Method and apparatus for planning local path in dynamic grid containing reference local path for autonomous driving

Abstract

The present invention relates to a method and an apparatus for planning a local path in a dynamic grid containing a reference local path for autonomous driving. A method for planning a local path in a dynamic grid containing a reference local path for autonomous driving, according to one embodiment of the present invention, comprises the steps of: setting a reference distance in consideration of a preset maximum speed of a robot for autonomous driving; checking whether there is an obstacle within a drivable area; checking whether there is an obstacle within or after the set reference distance; if there is an obstacle after the set reference distance, calculating a cost function in consideration of the surrounding environment and deriving a reference area path moved to the lateral coordinates of a minimum cost as a local path; and if there is an obstacle within the set reference distance, dynamically generating a unit grid, calculating a cost function in consideration of the surrounding environment and the available operation range of the robot, and deriving an area path composed of the lateral coordinates of the minimum cost as a local path.

Description

Method and device for local path planning in a dynamic grid containing reference local paths for autonomous driving

[0001] The present invention relates to a local path planning method, and more particularly, to a local path planning method and device in a dynamic grid containing a reference local path for autonomous driving.

[0002] With the recent advancement of autonomous driving technology, autonomous driving is being attempted on various platforms, including automobiles, aircraft, drones, robots, and ships. To precisely control the movements of four-wheeled mobile robots, robust response to the numerous hazards posed by the surrounding environment is crucial. Autonomous vehicles primarily operate on roads, with regular obstacle and driving areas. In contrast, outdoor delivery robots navigate complex and diverse environments, such as sidewalks, crosswalks, and indoor spaces. Therefore, the robot's route, based on a semantic map for driving, must accommodate the diverse driving areas it encounters and address areas with shadowed location information.

[0003] In the case of maps utilizing GNSS (Global Navigation Satellite System) information for outdoor autonomous driving of delivery robots, the type and scope of the drivable area can change depending on the route in local coordinates. While highly accurate location information can handle all these situations, vision sensor information should be leveraged to provide overlapping detection results to accommodate situations where location information is obscured, such as when there are densely packed tall buildings or poor internet connectivity. Furthermore, camera-based autonomous driving systems must be able to handle low-accuracy vision sensor information.

[0004] Through this, the algorithms commonly used for path planning in autonomous mobile robots utilize grids to map perceived information. Smaller grid units and larger overall grid sizes lead to greater computational demands. Furthermore, larger grid units and smaller overall grid sizes lead to greater loss of perceived information. Hardware capable of high computational demands inevitably becomes less economical.

[0005] Embodiments of the present invention aim to provide a method and device for local path planning in a dynamic grid containing a reference local path for autonomous driving, in order to reduce the unit cost of a robot and perform robust and continuous lateral control.

[0006] However, the problem to be solved by the present invention is not limited to this, and may be expanded in various ways in environments that do not deviate from the spirit and scope of the present invention.

[0007] According to one embodiment of the present invention, a local path planning method performed by a local path planning device may be provided, the method comprising: setting a reference distance in consideration of a preset maximum speed of a robot for autonomous driving; checking whether there is an obstacle within a drivable area; checking whether there is an obstacle after or within the preset reference distance; if there is an obstacle after the preset reference distance, calculating a cost function in consideration of the surrounding environment and deriving a reference local path moved to a minimum cost horizontal coordinate as a local path; and if there is an obstacle within the preset reference distance, dynamically generating a unit grid and then calculating a cost function in consideration of the surrounding environment and an available range of motion of the robot and deriving a local path composed of minimum cost grid coordinates as a local path.

[0008] The method may further include a step of deriving a predetermined reference area path as a local path if there is no obstacle within the candidate reference area path of the edge of the drivable area or if there is no obstacle within or beyond the set reference distance.

[0009] The step of checking whether there is an obstacle after or within the above-mentioned set reference distance can search for the closest obstacle and the closest obstacle with the highest movement risk through relative position information and relative speed information of the obstacle among at least one obstacle.

[0010] The step of deriving a reference area path moved to the minimum cost horizontal coordinate as a local path is such that the longitudinal distance of the initial optimal path is fixed through the drivable area, and the longitudinal unit grid length can be determined using the fixed longitudinal distance and the speed of the robot.

[0011] The step of deriving a reference area path moved to the minimum cost horizontal coordinate as a local path can calculate a cost function using the horizontal relative distance of the obstacle, the horizontal relative distance to the upper and lower edges of the drivable area, and the coefficient of the reference path to be traced.

[0012] The step of deriving a local path composed of the minimum cost grid coordinates as a local path may be performed by fixing the longitudinal distance of the initial optimal path through the drivable area, and determining the longitudinal unit grid length based on the fixed longitudinal distance and the relative speed with respect to the obstacle.

