JP7333565B1 - Aircraft and method of controlling the aircraft - Google Patents

Abstract

[Problem] To provide a flying object and a control method for the flying object that can fly stably even when moving between GNSS and non-GNSS environments. [Solution] A flight that autonomously flies based on position information of the flying object UAV. A method for controlling an aircraft UAV, wherein the aircraft UAV has an RTK-GNSS positioning function, a sensor positioning function, a range measurement function, a direction acquisition function, and a flight control method for controlling the flight of the aircraft based on these information. The control unit C constantly grasps the position of the flying object UAV using the RTK-GNSS positioning function and the sensor positioning function, and the RTK-GNSS positioning function obtains a FIX solution. In this case, the control unit C controls the flight of the aircraft based on the position information of the aircraft UAV obtained from the RTK-GNSS positioning function, and when the RTK-GNSS positioning function cannot obtain a FIX solution, , the control unit C controls the flight of the aircraft based on the position information of the control unit C obtained from the sensor positioning function. [Selection diagram] Figure 1

Description

The present invention relates to an aircraft and a control method for the aircraft.

In recent years, small unmanned aerial vehicles have been used to inspect bridges, structures, etc. (hereinafter these may be referred to as structures, etc.). For example, small unmanned aerial vehicles are also used to inspect structures for cracks and paint peeling. Small unmanned aerial vehicles are also used to assess the health of infrastructure structures. For example, a 3D model and a developed (orthorectified) image are created from images captured using an imaging device (for example, a digital camera, an infrared camera, etc.) mounted on a multicopter. If such a 3D model and developed (ortho) image can be created, it will be possible to improve the efficiency and accuracy of inspection work for infrastructure structures.

In inspections using such small unmanned aerial vehicles, inspections by autonomous flight are being considered, and small unmanned aerial vehicles equipped with control devices are being developed and put into practical use. A small unmanned aerial vehicle equipped with an autonomous flight control device (hereinafter simply referred to as an autonomous flying small unmanned aerial vehicle) grasps its current position in cooperation with the Global Navigation Satellite System (GNSS), and aims Autonomous flight control is performed so as to autonomously reach the position. When using such autonomous flying small unmanned aerial vehicles for inspection of bridges, structures, etc., it is necessary to grasp the inspection position with high accuracy. An error of about ten meters occurs in the estimated accuracy. Therefore, when inspecting bridges, structures, etc., it is necessary to use an autonomously flying small unmanned aerial vehicle that employs the RTK-GNSS positioning method and has cm-order position estimation accuracy.

In addition, in the RTK-GNSS positioning method, the autonomous flying small unmanned aerial vehicle grasps its own position based on the radio waves emitted from four or more GNSS satellites and the signals from the reference station. When the strength is weak, when radio waves from four or more GNSS satellites cannot be received, when signals from the reference station cannot be obtained, etc., the autonomous flying small unmanned aerial vehicle cannot grasp its own position. For example, if the above-mentioned structures, such as bridges, are to be inspected, if there are places below the bridge or places surrounded by walls of the structure, etc., the radio waves from the GNSS satellite cannot reach them. , it may become impossible to grasp its own position using GNSS. As a result, the autonomously flying small unmanned aerial vehicle cannot accurately determine its own position, so there is a possibility that the accuracy of the inspection will be lowered or that the inspection will not be possible.

Therefore, in an autonomous flying small unmanned aerial vehicle, when grasping its own position, there is a positioning mode (RTK-GNSS positioning mode) that grasps its own position by the RTK-GNSS positioning method, and an image taken by a camera and an acceleration sensor Techniques for switching between a positioning mode (sensor positioning mode) for grasping the position of oneself using a sensor, etc. have also been developed (see Patent Documents 1 to 4 and Non-Patent Document 1, for example). Specifically, in an environment where one's own position can be grasped using the RTK-GNSS positioning method (hereinafter sometimes referred to as a GNSS environment), one's position is grasped by the RTK-GNSS positioning mode, and the RTK-GNSS positioning method is used. In an environment where self-position cannot be grasped (hereinafter sometimes referred to as a non-GNSS environment), a technique for grasping self-position in a sensor positioning mode has been developed. In addition, in Non-Patent Document 1, RTK-GNSS positioning solution is used to determine whether the autonomous flying small unmanned aerial vehicle is in a GNSS environment or a non-GNSS environment, and to switch the positioning mode. disclosed.

JP 2016-111414 A JP 2018-156660 A JP 2021-157493 A JP 2021-157494 A

Ichiyo Saito et al., “Seamless Indoor/Outdoor Positioning Using RTK-GNSS Positioning Solution for Switching to Positioning Mode”, Proceedings of Annual Conference of Japan Photogrammetry Society, May 2022

However, when the autonomously flying small unmanned aerial vehicle moves between the GNSS environment and the non-GNSS environment, there is a difference between the own position grasped by the RTK-GNSS positioning mode and the self position grasped by the sensor positioning mode. If there is a deviation, there is a possibility that an abnormality will occur in the movement of the autonomous flying small unmanned aerial vehicle when the positioning mode is switched. For example, in the sensor positioning mode, even if the self-position in the absolute coordinate system of the autonomous flying small unmanned aerial vehicle and the self-position in the absolute coordinate system grasped in the sensor positioning mode match in the initial state, after a certain period of time has passed, Accumulated errors result in deviations between the self-position in the absolute coordinate system of the autonomously flying small unmanned aerial vehicle and the self-position in the absolute coordinate system ascertained in the sensor positioning mode. When switching from the sensor positioning mode to the RTK-GNSS positioning mode with the cumulative error accumulated, the flight controller of the autonomous flying small unmanned aerial vehicle flies the small unmanned aerial vehicle to the position in the absolute coordinate system grasped in the RTK-GNSS positioning mode. Since it is trying to move, there is a possibility that the small unmanned aerial vehicle will move abnormally due to the sudden change in position, and in the worst case, the small unmanned aerial vehicle may crash.

SUMMARY OF THE INVENTION In view of the above circumstances, an object of the present invention is to provide a flying object capable of stably flying even when moving between a GNSS environment and a non-GNSS environment, and a control method for the flying object.

