Autonomous travel control apparatus and method therefor

The autonomous driving control device optimizes data processing and storage by adjusting sampling rates and setting weight values for sensing data, addressing efficiency and accuracy issues in mobile robot navigation.

JP2024164787APending Publication Date: 2024-11-27HYUNDAI MOTOR CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023186241
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-15
Filing Date
2023-10-31
Publication Date
2024-11-27

AI Technical Summary

Technical Problem

Existing autonomous driving systems for mobile robots face challenges in efficient data processing and storage due to load generation and insufficient capacity, particularly when handling large amounts of sensing data for location identification and map data.

Method used

An autonomous driving control device that adjusts sampling rates for sensing data based on identified objects, generates voxel data with position information, and sets weight values for different portions of the data to optimize storage and accuracy.

Benefits of technology

This approach enhances storage capacity and accuracy of location information by efficiently sampling and storing point cloud data, reducing accuracy degradation and improving object identification during autonomous travel.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024164787000001_ABST
    Figure 2024164787000001_ABST
Patent Text Reader

Abstract

To provide an autonomous travel control apparatus capable of getting efficient and correct position information.SOLUTION: An autonomous travel control apparatus includes a sensor unit, a memory storing an instruction and a control unit, so as to get first sensing data related to a travel route using a first sensor, identify a first-designated object, identify a first sampling rate to be applied to first point cloud data, get second sensing data related to the travel route using a second sensor, identify a second-designated object, identify a second sampling rate to be applied to second point cloud data based on a result of comparison between the first-designated object and the second-designated object, and perform sampling on the first and second sensing data using the first and second sampling rates.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an autonomous driving control device and method, and more particularly to a technology for acquiring various data for driving a driving device (e.g., a mobile robot). [Background technology]

[0002] As technology advances, various technologies related to traveling devices have been developed. For example, various methods for data processing have been developed as more data is used to control the traveling of the traveling device to a destination. The traveling device may be embodied in various types. For example, various traveling devices such as an autonomous vehicle, a mobile robot, and an unmanned flying device may be embodied so that the traveling can be remotely controlled without the driver directly controlling the traveling.

[0003] In particular, mobile robots may be used for various purposes in various environments. For example, a mobile robot may move inside a specified building (e.g., a hotel, a restaurant, etc.) based on an autonomous driving technique. For example, a mobile robot may identify information on its current location and surrounding conditions in real time using at least one sensor, and autonomously drive based on at least a part of the identified information. However, when performing autonomous driving for a mobile robot, a large amount of data must be calculated to identify location information, or data required for driving (e.g., map data) must be stored, which may cause a load during processing or may cause difficulties in normal operation due to insufficient storage capacity. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 3-192407 Summary of the Invention [Problem to be solved by the invention]

[0005] One embodiment of the present invention provides an autonomous driving control device and method that distinguishes different sampling rates for sensing data acquired using different types of sensors (e.g., cameras and Lidar) depending on whether a specified object can be identified.

[0006] One embodiment of the present invention provides an autonomous driving control device and method that, when an object satisfying specified conditions is identified in data (e.g., point cloud data) acquired using a sensor, identifies a sampling rate to be applied to point cloud data corresponding to the identified object based on predefined conditions, and then performs sampling on the sensing data.

[0007] One embodiment of the present invention provides an autonomous driving control device and method configured to generate voxel data including location information of a specified object based on at least a portion of sampled sensing data, and store map data including the voxel data in a memory.

[0008] One embodiment of the present invention provides an autonomous driving control device and method configured to set different weighting values ​​for parts of voxel data that do not satisfy a specified condition (or parts that do not include a specified object) and parts that satisfy a specified condition (or parts that include a specified object) and compare voxel data based on the set weighting values.

[0009] The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0010] According to an embodiment of the present invention, an autonomous driving control device includes a sensor unit including at least one sensor, a memory storing at least one instruction, and a control unit electrically connected to the sensor unit and the memory. When the instruction is executed by the control unit, the autonomous driving control device is configured to: acquire first sensing data related to a driving path of a driving device using a first sensor of the at least one sensor; identify a first designated object included in the first sensing data and satisfying a specified condition; identify a first sampling rate to be applied to first point cloud data corresponding to the first designated object; acquire second sensing data related to the driving path using a second sensor of the at least one sensor; identify a second designated object included in the second sensing data; identify a second sampling rate to be applied to second point cloud data corresponding to the second designated object based on a comparison result between the first designated object and the second designated object; and sample the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively.

[0011] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to sample the first point cloud data corresponding to the first designated object in the first sensing data using the first sampling rate, and to sample the second point cloud data corresponding to the second designated object in the second sensing data using the second sampling rate.

[0012] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to identify the second sampling rate as being equal to the first sampling rate or as being greater than the first sampling rate when the first designated object and the second designated object are identified as being the same.

[0013] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to sample at least a portion of the second sensing data using a predefined sampling rate if the first designated object and the second designated object are not identical or if the second sensing data does not include the second designated object.

