A drone system that can safely land a drone by changing the angle of the 3D lidar

Abstract

The present invention relates to a drone system capable of safely landing a drone by varying the angle of a 3D LiDAR, and is characterized by comprising: a drone capable of autonomous flight and take-off and landing using a plurality of propellers; a sensing unit formed at the bottom of the drone and configured to collect 3D LiDAR information and image information to determine surrounding terrain and obstacles when the drone is autonomously flying and landing; and a variable unit formed between the sensing unit and the drone and configured to control the angle of the sensing unit to adjust the direction of collection of the 3D LiDAR information and the image information.

Description

Technology Field

[0001] The present invention relates to a drone system capable of safely landing a drone by varying the angle of a 3D lidar, and more specifically, to a drone system capable of safely landing a drone by varying the angle of a 3D lidar to obtain high-resolution ground information by increasing the density of cloud points for the landing point while varying the angle of the 3D lidar when obtaining ground information for the landing point of the drone. Background Technology

[0002] In recent years, various studies have been conducted on ways to utilize small unmanned aerial vehicles, such as multi-copters.

[0003] An unmanned aerial vehicle refers to an aircraft capable of automatically taking off, flying, and landing to achieve a specific purpose without human intervention, based on a programmed control method or control signals transmitted from an external source; it is also called a drone.

[0004] The applications of drones are endless, ranging from military reconnaissance drones to conventional fields such as unmanned cameras and unmanned parcel delivery services.

[0005] In particular, with the advancement of technology, the weight of the drones themselves has decreased, their travel distance has increased, and the roles of drones are becoming more diversified as they are equipped with various sensors.

[0006] When such drones perform designated missions through autonomous flight, they are configured to avoid obstacles by precisely analyzing various obstacles present in the flight path or landing point through video and 3D LiDAR.

[0007] In particular, 3D LiDAR is one of the important factors for determining the shape, size, and distance of obstacles, and there are various types of LiDAR sensors ranging from low resolution to high resolution.

[0008] While using high-resolution LiDAR sensors enables precise obstacle detection, allowing drones to fly or land stably, it is not only expensive but also presents the problem of requiring complex algorithms and long processing times to interpret the data.

[0009] On the other hand, when using equipment that includes a low-resolution LiDAR sensor, the cost is low and objects close to the LiDAR sensor have a high density of cloud points, allowing the shape of the object to be detected and recognized. However, as the distance from the LiDAR sensor increases, the density of cloud points decreases, resulting in a problem of insufficient output data for recognizing and detecting objects.

[0010] Therefore, while drones equipped with high-resolution LiDAR sensors clearly detect the shape of an object due to their high resolution, drones equipped with low-resolution LiDAR sensors have a problem in that they cannot clearly detect the shape of an object as the object's distance decreases.

[0011] As a result, when the drone attempts to land, it determines that the ground information of the landing point is suitable for landing from a distance and attempts to land, but as the distance gets closer, the shape and size of the obstacle change, causing it to raise its altitude again to find another landing point or fail to recognize the obstacle and collide with it.

[0012] In addition, conventional drone landing guidance systems guide the drone to descend and land while maintaining a horizontal position when an infrared sensor detects the infrared light from the landing base, which had the problem of requiring such infrastructure or equipment to be installed in advance. Prior art literature

[0013] Korean Patent Publication No. 10-2019-0092752 The problem to be solved

[0014] The objective of the present invention to solve the above-mentioned problems is to provide a drone system that can safely land a drone by varying the angle of a 3D LiDAR, which enables safe landing even in environments where there is no landing station, by automatically determining a landing point by analyzing terrain features.

[0015] In addition, another objective of the present invention is to provide a drone system capable of safely landing a drone by varying the angle of a 3D lidar, which can acquire high-resolution obstacle and ground information at a landing site by changing the position where a laser is irradiated at the same altitude by varying the angle of a low-resolution lidar sensor, thereby improving the cloud point density of objects present during flight or at the landing site.

[0016] In addition, another objective of the present invention is to provide a drone system capable of safely landing a drone by varying the angle of a 3D lidar, which can replace expensive high-resolution lidar sensors by acquiring high-resolution data using an inexpensive low-resolution lidar sensor, thereby enabling safe autonomous flight, landing, and mission execution in complex environments using an inexpensive drone. means of solving the problem

[0017] A drone system of the present invention capable of safely landing a drone by varying the angle of a 3D LiDAR to solve the above problem comprises: a drone capable of autonomous flight and take-off and landing using a plurality of propellers; a sensing unit formed at the bottom of the drone and configured to collect 3D LiDAR information and image information to determine surrounding terrain and obstacles when the drone is autonomously flying and landing; and a variable unit formed between the sensing unit and the drone and configured to control the angle of the sensing unit to adjust the direction of collection of the 3D LiDAR information and the image information.