<Control method of flying object>
A control method for a flying object according to a first aspect of the invention is a method for controlling a flying object that autonomously flies along a predetermined flight route based on position information of the flying object , wherein the flying object controls RTK-1 based on GNSS satellite signals. RTK-GNSS positioning function that estimates the position of the flying object by GNSS positioning, sensor positioning function that estimates the position of the flying object by positioning with the optical sensor mounted on the flying object, and the optical sensor mounted on the flying object A range finding function for measuring the distance to an object, a direction acquisition function for estimating the nose direction of the aircraft by a magnetic direction sensor mounted on the aircraft, and the RTK-GNSS positioning function or the sensor positioning function. a flight control function for controlling the flight of the aircraft based on the obtained position information of the aircraft, the distance information obtained by the range finding function, and the nose direction information obtained by the heading acquisition function; and the RTK-GNSS positioning function and the sensor positioning function both have a function of providing position information of the flying object to the flight control function in the NMEA format, When the RTK-GNSS positioning function obtains a FIX solution, the flight control function of the control unit controls the flight of the aircraft based on the position information of the aircraft obtained from the RTK-GNSS positioning function, and the RTK - if the GNSS positioning function cannot obtain a FIX solution, the flight control function of said control unit controls the flight of the aircraft based on the position information of the aircraft obtained from said sensor positioning function , and said RTK-GNSS; When transitioning from a GNSS environment in which the positioning function can obtain FIX solutions to a non-GNSS environment in which the RTK-GNSS positioning function cannot obtain FIX solutions, the flight control function of the control unit changes the measurement origin in the sensor positioning function. The timing at which the RTK-GNSS positioning function can no longer obtain a FIX solution by applying the position acquired by the RTK-GNSS positioning function immediately before the RTK-GNSS positioning function cannot obtain a FIX solution. In a short time, the flight of the aircraft is controlled based on the position information of the aircraft obtained from the sensor positioning function, and the GNSS positioning function is moved from a non-GNSS environment in which the GNSS positioning function cannot obtain a FIX solution. in the absolute coordinate system estimated by the sensor positioning function just before the GNSS positioning function can obtain the FIX solution. The position of the flying object in the absolute coordinate system estimated by the sensor positioning function so as to eliminate the difference between the position of the flying object and the position of the flying object in the absolute coordinate system estimated by the GNSS positioning function over a certain period of time. is characterized by adjusting the
A flying object control method according to a second invention is the flying object control method according to the first invention , wherein the optical sensor comprises six IMU stereo cameras for capturing images in the longitudinal direction, the lateral direction, and the vertical direction of the flying object, respectively; The function is characterized by estimating the position of each flying object based on the information of six IMU stereo cameras by Visual Odometry, and estimating the position of the flying object based on the median value of the estimated positions of multiple flying objects. and
A flying object control method according to a third aspect of the invention is the method according to the second aspect, wherein the control unit has a camera selection function for changing the signal of the IMU stereo camera used for estimating the position of the flying object in accordance with the flight scene of the flying object. In the GNSS environment, the camera selection function estimates the position of the flying object based on the position of the flying object obtained based on the information of the IMU stereo camera that shoots in the front-back direction and the left-right direction, and estimates the position of the flying object in a non-GNSS environment. In the environment, in estimating the position of six flying objects, the position of the flying object is estimated by increasing the weight of the position of the flying object obtained based on the information of the IMU stereo camera that shoots up and down. .
<Aircraft>
An aircraft according to a fourth aspect of the invention comprises an airframe, a propulsion rotor provided in the airframe, a GNSS receiver for receiving GNSS satellite signals, an optical sensor for measuring conditions around the airframe, and the GNSS receiver. a controller for controlling the operation of the propulsion rotors to control autonomous flight of the vehicle based on received GNSS satellite signals and/or information obtained by the optical sensors, wherein the controller controls the GNSS; RTK-GNSS positioning function for estimating the position of the aircraft by RTK-GNSS positioning based on GNSS satellite signals received by the receiver, and sensor positioning for estimating the position of the aircraft by positioning by the optical sensor mounted on the aircraft. a range finding function for measuring the distance to surrounding objects using an optical sensor mounted on the airframe; and a direction acquisition function for estimating the nose direction of the aircraft using a magnetic direction sensor mounted on the airframe; Based on the position information of the aircraft obtained by the RTK-GNSS positioning function or the sensor positioning function, the distance information obtained by the range finding function, and the nose direction information obtained by the bearing acquisition function and a flight control function for controlling the flight of the aircraft , wherein the RTK-GNSS positioning function and the sensor positioning function both provide position information of the aircraft to the flight control function in NMEA format. The control unit constantly grasps the position of the aircraft by the RTK-GNSS positioning function and the sensor positioning function, and when the RTK-GNSS positioning function obtains a FIX solution The control unit controls the flight of the aircraft based on the position information of the aircraft obtained from the RTK-GNSS positioning function, and if the RTK-GNSS positioning function cannot obtain a FIX solution, the control The unit has a function of controlling the flight of the aircraft based on the position information of the aircraft obtained from the sensor positioning function , and the flight control function of the control unit is such that the RTK-GNSS positioning function provides a FIX solution. When shifting from the obtained GNSS environment to a non-GNSS environment in which the RTK-GNSS positioning function cannot obtain a FIX solution, at the timing when the RTK-GNSS positioning function cannot obtain a FIX solution, from the sensor positioning function A function to shift to a state of controlling the flight of the aircraft based on the obtained position information of the aircraft, and a GNSS in which the GNSS positioning function can obtain a FIX solution from a non-GNSS environment in which the GNSS positioning function cannot obtain a FIX solution. When moving to the environment, the position of the flying object in the absolute coordinate system estimated by the sensor positioning function immediately before the GNSS positioning function can obtain a FIX solution, and the flying object estimated by the GNSS positioning function and a function of adjusting the position of the flying object in the absolute coordinate system estimated by the sensor positioning function so as to eliminate the difference from the position in the absolute coordinate system over a certain period of time. .
A flying object control method of a fifth invention is the flying object control method according to the fourth invention, wherein the optical sensor comprises six IMU stereo cameras for photographing the flying object in the longitudinal direction, the lateral direction, and the vertical direction, respectively, and the sensor The positioning function estimates the position of each flying object based on information from six IMU stereo cameras using Visual Odometry, and estimates the position of the flying object based on the median value of the estimated positions of multiple flying objects. Characterized by
A control method for a flying object according to a sixth aspect of the invention is the method according to the fifth aspect, wherein the control unit has a camera selection function of changing the signal of the IMU stereo camera used for estimating the position of the flying object in accordance with the flight scene of the flying object. In the GNSS environment, the camera selection function estimates the position of the flying object based on the position of the flying object obtained based on the information of the IMU stereo camera that shoots in the front-back direction and the left-right direction, and estimates the position of the flying object in a non-GNSS environment. In the environment, when estimating the position of six flying objects, it has a function to estimate the position of the flying object by increasing the weight of the position of the flying object obtained based on the information of the IMU stereo camera that shoots up and down. It is characterized by