[0013] The step of deriving a local path composed of the minimum cost grid coordinates as a local path can be performed by calculating a cost function through a lateral distance and selecting a range of motion of lateral movement using the speed and angular velocity of the robot to determine a lateral unit grid length.

[0014] The step of deriving a local path composed of the minimum cost grid coordinates as a local path can calculate a cost function using the longitudinal and lateral relative distances of the obstacle, the lateral relative distances to the upper and lower edges of the drivable area, the coefficient of the reference path to be traced, the range of motion of the lateral movement of the robot, and the estimated range of motion of the obstacle.

[0015] Meanwhile, according to another embodiment of the present invention, a local path planning device in a dynamic grid containing a reference local path for autonomous driving may be provided, including a memory storing one or more programs; and a processor executing the one or more stored programs, wherein the processor sets a reference distance in consideration of a preset maximum speed of a robot for autonomous driving, checks whether there is an obstacle within a drivable area, checks whether there is an obstacle after or within the set reference distance, and if there is an obstacle after the set reference distance, calculates a cost function in consideration of the surrounding environment and derives a reference local path moved to a minimum cost horizontal coordinate as a local path, and if there is an obstacle within the set reference distance, dynamically generates a unit grid, calculates a cost function in consideration of the surrounding environment and the operation available range of the robot, and derives a local path composed of minimum cost grid coordinates as a local path.

[0016] The processor may derive a predetermined reference area path as a local path if there is no obstacle within the candidate reference area path of the edge of the drivable area or if there is no obstacle within or beyond the set reference distance.

[0017] The above processor can search for the closest obstacle and the closest obstacle with the highest behavioral risk through relative position information and relative velocity information of the obstacle among at least one obstacle.

[0018] The processor may determine a longitudinal unit grid length based on a fixed longitudinal distance of an initial optimal path through the drivable area and the fixed longitudinal distance and the speed of the robot.

[0019] The processor can calculate a cost function using a coefficient of a lateral relative distance of the obstacle, a lateral relative distance from the upper and lower edges of the drivable area, and a reference path to be tracked.

[0020] The processor may determine a longitudinal unit grid length based on a fixed longitudinal distance of an initial optimal path through the drivable area and a relative speed with respect to the obstacle.

[0021] The above processor can calculate a cost function through a lateral distance and select a range of motion of lateral movement using the speed and angular velocity of the robot to determine a lateral unit grid length.

[0022] The processor can calculate a cost function using the longitudinal and lateral relative distances of the obstacle, the lateral relative distances to the upper and lower edges of the drivable area, the coefficients of the reference path to be tracked, the range of motion of the lateral movement of the robot, and the estimated range of motion of the obstacle.

[0023] The disclosed technology may have the following effects. However, this does not mean that a particular embodiment must include all or only the following effects, and therefore the scope of the disclosed technology should not be construed as being limited thereby.

[0024] Embodiments of the present invention can reduce the unit cost of robots. Embodiments of the present invention utilize information acquired through relatively inexpensive vision sensors (compared to Lidar) and low-cost global sensors (e.g., GPS, IMU). Furthermore, they benefit from increased computational capacity, allowing the selection of relatively inexpensive computing hardware.

[0025] Embodiments of the present invention can perform robust and continuous lateral control. These embodiments utilize data extracted from the robot's local sensors, enabling safer and more continuous control than methods that extract a reference local path solely through position information processed by global sensors such as GPS and IMU. Furthermore, these embodiments can induce smoother movement by considering cases where preemptive avoidance is possible.

[0026] Embodiments of the present invention enable general-purpose autonomous driving that does not rely on high-accuracy position information and recognition results. Driving solely based on position information can struggle to achieve high accuracy in urban areas, particularly those with tall, dense buildings or glass walls. Positional errors as small as tens of centimeters or meters can make safe and smooth control difficult. Furthermore, vision sensor information is probabilistic, making it difficult to consistently achieve high-accuracy recognition results. Low-accuracy recognition results can make the driving area appear wider or narrower than it actually is. This also makes it difficult to trust the recognition results, making safe driving difficult. Therefore, a reference area path must be derived using overlapping sensors. Embodiments of the present invention address these issues and enable robust and smooth driving control, even in specific urban areas or in areas with occasional low-accuracy recognition results.

[0027] Embodiments of the present invention can derive comfortable behavior and obtain benefits in computational amount by utilizing a dynamic grid rather than a fixed grid.

[0028] FIG. 1 is a drawing showing the configuration of a robot including a local path planning device according to one embodiment of the present invention.

[0029] FIG. 2 is a diagram showing a drivable area and a dynamic grid used in one embodiment of the present invention.