[0018] In addition, the sensing unit of the drone system capable of safely landing the drone by varying the angle of the 3D lidar of the present invention includes a lidar sensor formed to recognize and detect objects and acquire 3D lidar information by irradiating a laser in a horizontal direction toward the front, and an image sensor formed to acquire image information by capturing images of the front, and the drone is characterized by determining terrain features and obstacles using information collected from the lidar sensor and the image sensor to set a flight path or landing point.

[0019] In addition, the variable part of the drone system capable of safely landing a drone by varying the angle of the three-dimensional lidar of the present invention is characterized by comprising: a rotating member formed on the upper part of the sensing member and supporting the pitch angle of the sensing member so as to vary; a support member formed on both upper sides of the rotating member and supporting the rotating member so as to vary the angle of the rotating member; and a rotating motor formed on the side of the support member to control the pitch angle of the rotating member and to rotate repeatedly.

[0020] In addition, the sensing unit of the drone system capable of safely landing the drone by varying the angle of the 3D lidar of the present invention is characterized by increasing the density of cloud points by changing the position of the laser reaching the object while adjusting the pitch angle by the variable unit when the cloud point density of the object included in the 3D lidar information is low.

[0021] In addition, the sensing unit of the drone system capable of safely landing the drone by varying the angle of the 3D lidar of the present invention is capable of rotating while varying the pitch angle by the variable unit, and is characterized by irradiating a laser to collect the 3D lidar information whenever the pitch angle is varied, collecting the cloud points of the object by angle, and then merging them to improve the vertical resolution by increasing the density of the cloud points of the object.

[0022] In addition, the sensing unit of the drone system capable of safely landing the drone by varying the angle of the 3D lidar of the present invention is characterized by setting the distance value at which the laser reaches the object in an initial state where the pitch angle is not varied as zero, and correcting the distance value by converting the distance value of the laser at which the object reaches the object in a state where the pitch angle is varied by the variable unit to zero. Effects of the invention

[0023] As described above, according to the drone system that can safely land a drone by varying the angle of the 3D lidar according to the present invention, the drone analyzes terrain features and automatically determines a landing point, thereby enabling a safe landing even in an environment where there is no landing station for the drone.

[0024] In addition, according to the drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention, by varying the angle of a low-resolution lidar sensor, the position where a laser is irradiated at the same altitude is changed, thereby improving the density of cloud points of objects present during flight or at the landing point, and thus enabling the acquisition of high-resolution obstacle and ground information at the landing point.

[0025] In addition, according to the drone system of the present invention, which can safely land a drone by varying the angle of a 3D lidar, high-resolution data can be acquired using an inexpensive low-resolution lidar sensor, thereby replacing an expensive high-resolution lidar sensor and enabling the safe performance of autonomous flight, landing, and missions in complex environments using an inexpensive drone. Brief explanation of the drawing

[0026] FIG. 1 is a front view showing a drone system capable of safely landing a drone by varying the angle of a three-dimensional lidar according to the present invention. FIG. 2 is a perspective view showing a variable part and a sensing part of a drone system that can safely land a drone by varying the angle of a three-dimensional lidar according to the present invention. FIG. 3 is a front view showing a variable part and a sensing part of a drone system that can safely land a drone by varying the angle of a three-dimensional lidar according to the present invention. FIG. 4 is a perspective view showing the angle of a variable part of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention in the pitch direction. FIG. 5 is a perspective view showing the variable part of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention, with the angle varied in the yaw direction. FIG. 6 is an exemplary diagram showing the appearance of a laser being first irradiated from a sensing unit of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention. FIG. 7 is an exemplary diagram showing a second laser irradiation in a sensing unit of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention. FIG. 8 is an exemplary diagram showing the third irradiation of a laser in a sensing unit of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention. FIG. 9 is an exemplary diagram showing a combination of lasers irradiated from a sensing unit of a drone system that can safely land a drone by varying the angle of a 3D lidar according to the present invention. Specific details for implementing the invention

[0027] The specific features and advantages of the present invention will be described in detail below with reference to the accompanying drawings. Prior to this, if it is determined that a detailed description of the functions and configurations related to the present invention may unnecessarily obscure the essence of the invention, such detailed description will be omitted.