<Control method of flying object>
According to the first invention, in a GNSS environment, the RTK-GNSS positioning function can estimate the accurate position of the flying object, and in a non-GNSS environment, the sensor positioning function can estimate the position of the flying object based on the signal of the optical sensor. The control unit's flight control functions allow the vehicle to fly autonomously in both GNSS and non-GNSS environments. In addition, the control unit always estimates the position of the flying object by the RTK-GNSS positioning function and the sensor positioning function, and furthermore, the position information to be used is determined by whether the RTK-GNSS positioning function can obtain a FIX solution. Because of the change, autonomous flight of the vehicle can be maintained seamlessly even when the vehicle moves between GNSS and non-GNSS environments. In addition, since the control based on the position information of the RTK-GNSS positioning function is switched to the control based on the position information of the sensor positioning function in a short time, it is possible to prevent the accuracy of the position control of the flying object from deteriorating. Furthermore, it is possible to prevent a decrease in accuracy of position control of the flying object. Since it takes a certain amount of time to switch from the control based on the position information of the sensor positioning function to the control based on the position information of the RTK-GNSS positioning function, it is possible to switch from the control based on the position information of the sensor positioning function to the control based on the RTK-GNSS positioning function. Abnormal movement of the flying object can be suppressed when switching to control based on position information.
According to the second invention, measurement errors can be eliminated simply and in real time, so the robustness of the position estimation of the flying object UAV by the sensor positioning function can be improved.
According to the third invention, it is possible to increase the accuracy of estimating the position of the flying object obtained by the sensor positioning function.
<Aircraft>
According to the fourth invention, in a GNSS environment, the RTK-GNSS positioning function can estimate the accurate position of the flying object, and in a non-GNSS environment, the sensor positioning function can estimate the position of the flying object based on the signal of the optical sensor. The control unit's flight control functions allow the vehicle to fly autonomously in both GNSS and non-GNSS environments. In addition, the control unit always estimates the position of the flying object by the RTK-GNSS positioning function and the sensor positioning function, and furthermore, the position information to be used is determined by whether the RTK-GNSS positioning function can obtain a FIX solution. Because of the change, autonomous flight of the vehicle can be maintained seamlessly even when the vehicle moves between GNSS and non-GNSS environments. In addition, since the control based on the position information of the RTK-GNSS positioning function is switched to the control based on the position information of the sensor positioning function in a short time, it is possible to prevent the accuracy of the position control of the flying object from deteriorating. Furthermore, it is possible to prevent a decrease in accuracy of position control of the flying object. Since it takes a certain amount of time to switch from the control based on the position information of the sensor positioning function to the control based on the position information of the RTK-GNSS positioning function, it is possible to switch from the control based on the position information of the sensor positioning function to the control based on the RTK-GNSS positioning function. Abnormal movement of the flying object can be suppressed when switching to control based on position information.
According to the fifth invention, it is possible to improve the robustness of position estimation of the flying object UAV by the sensor positioning function.
According to the sixth invention, it is possible to improve the accuracy of estimating the position of the flying object obtained by the sensor positioning function.

FIG. 2 is a schematic explanatory diagram of flight control of the aircraft UAV of the present embodiment; 1 is a schematic block diagram of an air vehicle UAV of this embodiment; FIG. It is a schematic explanatory drawing of the flying object UAV of this embodiment. FIG. 4 is a schematic control flow diagram when moving from a GNSS environment to a non-GNSS environment; FIG. 4 is a schematic control flow diagram when moving from a non-GNSS environment to a GNSS environment; It is a schematic photograph of Shikoku Saburo Bridge where the flight experiment was conducted. It is the figure which showed the passing point of the flying object. The flight trajectory is plotted on the satellite image, (A) is the flight trajectory estimated by the flight controller, and (B) is the flight trajectory obtained by motion capture, that is, the flight trajectory of the measured data. .

Next, embodiments of the present invention will be described with reference to the drawings.
A control method for a flying object according to the present invention is a method for controlling a flying object that autonomously flies based on the position information of the UAV, and enables smooth movement even when moving between a GNSS environment and a non-GNSS environment. It is characterized by making it possible.

The GNSS environment referred to in this specification means an environment in which the aircraft can grasp its own position using the RTK-GNSS positioning method, and the non-GNSS environment is difficult to grasp the position by the RTK-GNSS positioning method. It means an environment (or an environment in which the accuracy of positioning is degraded).

Further, the flying object control method of the present invention is suitable for autonomous flight of a flying object used for inspection of bridges, structures, etc. (hereinafter these may be referred to as structures, etc.). In other words, it is suitable for controlling a flying object that performs autonomous flight that requires position estimation accuracy on the order of centimeters. When conducting such inspections of structures, etc., measuring instruments for measuring objects to be inspected in structures, etc. are installed on the aircraft. For example, an inspection camera, an infrared camera, a laser scanner, or the like for photographing a structure or the like is provided on the aircraft.

It should be noted that the application for autonomous flight of a flying object by the control method of the flying object of the present invention is not limited to inspection of structures, etc., but can also be used as a control method for controlling the autonomous flight of a flying object used for disaster observation, environmental survey, etc. can do. In this case, an inspection camera, an infrared camera, a laser scanner, etc. are provided on the aircraft.

<Aircraft UAV>
First, before describing the control method of the flying object of the present embodiment, the flying object UAV of the present embodiment will be described.
In the following description, except for the basic configuration of the flying object UAV of the present embodiment, equipment necessary for controlling autonomous flight will be described, and equipment not particularly necessary for controlling autonomous flight (such as inspection equipment) will be described as appropriate. I omit the explanation.

<Aircraft B>
As shown in FIG. 3, the air vehicle UAV has a body B. As shown in FIG. The body B includes a body body BB and legs BL for placing the aircraft UAV on the ground provided below the body body BB. The airframe main body BB is provided with a propulsion rotor R, a control unit C, inspection equipment, a battery for supplying electric power to these equipment, and the like.

<Propulsion rotor R>
As shown in FIG. 3, the fuselage B is provided with three arms BA extending laterally from the fuselage body BB, and each arm BA is provided with a propulsion rotor R. Each propulsion rotor R has a propeller P and a motor for rotating the propeller P, and is attached to the arm BA via an attitude adjustment mechanism RT that adjusts the attitude with respect to the arm BA.

The number of propulsion rotors R provided in the aircraft UAV of this embodiment is not particularly limited. The aircraft UAV of this embodiment may have four or more propulsion rotors R as long as it can fly stably.
Further, the propulsion rotor R does not necessarily have to be attached to the arm BA via the attitude adjustment mechanism RT, and the propulsion rotor R may be fixed to the arm BA.

