JP7368966B2 - Autonomous control device, autonomous control method, autonomous control program - Google Patents

Description

The present invention relates to an autonomous control device, an autonomous control method, and an autonomous control program for controlling a mobile device.

BACKGROUND ART Conventionally, in automatic running of a working machine such as a tractor, a technique has been proposed for controlling a steering angle by eliminating a deviation between an actual running position and a scheduled running position (for example, Patent Document 1).

JP 2019-41593 Publication

In the case where the working machine automatically travels, it is assumed that the planned route A1 is a straight route, as shown in FIG. 29, for example. The actual traveling route of the work equipment is affected by the unevenness of the ground, and if the traveling direction is not corrected, it will deviate from the planned route A1, so the work equipment must proceed so as to return to the planned route A1 at a predetermined timing. Correct direction. Therefore, for example, the actual route of the working machine is a bent route, such as route A1a. When positioning is performed by receiving positioning radio waves from a navigation satellite as in the above technology, the positioning results are output at predetermined time intervals such as time t0, t1, t2, etc. Ru. In this case, the direction of travel cannot be corrected while the positioning results are being output. As a result, the actual traveling route deviates from the planned route A1 by a predetermined width W1, for example.

The present invention attempts to solve this problem, and provides an autonomous control device, an autonomous control method, and an autonomous control program that can reduce the deviation between the planned route and the actual route than before. With the goal.

A first invention is an autonomous control device for controlling a mobile device capable of autonomous movement, comprising: a positioning means for positioning a current position using positioning radio waves from a navigation satellite; and a positioning result obtained by the positioning means. a direction change measuring means for measuring a change in the nose direction of the moving device at a time interval shorter than an output interval of the moving device; an autonomous control device comprising: a first correction means for correcting the nose direction; and a return means for adjusting the moving device to move in a planned direction on a planned route based on a positioning result by the positioning device. It is.

According to the configuration of the first aspect of the invention, the nose direction of the moving device can be corrected by the first correction means while the positioning result is being outputted so as to reduce the change in the nose direction. Even if the moving device is not moving in the planned direction on the planned route despite the correction of the nose direction by the first correction means, the moving device can be moved in the planned direction on the planned route by adjustment by the return device. adjusted to move to. Here, since the nose direction is corrected by the first correction means prior to adjustment by the return means, the deviation between the planned travel path of the moving device and the actual travel path can be made smaller than before. .

A second invention, in the configuration of the first invention, detects a change in the moving speed of the mobile device in a direction perpendicular to the traveling direction of the mobile device at a time interval shorter than an output interval of the positioning result by the positioning means. It has a speed change measuring means for measuring, and the first correcting means adjusts the movement position due to the change in the nose direction measured by the direction change measuring means and the change in the moving speed measured by the speed change measuring means. The autonomous control device is configured to correct the nose direction of the mobile device so as to reduce a change in the direction of the aircraft.

According to the structure of the second invention, since the speed change measuring means measures the change in the moving speed in the direction orthogonal to the nose direction, for example, if the moving device skids, it is possible to avoid parallel movement while maintaining the nose direction. Changes in speed can be measured even when there is no change in the nose direction, such as when the aircraft moves to The first correction means corrects the nose direction so as to reduce the change in the nose direction, and also corrects the change in the moving position due to the change in the moving speed measured by the speed change measuring means. The heading direction of the mobile device can be corrected. That is, while the positioning results are being output, even if the nose direction changes or even if the nose direction does not change, it is possible to correct the deviation from the planned traveling route.

In a third invention, in the configuration of the first invention or the second invention, in a time interval shorter than an output interval of positioning results by the positioning means, the movement of the mobile device in a relative relationship with the direction of geomagnetism is so as to reduce the difference between the relative direction measuring means for measuring the nose direction and the planned direction defined by the relative relationship with the direction of the earth's magnetic field and the nose direction measured by the relative direction measuring means. The autonomous control device includes a second correction means for correcting the nose direction of the moving device.

For example, even if the traveling direction of the moving device deviates from the planned direction, if the deviated state continues for a predetermined period of time, it cannot be detected as a change in the nose direction. In this regard, according to the configuration of the third invention, even if there is no change in the nose direction during a predetermined period of time during which the positioning result is output, the second correction means measures the relative direction measurement means. It is possible to reduce the deviation between the nose direction and the planned direction. Thereby, the deviation between the moving speed of the moving device and the planned speed can be made smaller than before.

In a fourth invention, in the configuration of the third invention, the nose of the mobile device is configured based on Doppler information indicating a Doppler effect of the positioning radio waves caused by relative movement between the navigation satellite and the mobile device. a nose direction calculation means for calculating a direction; a direction deviation calculation means for calculating a deviation between the nose direction calculated by the nose direction calculation means and the planned direction; and the deviation calculated by the direction deviation calculation means. and a third correcting means for correcting the nose direction of the moving device so as to reduce the amount of the moving device.