[0028] The present invention relates to a drone system capable of safely landing a drone by varying the angle of a 3D lidar, and more specifically, to a drone system capable of safely landing a drone by varying the angle of a 3D lidar to obtain high-resolution ground information by increasing the density of cloud points for the landing point while varying the angle of the 3D lidar when obtaining ground information for the landing point of the drone.

[0029] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0030] FIG. 1 is a front view showing a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, FIG. 2 is a perspective view showing a variable part (300) and a sensing part (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, FIG. 3 is a front view showing a variable part (300) and a sensing part (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention.

[0031] As illustrated in FIGS. 1 to 3, a drone (100) system capable of safely landing the drone (100) by varying the angle of a three-dimensional lidar according to the present invention is characterized by comprising: a drone (100) capable of autonomous flight and take-off and landing using a plurality of propellers; a sensing unit (200) formed at the bottom of the drone (100) and formed to collect three-dimensional lidar information and image information to determine surrounding terrain and obstacles when the drone (100) is autonomously flying and landing; and a variable unit (300) formed between the sensing unit (200) and the drone (100) and capable of controlling the angle of the sensing unit (200) to adjust the direction of collection of three-dimensional lidar information and image information.

[0032] Additionally, the detection unit (200) includes a LiDAR sensor (210) formed to recognize and detect objects and acquire three-dimensional LiDAR information by irradiating a laser in a horizontal direction in front, and an image sensor (220) formed to acquire image information by capturing a video image in front, and the drone (100) is characterized by determining terrain features and obstacles using information collected from the LiDAR sensor (210) and the image sensor (220) to set a flight path or landing point.

[0033] Additionally, the variable part (300) is characterized by including a rotating member (310) formed on the upper part of the sensing part (200) and supporting the pitch angle of the sensing part (200) so that it can be varied, a support member (320) formed on both upper sides of the rotating member (310) and supporting the rotating member (310) so that the angle of the rotating member (310) can be varied, and a rotating motor (322) formed on the side of the support member (320) to control the pitch angle of the rotating member (310) and to allow it to rotate repeatedly.

[0034] The drone (100) is capable of autonomous flight using multiple propellers and can perform missions such as firefighting, search, rescue, surveillance, and exploration by equipping a separate module, and is configured to fly by automatically setting a path to perform a predetermined mission automatically without user control.

[0035] This drone (100) can detect and analyze terrain features and obstacles existing on the flight path in real time through the detection unit (200) via autonomous flight, and can reset the path to avoid the detected terrain features and obstacles and perform a set mission.

[0036] In addition, the drone (100) can automatically set a landing point and land for reasons such as mission execution, battery discharge, or return. The landing point must have flat terrain unlike the flight path, and after analyzing to ensure that no collision occurs during descent or that the aerodynamic flow is obstructed by trees or bushes, it can attempt to land at a safe point.

[0037] The sensing unit (200) is formed on the lower part of the drone (100) so that the drone (100) can safely perform autonomous flight and landing, and is used to collect data necessary for the flight of the drone (100) by irradiating a laser and taking images toward the front of the drone (100) or the landing point.

[0038] To this end, the detection unit (200) is formed such that the LiDAR sensor (210) and the image sensor (220) are positioned facing the same direction, and the LiDAR sensor (210) irradiates a laser in the direction in which the drone (100) is moving to collect three-dimensional LiDAR information, and the image sensor (220) generates an image of the direction in which the drone (100) is moving to provide data that allows the drone (100) to determine a flight path or landing point based on image information.

[0039] The lidar sensor (210) can collect cloud points regarding the shape, location, and size of terrain features and obstacles by repeatedly irradiating a laser in a horizontal direction forward to collect three-dimensional lidar information and using the reflected laser.

[0040] The image sensor (220) is used to combine images with three-dimensional cloud points collected through the lidar sensor (210) to fill the empty spaces between cloud points with images, thereby distinguishing the actual shape, location, and size, and finally, the three-dimensional lidar information and image information can be merged and used in the drone (100).

[0041] That is, the drone (100) can search, set, avoid, and fly an autonomous flight path through information merged with 3D LiDAR information and image information, and can land safely by clearly distinguishing the landing point, thereby reducing battery consumption due to multiple landing attempts and the risk of collision with terrain features and obstacles.