[0030] Figure 3 is a drawing showing a reference distance and a safety distance used in one embodiment of the present invention.

[0031] FIG. 4 is a flowchart of a local path planning method in a dynamic grid containing a reference local path for autonomous driving according to one embodiment of the present invention.

[0032] FIG. 5 is a drawing specifically showing a flowchart in a case where there is an obstacle after a reference distance in a local path planning method according to one embodiment of the present invention.

[0033] FIG. 6 is a drawing specifically showing a flowchart in a case where there is an obstacle within a reference distance in a local path planning method according to one embodiment of the present invention.

[0034] FIG. 7 and FIG. 8 are diagrams showing a case where there is an obstacle beyond a reference distance and a case where there is an obstacle within a reference distance in one embodiment of the present invention.

[0035] FIG. 9 is a configuration diagram of a local path planning device in a dynamic grid containing a reference local path for autonomous driving according to one embodiment of the present invention.

[0036] The present invention is capable of various modifications and embodiments. Specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it is to be understood that all modifications, equivalents, and alternatives included within the technical spirit and scope of the present invention are included. In describing the present invention, if a detailed description of a related known technology is judged to obscure the gist of the present invention, the detailed description will be omitted.

[0037] Terms like "first" and "second" may be used to describe various components, but these terms do not limit the components themselves. These terms are used solely to distinguish one component from another.

[0038] The terminology used in this invention is solely for the purpose of describing specific embodiments and is not intended to limit the invention. The terminology used in this invention has been selected from widely used, current terms, taking into account the functions of the invention. However, this may vary depending on the intentions of those skilled in the art, precedents, or the emergence of new technologies. Furthermore, in certain cases, the applicant may arbitrarily select terms, in which case their meanings will be described in detail in the relevant description of the invention. Therefore, the terms used in this invention should not be defined simply as names of terms, but rather based on their meanings and the overall content of the invention.

[0039] Singular expressions include plural expressions unless the context clearly dictates otherwise. In the present invention, terms such as "comprise" or "have" are intended to indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but should be understood not to preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0040] [Explanation of symbols]

[0041] 10: Robot

[0042] 20: Obstacles

[0043] 110: Reference Path Planning Module

[0044] 120: Cognitive Module

[0045] 130: Mission Module

[0046] 100: Local path planning device

[0047] 210: Memory

[0048] 220: Processor

[0049] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. In describing with reference to the attached drawings, identical or corresponding components are assigned the same drawing numbers, and redundant descriptions thereof will be omitted.

[0050] FIG. 1 is a drawing showing the configuration of a robot including a local path planning device according to one embodiment of the present invention.

[0051] As illustrated in FIG. 1, a robot (10) including a local path planning device (100) according to an embodiment of the present invention includes a reference path planning module (110), a recognition module (120), a mission module (130), and a local path planning device (100). However, not all of the illustrated components are essential components. A robot (10) including a local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving may be implemented with more components than the illustrated components, and a robot (10) including a local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving may also be implemented with fewer components.

[0052] Hereinafter, the specific configuration and operation of each component of a robot (10) including a local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving of FIG. 1 will be described.

[0053] The robot (10) plans a local path in a dynamic grid containing a reference local path for autonomous driving through a reference path planning module (110), a recognition module (120), a mission module (130), and a local path planning device (100).

[0054] To this end, the reference path planning module (110) transmits the determined reference area path and the reference area path of the edge of the drivable area to the local path planning device (100).

[0055] The recognition module (120) transmits the relative distance and speed of the obstacle to the local path planning device (100).

[0056] The mission module (130) transmits the global coordinates of the destination to the local path planning device (100).

[0057] And the local path planning device (100) performs a local path planning method using a dynamic grid containing a reference path for autonomous driving by using the transmitted determined reference local path and the reference local path of the edge of the drivable area, the relative distance and speed of an obstacle, and the global coordinates of the destination.

[0058] When explaining the local path planning method, the local path planning device (100) sets a tractable reference distance considering the maximum speed of the robot (10). The local path planning device (100) performs path planning in two stages based on the set reference distance. Before performing path planning, the local path planning device (100) extracts two reference paths: a determined reference path and an edge candidate reference path of a drivable area. The local path planning device (100) checks whether there is an obstacle (20) within the two edge candidate reference paths.

[0059] To explain from step 1, if there is an obstacle (20) after the reference distance, the local path planning device (100) calculates a cost function considering the surrounding environment. The local path planning device (100) switches the reference path in advance to the point with the minimum cost to smoothly induce lateral control in advance. If there is no obstacle (20) after the reference distance, the local path planning device (100) smoothly induces lateral control to the reference path.