<Control section C>
A body main body BB of the aircraft UAV is provided with a controller C for controlling the operation of the aircraft UAV. The control unit C has a flight controller FC, a companion computer CC, a GNSS receiver GR, an IMU stereo camera IC, a magnetic orientation sensor MS, and a storage unit MR (see FIG. 2).

<Memory unit MR>
The storage unit MR stores information necessary for autonomous flight of the flying object UAV and work such as inspection. For example, the storage unit MR contains information about the flight route of the flying object UAV autonomously flying, information about the object to be inspected and the position to be inspected when inspecting a structure, etc., the angle of the gimbal, the camera shooting information and the like are stored. Information such as the flight route for autonomous flight, the object to be inspected, and the position is stored in the storage unit MR as information in the WGS84 coordinate system, but the information is not necessarily limited to the information in the WGS84 coordinate system. For example, the information of the PZ coordinate system or the plane orthogonal coordinate system may be used.

<Flight controller FC>
The flight controller FC has the function of controlling the flight of the air vehicle UAV. This flight controller FC controls the attitude, moving direction, moving speed, etc. of the flying object UAV based on information supplied from the companion computer CC. In other words, the flight controller FC controls the number of rotations of the motors of the propulsion rotors R and the attitude adjustment mechanism to control the attitude, movement direction, movement speed, etc. of the aircraft UAV so that the aircraft UAV is determined in advance. It is controlled to move along a predetermined flight route (see FIG. 1(B)).

<Companion computer CC>
The companion computer CC has a function of providing information necessary for flight control to the flight controller FC. Specifically, the companion computer CC carries out inspections and information from the GNSS receiver GR, the IMU stereo camera IC, and the magnetic direction sensor MS, as well as the autonomous flight route of the flying object UAV stored in the storage unit MR. The flight controller FC has a function of creating and providing information for adjusting the attitude, movement direction, and movement speed of the flying object UAV based on information such as the object and position. Then, the companion computer CC has a function of grasping the position of the flying object UAV based on the RTK-GNSS positioning function and the sensor positioning function, and a flying object UAV based on any function of the RTK-GNSS positioning function and the sensor positioning function. It also has a function (switching function) to decide whether to create information for controlling the flight of the aircraft. The companion computer CC also has a range finding function for measuring the distance between the flying object UAV and surrounding objects, and an azimuth acquisition function for estimating the heading direction of the flying object UAV.

<GNSS receiver GR>
The GNSS receiver GR has a function of receiving GNSS satellite signals transmitted from GNSS satellites ST, and has a function of receiving a plurality of GNSS satellite signals transmitted from a plurality of GNSS satellites. The GNSS receiver GR also has the function of receiving information transmitted from the reference station BS, that is, information regarding a plurality of GNSS satellite signals received by the reference station BS. The GNSS receiver GR then has the function of supplying the received signal to the companion computer CC.

<IMU stereo camera IC>
The IMU stereo camera IC is a stereo camera mounted with an Inertial Measurement Unit (inertial measurement unit: IMU). The flying object UAV of this embodiment has six IMU stereo camera ICs (see FIG. 2). That is, the aircraft UAV of this embodiment is provided with IMU stereo camera ICs on the front, rear, left, right, and top and bottom of the aircraft UAV.

The IMU is equipped with an acceleration sensor and an angular velocity sensor, and has a function of detecting three-dimensional inertial motion (translational motion and rotational motion in orthogonal three-axis directions) of the flying object UAV. Also, the stereo camera has a function of photographing the surroundings of the flying object UAV. The IMU stereo camera IC has a function of supplying the detected three-dimensional inertial motion of the UAV and the captured images to the companion computer CC.

Note that six IMU stereo camera ICs do not necessarily have to be provided, and at least one unit may be provided in front of the airframe B of the aircraft UAV. However, if six IMU stereo camera ICs are provided, the accuracy of estimating the position of the flying object UAV can be increased when estimating the position of the flying object UAV by visual odometry.

Specifically, the sensor positioning function of the companion computer CC obtains the median value of the position and orientation of the flying object UAV estimated based on the information of the six IMU stereo camera ICs. Then, measurement errors can be removed easily and in real time, so that the robustness of position estimation of the flying object UAV by the sensor positioning function of the companion computer CC can be improved.

In addition, the companion computer CC judges the flight scene of the flying object UAV, and the combination method of the six IMU stereo camera ICs, that is, the signal of the IMU stereo camera IC used for the position estimation of the flying object UAV. It has a function to change according to the scene (camera selection function). By providing a camera selection function, it is possible to improve the accuracy of position estimation by Visual Odometry. Determination of the flight scene of the flying object UAV by the camera selection function, as described later, is the determination of the GNSS environment and the non-GNSS environment depending on whether the RTK-GNSS positioning function has obtained a FIX solution, and the estimation by Visual Odometry It consists of determining the flight direction (front, back, left, right, up and down) relative to the nose. In the GNSS environment, it is possible to improve the position estimation accuracy of visual odometry by combining front, back, left, and right cameras, assuming that the estimation accuracy of visual odometry is degraded because the upward camera captures only the sky. When the altitude above the ground is high, the position estimation accuracy of Visual Odometry can be improved by combining a downward facing camera. In a non-GNSS environment, the imaging range in the vertical direction has more image features than in the front, rear, left, and right directions. Therefore, by increasing the weight of the estimation result of the upper and lower cameras, the position estimation accuracy by Visual Odometry can be improved.

In addition, if the surrounding image is measured by the IMU stereo camera IC, the distance to objects around the flying object UAV, such as a structure to be inspected, can be estimated based on the image using the range survey function of the companion computer CC. You can also Then, based on the estimated distance information (distance information), the flying object UAV can fly while maintaining a predetermined distance from the structure or the like. For example, it is necessary to prevent the flying object UAV from approaching a structure more than a certain amount, or to keep the distance from the structure to be inspected at a distance suitable for the work when performing work such as inspection. , it is possible to control the flight of the vehicle UAV based on the measured distance.

The stereo camera described above corresponds to the optical sensor recited in the claims. Note that the optical sensor does not necessarily have to be a stereo camera unless the position of the flying object UAV is estimated by visual odometry as will be described later. For example, instead of the stereo camera, a monocular camera, a TOF camera, or the like may be provided as an optical sensor to estimate the position of the flying object UAV.

Note that the control unit C may include a laser scanner, an ultrasonic sensor, a millimeter wave radar, etc., in addition to the sensors described above. If such sensors are provided, the ranging function can also measure the distance to objects, such as structures to be inspected, in the surroundings of the air vehicle UAV based on signals from these sensors.