[0042] It is preferable that the LiDAR sensor (210) and the image sensor (220) be formed as a single module, but it is also possible to form them independently as needed and then merge only the collected information.

[0043] The variable part (300) is used to vary the angle in the yaw and pitch directions so that the sensing part (200) can collect three-dimensional LiDAR information and image information in the forward direction where the drone (100) is moving, and is automatically controlled by the drone (100) to change the position where the sensing part (200) irradiates a laser or takes a picture.

[0044] That is, the variable part (300) can vary the angle of the sensing part (200) toward the path set by the drone (100) when performing a mission through the autonomous flight of the drone (100), and when the drone (100) lands, the angle can be adjusted toward the lower direction of the drone (100) so that the sensing part (200) faces the ground.

[0045] The operation of the sensing unit (200) and the variable unit (300) will be explained in detail.

[0046] The LiDAR sensor (210) formed in the detection unit (200) scans surrounding objects or terrain features by rotating a laser beam formed inside in a horizontal direction, and calculates the distance using the time it takes for the irradiated laser to hit an object, reflect, and return, and can determine the position and shape of the object through the reflected position.

[0047] The lidar sensor (210) is formed in a cylindrical shape and has a window formed on the front for the laser to pass through. The size and width of the window may vary depending on the angle of view for scanning the laser beam in the horizontal direction.

[0048] Additionally, an image sensor (220) may be formed on the front of the lidar sensor (210), and it is desirable that the lidar sensor (210) and the image sensor (220) be able to collect information in the same direction, so that when the lidar sensor (210) is varied by the variable part (300), the image sensor (220) can also be varied in the same direction.

[0049] The lidar sensor (210) used in the present invention is described as a low-resolution, inexpensive lidar sensor (210) having four laser oscillation modules.

[0050] The lidar sensor (210) is formed so that it can be used by being coupled to the upper or lower part of the drone (100).

[0051] The pivoting device (310) is used to allow the pitch direction angle of the lidar sensor (210) to be varied, and the pitch refers to the direction in which the front and rear of the lidar sensor (210) rotate in the forward or rear direction with respect to the side of the lidar sensor (210).

[0052] In the present invention, the explanation is based on the assumption that the lidar sensor (210) is formed at the bottom of the drone (100), and as a result, the pivot (310) is coupled to the top of the lidar sensor (210).

[0053] The lower part of the pivoting member (310) is formed flat so that it can be closely attached to the upper part of the lidar sensor (210), and each side of the pivoting member (310) is protruded in an upward direction so that a pivoting member (311) is provided to serve as an axis for rotating the lidar sensor (210).

[0054] It is preferable that the pivoting member (311) protrudes in a triangular shape and has a rounded top, and that a hole is formed so that an axis can be joined to allow the pivoting member (310) to pivot.

[0055] In addition, a fixing bolt (312) for coupling with the lidar sensor (210) is formed in the center of the pivoting member (310), and the lidar sensor (210) and the lower part of the pivoting member (310) can be fixed in close contact by the fixing bolt (312).

[0056] The support member (320) is formed on the upper part of the pivot member (310) and is axially coupled with the pivot member (311) formed on the pivot member (310) to support the pivot member (310) and the lidar sensor (210) so that they can rotate relative to the support member (320).

[0057] At this time, the support member (320) is formed with a circular upper portion so that it can be directly connected to the drone (100), and the lower sides are provided with support members (321) that protrude outward toward the lower direction.

[0058] The support member (321) is formed in the shape of an inverted triangle, and the bottom is rounded and has a through hole formed so that it can be axially connected to the pivot member (311).

[0059] That is, the pivoting member (311) of the pivoting member (310) is positioned on the inner surface of the support member (321) formed on both lower sides of the support member (320), and the support member (321) and the pivoting member (311) can be connected to each other by connecting a shaft to each hole, and it is preferable that a bearing is formed in the hole formed in the support member (321) and a key for fixing the shaft is formed in the hole formed in the pivoting member (311) so that the pivoting member (311) can rotate according to the rotational direction of the shaft.

[0060] Through this, the pivoting member (311) can be rotated forward and backward along the rotational direction of the axis while supported by the axially coupled support member (321).

[0061] Additionally, a rotation motor (322) is coupled to the side of the support (320) to adjust the rotation angle of the rotation member (311) or to rotate it repeatedly, and the pitch angle of the rotation member (310) can be varied according to the rotation direction of the rotation motor (322).

