost (CDGE) is calculated using a pre-set path cost for a polygonal bypass node (Ng) formed for bypassing an optimized polygonal bypass path (Pg) that bypasses the obstacle (Obst) detected from the previous branching point and returns to the original path side, for example, the constraint path or the virtual connection path side, and polygonal link information (Lg), which is edge information.
[0137] Additionally, as illustrated in FIG. 24, the heuristic bypass expansion cost (CDHE) represents the cost for a straight non-bypass path that bypasses an obstacle. As illustrated in the figure, the heuristic bypass expansion cost (CDHE) is an obstacle penetration path that bypasses an obstacle and penetrates straight through, from the point where the multi-angle bypass path branches off from the preceding path to the point where it merges again. The Manhattan distance between the two points is calculated through their link edge information and converted into a cost.
[0138] Here, a multi-angle bypass path is generated through the previously described bypass path generation unit (21631), and a curved bypass path (see FIG. 25 and FIG. 26) is generated through the path smoothing unit (21633). The Congestion Bypass Expansion Cost (CDCE) represents the cost incurred from obstacle detection for the curved bypass path calculated from the multi-angle bypass path, which is required to determine the degree of obstacles in the area through exploration along the curved bypass path when the presence of other obstacles or other mobile robot units is confirmed on the curved path.
[0139] At this time, the detection radius (ro1, ro2; ro1) according to obstacle detection in the detection unit (11) of the mobile robot unit (10) <ro2)을 차등화시킨 탐지 구조를 취하는 경우, 혼잡 우회 확장 비용(CDCE)는 다음과 같이 산출될 수 있다(도 25 및 도 27 참조).
[0140]
[0141] Here, w1 and w2 can be implemented as preset values as weights based on the obstacle detection range.
[0142] Additionally, the slope bypass expansion cost (CDSE) represents the cost due to the curve slope of the curve bypass path derived from the multi-angle bypass path. The slope is calculated using two points on the curve bypass path, and the cost due to the curve slope of the curve bypass path is calculated using a preset slope cost. For example, as illustrated in the drawings (see Figs. 26 and 28), in the case of a path smoothed curve bypass path, the cost may be derived by calculating the slope or gradient using two points for each point (S1, S2, S3, S4, ..., Sn) on the path.
[0143]
[0144] Here, ws represents the slope bypass extension weights, and sn+1 and sn represent vectors for points on each curve bypass path.
[0145] When multiple hybrid path candidates (Pcnd) generated by the mobile robot path generation unit (21) are formed through such a series of cost calculations, the hybrid path information determination unit (218) selects a final hybrid path candidate according to preset criteria, such as selecting a path with the minimum cost according to preset criteria.
[0147] Hereinafter, a control method for a hybrid path-driving mobile robot system according to the present invention will be described with reference to the drawings.
[0148] First, the hybrid path driving mobile robot system control method of the present invention generates and determines a hybrid path for driving the mobile robot, and this series of generation processes is executed in a mobile robot path generation unit (21).
[0149] That is, a hybrid path for driving a mobile robot is generated and determined in the mobile robot path generation unit (21), and the hybrid path for driving a mobile robot is derived using the current location information and destination information of the mobile robot unit and a constraint path. As described above, the current location information and destination information represent location information for executing driving in a work environment that includes pre-set node and edge information, and the constraint path refers to a pre-set path to be used that consists of node information and edge information necessary to move from the current location information to the destination information.
[0150] In such cases, when the current location information, destination information, and the constraint path are discontinuous, a virtual connected path is generated using the constraint path, current location information, and destination information, and a hybrid path for the mobile robot's driving is generated from the constraint path and the virtual connected path.
[0151] More specifically, as illustrated in FIG. 35, the control method for a hybrid path driving mobile robot system of the present invention includes a position information verification step (S10), a constraint path verification step (S20), a path connectivity verification step (S30), a virtual connection path information generation step (S40), and a hybrid path information determination step (S80).
[0152] The location information verification step (S10) is executed in the previously described location information verification unit (211), and in the location information verification step (S10), the current location information and destination information of a robot performing driving in a work environment including pre-set node and edge information are verified.
[0153] The constraint path verification step (S20) is executed in the path attribute verification unit (212). In the constraint path verification step (S20), the constraint path is verified. As previously described, the constraint path is one of the pre-set paths within the work environment and consists of node information and edge information necessary to move from the current location information to the destination information. In this case, depending on the circumstances, the constraint path verification step (S20) may verify not only the constraint path but also the continuity of the constraint path; however, this embodiment focuses on the case where it is executed in a separate path connectivity verification step.
[0154] After the constraint path verification step (S20) is executed, the path connectivity verification step (S30) is executed, and the path connectivity verification step (S30) is executed in the path connectivity verification unit (213). In the path connectivity verification step (S30), the connectivity between the current location information, destination information, and the constraint path is verified.
[0155] After the path connectivity verification step (S30) is executed, the virtual connection path information generation step (S40) is executed, and the virtual connection path information generation step (S40) is executed in the virtual connection path information generation unit (214). In the virtual connection path information generation step (S40), when a non-connection is confirmed in the path connectivity verification unit (213), a virtual node and a virtual link are generated to connect the current location information, destination information, and the constraint path.
[0156] After this process is completed, a hybrid path information determination step (S80) is executed, and the hybrid path information determination step (S80) is executed through a hybrid path information determination unit (218). In the hybrid path information determination step (S80), a hybrid path is generated from a constraint path and a virtual connection path.