[0014] According to an embodiment, the predefined sampling rate may be less than the first sampling rate.

[0015] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to generate first voxel data including position information of at least one of the first designated object, the second designated object, or any combination thereof based on at least one of the sampled first sensing data, the sampled second sensing data, or any combination thereof, and to store map data including the first voxel data in the memory.

[0016] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to generate second voxel data based on third sensing data acquired using the second sensor, and identify a current position of the driving device based on a matching result between the first voxel data and the second voxel data included in the map data.

[0017] According to one embodiment, when the instruction is executed by the control unit, the autonomous driving control device may be configured to set a weight value for a portion of the first voxel data and the second voxel data that does not satisfy the specified condition to a first value and perform matching, and set a weight value for a portion of the first voxel data and the second voxel data that satisfies the specified condition, including at least one of the first specified object, the second specified object, or any combination thereof, to a second value greater than the first value and perform matching.

[0018] According to an embodiment, the second voxel data may include voxel coordinates of a blob corresponding to at least one object that satisfies the specified condition.

[0019] According to one embodiment, when the instructions are executed by the control unit, the autonomous driving control device may be configured to apply a point cloud registration algorithm to the first voxel data and the second voxel data to generate the matching result.

[0020] According to an embodiment of the present invention, an autonomous driving control method includes a step of a control unit acquiring first sensing data related to a driving path of a driving device using a first sensor; a step of the control unit identifying a first designated object included in the first sensing data and satisfying a specified condition, and identifying a first sampling rate to be applied to first point cloud data corresponding to the first designated object; a step of the control unit acquiring second sensing data related to the driving path using a second sensor; a step of the control unit identifying a second designated object included in the second sensing data, and identifying a second sampling rate to be applied to second point cloud data corresponding to the second designated object based on a comparison result between the first designated object and the second designated object; and a step of the control unit sampling the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively.

[0021] According to one embodiment, the step of the control unit sampling the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively, may include the step of the control unit sampling the first point cloud data corresponding to the first designated object in the first sensing data using the first sampling rate and the step of the control unit sampling the second point cloud data corresponding to the second designated object in the second sensing data using the second sampling rate.

[0022] According to one embodiment, the step of the control unit identifying the second sampling rate may include the step of the control unit identifying the second sampling rate as being equal to the first sampling rate or as being greater than the first sampling rate when the first designated object and the second designated object are identified as being the same.

[0023] According to one embodiment, the step of the control unit sampling the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively, may include a step of the control unit sampling at least a portion of the second sensing data using a predefined sampling rate when the first designated object and the second designated object are not identical or the second sensing data does not include the second designated object.

[0024] According to an embodiment, the predefined sampling rate may be less than the first sampling rate.

[0025] According to one embodiment, the autonomous driving control method may include a step in which the control unit generates first voxel data including position information of at least one of the first designated object, the second designated object, or any combination thereof based on at least one of the sampled first sensing data, the sampled second sensing data, or any combination thereof, and a step in which the control unit stores map data including the first voxel data in a memory.

[0026] According to one embodiment, the autonomous driving control method may include a step in which the control unit generates second voxel data based on third sensing data acquired using the second sensor, and a step in which the control unit identifies a current position of the traveling device based on a matching result between the first voxel data and the second voxel data included in the map data.

[0027] According to one embodiment, the step of the control unit identifying the current position of the traveling device based on the matching result between the first voxel data and the second voxel data included in the map data may include the step of the control unit setting a weight value for a portion of the first voxel data and the second voxel data that does not satisfy the specified condition to a first value and performing matching, and the step of the control unit setting a weight value for a portion that satisfies the specified condition, including at least one of the first designated object, the second designated object, or a combination thereof, to a second value greater than the first value and performing matching.

[0028] According to an embodiment, the second voxel data may include voxel coordinates of a blob corresponding to at least one object that satisfies the specified condition.

[0029] According to one embodiment, the step of the control unit identifying the current position of the traveling device based on the matching result between the first voxel data and the second voxel data included in the map data may include the step of the control unit applying a point cloud registration algorithm to the first voxel data and the second voxel data to generate the matching result. Effect of the Invention

[0030] The effects of the autonomous driving control device and method according to the present invention are as follows. According to at least one of the embodiments of the present invention, in the process of processing data for the autonomous driving of a driving device (or a mobile robot), a weight value is set according to the presence or absence of an object satisfying a specified condition, and map data is generated using at least a part of point cloud data, thereby obtaining more efficient and accurate position information. According to at least one of the embodiments of the present invention, among data obtained using a sensor, point cloud data is sampled and stored based on a sampling rate of a specified value, thereby obtaining position information while reducing storage capacity. According to at least one of the embodiments of the present invention, when performing a positioning function for a driving device, if a specified object providing relatively more accurate information is identified, a sampling rate for data (e.g., point cloud data) corresponding to the identified specified object is reduced, thereby preventing a problem of a decrease in accuracy that may occur due to sampling. In addition, various effects that can be directly or indirectly understood through this document may be provided. [Brief description of the drawings]