<Magnetic direction sensor MS>
The magnetic direction sensor MS is, for example, a sensor such as an MR sensor or an MI sensor that detects the magnetic force of the earth. The magnetic direction sensor MS is provided to detect the orientation (orientation of the nose) of the UAV by detecting the orientation of the earth's magnetic force. The magnetic direction sensor MS also has the function of providing information about the direction of the detected earth's magnetic force to the companion computer CC. The companion computer CC has an azimuth acquisition function, which estimates the nose direction of the flying object UAV, that is, the nose direction in the WGS84 coordinate system, etc., based on the information on the detected direction of the earth's magnetic force. and create heading information. For example, if the magnetic direction sensor MS is installed in the body B of the flying object UAV so that the direction of the nose of the flying object UAV and the x-axis of the magnetic direction sensor MS match, the magnetic force detected by the magnetic direction sensor MS can estimate the orientation of the nose in the WGS84 coordinate system or the like.

<Explanation of functions of companion computer CC>
As noted above, companion computer CC performs the various functions described above. That is, it has a function of executing an RTK-GNSS positioning function, a sensor positioning function, a switching function, a range finding function, and an azimuth acquisition function (see FIG. 2). The ranging function and the azimuth acquisition function have the functions described above. In the following, the RTK-GNSS positioning function, the sensor positioning function and the switching function are described in detail.

<RTK-GNSS positioning function>
The RTK-GNSS positioning function has a function of calculating the current position of the flying object UAV, that is, the position of the flying object UAV on the WGS84 coordinate system, based on the GNSS satellite signals received by the GNSS receiver GR. Also, the RTK-GNSS positioning function has a function of calculating a positioning solution based on the received GNSS satellite signals. This positioning solution is a FIX solution, a FLOAT solution, a DGPS solution, or a single positioning solution. It has a function to create the position as NMEA format information. Below, the position information of the flying object UAV in the NMEA format may simply be referred to as the position information of the flying object UAV.

<Sensor positioning function>
The sensor positioning function is based on the images obtained by the stereo camera of the IMU stereo camera IC and the acceleration measured by the IMU. It is a function to calculate For example, as shown in FIG. 1B, when an aircraft UAV moves from a position P3 (original position) to a position P4 (current position), an image obtained by a stereo camera at position P3 and a stereo camera at position P4 By comparing the image obtained by the camera, the sensor positioning function calculates a motion vector from position P3 to position P4. Then, the sensor positioning function calculates the coordinates of the current position of the flying object UAV in the local coordinate system based on the coordinates of the original position, based on the calculated movement vector information. The coordinates of the current position of the air vehicle UAV in the local coordinate system can be created from information obtained by adding the motion vector to the coordinates of the original position. In addition, the sensor positioning function converts the position information of the flying object UAV in the local coordinate system into the position information of the flying object UAV in the WGS84 coordinate system by a known method, and based on the position information of the flying object UAV in this WGS84 coordinate system It has a function to create NMEA format information. Below, the position information of the flying object UAV in the NMEA format may simply be referred to as the position information of the flying object UAV.

Note that the method of comparing the images measured at each position by the sensor positioning function is not particularly limited. It can be implemented using common image matching methods. When comparing the images measured at each position, information such as the attitude and orientation of the flying UAV at positions P3 and P4 obtained based on information such as acceleration measured by the IMU of the IMU stereo camera IC is used. image matching.

In addition, the sensor positioning function uses the RTK- It also has the ability to use the coordinates obtained by the GNSS positioning function.

<Switching function>
The companion computer CC provides position control based on the position information of the UAV obtained by the RTK-GNSS positioning function and the position information of the UAV obtained by the sensor positioning function according to the environment in which the UAV is flying. It has a function of switching between position control based on Specifically, in the GNSS environment, position control is performed based on the position information of the flying object UAV obtained by the RTK-GNSS positioning function, and in the non-GNSS environment, the positioning is performed based on the position information of the flying object UAV obtained by the sensor positioning function. control. The switching function uses the position of the aircraft UAV obtained by the RTK-GNSS positioning function to perform position control when transitioning from the GNSS environment to the non-GNSS environment and from the non-GNSS environment to the GNSS environment. It has a function to switch between the information and the position information of the air vehicle UAV obtained by the sensor positioning function.

The switching function performs this switching based on the following information.
First, the companion computer CC constantly calculates the positioning solution by the RTK-GNSS positioning function and creates the position information of the flying object UAV in the WGS84 coordinate system obtained by the sensor positioning function.

If the positioning solution calculated by the RTK-GNSS positioning function is a FIX solution, the switching function provides the position information of the vehicle UAV obtained from the RTK-GNSS positioning function to the flight controller FC. That is, the switching function controls the information supplied to the flight controller FC such that the flight controller FC controls the flight of the vehicle UAV based on the vehicle UAV position information in NMEA format obtained from the RTK-GNSS positioning function. .

On the other hand, if the RTK-GNSS positioning function cannot calculate a FIX solution, that is, if the RTK-GNSS positioning function can only calculate a FLOAT solution, a DGPS solution, or a single positioning solution, the switching function will switch to the flight It feeds the position information of the body UAV to the flight controller FC. That is, the switching function controls the information supplied to the flight controller FC such that the flight controller FC controls the flight of the vehicle UAV based on the position information of the vehicle UAV in NMEA format obtained from the sensor positioning function.

<Control switching time>
Also, when shifting from the GNSS environment to the non-GNSS environment, the switching function converts the position information of the flying object UAV supplied to the flight controller FC from the position information of the flying object UAV of the RTK-GNSS positioning function to the sensor positioning function. position information of the air vehicle UAV. In this case, the switching function has a function of switching the position information of the flying object UAV to the position information of the flying object UAV of the sensor positioning function at the timing when the RTK-GNSS positioning function cannot obtain a FIX solution. That is, during a GNSS environment, the switching function provides position information of the vehicle UAV based on the RTK-GNSS positioning function to the flight controller FC, but when the situation arises where the RTK-GNSS positioning function cannot obtain a FIX solution, the switching function The function switches to provide position information of the air vehicle UAV based on sensor positioning functions to the flight controller FC in a short period of time (eg, within 1 second). Then, the difference between the position information of the flying object UAV by the RTK-GNSS positioning function and the position information of the flying object UAV by the sensor positioning function at the time of switching becomes small. Therefore, it is possible to prevent abrupt changes in the attitude, position, etc. of the flying object UAV due to the switching of information, thereby preventing deterioration in the accuracy of position control of the flying object UAV by the flight controller FC.