[0062] By means of the rotation motor (322), the rotation platform (310) and the lidar sensor (210) can have their pitch angle varied, and the position of the laser beam irradiated in the vertical direction can be varied to increase the vertical resolution of the object in front.

[0063] The principle of how vertical resolution is improved will be described later with reference to the attached drawings.

[0064] Additionally, it is preferable that the rotation motor (322) be formed to control the rotation angle using a step motor, thereby allowing the rotation base (310) to rotate only up to a set pitch angle when rotating it in the forward or backward direction.

[0065] In the present invention, a stepper motor was used as an example, but the pitch angle can also be controlled using an inverter, a servo motor, or an encoder.

[0066] In addition, a separate gearbox is formed in the rotation motor (322) so that the rotation motor (322) can repeatedly rotate the rotation stage (310), and the gearbox is connected to a shaft that is key-coupled to a hole formed in the rotation stage (311) so that the shaft can be rotated.

[0067] Through this, the rotation motor (322) can rotate the rotation platform (310) while rotating only in a certain direction, and the pitch angle at which the rotation platform (310) and the LiDAR sensor (210) rotate inside the gearbox can also be set.

[0068] Additionally, the support (320) may be formed such that its angle varies in the yaw direction, allowing the lidar sensor (210) to rotate in the horizontal direction.

[0069] To this end, a fixing member (330) may be provided on the upper part of the support member (320), and the fixing member (330) may be pivotally connected to the center of the support member (320) so that the support member (320) can rotate with respect to the fixing member (330).

[0070] When the fixed member (330) is formed, the support member (320) is connected to the fixed member (330), and it is desirable that the fixed member (330) be connected to the drone (100) so that it can move along with the drone (100).

[0071] The fixed member (330) is formed so that it can be coupled to the lower part of the drone (100) by having fixed ends (331) protruding in a triangular shape on both upper sides, and a rotary motor (332) is fixed in the center and coupled thereto, and then the drive shaft of the rotary motor (332) is coupled to the support member (320) so that the support member (320) can be rotated in the yaw direction.

[0072] To this end, a through hole is provided in the center of the fixed member (330) and a bearing is inserted therein to prevent rotation by the drive shaft of the rotary motor (332), and a key is provided in the hole formed in the center of the support member (320) to be coupled with the drive shaft.

[0073] Depending on the rotation direction of the drive shaft, the angle of the support (320) varies, and the pivot (310) and the lidar sensor (210) can rotate in the left and right directions. Therefore, even if the low-cost lidar sensor (210) has a narrow field of view, the field of view can be expanded to a wider range by the rotation of the support (320).

[0074] It is preferable to use a stepper motor for the rotary motor (332) to vary the yaw angle of the support (320) and to numerically verify the varied angle, and the yaw angle may also be controlled using an inverter, a servo motor, or an encoder.

[0075] FIG. 4 is a perspective view showing the variable part (300) of a drone (100) system, which can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, with the angle varied in the pitch direction.

[0076] As illustrated in FIG. 4, the drone (100) system, which can safely land the drone (100) by varying the angle of the three-dimensional lidar according to the present invention, allows the pitch angle to be varied by the rotation of the rotation motor (322), thereby enabling the rotation platform (310) and the lidar sensor (210) to rotate in the forward and rear directions.

[0077] As the pitch angle varies, the laser irradiated from the lidar sensor (210) can be irradiated to different positions in the vertical direction compared to when the pitch angle is 0.

[0078] That is, even when using a low-cost LiDAR sensor (210), the laser can be irradiated while varying the pitch angle, so the cloud point density for recognizing objects at a distance can be improved.

[0079] FIG. 5 is a perspective view showing the variable part (300) of a drone (100) system, which can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, with the angle varied in the yaw direction.

[0080] As illustrated in FIG. 5, the drone (100) system, which can safely land the drone (100) by varying the angle of the three-dimensional lidar according to the present invention, allows the rotating platform (310) and the lidar sensor (210) to rotate in the left and right directions as the yaw angle is varied by the rotation of the rotary motor (332).

[0081] Through this, even a low-cost LiDAR sensor (210) with a narrow field of view when the laser is irradiated in a horizontal direction can have its field of view widened by varying the angle, thereby allowing for the acquisition of a larger amount of data.