[0060] Next, in step 2, if there is an obstacle (20) within the reference distance, the local path planning device (100) dynamically generates a unit grid based on the current speed of the robot (10) and the longitudinal relative distance and relative speed with respect to the obstacle (20). The local path planning device (100) calculates a cost function by considering the surrounding environment and the available operation range of the robot (10). The local path planning device (100) smoothly induces lateral control along a path composed of points with the minimum cost. If there is no obstacle (20) within the reference distance, the local path planning device (100) smoothly induces lateral control along the reference path.

[0061] The above process is performed when the destination is not within the detection range, and when the destination is within the detection range, the local path planning device (100) can add a cost function that takes into account the relative distance to the destination to the above process.

[0062] Hereinafter, the local path planning method performed by the local path planning device (100) will be described in more detail.

[0063] The local path planning device (100) sets a tractable reference distance considering the maximum speed of the robot (10). The local path planning device (100) performs path planning in two stages based on the set reference distance. Before performing path planning, the local path planning device (100) extracts two reference paths: a determined reference path and an edge candidate reference path of the drivable area. The local path planning device (100) checks whether there is an obstacle (20) within the two edge candidate reference paths.

[0064] First, to explain step 1, if there is an obstacle (20) after the reference distance, the local path planning device (100) calculates and adds a total of three costs. The local path planning device (100) calculates a cost function using the lateral relative distance and relative velocity with respect to the obstacle (20), and calculates the cost function using the lateral relative distance between two edges. In addition, the local path planning device (100) calculates the cost function using the coefficient of the determined reference path. The local path planning device (100) induces smooth lateral control in advance by switching the reference path in advance to the point of the minimum cost. The local path planning device (100) induces smooth lateral control to the reference path if there is no obstacle (20) after the reference distance. The algorithm corresponding to step 1 is when the destination is not within the detection range. If the destination is within the detection range, the local path planning device (100) can add a cost function that considers the relative distance to the destination to the above process.

[0065] Next, in step 2, if there is an obstacle (20) within the reference distance, the local path planning device (100) dynamically generates a unit grid based on the current speed of the robot (10) and the longitudinal relative distance and relative speed with the obstacle (20). This is generated as a relative coordinate with the robot (10) for generating a path by applying a cost with the surrounding information, not for the purpose of mapping the recognized surrounding information. The local path planning device (100) calculates and adds a total of four costs based on the generated coordinates. The local path planning device (100) calculates a cost function through the relative distance and relative speed with the obstacle (20) and calculates the cost function with the lateral relative distance between the two edges. In addition, the local path planning device (100) calculates the cost function with the coefficient of the determined reference path. The local path planning device (100) calculates the cost function with the relative angle and relative distance with the previously searched coordinates so that the robot (10) can search forward by considering the available operation range based on the searched coordinates. The local path planning device (100) smoothly induces lateral control along a path composed of points with the lowest cost. The local path planning device (100) smoothly induces lateral control along the reference path if there is no obstacle (20) within the reference distance. The algorithm corresponding to step 2 is for cases where the destination is not within the detection range. If the destination is within the detection range, the local path planning device (100) can add a cost function that considers the relative distance to the destination to the above process.

[0066] The local path planning device (100) guides driving along a local path derived through a series of processes of steps 1 and 2 above.

[0067] FIG. 2 is a diagram showing a drivable area and a dynamic grid used in one embodiment of the present invention.

[0068] The drivable area (101) in which the robot (10) can drive is depicted as a rectangular area as shown in (a) of Fig. 2.

[0069] The reference path represents the path that the robot (10) must trace, based on the upper and lower edges of the edges of the robot's (10) drivable area in the direction in which the robot (10) is facing.

[0070] A grid divides the drivable area into specific intervals to derive local paths within the drivable area.

[0071] The dynamic grid (102) is a grid whose spacing is adjusted to reflect the speed of the robot (10), the status of obstacles in front, and the drivable area, and is depicted as a dynamic area as shown in (b) of Fig. 2.

[0072] Figure 3 is a drawing showing a reference distance and a safety distance used in one embodiment of the present invention.

[0073] The reference distance (103) represents a reliable distance that the robot (10) can recognize in the drivable area, and is shown as the total distance to the pedestrian in Fig. 3.

[0074] As illustrated in Fig. 3, the safety distance (104) represents the distance obtained by subtracting the distance derived from a formula that satisfies string stability, which can satisfy the longitudinal relative distance between the robot (10) and surrounding obstacles (front / rear) from the distance from the pedestrian in front. Here, the unit grid represents one grid in an area proportionally divided according to the speed of the robot (10) up to the safety distance (104).

[0075] FIG. 4 is a flowchart of a local path planning method in a dynamic grid containing a reference local path for autonomous driving according to one embodiment of the present invention.