On the other hand, when shifting from the non-GNSS environment to the GNSS environment, the switching function converts the position information of the flying object UAV supplied to the flight controller FC from the position information of the flying object UAV of the sensor positioning function to the RTK-GNSS positioning function. position information of the air vehicle UAV. In this case, after the RTK-GNSS positioning function can obtain a FIX solution, the position information of the flying object UAV of the sensor positioning function and the position information of the flying object UAV of the RTK-GNSS positioning function will be The switching function has the ability to adjust the information supplied to the air vehicle UAV so that . In other words, when the switching function changes from the non-GNSS environment to the GNSS environment, the position information of the air vehicle UAV used for controlling the air vehicle UAV is immediately changed to the position information of the air vehicle UAV of the RTK-GNSS positioning function. I can't switch. Aircraft calculated by the sensor positioning function so that the position calculated by the sensor positioning function and the position calculated by the RTK-GNSS positioning function match during a certain transition period (including entering a certain range) Adjust UAV location information. Then, when the position information of the flying object UAV calculated by the adjusted sensor positioning function and the position information of the flying object UAV calculated by the RTK-GNSS positioning function match, the switching function is to supply the flying object UAV to the flight controller FC. The position information is switched to the position information of the air vehicle UAV of the RTK-GNSS positioning function.

The reason why a certain transition period is provided and the switching function adjusts the position information of the air vehicle UAV calculated by the sensor positioning function during this transition period is as follows.

In the non-GNSS environment, the sensor positioning function calculates the movement vector by Visual Odometry and grasps the current position by integrating the movement vector. However, the movement vector has an error with respect to the actual movement of the flying object UAV, and the error is also accumulated by accumulating the movement vector. In other words, the position information of the flying object UAV calculated by the sensor positioning function includes accumulated errors, and an error occurs between the flight route of the flying object UAV and the predetermined flight route by the sensor positioning function. (See FIG. 1(B)). Then, when moving from the non-GNSS environment to the GNSS environment, the information provided to the flight controller FC is simply changed from the position information of the flying object UAV of the sensor positioning function to the position information of the flying object UAV of the RTK-GNSS positioning function. When switched, the flight controller FC attempts to make large corrections to the position of the air vehicle UAV in a short period of time, which may cause the air vehicle UAV to become uncontrollable. Therefore, the switching function provides a certain transition period (for example, about several seconds) so that the position information of the flying object UAV calculated by the sensor positioning function and the position information of the flying object UAV calculated by the RTK-GNSS positioning function match. After entering the state, the position information of the flying object UAV supplied to the flight controller FC is switched to the information of the RTK-GNSS positioning function. Then, when the control based on the position information of the airborne UAV of the sensor positioning function is switched to the control based on the positional information of the airborne UAV of the RTK-GNSS positioning function, the occurrence of abnormal movement of the airborne UAV can be suppressed.

A method for matching the position information of the flying object UAV calculated by the sensor positioning function and the position information of the flying object UAV calculated by the RTK-GNSS positioning function is not particularly limited. For example, by gradually correcting the position information of the flying object UAV calculated by the sensor positioning function using the position information of the flying object UAV calculated by the RTK-GNSS positioning function, both can be matched. For example, correct the position information of the UAV so that the position information of the UAV calculated by the sensor positioning function is closer to the position information of the UAV calculated by the RTK-GNSS positioning function by a certain ratio. The corrected position information is created, and the corrected position information is supplied to the flight controller FC. By repeating this process, the difference between the position calculated by the sensor positioning function and the position calculated by the RTK-GNSS positioning function becomes smaller. The position can be matched with the position calculated by the RTK-GNSS positioning function.

<Flight control of the flying object UAV of the present embodiment>
If the flying object UAV of this embodiment has the configuration as described above, the flying object UAV can inspect structures while moving smoothly between the GNSS environment and the non-GNSS environment. can. In the following description, the case of flying a flight route that passes under the bridge BR will be described as a representative example.

<Movement in GNSS environment>
As shown in FIG. 4, when the air vehicle UAV flies from the GNSS environment, the companion computer CC first acquires the current position of the air vehicle UAV through the RTK-GNSS positioning function. That is, when the RTK-GNSS positioning function obtains the FIX solution, the companion computer CC determines the position in the WGS84 coordinate system calculated by the RTK-GNSS positioning function as the current position of the air vehicle UAV. The companion computer CC also has an orientation acquisition function that estimates the orientation of the air vehicle UAV based on the signal from the magnetic orientation sensor MS. When the current position and orientation of the air vehicle UAV are obtained, the companion computer CC calculates the position information of the air vehicle UAV of the RTK-GNSS positioning function, the information of the air vehicle UAV orientation, and the flight information stored in the storage unit MR. route information to the flight controller FC. Then, based on the information, the flight controller FC controls the rotation speed of the motor of each propulsion rotor R and the attitude adjustment mechanism so that the aircraft UAV moves along a predetermined flight route. The companion computer CC measures the distance between the UAV and the surrounding objects by the ranging function while the UAV is in flight.

While the air vehicle UAV is moving, the companion computer CC continuously acquires the current position of the air vehicle UAV by the RTK-GNSS positioning function. Here, "continuously acquiring the current position" means measuring the current position of the flying object UAV at a predetermined cycle (time interval). For example, the current position is acquired at about 5 to 10 Hz, and the obtained position information of the flying object UAV of the RTK-GNSS positioning function and the information of the direction of the flying object UAV (head direction information) are supplied to the flight controller FC. . Then, the flight controller FC confirms whether or not the acquired current position of the flying object UAV is located on the predetermined flight route. Move the air vehicle UAV back. Then, the flying object UAV can stably and autonomously fly on a predetermined flight route.

Here, while the flying object UAV is moving, the companion computer CC acquires the current position of the flying object UAV by the RTK-GNSS positioning function, and also acquires the current position of the flying object UAV by the sensor positioning function. . For example, a movement vector from the position (reference position) of the flying object UAV at a certain time T1 to the current position of the flying object UAV is calculated, and the current position of the flying object UAV is calculated by integrating the movement vector with the reference position. . In this case, the sensor positioning function calculates the current position of the air vehicle UAV using the value obtained by integrating the position in the WGS84 coordinate system calculated by the RTK-GNSS positioning function at time T1 with the movement vector as the position in the local coordinate system. Here, in the GNSS environment, assuming that the estimation accuracy of Visual Odometry is degraded because the upward camera captures only the sky, the position estimation accuracy of Visual Odometry is improved by combining front, rear, left, and right cameras. When the altitude above the ground is high, the position estimation accuracy of Visual Odometry is improved by combining a downward facing camera. The sensor positioning function performs similar processing while the position information of the flying object UAV of the RTK-GNSS positioning function is being supplied to the flight controller FC. However, the companion computer CC does not provide the air vehicle UAV position information for the sensor positioning function to the flight controller FC.