[0158] Meanwhile, as illustrated in FIG. 36 in more detail, the virtual connection path information generation step (S40) includes a virtual circle formation step (S41), a virtual node generation step (S43), and a virtual link generation step (S45), and the virtual connection path information generation step (S40) is executed through a virtual connection path information generation unit (214).
[0159] The virtual circle formation step (S41) is executed in the virtual circle formation unit (2141). When a non-connection is confirmed in the path connectivity verification step (S30) during the virtual circle formation step (S41), a virtual circle is formed from at least one of the current location information, destination information, and the end of the constraint path.
[0160] At this time, the virtual circle formed in the virtual circle formation step (S41) can have its radius increased by a preset interval size centered on a selected center point.
[0161] At this time, the virtual circle formed in the virtual circle formation step (S41) can be formed around the current location information or destination information. That is, as shown in FIG. 10, when the current location information or destination information is not located on the constraint path (Pc) but is separated and located at a distanced point, a virtual circle is required, and the virtual circle formation step (S41) forms a virtual circle (Cv) around the current location information and / or destination information.
[0162] In addition, the virtual circle formed in the virtual circle formation step (S41) is not limited to being formed around current location information or destination information. Even if current location information or destination information exists on the constraint path (Pc), if the constraint path (Pc) is fragmented and separated, that is, formed into multiple separated paths, it is formed around one of the end nodes of any of the separated constraint paths. As illustrated in the drawing, one of the end nodes of the discontinuous constraint paths can be formed as the center point of the virtual circle (Cv), and the radius of the virtual circle (Rv) can be gradually changed.
[0163] After the virtual circle formation step (S41) is completed, the virtual node creation step (S43) is executed. The virtual node creation step (S43) is executed in the virtual node creation unit (2143), and in the virtual node creation step (S43), the contact point between the virtual circle (Cv) and the constraint path (Pc) is formed as a virtual node (Nv) through the virtual node creation unit (2143).
[0164] That is, as described above, when the virtual circle formed in the virtual circle formation step (S41) is formed around current location information or destination information, a point that comes into contact with the constraint path (Pc) when the virtual circle radius (Rv) increases around the current location information or destination information is formed as a virtual node (Nv). At this time, the virtual node may be formed only at the initial contact point, but additional contact points, such as a second or third point, may be formed as additional virtual nodes (Nv) according to the preset method and utilized as elements constituting multiple hybrid path candidates (Pcnd) described later.
[0165] In addition, when a constraint path (Pc) is fragmented and separated, that is, formed into multiple separated paths, when the virtual circle radius (Rv) increases around any terminal node of any of the separated constraint paths, the point in contact with the constraint path (Pc) is formed as a virtual node (Nv). It is evident that in these cases as well, additional contact points, such as a second or third one, can be formed as additional virtual nodes (Nv) according to the preset method described above and utilized as elements constituting multiple hybrid path candidates (Pcnd) described later.
[0166] Then, a virtual link creation step (S45) is executed. The virtual link creation step (S45) is executed through a virtual link creation unit (2145), and in the virtual link creation step (S45), a virtual link is created connecting a virtual node (Nv), current location information (Ip), destination information (Io), and at least one terminal node (Ne) among the terminals of the constraint path. That is, in the virtual link creation step (S45), when the current location information (Ip) and destination information (Io) are central, the virtual link (Lv) is formed by setting the virtual node (Nv) on the constraint path (Pc), the current location information (Ip), and the destination information (Io) as terminals.
[0167] Additionally, in the virtual link creation step (S45), when the current location information (Ip) and destination information (Io) are central, the virtual link (Lv) is formed by setting the virtual node (Nv) on the constraint path (Pc) and at least one terminal node (Ne) among the terminals of the constraint path as ends.
[0168] After the virtual link creation step (S45) is completed in this manner, the virtual connection path creation step (S47) is executed. The virtual connection path creation step (S47) is executed through the virtual connection path creation unit (2147), and in the virtual connection path creation step (S47), a virtual connection path (Pv) is created using a virtual node (Nv) and a virtual link (Lv). Through this process, a virtual connection path (Pv) is created by connecting the current location information (Ip), the destination information (Io), and the constraint path (Pc).
[0169] When a virtual connection path (Pv) is formed, a hybrid path information determination step (S80) may be executed, and the hybrid path information determination step (S80) is executed in the hybrid path information determination unit (218). In the hybrid path information determination step (S80), a hybrid path is generated from the constraint path and the virtual connection path. At this time, the hybrid path information determination unit (218) forms a single path that excludes overlapping areas rather than a simple overlap of the constraint path (Pc) and the virtual connection path (Pv), but when multiple virtual connection paths are generated as described above, one path among the multiple candidates may be selected through the hybrid path information determination unit (218).
[0171] Meanwhile, in the case of the present invention, a configuration including a hybrid path candidate generation step (S50) between the virtual connection path information generation step (S40) and the hybrid path information determination step (S80) may be adopted. In this case, the hybrid path candidate generation step (S50) is executed through the hybrid path candidate generation unit (215). In the hybrid path candidate generation step (S50), a hybrid path candidate (Pcnd) is generated using a virtual connection path (Pv) and a constraint path (Pc). If only one single driving path is formed through the generated virtual connection path (Pv) and the pre-set constraint path (Pc), a separate selection process may be unnecessary for a single candidate path. However, depending on the case, in the present invention, multiple single paths may be formed by the virtual connection path and the constraint path. In this case, in the hybrid path candidate generation step (S50), multiple hybrid path candidates (Pcnd) may be generated using the virtual connection path and the constraint path generated by the virtual connection path information generation unit (214) through the hybrid path candidate generation unit (215).