[0031] [Figure 1] 1 is a block diagram showing components of an autonomous driving control device according to an embodiment of the present invention. [Diagram 2] 1 is an operational conceptual diagram of an autonomous driving control device according to an embodiment of the present invention; [Diagram 3] 1 is an operational conceptual diagram of an autonomous driving control device according to an embodiment of the present invention; [Figure 4] FIG. 2 is an operation flow diagram of the autonomous driving control device according to one embodiment of the present invention. [Diagram 5] FIG. 2 is an operation flow diagram of the autonomous driving control device according to one embodiment of the present invention. [Figure 6] FIG. 2 is an operation flow diagram of the autonomous driving control device according to one embodiment of the present invention. [Figure 7] 1 illustrates a computer system relating to an autonomous driving control method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] Some embodiments of the present invention will now be described in detail with reference to the illustrative drawings.

[0033] It should be noted that when reference numerals are attached to components in each drawing, the same numerals are used for the same components. In addition, in describing the embodiments of the present invention, if a detailed description of related known configurations or functions is deemed to hinder understanding of the embodiments of the present invention, the detailed description thereof will be omitted.

[0034] In describing the components of the embodiments of the present invention, terms such as 1, 2, A, B, (a), (b) and the like can be used. Such terms are used only to distinguish the components from other components, and do not limit the essence, order or sequence of the components. In addition, unless otherwise defined, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and should not be interpreted as idealized or overly formal unless expressly defined in this application.

[0035] Hereinafter, an embodiment of the present invention will be described in detail with reference to FIGS.

[0036] FIG. 1 is a block diagram showing components of an autonomous driving control device according to an embodiment of the present invention. According to an embodiment, the autonomous driving control device 100 includes at least one of a sensor unit 110, a memory 120, a control unit 130, or any combination thereof. The configuration of the autonomous driving control device 100 in FIG. 1 is exemplary, and the embodiment of the present invention is not limited thereto. For example, the autonomous driving control device 100 further includes components (e.g., an interface unit, a display unit, an audio output unit, a communication unit, etc.) not shown in FIG. 1. According to an embodiment, the autonomous driving control device 100 controls a traveling device (e.g., a mobile robot) using at least some of the above-mentioned components. According to an embodiment, the sensor unit 110 acquires (or senses) various information used for traveling of the mobile device. For example, the sensor unit 110 includes at least one sensor including at least one of a camera, a radar, a LiDAR, or any combination thereof.

[0037] For example, the sensor unit 110 obtains information about an external object (eg, at least one of a person, another vehicle, a building, a structure, or any combination thereof) using at least one sensor.

[0038] As an example, the sensor unit 110 includes a first sensor including various types of cameras. For example, the first sensor may include an RGB (red-green-blue) camera. The sensor unit 110 obtains first sensing data including an RGB image using the RGB camera. The first sensing data includes, for example, image data related to a traveling path of the traveling device. As an example, the sensor unit 110 includes a second sensor including various types of LIDAR. For example, the second sensor may include a 3D LIDAR. The sensor unit 110 obtains second sensing data using the 3D LIDAR. The second sensing data includes, for example, LIDAR data related to a traveling path of the traveling device.

[0039] According to an embodiment, the memory 120 stores instructions or data. For example, the memory 120 stores one or more instructions that, when executed by the controller 130, cause the autonomous driving control device 100 to perform various operations. For example, the memory 120 may be embodied in a chipset together with the controller 130. The controller 130 includes at least one of a communication processor or a modem.

[0040] For example, the memory 120 stores various information associated with the autonomous driving control device 100. As an example, the memory 120 may store information related to the operation history of the control unit 130. As an example, the memory 120 stores information associated with the status and / or operation of components (e.g., a drive unit (or a motor)) of a mobile device (e.g., a mobile robot) controlled by the autonomous driving control device 100 and / or components of the autonomous driving control device 100 (e.g., at least one of the sensor unit 110, the control unit 130, or any combination thereof).

[0041] According to an embodiment, the control unit 130 may be operatively connected to at least one of the sensor unit 110, the memory 120, or any combination thereof. For example, the control unit 130 controls the operation of at least one of the sensor unit 110, the memory 120, or any combination thereof. For example, the control unit 130 may control the autonomous driving of the traveling device. The control unit 130 obtains various sensing data regarding the traveling route of the traveling device, for example, using the sensor unit 110.

[0042] For example, the control unit 130 obtains first sensing data regarding the traveling route of the traveling device using a first sensor (e.g., RGB camera) included in the sensor unit 110. For example, the control unit 130 obtains second sensing data regarding the traveling route of the traveling device using a second sensor (e.g., 3D lidar) included in the sensor unit 110.

[0043] For example, the control unit 130 identifies objects included in the sensing data and determines whether a designated object that satisfies a designated condition is present among the identified objects. As an example, the control unit 130 identifies a first designated object that is included in the first sensing data and satisfies a designated condition, and identifies a first sampling rate to be applied to first point cloud data corresponding to the first designated object from the first sensing data.