[0076] In step S101, the local path planning device (100) sets the reference distance and robot status information.

[0077] In step S102, the local path planning device (100) checks whether there is an obstacle (20) within the reference area path at the edge of the drivable area.

[0078] In step S103, the local path planning device (100) plans the determined reference area path as a local path when there is no obstacle (20) within the reference area path at the edge of the drivable area.

[0079] In step S104, the local path planning device (100) checks whether there is an obstacle (20) within the reference area path at the edge of the drivable area after the reference distance.

[0080] In step S105, the local path planning device (100) calculates a cost function of the transverse coordinates within the edge of the drivable area at the longitudinal distance of the nearest obstacle, if there is an obstacle (20) after the reference distance.

[0081] In step S106, the local path planning device (100) derives a reference local path moved to the minimum cost horizontal coordinate as a local path.

[0082] Meanwhile, in step S107, the local path planning device (100) checks whether there is an obstacle (20) within the reference distance, if there is no obstacle (20) after the reference distance.

[0083] In step S108, the local path planning device (100) determines the size of the unit grid / entire grid based on the relative distance and speed of the nearest obstacle and the status of the robot (10) if there is an obstacle (20) within the reference distance, and then calculates a cost function.

[0084] In step S109, the local path planning device (100) derives a local path composed of minimum cost grid coordinates as a local path.

[0085] In step S110, the local path planning device (100) plans the determined reference local path as a local path if there is no obstacle (20) within the reference distance.

[0086] FIG. 5 is a drawing specifically showing a flowchart in a case where there is an obstacle after a reference distance in a local path planning method according to one embodiment of the present invention.

[0087] In step S201, the local path planning device (100) checks for an obstacle (20) within a reference distance from the edge of the drivable area. Here, since the drivable area is derived through the shape of the reference path, the local path planning device (100) checks for the presence or absence of an obstacle (20) within the drivable area.

[0088] In step S202, the local path planning device (100) checks for an obstacle (20) after a reference distance (a recognition confidence distance). The reference distance is a reliable recognition distance, and the local path planning device (100) checks for an obstacle (20) after the reference distance. After step S202, since the reliability of recognition data for obstacles located after the reference distance is low, the purpose is to minimize lateral movement when an obstacle (20) that is expected to approach approaches rather than to avoid it.

[0089] In step S203, the local path planning device (100) searches for the closest obstacle and the one with the highest behavioral risk using the relative position information and relative speed information of the obstacle (20) among at least one obstacle. That is, the local path planning device (100) searches for the closest obstacle and the one with the most dangerous behavior using the relative position information and relative speed information of the closest obstacle.

[0090] In step S204, the local path planning device (100) fixes the longitudinal distance of the initial optimal path through the drivable area, and determines the longitudinal unit grid length based on the fixed longitudinal distance and the speed of the robot (10). Here, the drivable area at the derived longitudinal distance from the nearest obstacle represents the drivable area of ​​FIG. 3. The longitudinal distance of the initial optimal path can be fixed through this drivable area. The longitudinal unit grid length is determined based on the fixed longitudinal distance and the speed of the robot (10).

[0091] In step S205, the local path planning device (100) calculates a cost function using the transverse relative distance of the obstacle (20), the transverse relative distance to the upper and lower edges of the drivable area, and the coefficient of the reference path to be tracked. The local path planning device (100) can calculate the cost function using the transverse distance. The factors considered in this cost function can be the transverse relative distance of the obstacle (20), the transverse relative distance to the upper and lower edges of the drivable area, and the coefficient of the reference path to be tracked.

[0092] In step S206, the local path planning device (100) derives a reference local path moved with fixed longitudinal coordinates and minimum cost transverse coordinates as a local path. Here, since the local path planning device (100) has fixed longitudinal coordinates, longitudinal and transverse coordinates can be derived. Based on the derived longitudinal and transverse coordinates, the reference path moved can be derived as a local path.

[0093] FIG. 6 is a drawing specifically showing a flowchart in a case where there is an obstacle within a reference distance in a local path planning method according to one embodiment of the present invention.

[0094] In step S301, the local path planning device (100) identifies an obstacle (20) within a reference distance (a recognition confidence distance). Here, the local path planning device (100) determines whether an obstacle (20) is within the reference distance, considering that the reference distance is a reliable recognition distance. Obstacles within the reference distance have reliable recognition data.

[0095] In step S302, the local path planning device (100) searches for the closest obstacle and the one with the highest behavioral risk using the relative position information and relative speed information of the obstacle (20) among at least one obstacle. The local path planning device (100) can search for the closest obstacle and the one with the most dangerous behavior using the relative position information and relative speed information of the closest obstacle.