[0172] Between the hybrid path candidate generation step (S50) and the hybrid path information determination step (S80), an obstacle path verification step (S60) may be further provided in addition to the hybrid path candidate generation step (S50). That is, as illustrated in FIGS. 37 and 38, the obstacle path verification step (S60) is executed in the obstacle path verification unit (216). In the obstacle path verification step (S60), among the hybrid path candidates generated in the hybrid path candidate generation step (S50), obstacle information is verified, and if obstacle information (Iobst) is verified, an obstacle bypass path (Pobst) is generated.
[0173] The obstacle path verification step (S60) includes an obstacle information verification step (S61) and an obstacle bypass path generation step (S63), as illustrated in FIG. 37. In the obstacle information verification step (S61), obstacle information (Iobst) among the hybrid path candidates (Pcnd) is verified. At this time, the obstacle information (Iobst) may be a fixed static obstacle that is a component included in the preset data for the work environment, or a dynamic obstacle caused by a mobile robot unit different from the mobile robot unit currently being searched for paths, and various types of obstacle information may be verified.
[0174] Then, an obstacle bypass path generation step (S63) is executed (see FIG. 38), in which an obstacle bypass path generation step (S63) is generated using previously identified obstacle information (Iobst). At this time, the obstacle bypass path generation step (S63) is executed through an obstacle bypass path generation unit (2163), and as previously described, the obstacle bypass path generation unit (2163) may include a bypass path generation unit (21631) and a path smoothing unit (21633).
[0175] The obstacle bypass path generation step (S63) includes the obstacle bypass path calculation step (S631) and the obstacle bypass path smoothing step (S633). The obstacle bypass path calculation step (S631) is executed through the bypass path generation unit (21631) in this embodiment. Although it is described as generating a path to bypass obstacles discovered using the A* path search algorithm, it is the same as previously described that various options are possible, such as using the Dijkstra algorithm when searching for obstacle bypass paths. Then, the obstacle bypass path smoothing step (S633) is executed so that the A* path as the discovered obstacle bypass path is smoothed through the bypass path smoothing unit (21633), and an optimized curve path can be derived through interpolation methods such as spline curve interpolation, methods such as Bézier curves, or connecting curve paths of key points using the least squares method or gradient descent method.
[0177] The mobile robot path generation unit (21) may include a hybrid path candidate cost calculation unit (217), and the hybrid path candidate cost calculation step (S70, see FIG. 39) is executed through the hybrid path candidate cost calculation unit (217). In the hybrid path candidate cost calculation step (S70), a hybrid path candidate cost formed according to a preset cost standard is calculated for the hybrid path candidate. If there are multiple hybrid path candidates identified in the preceding step, the optimized hybrid path among the hybrid path candidates is determined by comparing the hybrid path candidate costs calculated in the hybrid path candidate cost calculation step (S70).
[0178] The hybrid path candidate cost calculation unit (217) includes a constraint cost calculation unit (2171). As illustrated in FIGS. 39 to 41, the hybrid path candidate cost calculation step (S70) includes a constraint cost calculation step (S71), and the constraint cost calculation step (S71) is executed through the constraint cost calculation unit (2171) to calculate the constraint link node cost for link driving and node rotation driving from the link information and node information of the constraint path (Pc) of the hybrid path candidate.
[0179] More specifically, the constraint cost calculation step (S71) includes a link cost calculation step (S711) and a node cost calculation step (S713). In the link cost calculation step (S711), a link cost (CCL) for link driving is calculated, and the link cost (CCL) for link driving is calculated from the link information of the constraint path of the hybrid path candidate.
[0180] In the node cost calculation step (S713), the node cost (CCR) for turning is calculated. The node cost (CCR) for turning is calculated from the node information of the constraint path (Pc) of the hybrid path candidate (Pcnd). The cost calculation follows a preset method. For example, various methods of cost calculation can be performed, such as adopting a structure where the cost increases as the angle of rotation deviating from the previous direction of travel in the corresponding node information increases.
[0181] Additionally, the hybrid path candidate cost calculation unit (217) further comprises a virtual cost calculation unit (2173), and the hybrid path candidate cost calculation step (S70) may include a virtual cost calculation step (S73), and the virtual cost calculation step (S73) is executed through the virtual cost calculation unit (2173). In the virtual cost calculation step (S73), the virtual link node cost for link driving and node rotation driving is calculated from the virtual link information (Lv) and virtual node information (Nv) of the virtual connection path (Pv) of the hybrid path candidate (Pcnd).
[0182] That is, the virtual cost calculation step (S73) includes a virtual link cost calculation step (S731) and a virtual node cost calculation step (S733). In the virtual link cost calculation step (S731), a link cost (CVL) for link driving is calculated from the link information (Lv) of the virtual connection path (Pv) of the hybrid path candidate (Pcnd). In the virtual node cost calculation step (S733), a node cost (CVR) for turning driving is calculated. The node cost (CVR) for turning driving is calculated from the node information (Nv) of the virtual connection path (Pv) of the hybrid path candidate (Pcdn). For example, the cost calculation follows a preset method. It is the same as previously described that various cost calculation methods may be performed, such as adopting a structure where the cost increases as the angle of rotation deviating from the previous direction of travel in the corresponding node information increases.