[0044] For example, the specified condition includes a condition associated with at least one of the size, shape, color, volume, weight, external shape, or any combination thereof of the object. For example, the specified object (e.g., the first specified object and / or the second specified object) may be a fixed object relatively useful for positioning for autonomous driving control of a traveling device traveling in a specific space. The specified object satisfying the specified condition includes, for example, at least one of a pillar, a desk, a fire extinguisher, a sofa, or any combination thereof contained (or placed) in a building.

[0045] For example, the control unit 130 identifies a first sampling rate to be applied to the first point cloud data corresponding to the first designated object. As an example, the control unit 130 identifies a first sampling rate required for sampling the first point cloud data from the first sensing data. The control unit 130 identifies the first sampling rate using, for example, sampling rate information (e.g., a table) stored in the memory 120 and information on the identified first designated object.

[0046] For example, the control unit 130 identifies a second designated object included in the second sensing data. For example, the control unit 130 compares the first designated object included in the first sensing data with the second designated object included in the second sensing data. As an example, the control unit 130 determines whether the first designated object is substantially identical to the second designated object based on the comparison result. For example, the control unit 130 identifies a second sampling rate to be applied to the second point cloud data corresponding to the second designated object based on the comparison result.

[0047] For example, when the first designated object and the second designated object are identified as the same object, the control unit 130 identifies the second sampling rate as a value equal to or greater than the first sampling rate. In other words, when the first designated object and the second designated object are identified as the same object, the control unit 130 may identify the second sampling rate as a value equal to or greater than the first sampling rate.

[0048] For example, if the first designated object and the second designated object are not the same or the second sensing data does not include the second designated object, the control unit 130 identifies a predefined sampling rate (e.g., a sampling rate stored in the memory 120) as the second sampling rate. The control unit 130 may sample the second sensing data including the second designated object using the predefined sampling rate. For example, the predefined sampling rate may be lower than the first sampling rate.

[0049] For example, the control unit 130 may sample the first sensing data and the second sensing data using a first sampling rate and a second sampling rate, respectively. As an example, the control unit 130 may sample first point cloud data corresponding to a first designated object in the first sensing data using the first sampling rate.

[0050] For example, the control unit 130 may sample second point cloud data corresponding to a second designated object in the second sensing data using a second sampling rate. For example, the sampling rate for point cloud data corresponding to a designated object in the sensing data may be higher than a sampling rate corresponding to an area not corresponding to the designated object (or an area where the designated object does not exist).

[0051] For example, the control unit 130 generates voxel data using at least a portion of the sampling results. As an example, the control unit 130 generates first voxel data including position information of at least one of a first designated object, a second designated object, or a combination of any of them based on at least one of the sampled first sensing data, the sampled second sensing data, or a combination of any of them. The control unit 130 stores at least a portion of map data including the first voxel data in the memory 120.

[0052] For example, the control unit 130 acquires the first sensing data and the second sensing data, acquires the third sensing data using a second sensor among at least one sensor included in the sensor unit 110, and generates the second voxel data using the third sensing data sampled based on a designated sampling rate. The second voxel data includes, for example, voxel coordinates of a blob corresponding to at least one object satisfying a designated condition.

[0053] For example, the control unit 130 may identify the current position of the traveling device based on a matching result between first voxel data and second voxel data included in the map data. The control unit 130 may generate a matching result between the first voxel data and the second voxel data by applying, for example, various types of point cloud registration (e.g., ICP (iteractive closest point)) algorithms.

[0054] As an example, the control unit 130 sets a weighting value for a portion of the first voxel data and the second voxel data that does not satisfy a specified condition as a first value and performs matching. The first value may be, for example, 1. As an example, the control unit 130 sets a weighting value for a portion that satisfies a specified condition including at least one of a first designated object, a second designated object, or a combination of either of them as a second value greater than the first value and performs matching. The second value may be, for example, a number greater than the first value (e.g., 1). For example, the control unit 130 stores various data generated through the above-mentioned operations in the memory 120.

[0055] FIG. 2 is a conceptual diagram of the operation of the autonomous driving control device according to one embodiment of the present invention. According to one embodiment, the autonomous driving control device 200 (e.g., the autonomous driving control device 100 of FIG. 1) generates voxel data based on sensing data acquired using at least one sensor (212, 214) and stores the voxel data in a memory 220 (e.g., the memory 120 of FIG. 1).

[0056] For example, the autonomous driving control device 200 acquires first sensing data using a first sensor 212 (e.g., an RGB camera). The first sensing data includes data related to a driving path of the driving device. With reference to reference number S210, for example, the autonomous driving control device 200 detects an object based on an image included in the first sensing data. As an example, the autonomous driving control device 200 identifies a first designated object that is included in the first sensing data and satisfies a specified condition. As an example, the autonomous driving control device 200 identifies first point cloud data corresponding to the first designated object.