[0096] In step S303, the local path planning device (100) fixes the longitudinal distance of the initial optimal path through the drivable area, and determines the longitudinal unit grid length based on the fixed longitudinal distance and the relative speed with respect to the obstacle (20). Here, the drivable area at the derived longitudinal distance with respect to the nearest obstacle represents the drivable area of ​​FIG. 3. The longitudinal distance of the initial optimal path can be fixed through this drivable area. The longitudinal unit grid length is determined based on the fixed longitudinal distance and the relative speed with respect to the obstacle (20).

[0097] In step S304, the local path planning device (100) selects a range of motion for lateral movement using the speed and angular velocity of the robot (10) and determines a lateral unit grid length. The local path planning device (100) can select a range of motion for lateral movement using the speed and angular velocity (yaw rate) of the robot (10) and determine a lateral unit grid length.

[0098] In step S305, the local path planning device (100) calculates a cost function using the longitudinal and transverse relative distances of the obstacle (20), the transverse relative distances with the upper and lower edges of the drivable area, the coefficients of the reference path to be tracked, the lateral movement range of the robot (10), and the estimated lateral movement range of the obstacle (20). Here, the local path planning device (100) calculates the cost function using the lateral distance. The elements considered in the cost function can be the longitudinal and transverse relative distances of the obstacle (20), the transverse relative distances with the upper and lower edges of the drivable area, the coefficients of the reference path to be tracked, the lateral movement range of the robot (10), and the estimated lateral movement range (relative distance / relative angle) of the obstacle (20).

[0099] In step S306, the local path planning device (100) derives the fixed longitudinal coordinates and the minimum cost transverse coordinates as the initial coordinates of the local path. As described above, since the longitudinal coordinates are fixed, the longitudinal and transverse coordinates can be derived. The derived longitudinal and transverse coordinates can become the initial coordinates of the local path.

[0100] In step S307, the local path planning device (100) derives coordinates from the initial coordinates up to a pre-specified local path search distance by repeating the above process, thereby deriving a local path. That is, the local path planning device (100) can derive a local path by deriving coordinates up to a pre-specified local path search distance by repeating the process from steps S302 to S307 from the initial coordinates.

[0101] FIG. 7 and FIG. 8 are diagrams showing a case where there is an obstacle beyond a reference distance and a case where there is an obstacle within a reference distance in one embodiment of the present invention.

[0102] Fig. 7 illustrates a case where an obstacle (20) is present after a reference distance, and lateral control is performed along a reference area path moved to the lateral coordinate with the minimum cost. Here, the shaded area illustrated in Fig. 7 is for the purpose of showing the process of deriving the lateral coordinate while the longitudinal coordinate is fixed.

[0103] Fig. 8 illustrates a case where there is an obstacle (20) within a reference distance, and lateral control is performed using a local path composed of minimum cost grid coordinates. Here, the shaded area illustrated in Fig. 8 is assumed to be a drivable area where a local path will be generated.

[0104] The shaded areas shown in FIGS. 7 and 8 are expressed as [Mathematical Formula 1] and [Mathematical Formula 2] below, respectively.

[0105]

[0106] Here, represents the minimum distance to the nearest front object (obstacle) within the drivable area, represents the lateral spacing of the drivable area.

[0107]

[0108] Here, Is It represents the longitudinal distance to .

[0109] FIG. 9 is a configuration diagram of a local path planning device in a dynamic grid containing a reference local path for autonomous driving according to one embodiment of the present invention.

[0110] As illustrated in FIG. 9, a local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving according to an embodiment of the present invention includes a memory (210) and a processor (220). However, not all of the illustrated components are essential components. The local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving may be implemented with more components than the illustrated components, or the local path planning device (100) in a dynamic grid containing a reference local path for autonomous driving may be implemented with fewer components.

[0111] Below, the specific configuration and operation of each component of the local path planning device (100) in the dynamic grid containing the reference local path for autonomous driving of FIG. 9 are described.

[0112] The memory (210) stores one or more programs related to a local path planning method in a dynamic grid containing a reference local path for autonomous driving.

[0113] The processor (220) executes one or more programs stored in the memory (210). The processor (220) sets a reference distance considering a preset maximum speed of the robot (10) for autonomous driving, checks whether there is an obstacle (20) within the driving area, checks whether there is an obstacle (20) after or within the set reference distance, and if there is an obstacle (20) after the set reference distance, calculates a cost function considering the surrounding environment and derives a reference local path moved to the horizontal coordinate of the minimum cost as a local path, and if there is an obstacle (20) within the set reference distance, dynamically generates a unit grid and then calculates a cost function considering the surrounding environment and the available operation range of the robot (10) and derives a local path composed of the grid coordinates of the minimum cost as a local path.