While the air vehicle UAV is flying along a predetermined flight route, the reception state of the GNSS satellite signals received by the GNSS receiver GR is not constant and changes. However, while the FIX solution is being obtained, the companion computer CC determines the position in the WGS84 coordinate system calculated by the RTK-GNSS positioning function as the current position of the air vehicle UAV.

<Transfer from GNSS environment to non-GNSS environment>
Eventually, when the air vehicle UAV tries to enter under the bridge BR (see FIG. 1), the reception condition of the GNSS satellite signal changes and a FIX solution cannot be obtained. Then, as shown in FIG. 4, the switching function of the companion computer CC supplies the position information of the flying object UAV of the sensor positioning function to the flight controller FC instead of the positional information of the flying object UAV of the RTK-GNSS positioning function. become. For example, within a short period of time (eg, within 1 second), the switching function of the companion computer CC supplies the position information of the air vehicle UAV of the RTK-GNSS positioning function to the position information of the air vehicle UAV of the sensor positioning function to the flight controller FC. become. In other words, the position information of the flying object UAV supplied to the flight controller FC can be changed with almost no time lag. Moreover, when the information of the flying object UAV of the sensor positioning function is supplied to the flight controller FC, the sensor positioning function will be the flight calculated by the RTK-GNSS positioning function when the FIX solution was finally obtained. Position information of the current position of the flying object UAV is calculated based on the position information of the position (reference position) of the UAV. Then, the deviation between the reference position and the current position can be reduced, so the stability of flight control can be maintained.

Moreover, the position information of the flying object UAV of the sensor positioning function is supplied to the flight controller FC as information in the same NMEA format as the position information of the flying object UAV of the RTK-GNSS positioning function. Then, the flight controller FC considers that the position information of the flying object UAV of the sensor positioning function is continuously supplied with the position information of the flying object UAV of the RTK-GNSS positioning function, and controls the flight of the flying object UAV. Therefore, stable flight control of the air vehicle UAV by the flight controller FC can be maintained.

<Movement in a non-GNSS environment>
When an air vehicle UAV enters a non-GNSS environment, such as under a bridge BR, the RTK-GNSS positioning function fails to obtain a FIX solution. Thus, as shown in FIG. 1, while an air vehicle UAV is moving, such as under a bridge, the companion computer CC provides position information of the air vehicle UAV for the sensor positioning function to the flight controller FC, allowing it to follow the flight route. Control the air vehicle UAV to move. Even in this case, the sensor positioning function calculates the position of the flying object UAV by Visual Odometry based on the information of the six IMU stereo camera ICs, so the calculated position of the flying object UAV can be highly accurate. Here, in a non-GNSS environment, image features are more abundant in the vertical direction than in the front, back, left, and right directions, so the camera selection function increases the weight of the estimation result of the upper and lower cameras, so that position estimation by Visual Odometry Improve accuracy. Note that the RTK-GNSS positioning function always calculates the positioning solution even in the non-GNSS environment.

<Moving from a non-GNSS environment to a GNSS environment>
When the air vehicle UAV escapes from under the bridge BR, the reception condition of GNSS satellite signals changes and becomes a condition in which a FIX solution can be obtained. In this state, there may be a large difference between the current position of the air vehicle UAV and the position of the air vehicle UAV immediately before switching. That is, there may be position differences due to accumulated errors in the sensor positioning function. In that state, if the UAV is controlled to move along the flight route using the position information of the UAV of the RTK-GNSS positioning function, there is a possibility that the abnormal movement of the UAV will occur due to a sudden position change. There is

Therefore, as shown in FIG. 5, the switching function of the companion computer CC ensures that the position information of the flying object UAV calculated by the sensor positioning function matches the position information of the flying object UAV calculated by the RTK-GNSS positioning function. Until then, the corrected position information of the flying object UAV calculated by the sensor positioning function is supplied to the flight controller FC. Eventually, when the position calculated by the sensor positioning function and the position calculated by the RTK-GNSS positioning function match, the switching function of the companion computer CC supplies the position information of the aircraft UAV of the RTK-GNSS positioning function to the flight controller FC. Then, the flight controller FC uses the position information of the flying object UAV of the RTK-GNSS positioning function to control the flying object UAV to move along the flight route.

By controlling as described above, even when the UAV moves between the GNSS environment and the non-GNSS environment, the UAV can smoothly fly autonomously.

It has been confirmed that the vehicle control method of the present invention allows the vehicle to smoothly transition between a GNSS environment and a non-GNSS environment.

In the experiment, we used an aircraft (test tricopter) in which three propulsion rotors were arranged at equal angular intervals on the circumference centered on the center of gravity of the aircraft (see Fig. 3).

The test tricopter was equipped with six IMU stereo cameras. In this experiment, a tracking camera (Intel Real sense T265) equipped with a visual odometry processing function was used as an IMU stereo camera.

The RTK-GNSS positioning module used mosaic-X5 (Septentrio) and an antenna.
IST8310 (CUAV NEOv2) was used for the magnetic orientation sensor.

A Jetson Nano (Nvidia) was used as the companion computer. Nora (CUAV) customized ArduCopter-4.0.7 was used as the flight controller.

As shown in Fig. 6, in the experiment, the test tricopter was allowed to fly autonomously so as to enter under the Shikoku Saburo Bridge from the outside of the Shikoku Saburo Bridge in Tokushima, and then leave from under the Shikoku Saburo Bridge. In addition, as shown in Fig. 7, a waypoint was prepared between the abutment and the pier of the bridge, and an autonomous flight route was set so that the test tricopter would pass through this waypoint.

An optical marker was attached to the flying object, and the flight trajectory and attitude of the flying object were obtained as actual measurement data. The flight trajectory of the flying object was acquired by a motion capture system.

Experimental results are shown below.
FIG. 8B shows the flight trajectory obtained by motion capture, that is, the flight trajectory of the actually measured data. Match. In other words, it was confirmed that the flight controller seamlessly estimated the self-position and trajectory data in the GNSS environment and the non-GNSS environment by the flight control method of the present invention.

The flying object control method of the present invention is also suitable as a control method for controlling the autonomous flight of flying objects used for inspection of bridges, structures, etc., disaster observation, environmental surveys, and the like.