[0057] With reference to reference number S220, for example, the autonomous driving control device 200 identifies a sampling rate for sampling the point cloud data. As an example, the autonomous driving control device 200 identifies a first sampling rate to be applied for sampling to the first point cloud data included in the first sensing data. For example, the autonomous driving control device 200 acquires the second sensing data using the second sensor 214 (e.g., 3D lidar). The second sensing data includes data related to the traveling route of the traveling device.

[0058] With reference to reference number S230, for example, the autonomous driving control device 200 may sample point cloud data. As an example, the autonomous driving control device 200 samples at least a portion of first sensing data including first point cloud data. For example, the autonomous driving control device 200 performs sampling by setting a first sampling rate for the first point cloud data corresponding to a first designated object to be higher than a sampling rate for sensing data corresponding to an area where the designated object does not exist. As an example, the autonomous driving control device 200 stores at least a portion of the sampled first sensing data in the memory 220.

[0059] For example, the autonomous driving control device 200 performs sampling by setting a second sampling rate for the second point cloud data corresponding to the second designated object to be higher than a sampling rate for the sensing data corresponding to an area where the designated object does not exist. For example, the autonomous driving control device 200 determines the second sampling rate based on a comparison result between the first designated object and the second designated object.

[0060] Referring to reference number S240, the autonomous driving control device 200 generates voxel data based on at least a portion of the sampled first sensing data and second sensing data. As an example, the autonomous driving control device 200 generates voxel data through a voxelize operation on at least a portion of the sampled first sensing data and second sensing data. The autonomous driving control device 200 generates first voxel data including position information of at least one of a first designated object, a second designated object, or a combination of any of them, based on at least one of the sampled first sensing data, the sampled second sensing data, or a combination of any of them. As an example, the autonomous driving control device 200 stores at least a portion of the first voxel data in the memory 220.

[0061] FIG. 3 is a conceptual diagram of the operation of the autonomous driving control device according to one embodiment of the present invention. According to an embodiment, the autonomous driving control device 300 (e.g., the autonomous driving control device 100, 200 in FIG. 1 and / or FIG. 2) generates voxel data based on sensing data acquired using at least one sensor and stores the voxel data in the memory 320 (e.g., the memory 120, 220 in FIG. 1 and / or FIG. 2). For example, the autonomous driving control device 300 acquires third sensing data using the second sensor 314 (e.g., 3D lidar). The third sensing data may be sensing data acquired using the second sensor 314 after the autonomous driving control device 300 acquires the second sensing data described above in the description of FIG. 2. For example, the third sensing data includes at least one point cloud data.

[0062] With reference to reference number S310, for example, the autonomous driving control device 300 generates second voxel data based on the third sensing data. As an example, the autonomous driving control device 300 generates the second voxel data through a voxel operation on at least a portion of the third sensing data. With reference to reference number S320, for example, the autonomous driving control device 300 identifies current position information of the traveling device. As an example, the autonomous driving control device 300 identifies current position information of the traveling device based on a matching result generated by matching the first voxel data and the second voxel data. As an example, the autonomous driving control device 300 matches the first voxel data stored in the memory 320 with the generated second voxel data.

[0063] For example, the autonomous driving control device 300 may apply various types of point cloud registration (e.g., ICP (iteractive closest point)) algorithms to generate matching results between the first voxel data and the second voxel data.

[0064] As an example, the autonomous driving control device 300 performs matching by setting a weight value for a portion of the first voxel data and the second voxel data that does not satisfy a specified condition to a first value. The first value may be, for example, 1. As an example, the autonomous driving control device 300 performs matching by setting a weight value for a portion that satisfies a specified condition including at least one of a first specified object, a second specified object, or a combination of either of them to a second value greater than the first value. The second value may be, for example, a number greater than the first value (e.g., 1).

[0065] The following formula may be a formula used by the autonomous driving control device 300 to obtain the current position information of the driving device by modifying a part of the ICP algorithm using different weighting values.

[0066]

number

[0067] For example, s i may be point cloud data and / or voxel data currently acquired by the autonomous driving control device 300. For example, d i may be point cloud data and / or voxel data included in the map data. For example, i d i For example, w i may be a weighting value. For example, the autonomous driving control device 300 sets the weighting value to a different value depending on whether the designated object is related or not. For example, the autonomous driving control device 300 stores various data generated through the above-mentioned operations in the memory 320.

[0068] Fig. 4 is an operation flow diagram of an autonomous driving control device according to an embodiment of the present invention. According to one embodiment, the autonomous driving control device (e.g., the autonomous driving control device 100 in Fig. 1) performs the operations disclosed in Fig. 4. For example, at least some of the components included in the autonomous driving control device (e.g., at least one of the sensor unit 110, memory 120, control unit 130, or any combination thereof in Fig. 1) may be configured to perform the operations in Fig. 4.

[0069] In the following embodiments, the operations of S410 to S480 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In addition, the contents that correspond to or overlap with the contents described above with reference to FIG. 4 will be briefly described or omitted.