[0183] Meanwhile, the hybrid path candidate cost calculation step (S70) further includes an obstacle bypass expansion cost calculation step (S75). The hybrid path candidate cost calculation unit (217) includes an obstacle bypass expansion cost calculation unit (2175), and the obstacle bypass expansion cost calculation step (S75) is executed through the obstacle bypass expansion cost calculation unit (2175). In the obstacle bypass expansion cost calculation step (S75), the obstacle bypass expansion cost (CDE) for the obstacle bypass path is calculated.
[0184] More specifically, the obstacle bypass expansion cost calculation step (S75) includes the gradient bypass expansion cost calculation step (S751), the heuristic bypass expansion cost calculation step (S753), the congestion bypass expansion cost step (S755), and the congestion bypass expansion cost calculation step (S757).
[0185] In the step of calculating the gradient bypass expansion cost (S751), the gradient bypass expansion cost (CDGE) is calculated. The gradient bypass expansion cost (CDGE) is the cost for a multi-angle bypass path formed to bypass an obstacle, and refers to the cost incurred according to a preset standard when driving along it. As described above, the multi-angle bypass path is composed of a combination of link information as edges that are newly set up as nodes other than the constraint path within the grid map in the work environment, or as edges that connect the nodes in a straight line. The cost can be differentiated according to the angle between the link information and the rotation angle of the connected node information. It can take a cost structure that increases as the angle between the straight driving direction of the previous link information and the straight driving direction of the next link information increases, and the cost increases as the length of the link information increases. In some cases, if speed information is included in the link information, the cost may fluctuate according to the speed, and various variations are possible.
[0186] In the heuristic bypass expansion cost calculation step (S753), the heuristic bypass expansion cost (CDHE) is calculated, and the heuristic bypass expansion cost (CDHE) is the cost for a straight non-bypass path that does not bypass an obstacle, and the calculation process is carried out accordingly. Here, the heuristic bypass expansion cost (CDHE) can practically be formed by the points where the multi-angle bypass path branches off and the points where it rejoins. An increase in this heuristic bypass expansion cost (CDHE) means that the lengths of the branching points and the rejoining points increase. By simultaneously using the gradient bypass expansion cost (CDGE) for the multi-angle bypass path, the overall size of the obstacle can be determined through the distance in the direction of penetration of the mobile robot unit that must bypass the obstacle on the path and the distance in the lateral direction perpendicular thereto, and it is possible to calculate a more accurate bypass cost.
[0187] Additionally, in the congestion bypass expansion cost step (S755), the congestion bypass expansion cost (CDCE) is calculated. The congestion bypass expansion cost (CDCE) in the congestion bypass expansion cost step (S755) is the cost incurred due to obstacle detection for the curved bypass path calculated from the previously identified multi-angle bypass path, and the calculation process for this is executed. The congestion bypass expansion cost (CDCE) refers to the cost incurred as a detection process is executed to verify the presence or absence of additional static obstacles expected when driving the multi-angle bypass path, or the presence or absence of additional dynamic obstacles expected to affect other mobile robot units at that point in time. Through this process, the occurrence of congestion on the obstacle bypass path can be prevented.
[0188] Subsequently, the slope bypass expansion cost calculation step (S757) is executed. In the slope bypass expansion cost calculation step (S757), the slope bypass expansion cost (CDSE) is calculated, and the slope bypass expansion cost (CDSE) represents the cost resulting from the curve slope of the curve bypass path calculated from the multi-angle bypass path. The slope bypass expansion cost (CDSE) is calculated based on the slope gradient connecting two points on the curve bypass path, through comparison with a preset reference value or a preset arithmetic formula. Various variations are possible, such as using the slope gradient directly or assigning cost weights by identifying inflection points through the rate of change of the slope gradient.
[0189] FIGS. 29 to 34 illustrate an example of a hybrid path search process executed through the method described above, wherein the hybrid path search process is illustrated for the case where both the current location information (Ip) and the destination information (Io) are formed at a location separated from the constraint path (Pc), and FIGS. 29 to 34 illustrate a process of searching for a single hybrid path in the case where the current location information (Ip) and / or the destination information (Io) do not exist on the constraint path (Pc) and the constraint path is also formed as multiple paths rather than a single path.
[0190] First, FIG. 29 illustrates the formation state of virtual circles (Cv1, Cv2) for hybrid path search when the current location information (Ip) is separated from the constraint path (Pc) and the destination information (Io) is located on the constraint path (Pc). At this time, two contact points are formed from the current location information (Ip) to the constraint path (Pc), virtual nodes (Nv1, Nv2) are identified, and two virtual connection paths are formed through virtual links (Lv1, Lv2) connecting them. At this time, two hybrid path candidates are formed for each of the two virtual connection paths, and the hybrid path candidate (Pcand2) with the smaller cost value is selected as the final hybrid path (Phyb).
[0191] FIGS. 31 and 32 illustrate the formation state of virtual circles (Cv1, Cv2) for hybrid path search when current location information (Ip) and destination information (Io) are separated from the constraint path (Pc). In this case, two contact points are formed from the current location information (Ip) to the constraint path (Pc), virtual nodes (Nv1, Nv2) are identified, and two virtual connection paths are formed through virtual links (Lv1, Lv2) connecting them. Additionally, two contact points are formed from the destination information (Io) to the constraint path (Pc), virtual nodes (Nv3, Nv4) are identified, and two virtual connection paths are formed through virtual links (Lv3, Lv4) connecting them. At this time, four hybrid path candidates are formed for each of the four virtual connection paths, and the hybrid path candidate (Pcand3) passing through Nv1 and Nv4 with the lowest cost value is selected as the final hybrid path (Phyb).