[0082] FIG. 6 is an exemplary diagram showing a first irradiation (211) of a laser in a sensing unit (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, FIG. 7 is an exemplary diagram showing a second irradiation (212) of a laser in a sensing unit (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, FIG. 8 is an exemplary diagram showing a third irradiation (213) of a laser in a sensing unit (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention, FIG. 9 is an exemplary diagram showing a combination of lasers irradiated in a sensing unit (200) of a drone (100) system that can safely land a drone (100) by varying the angle of a three-dimensional lidar according to the present invention.

[0083] As illustrated in FIGS. 6 to 9, the sensing unit (200) of the drone (100) system, which can safely land the drone (100) by varying the angle of the three-dimensional lidar according to the present invention, is capable of rotating while varying the pitch angle by means of a variable unit (300), and is characterized by irradiating a laser to collect three-dimensional lidar information whenever the pitch angle is varied, collecting cloud points of an object by angle, and then merging them to improve vertical resolution by increasing the density of the object's cloud points.

[0084] Generally, inside the lidar sensor (210), oscillation modules for irradiating lasers are arranged at regular intervals in the vertical direction, and lasers are irradiated sequentially in the horizontal direction by an internal prism to scan and recognize surrounding objects.

[0085] The more expensive the LiDAR sensor (210), the more laser oscillation modules are arranged in the vertical direction, so the lasers irradiated in the vertical direction can be irradiated densely, allowing for the collection of a large amount of information for recognizing objects, and thus the recognition rate of objects located at a distance is also high.

[0086] However, in the case of low-cost laser oscillation modules, the number of vertically arranged laser oscillation modules is small, so for objects located at close range, the lasers emitted from all laser oscillation modules can reach and be reflected from the object, but for objects located at long range, only some of the laser oscillation modules reach the object, so there was a problem that the object recognition rate was very low.

[0087] In addition, while horizontal resolution can be easily obtained as the laser scans horizontally, vertical resolution is composed of only a few cloud points, making it difficult to accurately determine the shape.

[0088] That is, when the drone (100) lands, the 3D LiDAR information collected at a high altitude has a low cloud point density for distant objects, so when looking only at the initially collected information, the object resolution for terrain features or obstacles is low and it is determined that landing is possible, but when the altitude is lowered for landing, nearby objects are clearly recognized and it is determined that landing is impossible, so the altitude is raised and a new landing point is found again, and problems occur.

[0089] Therefore, there is a need to improve the resolution of 3D LiDAR information even from a long distance to clearly analyze the landing point at once, thereby inducing a safe landing of the drone (100).

[0090] As shown in FIG. 6, when a laser is first irradiated (211) from the lidar sensor (210), the first object (10) close to the lidar sensor (210) receives 4 lasers and can acquire 4 cloud points, but the second object (20) located at a distance receives only 2 lasers and thus acquires 2 cloud points.

[0091] That is, when the LiDAR sensor (210) is irradiated without varying the pitch angle, the shape, size, and location of the second object (20) located at a distance become unclear, and clear data of the second object (20) can only be obtained when it is brought closer.

[0092] For safe autonomous flight and landing, it is important to identify a second object (20) located at a distance in advance, so it is necessary to increase the cloud point density of the second object (20) located at a distance using a low-cost lidar sensor (210).

[0093] To this end, the LiDAR sensor (210) rotates forward and backward while varying the pitch angle by the rotation motor (322), and the vertical position where the laser is irradiated can be changed by irradiating the laser n times each time the angle is varied n times.

[0094] As shown in FIG. 7, the lidar sensor (210) is rotated to the rear, which is a different position from the first irradiation (211), and a laser is irradiated secondarily (212) to scan the first object (10) and the second object (20), and two additional cloud points for the first object (10) and the second object (20) can be acquired.

[0095] As shown in FIG. 8, the lidar sensor (210) is rotated forward to a position different from the first irradiation (211) and second irradiation (212) and can scan the first object (10) and the second object (20) by irradiating the laser a third time (213), and can additionally acquire two cloud points for the first object (10) and the second object (20).

[0096] At this time, the LiDAR sensor (210) irradiates a laser while varying the angle by a set pitch angle, the pitch angle can be set between -10 and 10 degrees, and the LiDAR sensor (210) can be configured to irradiate a laser whenever the pitch angle varies in increments of 0.1 to 1 degree.

[0097] The initial state of the lidar sensor (210) is preferably a state where the pitch angle is 0 degrees when flying, and when the drone (100) lands, the initial state can be defined as a state where the pitch angle is 90 degrees so that it faces the ground from the air.