[0070] According to an embodiment, the autonomous driving control device performs object detection based on an image (S410). For example, the autonomous driving control device detects an object based on image data acquired using at least one camera. As an example, the autonomous driving control device identifies a bounding box (Bbox) corresponding to the object.

[0071] According to one embodiment, the autonomous driving control device determines a sampling rate to be applied for sampling for each object (S420). For example, if the object satisfies a specified condition, the autonomous driving control device performs sampling using a relatively high sampling rate. According to one embodiment, the autonomous driving control device detects blobs based on point cloud data among data acquired using at least one LIDAR (S430).

[0072] According to one embodiment, the autonomous driving control device matches a blob and a bounding box (Bbox) (S440). For example, the autonomous driving control device determines whether the blob and the bounding box match (or the degree of identity). According to one embodiment, the autonomous driving control device determines whether an object of interest blob is detected (S450).

[0073] For example, the autonomous driving control device compares a bounding box that satisfies a specified condition with at least one blob, and then detects a blob that is determined to be the same as the bounding box that satisfies the specified condition (or is determined to be an object of interest) as a result of the comparison. For example, if an object of interest blob is detected (e.g., step S450-Yes), the autonomous driving control device performs step S468. For example, if an object of interest blob is not detected (e.g., step S450-No), the autonomous driving control device performs step S464. According to one embodiment, the autonomous driving control device applies a general sampling rate (S464).

[0074] For example, if there is no object blob of interest, the autonomous driving control device performs sampling on all sensing data acquired using the LIDAR using a general sampling rate. The autonomous driving control device identifies the general sampling rate using table data related to the sampling rate stored in the memory.

[0075] According to one embodiment, the autonomous driving control device applies the sampling rate for the object of interest to the data corresponding to the object of interest (S468). For example, the autonomous driving control device identifies a sampling rate for sampling the data corresponding to the object of interest (e.g., point cloud data). As an example, the sampling rate for the object of interest may be equal to or greater than the general sampling rate of step S464.

[0076] According to one embodiment, the autonomous driving control device performs voxelization on at least a portion of the sampling results (S470). According to one embodiment, the autonomous driving control device stores at least a portion of the voxelization results in a memory (S480). For example, the autonomous driving control device stores map data including first voxel data generated as a result of the voxelization in a memory.

[0077] Fig. 5 is an operation flow diagram of an autonomous driving control device according to an embodiment of the present invention. According to one embodiment, the autonomous driving control device (e.g., the autonomous driving control device 100 in Fig. 1) performs the operations disclosed in Fig. 5. For example, at least some of the components included in the autonomous driving control device (e.g., at least one of the sensor unit 110, memory 120, control unit 130, or any combination thereof in Fig. 1) may be configured to perform the operations in Fig. 5.

[0078] In the following embodiments, the operations from S510 to S530 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation may be changed, or at least two operations may be performed in parallel. In addition, the contents that correspond to or overlap with the contents described above with reference to FIG. 5 will be briefly described or omitted. In addition, in the following embodiments, the operations from S510 to S530 may be operations that follow the operation of S480 in FIG. 4 described above, but the embodiments of the present invention are not limited thereto.

[0079] According to one embodiment, the autonomous driving control device voxelizes data acquired using a LIDAR (S510). For example, the autonomous driving control device stores map data including the first voxel data in a memory as shown in FIG. 4, and then generates second voxel data based on sensing data acquired using a LIDAR. According to one embodiment, the autonomous driving control device matches the map data and the voxelized data (e.g., the second voxel data) using an algorithm based on a specified weight value (S520).

[0080] For example, the autonomous driving control device may apply various types of point cloud registration (e.g., ICP (iteractive closest point)) algorithms to generate matching results between the first voxel data and the second voxel data.

[0081] As an example, the autonomous driving control device performs matching by setting a weighting value for a portion of the first voxel data and the second voxel data that does not satisfy a specified condition as a first value. The first value may be, for example, 1. As an example, the autonomous driving control device performs matching by setting a weighting value for a portion that satisfies a specified condition including at least one of a first designated object, a second designated object, or a combination thereof as a second value greater than the first value. The second value may be, for example, a number greater than the first value (e.g., 1). According to an embodiment, the autonomous driving control device acquires position information of the traveling device based on the matching result (S530).

[0082] Fig. 6 is an operation flow diagram of an autonomous driving control device according to an embodiment of the present invention. According to one embodiment, the autonomous driving control device (e.g., the autonomous driving control device 100 in Fig. 1) performs the operations disclosed in Fig. 6. For example, at least some of the components included in the autonomous driving control device (e.g., at least one of the sensor unit 110, memory 120, control unit 130, or any combination thereof in Fig. 1) may be configured to perform the operations of Fig. 6.

[0083] In the following embodiments, the operations of S610 to S650 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. In addition, the contents that correspond to or overlap with the contents described above with reference to FIG. 6 will be briefly described or omitted.