[0192] FIGS. 33 and 34 illustrate a path search case in which the current location information (Ip) is separated from the constraint path (Pc1) and the two constraint paths (Pc1, Pc2) are separated. In FIG. 33, a virtual circle is formed from the current location information (Ip) to form a contact point that contacts the constraint path (Pc1) as a virtual node (Nv1), and a virtual circle is formed from the end node of the constraint path (Pc2) to the constraint path (Pc2) to form a contact point as a virtual node (Nv2), and a virtual link (Lv2) connecting the end node and the virtual node (Nv2) is formed. As shown in FIG. 34, a hybrid path candidate (Pcand) is formed, and a final hybrid path (Phyb) can be determined through cost calculation and verification.
[0194] Although the present invention has been illustrated and described above in relation to preferred embodiments for illustrating the principles of the invention, the invention is not limited to the configuration and operation as illustrated and described.
[0195] For example, although the mobile robot path generation unit is described primarily as being mounted on a mobile robot server, it may also be configured to be mounted on a mobile robot unit (10) to enable independent operation, and it is evident from the description above that such an independent structure also falls within the scope of the present invention. Rather, those skilled in the art will understand that numerous changes and modifications to the present invention are possible without departing from the spirit and scope of the appended claims. Explanation of the symbols
[0197] * Explanation of symbols for major parts of the drawing * 1...Mobile robot system 10...Mobile robot unit
Claims
Claim 1 A mobile robot path generation unit (21) comprises: a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a constraint path composed of node information and edge information required to move from at least the current position information to the destination information, and a virtual connection path generated using the constraint path, the current position information, and the destination information when the current position information and destination information and the constraint path are separated or the constraint path is discontinuous, thereby generating and confirming a hybrid path for driving a mobile robot; wherein the mobile robot path generation unit (21) comprises: a path connectivity verification unit (213) that verifies connectivity between the current position information, the destination information, and the constraint path, and a virtual connection path information generation unit (214) that generates a virtual node and a virtual link to connect the current position information, the destination information, and the constraint path when a non-connection is confirmed in the path connectivity verification unit (213); and the virtual connection path information A hybrid path driving mobile robot system (1) is characterized by comprising: a virtual circle forming unit (2141) that forms a virtual circle from at least one of the current location information and destination information and end nodes of the constraint path that are not connected from the constraint path when a non-connection is confirmed in the path connectivity checking unit (213); a virtual node generating unit (2143) that forms a virtual node at a contact point between the virtual circle and the constraint path; a virtual link generating unit (2145) that generates a virtual link connecting the virtual node and at least one of the current location information and destination information or connecting a non-continuous constraint path; and a virtual connection path generating unit (2147) that generates a virtual connection path using the virtual node and the virtual link. Claim 2 A hybrid path driving mobile robot system (1), wherein the mobile robot path generation unit (21) comprises: a position information verification unit (211) for verifying current position information and destination information of a robot performing driving in a work environment including preset node and edge information; a path attribute verification unit (212) for verifying a constraint path composed of node information and edge information required to move from the current position information to the destination information; and a hybrid path information determination unit (218) for generating a hybrid path from the constraint path and the virtual connection path. Claim 3 delete Claim 4 A hybrid path driving mobile robot system (1) characterized in that, in the first paragraph, the virtual circle formed by the virtual circle forming unit (2141) is formed around the current location information or the destination information. Claim 5 A hybrid path driving mobile robot system (1) characterized in that, in the first paragraph, the virtual circle formed by the virtual circle forming unit (2141) is formed around one end node of the constraint path. Claim 6 In claim 2, the mobile robot path generation unit (21) comprises: a hybrid path candidate generation unit (215) that generates a hybrid path candidate using the virtual connection path generated by the virtual connection path information generation unit (214) and the constraint path. A hybrid path driving mobile robot system (1). Claim 7 In claim 6, the mobile robot path generation unit (21) further comprises: an obstacle path verification unit (216) that verifies obstacle information among the hybrid path candidates and generates an obstacle bypass path, characterized in that the hybrid path driving mobile robot system (1). Claim 8 In claim 7, the obstacle path verification unit (216) comprises: an obstacle information verification unit (2161) that verifies obstacle information among the hybrid path candidates, and an obstacle bypass path generation unit (2163) that generates an obstacle bypass path using the obstacle information, characterized in that the hybrid path driving mobile robot system (1). Claim 9 In claim 7, the mobile robot path generation unit (21) comprises: a hybrid path candidate cost calculation unit (217) that calculates a hybrid path candidate cost formed according to a preset cost standard for the hybrid path candidate, characterized in that the hybrid path driving mobile robot system (1). Claim 10 In claim 9, the hybrid path candidate cost calculation unit (217) comprises: a constraint cost calculation unit (2171) that calculates a constraint link node cost for link driving and node rotation driving from the link information and node information of the constraint path of the hybrid path candidate, characterized in that the hybrid path driving mobile robot system (1). Claim 11 A mobile robot path generation unit (21) comprises: a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and, when the current position information and destination information and the constraint path are separated or the constraint path is discontinuous, a virtual connection path is generated using the constraint path and the current position information and destination information to generate and confirm a hybrid path for driving a mobile robot. The mobile robot path generation unit (21) comprises: a position information verification unit (211) that verifies the current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a path attribute verification unit (212) that verifies a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and a path connectivity unit that verifies the connectivity between the current position information, the destination information, and the constraint path. The mobile robot path generation unit (21) comprises a verification unit (213), a virtual connection path information generation unit (214) that generates a virtual node and a virtual link to connect the current location information and the destination information and the constraint path when a non-connection is confirmed in the path connectivity verification unit (213), and a hybrid path information determination unit (218) that generates a hybrid path from the constraint path and the virtual connection path, and the mobile robot path generation unit (21) comprises: a hybrid path candidate generation unit (215) that generates a hybrid path candidate using the virtual connection path and the constraint path generated in the virtual connection path information generation unit (214).The mobile robot path generation unit (21) further includes an obstacle path verification unit (216) that checks obstacle information among the hybrid path candidates and generates an obstacle bypass path, and the mobile robot path generation unit (21) includes a hybrid path candidate cost calculation unit (217) that calculates a hybrid path candidate cost formed according to a preset cost standard for the hybrid path candidate, and the hybrid path candidate cost calculation unit (217) includes a constraint cost calculation unit (2171) that calculates a constraint link node cost for link driving and node turning driving from the link information and node information of the constraint path of the hybrid path candidate, and the constraint link node cost includes a link cost (CCL) for link driving calculated from the link information of the constraint path of the hybrid path candidate, and a node cost (CCR) for turning driving calculated from the node information of the hybrid path candidate. Claim 12 In claim 9, the hybrid path candidate cost calculation unit (217) comprises: a virtual cost calculation unit (2173) that calculates a virtual link node cost for link driving and node rotation driving from virtual link information and virtual node information of the virtual connection path of the hybrid path candidate, characterized in that the hybrid path driving mobile robot system (1). Claim 13 A mobile robot path generation unit (21) comprises: a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and, when the current position information and destination information and the constraint path are separated or the constraint path is discontinuous, a virtual connection path is generated using the constraint path and the current position information and destination information to generate and confirm a hybrid path for driving a mobile robot. The mobile robot path generation unit (21) comprises: a position information verification unit (211) that verifies the current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a path attribute verification unit (212) that verifies a constraint path composed of at least node information and edge information required to move from the current position information to the destination information, and a path connectivity unit that verifies the connectivity between the current position information, the destination information, and the constraint path. The mobile robot path generation unit (21) comprises a verification unit (213), a virtual connection path information generation unit (214) that generates a virtual node and a virtual link to connect the current location information and the destination information and the constraint path when a non-connection is confirmed in the path connectivity verification unit (213), and a hybrid path information determination unit (218) that generates a hybrid path from the constraint path and the virtual connection path, and the mobile robot path generation unit (21) comprises: a hybrid path candidate generation unit (215) that generates a hybrid path candidate using the virtual connection path and the constraint path generated in the virtual connection path information generation unit (214).The mobile robot path generation unit (21) further includes an obstacle path verification unit (216) that checks obstacle information among the hybrid path candidates and generates an obstacle bypass path, and the mobile robot path generation unit (21) includes a hybrid path candidate cost calculation unit (217) that calculates a hybrid path candidate cost formed according to a preset cost standard for the hybrid path candidate, and the hybrid path candidate cost calculation unit (217) includes a virtual cost calculation unit (2173) that calculates a virtual link node cost for link driving and node turning driving from the virtual link information and virtual node information of the virtual connection path of the hybrid path candidate, and the virtual link node cost includes a link cost (CVL) for link driving calculated from the link information of the virtual connection path of the hybrid path candidate, and a node cost (CVR) for turning driving calculated from the node information of the hybrid path candidate. Claim 14 A hybrid path-driving mobile robot system (1) characterized in that, in claim 13, the virtual link node cost further includes an obstacle bypass extension cost (CDE) for an obstacle bypass path. Claim 15 A hybrid path driving mobile robot system (1), wherein the obstacle bypass expansion cost (CDE) comprises: a gradient bypass expansion cost (CDGE) for a multi-angle bypass path formed to bypass an obstacle, a heuristic bypass expansion cost (CDHE) for a straight non-bypass path that does not bypass an obstacle, a congestion bypass expansion cost (CDCE) due to obstacle detection for a curve bypass path calculated from the multi-angle bypass path, and a slope bypass expansion cost (CDSE) due to the curve slope for a curve bypass path calculated from the multi-angle bypass path. Claim 16 Using a constraint path composed of current position information and destination information of a robot performing driving in a work environment including preset node and edge information, and node information and edge information necessary to move from the current position information to the destination information, if the current position information and destination information and the constraint path are discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information, and a hybrid path for driving a mobile robot is generated and confirmed from the constraint path and the virtual connection path; a position information verification step (S10) in which the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified; a constraint path verification step (S20) in which a constraint path composed of node information and edge information necessary to move from the current position information to the destination information is verified; a path connectivity verification step (S30) in which connectivity between the current position information, the destination information, and the constraint path is verified; and the path connectivity When a non-connection is confirmed in the verification unit (213), a virtual connection path information generation step (S40) is included in which a virtual node and a virtual link are generated to connect the current location information, the destination information, and the constraint path. The virtual connection path information generation step (S40) comprises: a virtual circle formation step (S41) in which a virtual circle is formed from at least one of the current location information, the destination information, and the end of the constraint path when a non-connection is confirmed in the path connectivity verification step (S30); a virtual node generation step (S43) in which a contact point between the virtual circle and the constraint path is formed as a virtual node;A method for controlling a hybrid path driving mobile robot system, characterized by including a virtual link generation step (S45) in which a virtual link is generated connecting the virtual node, the current location information, the destination information, and at least one of the ends of the constraints, and a virtual connection path generation step (S47) in which a virtual connection path is generated using the virtual node and the virtual link. Claim 17 A method for controlling a hybrid path driving mobile robot system, characterized in that, in claim 16, it includes a hybrid path information determination step (S80) in which a hybrid path is generated from the above-mentioned constraint path and the above-mentioned virtual connection path. Claim 18 delete Claim 19 A hybrid path driving mobile robot system control method according to claim 16, wherein the virtual circle formed in the virtual circle forming step (S41) is formed around the current location information or the destination information. Claim 20 A method for controlling a hybrid path-driving mobile robot system, characterized in that, in claim 16, the virtual circle formed in the virtual circle formation step (S41) is formed around a terminal node of the constraint path. Claim 21 A method for controlling a hybrid path driving mobile robot system according to claim 17, characterized by including a hybrid path candidate generation step (S50) in which a hybrid path candidate is generated using the virtual connection path and the constraint path between the virtual connection path information generation step (S40) and the hybrid path information determination step (S80). Claim 22 A method for controlling a hybrid path driving mobile robot system according to claim 21, characterized in that, between the hybrid path candidate generation step (S50) and the hybrid path information determination step (S80), an obstacle path verification step (S60) is further included in which obstacle information among the hybrid path candidates generated in the hybrid path candidate generation step (S50) is verified and an obstacle bypass path is generated. Claim 23 A method for controlling a hybrid path driving mobile robot system, wherein, in claim 22, the obstacle path verification step (S60) comprises: an obstacle information verification step (S61) in which obstacle information among the hybrid path candidates is verified, and an obstacle bypass path generation step (S63) in which an obstacle bypass path is generated using the obstacle information. Claim 24 A method for controlling a hybrid path driving mobile robot system, characterized in that, in claim 21, it includes a hybrid path candidate cost calculation step (S70) in which a hybrid path candidate cost formed according to a preset cost standard is calculated for the hybrid path candidate. Claim 25 A method for controlling a hybrid path driving mobile robot system, wherein, in claim 24, the hybrid path candidate cost calculation step (S70) comprises: a constraint cost calculation step (S71) in which a constraint link node cost for link driving and node rotation driving is calculated from the link information and node information of the constraint path of the hybrid path candidate. Claim 26 A position information verification step (S10) in which the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified, and a constraint path composed of node information and edge information necessary to move from the current position information to the destination information, wherein if the current position information and destination information and the constraint path are discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information, and a hybrid path for driving a mobile robot is generated and confirmed from the constraint path and the virtual connection path, and the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified; a constraint path verification step (S20) in which a constraint path composed of node information and edge information necessary to move from the current position information to the destination information is verified; a path connectivity verification step (S30) in which connectivity between the current position information, the destination information, and the constraint path is verified, and the path connectivity When a non-connection is confirmed in the verification unit (213), the method includes a virtual connection path information generation step (S40) in which a virtual node and a virtual link are generated to connect the current location information and the destination information and the constraint path, and a hybrid path information determination step (S80) in which a hybrid path is generated from the constraint path and the virtual connection path, and between the virtual connection path information generation step (S40) and the hybrid path information determination step (S80), a hybrid path candidate generation step (S50) in which a hybrid path candidate is generated using the virtual connection path and the constraint path is included.A method for controlling a hybrid path driving mobile robot system, comprising: a hybrid path candidate cost calculation step (S70) in which a hybrid path candidate cost formed according to a preset cost standard is calculated for the hybrid path candidate; wherein the hybrid path candidate cost calculation step (S70) comprises: a constraint cost calculation step (S71) in which a constraint link node cost for link driving and node rotation driving is calculated from the link information and node information of the constraint path of the hybrid path candidate; and wherein the constraint cost calculation step (S71) comprises: a link cost calculation step (S711) in which a link cost (CCL) for link driving calculated from the link information of the constraint path of the hybrid path candidate is calculated, and a node cost calculation step (S713) in which a node cost (CCR) for rotation driving calculated from the node information of the hybrid path candidate is calculated. Claim 27 A method for controlling a hybrid path driving mobile robot system, wherein, in claim 24, the hybrid path candidate cost calculation step (S70) comprises: a virtual cost calculation step (S73) in which a virtual link node cost for link driving and node rotation driving is calculated from the virtual link information and virtual node information of the virtual connection path of the hybrid path candidate. Claim 28 A position information verification step (S10) in which the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified, and a constraint path composed of node information and edge information necessary to move from the current position information to the destination information, wherein if the current position information and destination information and the constraint path are discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information, and a hybrid path for driving a mobile robot is generated and confirmed from the constraint path and the virtual connection path, and the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified; a constraint path verification step (S20) in which a constraint path composed of node information and edge information necessary to move from the current position information to the destination information is verified; a path connectivity verification step (S30) in which connectivity between the current position information, the destination information, and the constraint path is verified, and the path connectivity When a non-connection is confirmed in the verification unit (213), the method includes a virtual connection path information generation step (S40) in which a virtual node and a virtual link are generated to connect the current location information and the destination information and the constraint path, and a hybrid path information determination