[0098] That is, the variable part (300) can adjust the angle of the lidar sensor (210) so that the initial state of the pitch angle and yaw angle of the lidar sensor (210) changes according to the flight state and landing state, and the lidar sensor (210) irradiates a laser as the pitch angle changes to improve the density of the cloud point.

[0099] Accordingly, the LiDAR sensor (210) can be set to an initial state where the pitch angle is not varied, with the pitch angle being 0 degrees when flying and 90 degrees when landing, by the variable unit (300), and after correcting this initial state to the origin, the laser is irradiated while varying the pitch angle by the variable unit (300).

[0100] As the pitch angle unit of the laser irradiated from the lidar sensor (210) approaches 0.1 degrees, the number of cloud points that can be acquired increases, thereby enabling the acquisition of higher resolution information.

[0101] As shown in FIG. 9, the results of the laser irradiation (211), 2nd irradiation (212), and 3rd irradiation (213) are combined as the pitch angle of the lidar sensor (210) is varied, and the first object (10) at a short distance can acquire 12 cloud points, and the second object (20) at a long distance can acquire 6 cloud points.

[0102] In other words, compared to Figure 6, which was investigated without varying pitch angle, it can be seen that the cloud point has increased threefold.

[0103] In addition, the cloud points obtained in FIGS. 6 to 8 refer only to those where the laser is irradiated in a vertical direction at a specific location of the object. Since the laser is actually scanned in a horizontal direction, the number of cloud points in the vertical direction of the location scanned in the horizontal direction can be increased to obtain the shape of the object more precisely.

[0104] That is, the LiDAR sensor (210) scans by irradiating a laser at each set angle while the pitch angle is varied by the rotation motor (322), and by merging the cloud points collected at each pitch angle, the cloud point density for the first object (10) or the second object (20) can be increased to improve the vertical resolution.

[0105] Additionally, the sensing unit (200) is characterized by setting the distance value at which the laser reaches the object in an initial state where the pitch angle is not varied as zero, and correcting the distance value by converting the distance value of the laser at which the object reaches the object in a state where the pitch angle is varied by the variable unit (300) to zero.

[0106] When merging cloud points collected by irradiating a laser with a variable pitch angle, errors may occur in the distance values ​​for each cloud point depending on the pitch angle.

[0107] That is, the distance values ​​may differ between the laser irradiated at the initial position as in Fig. 6 and the laser irradiated with the pitch angle varied as in Fig. 7 or Fig. 8. To correct this, the initial position where the pitch angle is not varied is set as the zero point, and then the distance of the cloud point measured as the pitch angle varies is converted to the zero point.

[0108] This makes it possible to eliminate distance value errors for cloud points measured with each pitch angle varying.

[0109] In addition, to form a more accurate object shape as a cloud point, trigonometry is used to define the distance value at which the laser reaches the object with a variable pitch angle as the hypotenuse, and the horizontal distance, the base, can be calculated using hypotenuse x cosine(pitch angle), thereby allowing the distance value of the corresponding cloud point to be substituted for the base value.

[0110] In other words, since the distance can be changed to be equivalent to the horizontal laser irradiation distance using the base value, the error in distance values ​​for each cloud point is reduced when merging cloud points measured with varying pitch angles, thereby enabling stable securing of the object's shape and distance.

[0111] In addition, if necessary, both the distance measured in the initial state of the LiDAR sensor (210) and the method using trigonometry can be combined and used. In the case of trigonometry, if the difference between the distance measured in the initial state and the error range is 5% or more, it is desirable to apply the base derived through trigonometry.

[0112] Additionally, the detection unit (200) is characterized by increasing the density of cloud points by changing the position of the laser reaching the object while adjusting the pitch angle by the variable unit (300) when the cloud point density of the object included in the 3D LiDAR information is low.

[0113] The lidar sensor (210) scans horizontally when irradiating a laser, and may also be used to increase the cloud point density of a specific object to improve the processing speed of the data.

[0114] That is, in order to clearly recognize obstacles or terrain features existing in the direction in which the autonomous flying drone (100) is moving, the LiDAR sensor (210) has its pitch angle varied by the rotation motor (322) and intensively scans only the surroundings of the object until the unclear object can be clearly recognized.

[0115] At this time, it is desirable for the rotation motor (332) to rotate the support (320) according to the viewing angle of the lidar sensor (210) so that the lidar sensor (210) scans by intensively irradiating a narrow range centered on an unclear object with a laser, thereby increasing the cloud point density by intensively irradiating the laser in a short time and enabling the acquisition of clear data.