[0084] According to one embodiment, the autonomous driving control device acquires first sensing data related to a driving route of the driving device using a first sensor (S610). According to one embodiment, a first designated object that is included in the first sensing data and satisfies a designated condition is identified, and a first sampling rate to be applied to first point cloud data corresponding to the first designated object is identified (S620). For example, the autonomous driving control device identifies the first sampling rate using a table stored in a memory. According to one embodiment, the autonomous driving control device acquires second sensing data related to the driving route using a second sensor (S630).

[0085] According to one embodiment, the autonomous driving control device identifies a second designated object included in the second sensing data, and identifies a second sampling rate to be applied to the second point cloud data corresponding to the second designated object based on a comparison result between the first designated object and the second designated object (S640). For example, if the first designated object and the second designated object are identified as the same, the autonomous driving control device identifies the second sampling rate as a value equal to or greater than the first sampling rate. In other words, if the first designated object and the second designated object are identified as the same, the autonomous driving control device identifies the second sampling rate as a value equal to or greater than the first sampling rate.

[0086] For example, if the first designated object and the second designated object are not the same or the second sensing data does not include the second designated object, the autonomous driving control device identifies a predefined sampling rate (e.g., a sampling rate stored in the memory 120) as the second sampling rate. The autonomous driving control device may sample the second sensing data including the second designated object using the predefined sampling rate. For example, the predefined sampling rate may be lower than the first sampling rate.

[0087] According to one embodiment, the autonomous driving control device samples the first sensing data and the second sensing data using a first sampling rate and a second sampling rate, respectively (S650).

[0088] 7 illustrates a computer system relating to an autonomous driving control method according to an embodiment of the present invention. Referring to FIG. 7, a computer system 1000 relating to an autonomous driving control method includes at least one processor 1100, a memory 1300, a user interface input device 1400, a user interface output device 1500, a storage 1600, and a network interface 1700, which are connected via a bus 1200.

[0089] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that performs processing on instructions stored in the memory 1300 and / or the storage 1600. The memory 1300 and the storage 1600 may include various types of volatile or non-volatile storage media. For example, the memory 1300 may include a read only memory (ROM) and a random access memory (RAM).

[0090] Thus, the steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware executed by processor 1100, in a software module, or in a combination of the two. The software module may reside in a storage medium (i.e., memory 1300 and / or storage 1600) such as RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, etc.

[0091] An exemplary storage medium is coupled to the processor 1100 such that the processor 1100 can read information from and write information to the storage medium. Alternatively, the storage medium may be integral to the processor 1100. The processor and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. Alternatively, the processor and the storage medium may reside as discrete components in a user terminal.

[0092] The above description is merely an illustrative example of the technical idea of ​​the present invention, and those having ordinary knowledge in the technical field to which the present invention belongs can make various modifications and variations without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are for illustrative purposes only, and are not intended to limit the technical idea of ​​the present invention, and the scope of the technical idea of ​​the present invention is not limited by such embodiments. The scope of protection of the present invention should be interpreted according to the following claims, and all technical ideas within the scope equivalent thereto should be interpreted as being included in the scope of the present invention. [Explanation of symbols]

[0093] 100 Autonomous driving control device 110 Sensor section 120 Memory 130 Control section 200 Autonomous Driving Control Device 220 Memory 300 Autonomous Driving Control Device 314 Second Sensor 320 Memory 1000 Computer Systems 1100 Processor 1200 Bus 1300 Memory 1400 User Interface Input Device 1500 User interface output device 1600 Storage 1700 Network Interface

Claims

1. In the autonomous driving control device, a sensor unit including at least one sensor; a memory for storing at least one instruction; and a control unit electrically connected to the sensor unit and the memory, When the instruction is executed by the control unit, the autonomous driving control device: Obtaining first sensing data related to a travel path of a traveling device using a first sensor of the at least one sensor; identifying a first designated object included in the first sensing data and satisfying a designated condition; and identifying a first sampling rate to be applied to first point cloud data corresponding to the first designated object; acquiring second sensing data relating to the travel path using a second sensor of the at least one sensor; identifying a second designated object included in the second sensing data, and identifying a second sampling rate to be applied to second point cloud data corresponding to the second designated object based on a comparison result between the first designated object and the second designated object; An autonomous driving control device, configured to sample the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively.

2. When the instruction is executed by the control unit, the autonomous driving control device: Sampling the first point cloud data corresponding to the first designated object in the first sensing data using the first sampling rate; The autonomous driving control device of claim 1 , further comprising: a second sampling rate for sampling the second point cloud data corresponding to the second specified object in the second sensing data;

3. When the instruction is executed by the control unit, the autonomous driving control device:

2. The autonomous driving control device of claim 1, further comprising: a second sampling rate that is equal to or greater than the first sampling rate when the first designated object and the second designated object are identified as being the same object.