UAV Aircraft B Airframe BB Airframe main body BA Arm R Rotor RT Attitude adjustment mechanism C Control section CC Companion computer FC Flight controller GR GNSS receiver IC IMU stereo camera MS Magnetic direction sensor MR Storage section

Claims (6)

A control method for a flying object that autonomously flies along a predetermined flight route based on position information of the flying object ,
the aircraft is
An RTK-GNSS positioning function for estimating the position of the aircraft by RTK-GNSS positioning based on GNSS satellite signals;
A sensor positioning function that estimates the position of the flying object by positioning with an optical sensor mounted on the flying object;
A ranging function that measures the distance to surrounding objects with an optical sensor mounted on the flying object,
An azimuth acquisition function for estimating the nose direction of the aircraft using a magnetic heading sensor mounted on the aircraft, position information of the aircraft obtained by the RTK-GNSS positioning function or the sensor positioning function, and the range finding function. a flight control function for controlling the flight of the aircraft based on the distance information obtained by and the nose direction information obtained by the heading acquisition function;
is provided with a control unit having
The RTK-GNSS positioning function and the sensor positioning function are
Both have the function of providing the position information of the flying object to the flight control function in the NMEA format,
When the RTK-GNSS positioning function obtains a FIX solution, the flight control function of the control unit controls the flight of the aircraft based on the position information of the aircraft obtained from the RTK-GNSS positioning function,
When the RTK-GNSS positioning function cannot obtain a FIX solution, the flight control function of the control unit controls the flight of the aircraft based on the position information of the aircraft obtained from the sensor positioning function ,
When transitioning from a GNSS environment in which the RTK-GNSS positioning function can obtain a FIX solution to a non-GNSS environment in which the RTK-GNSS positioning function cannot obtain a FIX solution, the flight control function of the control unit performs the sensor positioning Applying the position obtained by the RTK-GNSS positioning function just before the situation where the RTK-GNSS positioning function cannot obtain the FIX solution as a measurement origin in the function, the RTK-GNSS positioning function obtains the FIX solution. quickly shift to a state in which the flight of the flying object is controlled based on the position information of the flying object obtained from the sensor positioning function at the timing when it is no longer possible,
When transitioning from a non-GNSS environment in which the GNSS positioning function cannot obtain FIX solutions to a GNSS environment in which the GNSS positioning functions can The difference between the position of the flying object in the absolute coordinate system estimated by the sensor positioning function and the position of the flying object in the absolute coordinate system estimated by the GNSS positioning function immediately before the situation becomes available, over a certain period of time adjust the position of the vehicle in the absolute coordinate system estimated by the sensor positioning function to eliminate
A control method for a flying object, characterized by: The optical sensor is
Equipped with six IMU stereo cameras that capture the forward and backward directions, the left and right directions, and the vertical direction of the aircraft ,
The sensor positioning function is
Visual Odometry estimates the position of each flying object based on information from six IMU stereo cameras, and estimates the position of the flying object based on the median value of the estimated positions of multiple flying objects.
2. The method for controlling a flying object according to claim 1, wherein: The control unit
It has a camera selection function that changes the signal of the IMU stereo camera used for estimating the position of the aircraft according to the flight scene of the aircraft.
The camera selection function is
In the GNSS environment, estimating the position of the flying object based on the position of the flying object obtained based on the information of the IMU stereo camera that shoots in the front and back direction and the left and right direction,
In a non-GNSS environment, in estimating the position of six flying objects, the position of the flying object is estimated by increasing the weight of the position of the flying object obtained based on the information of the IMU stereo camera that shoots up and down. > The control method of the flying object according to claim 2, characterized in that: Airframe and
a propulsion rotor provided on the fuselage;
a GNSS receiver for receiving GNSS satellite signals;
an optical sensor that measures the surroundings of the airframe;
a control unit that controls the operation of the propulsion rotor to control the autonomous flight of the aircraft based on the GNSS satellite signals received by the GNSS receiver and/or the information acquired by the optical sensor;
The control unit
An RTK-GNSS positioning function for estimating the position of the aircraft by RTK-GNSS positioning based on the GNSS satellite signals received by the GNSS receiver;
a sensor positioning function for estimating the position of the aircraft by positioning using an optical sensor mounted on the aircraft;
A ranging function that measures the distance to surrounding objects with an optical sensor mounted on the aircraft;
A direction acquisition function for estimating a nose direction of the aircraft using a magnetic direction sensor mounted on the aircraft;
Based on the position information of the aircraft obtained by the RTK-GNSS positioning function or the sensor positioning function, the distance information obtained by the range finding function, and the nose direction information obtained by the bearing acquisition function and a flight control function for controlling the flight of the aircraft,
The RTK-GNSS positioning function and the sensor positioning function are
Both have the function of providing the position information of the flying object to the flight control function in the NMEA format,
The control unit
Constantly grasping the position of the aircraft by the RTK-GNSS positioning function and the sensor positioning function,
When the RTK-GNSS positioning function obtains a FIX solution, the control unit controls the flight of the aircraft based on the position information of the aircraft obtained from the RTK-GNSS positioning function,
When the RTK-GNSS positioning function cannot obtain a FIX solution, the control unit has a function of controlling the flight of the aircraft based on the position information of the aircraft obtained from the sensor positioning function ,
The flight control function of the control unit is
When moving from a GNSS environment in which the RTK-GNSS positioning function can obtain a FIX solution to a non-GNSS environment in which the RTK-GNSS positioning function cannot obtain a FIX solution, the RTK-GNSS positioning function cannot obtain a FIX solution. A function of shifting to a state of controlling the flight of the aircraft based on the position information of the aircraft obtained from the sensor positioning function at the timing when the position of the aircraft is lost;
When moving from a non-GNSS environment in which the GNSS positioning function cannot obtain a FIX solution to a GNSS environment in which the GNSS positioning function can obtain a FIX solution, immediately before the GNSS positioning function can obtain a FIX solution, the The sensor positioning function is configured to eliminate the difference between the position of the flying object in the absolute coordinate system estimated by the sensor positioning function and the position of the flying object in the absolute coordinate system estimated by the GNSS positioning function over a certain period of time. and a function of adjusting the position of the aircraft in the estimated absolute coordinate system.
An aircraft characterized by: The optical sensor is
Equipped with six IMU stereo cameras that capture the forward and backward directions, the left and right directions, and the vertical direction of the aircraft ,
The sensor positioning function is
Visual Odometry estimates the position of each flying object based on information from six IMU stereo cameras, and estimates the position of the flying object based on the median value of the estimated positions of multiple flying objects.
5. The aircraft according to claim 4, characterized in that: The control unit
It has a camera selection function that changes the signal of the IMU stereo camera used for estimating the position of the aircraft according to the flight scene of the aircraft.
The camera selection function is
In the GNSS environment, estimating the position of the flying object based on the position of the flying object obtained based on the information of the IMU stereo camera that shoots in the front and back direction and the left and right direction,
In a non-GNSS environment, when estimating the position of six flying objects, it has a function to estimate the position of the flying object by increasing the weight of the position of the flying object obtained based on the information of the IMU stereo camera that shoots up and down. The aircraft according to claim 5, characterized in that :

Images