[0116] As described above, according to the drone system of the present invention, which can safely land a drone by varying the angle of a 3D lidar, the drone analyzes terrain features and automatically determines a landing point, thereby enabling safe landing even in environments where there is no landing station for the drone. Furthermore, by varying the angle of a low-resolution lidar sensor, the position where the laser is irradiated at the same altitude is changed, thereby improving the density of cloud points of objects present during flight or at the landing point, allowing for the acquisition of high-resolution information regarding obstacles and ground at the landing point. Since high-resolution data can be acquired using an inexpensive low-resolution lidar sensor, it is possible to replace expensive high-resolution lidar sensors and safely perform autonomous flight, landing, and missions in complex environments using an inexpensive drone.

[0117] As described above, although the present invention has been explained with reference to preferred embodiments, those skilled in the art may implement the present invention with various modifications or variations without departing from the technical spirit and scope described in the claims of the present invention. Accordingly, the scope of the present invention should be interpreted by the claims described to include such many variations. Explanation of the symbols

[0118] 10 : 1st object 20 : 2nd object 100 : Drone 200 : Sensing unit 210: LiDAR sensor 211: Primary investigation 212: 2nd investigation 213: 3rd investigation 220 : Image sensor 300 : Variable part 310 : Hoedongdae 311 : Hoedongdan 312 : Fixing bolt 320 : Support 321 : Support section 322 : Rotating motor 330 : Fixing bracket 331 : Fixing section 332 : Rotary motor

Claims

Claim 1 A drone system capable of autonomous flight and take-off and landing using multiple propellers; a sensing unit formed at the bottom of the drone and formed to collect 3D LiDAR information and image information to determine surrounding terrain and obstacles when the drone is autonomously flying and landing; and a variable unit formed between the sensing unit and the drone and capable of controlling the angle of the sensing unit to adjust the direction of collection of the 3D LiDAR information and the image information; wherein the sensing unit increases the density of cloud points by changing the position of the laser reaching the object while adjusting the pitch angle by the variable unit when the cloud point density of the object included in the 3D LiDAR information is low. Claim 2 A drone system capable of autonomous flight and take-off and landing using multiple propellers; a sensing unit formed at the bottom of the drone and formed to collect 3D LiDAR information and image information to determine surrounding terrain and obstacles when the drone is autonomously flying and landing; and a variable unit formed between the sensing unit and the drone and capable of controlling the angle of the sensing unit to adjust the direction of collection of the 3D LiDAR information and image information; wherein the sensing unit is capable of rotating as the pitch angle is varied by the variable unit, and whenever the pitch angle is varied, a laser is irradiated to collect the 3D LiDAR information to collect cloud points of an object by angle, and then merged to improve vertical resolution by increasing the density of the object's cloud points. Claim 3 A drone system capable of autonomous flight and take-off and landing using multiple propellers; a sensing unit formed at the bottom of the drone and configured to collect 3D LiDAR information and image information to determine surrounding terrain and obstacles when the drone is autonomously flying and landing; and a variable unit formed between the sensing unit and the drone and configured to control the angle of the sensing unit to adjust the direction of collection of the 3D LiDAR information and image information; wherein the sensing unit sets the distance value at which a laser reaches an object in an initial state where the pitch angle is not varied as zero, and corrects the distance value by converting the distance value of the laser reaching the object in a state where the pitch angle is varied by the variable unit to zero. Claim 4 A drone system capable of safely landing a drone by varying the angle of a 3D lidar, wherein, in any one of claims 1 to 3, the sensing unit comprises: a lidar sensor formed to recognize and detect an object by irradiating a laser in a horizontal direction toward the front and acquiring 3D lidar information; and an image sensor formed to acquire image information by capturing the front and acquiring a video image, wherein the drone is characterized by determining terrain features and obstacles using information collected from the lidar sensor and the image sensor to set a flight path or landing point. Claim 5 A drone system capable of safely landing a drone by varying the angle of a 3D LiDAR, characterized in that, in any one of claims 1 to 3, the variable part comprises: a pivot member formed on the upper part of the sensing part and supporting the pitch angle of the sensing part so as to vary; a support member formed on both upper sides of the pivot member and supporting the pivot member so as to vary the angle of the pivot member; and a rotation motor formed on the side of the support member to control the pitch angle of the pivot member and to rotate repeatedly. Claim 6 delete

Images