4. When the instruction is executed by the control unit, the autonomous driving control device:

2. The autonomous driving control device of claim 1, further comprising: a first sensing data unit configured to sample at least a portion of the second sensing data using a predefined sampling rate when the first designated object and the second designated object are not identical or when the second sensing data does not include the second designated object.

5. The autonomous driving control device according to claim 4 , wherein the predefined sampling rate is smaller than the first sampling rate.

6. When the instruction is executed by the control unit, the autonomous driving control device: generating first voxel data including position information of at least one of the first designated object, the second designated object, or a combination thereof, based on at least one of the sampled first sensing data, the sampled second sensing data, or a combination thereof; The autonomous driving control device according to claim 1 , further comprising: storing map data including the first voxel data in the memory.

7. When the instruction is executed by the control unit, the autonomous driving control device: generating second voxel data based on third sensing data acquired using the second sensor; The autonomous driving control device according to claim 6, further comprising: a controller configured to determine a current position of the traveling device based on a matching result between the first voxel data and the second voxel data included in the map data.

8. When the instruction is executed by the control unit, the autonomous driving control device: setting a weight value for a portion of the first voxel data and the second voxel data that does not satisfy the specified condition as a first value and performing matching; 8. The autonomous driving control device of claim 7, wherein the device is configured to set a weighting value for a part that satisfies the specified condition, including at least one of the first specified object, the second specified object, or a combination thereof, to a second value greater than the first value to perform matching.

9. The second voxel data is The autonomous driving control device of claim 7, further comprising voxel coordinates of a blob corresponding to at least one object that satisfies the specified condition.

10. When the instruction is executed by the control unit, the autonomous driving control device: The autonomous driving control device according to claim 7 , wherein the device is configured to generate the matching result by applying a point cloud registration algorithm to the first voxel data and the second voxel data.

11. In the autonomous driving control method, A step of the control unit acquiring first sensing data related to a travel path of the traveling device using a first sensor; The control unit identifies a first designated object included in the first sensing data and satisfying a designated condition, and identifies a first sampling rate to be applied to first point cloud data corresponding to the first designated object; The control unit acquires second sensing data regarding the travel route using a second sensor; The control unit identifies a second designated object included in the second sensing data, and identifies a second sampling rate to be applied to second point cloud data corresponding to the second designated object based on a comparison result between the first designated object and the second designated object; and The autonomous driving control method includes a step of: sampling the first sensing data and the second sensing data using the first sampling rate and the second sampling rate, respectively, by the control unit.

12. The step of sampling the first sensing data and the second sensing data by the control unit using the first sampling rate and the second sampling rate, respectively, further comprises: the control unit sampling the first point cloud data corresponding to the first designated object in the first sensing data using the first sampling rate; and 12. The autonomous driving control method of claim 11, further comprising: sampling the second point cloud data corresponding to the second specified object in the second sensing data using the second sampling rate.

13. The step of the control unit identifying the second sampling rate includes:

12. The autonomous driving control method of claim 11, further comprising: when the first designated object and the second designated object are identified as being the same, identifying the second sampling rate as being equal to or greater than the first sampling rate.

14. The step of sampling the first sensing data and the second sensing data by the control unit using the first sampling rate and the second sampling rate, respectively, further comprises:

12. The autonomous driving control method of claim 11, further comprising: sampling at least a portion of the second sensing data using a predefined sampling rate when the first designated object and the second designated object are not identical or the second sensing data does not include the second designated object.

15. The method of claim 14, wherein the predefined sampling rate is smaller than the first sampling rate.

16. The autonomous driving control method includes: generating first voxel data including position information of at least one of the first designated object, the second designated object, or a combination thereof, based on at least one of the sampled first sensing data, the sampled second sensing data, or a combination thereof, by the control unit; and The method of claim 11, further comprising: storing, in a memory, map data including the first voxel data, in the control unit.

17. The autonomous driving control method includes: generating second voxel data based on third sensing data acquired by the second sensor; and The autonomous driving control method of claim 16, further comprising: identifying a current position of the traveling device based on a matching result between the first voxel data and the second voxel data included in the map data by the control unit.

18. The step of the control unit identifying the current position of the traveling device based on the matching result between the first voxel data and the second voxel data included in the map data includes: the control unit sets a weight value for a portion of the first voxel data and the second voxel data that does not satisfy the specified condition as a first value, and performs matching; and 20. The autonomous driving control method of claim 17, further comprising: setting a weight value for a part that satisfies the specified condition, including at least one of the first specified object, the second specified object, or a combination thereof, to a second value greater than the first value, and performing matching.

19. The second voxel data is The autonomous driving control method of claim 17, further comprising: voxel coordinates of a blob corresponding to at least one object that satisfies the specified condition.

20. The step of the control unit identifying the current position of the traveling device based on the matching result between the first voxel data and the second voxel data included in the map data includes:

18. The method of claim 17, further comprising: generating the matching result by applying a point cloud registration algorithm to the first voxel data and the second voxel data.

Citation Information

Patent Citations

  • Guide controller for unmanned vehicle and controlled variable calculator for the same

    JP1991192407A