[0114] According to embodiments, the processor (220) may derive a predetermined reference area path as a local path if there is no obstacle (20) within the candidate reference area path of the edge of the drivable area or if there is no obstacle (20) within or after a set reference distance.

[0115] According to embodiments, the processor (220) can search for the closest obstacle and the closest obstacle with the highest behavioral risk through relative position information and relative velocity information of the obstacle among at least one obstacle.

[0116] According to embodiments, the processor (220) may determine a longitudinal unit grid length based on a fixed longitudinal distance of an initial optimal path through a drivable area and the fixed longitudinal distance and the speed of the robot (10).

[0117] According to embodiments, the processor (220) can calculate a cost function using the coefficients of the lateral relative distance of the obstacle (20), the lateral relative distance to the upper and lower edges of the drivable area, and the reference path to be tracked.

[0118] According to embodiments, the processor (220) may determine a longitudinal unit grid length based on a fixed longitudinal distance of an initial optimal path through a drivable area and a relative speed with respect to an obstacle (20).

[0119] According to embodiments, the processor (220) can calculate a cost function through a lateral distance and select a range of motion of the lateral movement using the speed and angular velocity of the robot (10) to determine a lateral unit grid length.

[0120] According to embodiments, the processor (220) can calculate a cost function using the longitudinal and lateral relative distances of the obstacle (20), the lateral relative distances to the upper and lower edges of the drivable area, the coefficients of the reference path to be tracked, the range of motion of the lateral movement of the robot (10) and the estimated range of motion of the obstacle (20).

[0121] Meanwhile, according to one embodiment of the present invention, the various embodiments described above can be implemented as software including instructions stored in a machine-readable storage medium that can be read by a machine (e.g., a computer). The device is a device that can call instructions stored from the storage medium and operate according to the called instructions, and may include an electronic device (e.g., electronic device (A)) according to the disclosed embodiments. When an instruction is executed by a processor, the processor can perform a function corresponding to the instruction directly or by using other components under the control of the processor. The instruction may include code generated or executed by a compiler or interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' means that the storage medium does not contain a signal and is tangible, but does not distinguish between data being stored semi-permanently or temporarily in the storage medium.

[0122] Furthermore, according to one embodiment of the present invention, the method according to the various embodiments described above may be provided as included in a computer program product. The computer program product may be traded as a commodity between sellers and buyers. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or online through an application store (e.g., Play Store™). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0123] Furthermore, according to one embodiment of the present invention, the various embodiments described above may be implemented in a computer-readable recording medium or a similar device using software, hardware, or a combination thereof. In some cases, the embodiments described herein may be implemented by the processor itself. In a software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein.

[0124] Meanwhile, computer instructions for performing processing operations of a device according to the various embodiments described above may be stored in a non-transitory computer-readable medium. The computer instructions stored in such a non-transitory computer-readable medium, when executed by a processor of a specific device, cause the specific device to perform processing operations in the device according to the various embodiments described above. A non-transitory computer-readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specific examples of non-transitory computer-readable media may include a CD, DVD, hard disk, Blu-ray disk, USB, memory card, or ROM.

[0125] In addition, each of the components (e.g., modules or programs) according to the various embodiments described above may be composed of a single or multiple entities, and some of the sub-components described above may be omitted, or other sub-components may be further included in various embodiments. Alternatively or additionally, some components (e.g., modules or programs) may be integrated into a single entity, which may perform the same or similar functions as those performed by each of the respective components prior to integration. Operations performed by modules, programs or other components according to various embodiments may be executed sequentially, in parallel, iteratively or heuristically, or at least some operations may be executed in a different order, omitted, or other operations may be added.

[0126] Although the preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be made by those skilled in the art without departing from the gist of the present invention as claimed in the claims. Furthermore, such modifications should not be understood individually from the technical idea or prospect of the present invention.

Claims

1. In a local path planning method performed by a local path planning device, A step for setting a reference distance by considering the preset maximum speed of the robot for autonomous driving; A step for checking whether there is an obstacle within the drivable area; A step of checking whether there is an obstacle within or beyond the set standard distance; If there is an obstacle beyond the above-mentioned set reference distance, a step of calculating a cost function by considering the surrounding environment and deriving a reference local path moved to the horizontal coordinate with the minimum cost as a local path; and A method for local path planning in a dynamic grid containing a reference local path for autonomous driving, comprising the steps of dynamically generating a unit grid if there is an obstacle within the above-described set reference distance, calculating a cost function by considering the surrounding environment and the operation availability range of the robot, and deriving a local path composed of grid coordinates with the minimum cost as a local path.