step (S80) in which a hybrid path is generated from the constraint path and the virtual connection path, and between the virtual connection path information generation step (S40) and the hybrid path information determination step (S80), a hybrid path candidate generation step (S50) in which a hybrid path candidate is generated using the virtual connection path and the constraint path is included.A method for controlling a hybrid path driving mobile robot system, comprising: a hybrid path candidate cost calculation step (S70) in which a hybrid path candidate cost formed according to a preset cost standard is calculated for the hybrid path candidate; wherein the hybrid path candidate cost calculation step (S70) comprises: a virtual cost calculation step (S73) in which a virtual link node cost for link driving and node turning driving is calculated from virtual link information and virtual node information of the virtual connection path of the hybrid path candidate; and wherein the virtual cost calculation step (S73) comprises: a virtual link cost calculation step (S731) in which a link cost (CVL) for link driving calculated from the link information of the virtual connection path of the hybrid path candidate is calculated, and a virtual node cost calculation step (S733) in which a node cost (CVR) for turning driving calculated from the node information of the virtual connection path of the hybrid path candidate is calculated. Claim 29 A method for controlling a hybrid path driving mobile robot system, wherein, in claim 24, the hybrid path candidate cost calculation step (S70) further includes an obstacle bypass extension cost calculation step (S75) for calculating an obstacle bypass extension cost (CDE) for an obstacle bypass path. Claim 30 In claim 29, the obstacle bypass expansion cost calculation step (S75) comprises: a gradient bypass expansion cost calculation step (S751) in which a gradient bypass expansion cost (CDGE) for a multi-angle bypass path formed to bypass an obstacle is calculated; a heuristic bypass expansion cost calculation step (S753) in which a heuristic bypass expansion cost (CDHE) for a straight non-bypass path that does not bypass an obstacle is calculated; a congestion bypass expansion cost step (S755) in which a congestion bypass expansion cost (CDCE) due to obstacle detection for a curve bypass path calculated from the multi-angle bypass path is calculated; and a slope bypass expansion cost calculation step (S757) in which a slope bypass expansion cost (CDSE) due to the curve slope for a curve bypass path calculated from the multi-angle bypass path is calculated. Claim 31 A mobile robot path generation unit comprises: a current position information and destination information of a robot performing driving in a work environment including preset node and edge information, a constraint path composed of node information and edge information necessary to move from at least the current position information to the destination information, and, when the current position information and destination information and the constraint path are separated or the constraint path is discontinuous, a virtual connection path is generated using the constraint path, the current position information, and the destination information to generate and confirm a hybrid path for driving a mobile robot; wherein the mobile robot path generation unit comprises: a path connectivity verification unit that verifies connectivity between the current position information, the destination information, and the constraint path; and a virtual connection path information generation unit that generates virtual nodes and virtual links to connect the current position information, the destination information, and the constraint path when a non-connection is confirmed by the path connectivity verification unit; and wherein the virtual connection path information generation unit comprises: when a non-connection is confirmed by the path connectivity verification unit, the current position information and the destination information that are not connected from the constraint path and the A hybrid path-driving mobile robot unit characterized by comprising: a virtual circle forming unit that forms a virtual circle from at least one of the end nodes of a constraint path; a virtual node generating unit that forms a contact point between the virtual circle and the constraint path as a virtual node; a virtual link generating unit that generates a virtual link connecting the virtual node and at least one of the current location information and the destination information or connecting a discontinuous constraint path; and a virtual connection path generating unit that generates a virtual connection path using the virtual node and the virtual link. Claim 32 Using a constraint path composed of current position information and destination information of a robot performing driving in a work environment including preset node and edge information, and at least node information and edge information required to move from said current position information to said destination information, if said current position information and destination information and said constraint path are discontinuous, a virtual connection path is generated using said constraint path, said current position information, and said destination information, and a hybrid path for driving a mobile robot is generated and confirmed from said constraint path and said virtual connection path, and a position information verification step in which the current position information and destination information of a robot performing driving in a work environment including preset node and edge information are verified, a constraint path verification step in which a constraint path composed of at least node information and edge information required to move from said current position information to said destination information is verified, a path connectivity verification step in which connectivity between said current position information, said destination information, and said constraint path is verified, and when non-connection is confirmed in said path connectivity verification unit, said current position information and said destination information, and said The method includes a virtual connection path information generation step in which virtual nodes and virtual links are generated to connect a constraint path, wherein the virtual connection path information generation step comprises: a virtual circle formation step in which a virtual circle is formed from at least one of the current location information, the destination information, and the end of the constraint path when a non-connection is confirmed in the path connectivity verification step; a virtual node generation step in which a contact point between the virtual circle and the constraint path is formed as a virtual node; and a virtual link generation step in which a virtual link is generated connecting the virtual node and at least one of the current location information, the destination information, and the end of the constraint path.A method for controlling a hybrid path-driving mobile robot unit, characterized by including a virtual connection path generation step in which a virtual connection path is generated using the virtual node and the virtual link.
Citation Information
Patent Citations
Systems and methods for optimizing route plans in operating environment
JP2022082419A
Own position estimating device
JP2022123568A
Moving path planning apparatus and method for robot
KR1020220124064A
Occupancy clustering according to radar data
US20220317302A1
Mobile robot unit, mobile robot system and method for controlling the same
KR1020240066049A