2. In paragraph 1, A method for local path planning in a dynamic grid containing reference local paths for autonomous driving, further comprising the step of deriving a predetermined reference local path as a local path if there is no obstacle within the candidate reference local path of the edge of the above-described drivable area or if there is no obstacle within or beyond the set reference distance.

3. In paragraph 1, The step of checking whether there is an obstacle within or beyond the above-mentioned standard distance is as follows: A local path planning method in a dynamic grid containing reference local paths for autonomous driving, which searches for the closest obstacle and the closest obstacle with the highest motion risk using relative position information and relative velocity information of the obstacle among at least one obstacle.

4. In paragraph 1, The step of deriving a reference local path moved to the horizontal coordinate of the minimum cost as a local path is as follows. A local path planning method in a dynamic grid containing a reference local path for autonomous driving, wherein the longitudinal distance of the initial optimal path is fixed through the above-mentioned drivable area, and the longitudinal unit grid length is determined by the fixed longitudinal distance and the speed of the robot.

5. In paragraph 1, The step of deriving a reference local path moved to the horizontal coordinate of the minimum cost as a local path is as follows. A local path planning method in a dynamic grid containing reference local paths for autonomous driving, wherein a cost function is calculated by calculating a coefficient of a lateral relative distance to the obstacle, a lateral relative distance to the upper and lower edges of the drivable area, and a reference path to be tracked.

6. In paragraph 1, The step of deriving a local path consisting of the above minimum cost grid coordinates as a local path is as follows. A local path planning method in a dynamic grid containing a reference local path for autonomous driving, wherein the longitudinal distance of the initial optimal path is fixed through the above-mentioned drivable area, and the longitudinal unit grid length is determined by the fixed longitudinal distance and the relative speed with respect to the obstacle.

7. In paragraph 1, The step of deriving a local path consisting of the above minimum cost grid coordinates as a local path is as follows. A local path planning method in a dynamic grid containing a reference local path for autonomous driving, wherein a cost function is calculated through a lateral distance, and a range of motion of the lateral movement is selected using the speed and angular velocity of the robot, thereby determining a lateral unit grid length.

8. In paragraph 1, The step of deriving a local path consisting of the above minimum cost grid coordinates as a local path is as follows. A local path planning method in a dynamic grid containing a reference local path for autonomous driving, wherein a cost function is calculated using the longitudinal and lateral relative distances of the obstacle, the lateral relative distances to the upper and lower edges of the drivable area, the coefficients of the reference path to be followed, the range of motion of the lateral movement of the robot, and the estimated range of motion of the obstacle.

9. Memory for storing one or more programs; and comprising a processor for executing one or more of the stored programs; The above processor, Set the reference distance by considering the preset maximum speed of the robot for autonomous driving, Check for any obstacles within the driving area, Check for the presence of obstacles within or beyond the above-mentioned standard distance, If there is an obstacle beyond the above-mentioned reference distance, the cost function is calculated by considering the surrounding environment, and the reference local path moved to the horizontal coordinate with the minimum cost is derived as the local path. A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which dynamically generates a unit grid when there is an obstacle within the above-mentioned set reference distance, calculates a cost function by considering the surrounding environment and the operation availability range of the robot, and derives a local path composed of grid coordinates with the minimum cost as a local path.

10. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which derives a predetermined reference local path as a local path if there is no obstacle within the candidate reference local path of the edge of the above-mentioned drivable area or if there is no obstacle within or beyond the set reference distance.

11. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which searches for the closest obstacle and the closest obstacle with the highest motion risk using relative position information and relative velocity information of the obstacle among at least one obstacle.

12. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, wherein the longitudinal distance of the initial optimal path is fixed through the above-mentioned drivable area, and the longitudinal unit grid length is determined by the fixed longitudinal distance and the speed of the robot.

13. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which calculates a cost function using the coefficients of the lateral relative distance to the obstacle, the lateral relative distance to the upper and lower edges of the drivable area, and the reference path to be tracked.

14. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, wherein the longitudinal distance of the initial optimal path is fixed through the above-mentioned drivable area, and the longitudinal unit grid length is determined by the fixed longitudinal distance and the relative speed with respect to the obstacle.

15. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which calculates a cost function through a lateral distance and determines a lateral unit grid length by selecting a range of motion of the lateral movement using the speed and angular velocity of the robot.

16. In paragraph 9, The above processor, A local path planning device in a dynamic grid containing a reference local path for autonomous driving, which calculates a cost function using the longitudinal and lateral relative distances of the obstacle, the lateral relative distances to the upper and lower edges of the drivable area, the coefficients of the reference path to be followed, the range of motion of the lateral movement of the robot, and the estimated range of motion of the obstacle.

Images