The autonomous control device can calculate the direction of movement, that is, the heading direction, based on Doppler information of positioning radio waves from a plurality of navigation satellites. To give a simple example, when Doppler information indicating that one navigation satellite A is approaching and Doppler information indicating that it is moving away from another navigation satellite B is obtained, the mobile device is approaching navigation satellite A. If the nose direction calculated based on Doppler information is more reliable than the nose direction calculated based on geomagnetism, controlling the aircraft nose direction using Doppler information is effective. Furthermore, since the Doppler information can be obtained earlier than the positioning results, the nose direction can be corrected during the nose direction control based on the reliable current position based on the positioning results. In this regard, according to the configuration of the fourth invention, since the third correction means can correct the nose direction based on Doppler information, it may be possible to control the nose direction with higher accuracy. . Thereby, the deviation between the moving speed of the moving device and the planned speed can be made smaller than before.

A fifth invention is the configuration of the second invention, wherein the speed change measuring means is further configured to measure the change in speed in the scheduled direction at a time interval shorter than an acquisition interval of Doppler information. further comprising: speed correction means for correcting the speed of the moving device so as to reduce the change in speed measured by the speed change measuring means; and calculating the moving speed of the moving device based on the Doppler information. and a speed return means for adjusting the speed of the mobile device so that the speed of movement of the mobile device matches a scheduled speed.

According to the configuration of the fifth aspect of the invention, the speed correcting means can correct the moving speed of the moving device so as to offset the change in speed measured by the speed change measuring means. Furthermore, even if the moving device is not moving at the planned speed despite the result of the movement speed correction by the speed correcting device, the moving device can be adjusted to move at the planned speed by the speed return device. be adjusted. Since the speed change measuring means measures changes in speed at time intervals shorter than the Doppler information acquisition interval, the moving speed of the mobile device can be controlled even while Doppler information is being acquired. That is, the moving speed can be corrected while the moving speed is being reliably controlled based on Doppler information. Thereby, the deviation between the moving speed of the moving device and the planned speed can be made smaller than before.

A sixth invention is a configuration according to any one of the first to fifth inventions, in which when a signal regarding the movement of the moving device is received from a work control means that controls the work performed by the moving device, The autonomous control device has priority means for prioritizing signals from the work control means.

For example, when a mobile device performs tilling work with a rotary attached, in order to perform the work effectively, the work control software adjusts the nose direction and movement speed according to the field conditions, etc. may be controlled. In this regard, according to the configuration of the sixth invention, when a signal regarding the movement of the moving device is received from the work control means, the priority means is provided for giving priority to the signal from the work control means, so that the planned route is changed at the planned speed. It is possible to not only move forward, but also to realize the movement route and movement speed to carry out the work effectively.

A seventh invention is the configuration of the third invention, in which the magnitude of the correction in the nose direction by the first correction means and/or the second correction means is determined based on the magnitude of the adjustment by the return means. This is an autonomous control device having a correction means for correction.

The larger the adjustment by the return means, the greater the extent to which the nose direction correction by the first correction means or the second correction means is insufficient. In this regard, according to the configuration of the seventh invention, the correction means corrects the magnitude of the correction of the nose direction by the first correction means and/or the second correction means based on the magnitude of the adjustment by the return means. can do. Thereby, while outputting the positioning results, the deviation between the moving speed of the mobile device and the planned speed can be made smaller than before.

An eighth invention is an autonomous control method using an autonomous control device for controlling a mobile device capable of autonomous movement, which has a positioning means for determining the current position using positioning radio waves from a navigation satellite, a direction change measuring step of measuring a change in the nose direction of the moving device at a time interval shorter than an output interval of positioning results by the positioning means; and reducing the change in the nose direction measured by the direction change measuring means. a first correction step of correcting the nose direction of the mobile device so as to make the adjustment; and a return step of adjusting the mobile device to move in a planned direction on a planned route based on the positioning result by the positioning means. , is an autonomous control method for implementing.

A ninth invention provides a positioning means for positioning a current position of a computer controlling an autonomous control device for controlling an autonomously movable mobile device using positioning radio waves from a navigation satellite; a direction change measuring means for measuring a change in the nose direction of the moving device at a time interval shorter than a result output interval; a direction change measuring means for measuring a change in the nose direction of the moving device; a first correction means for correcting the nose direction of the aircraft; and a return means for adjusting the moving device to move in a planned direction on a planned route based on the positioning result by the positioning device. It is an autonomous control program.

According to the present invention, the deviation between the planned traveling route and the actual traveling route can be made smaller than before.

It is a schematic diagram showing an operation of an autonomous control device concerning a first embodiment of the present invention. FIG. 2 is a schematic diagram showing the external configuration of a mobile device. FIG. 2 is a schematic diagram showing the internal configuration of a mobile device. FIG. 2 is a schematic diagram showing the functional configuration of a mobile device. FIG. 2 is a schematic diagram showing main information referenced by an autonomous control unit. FIG. 3 is a conceptual diagram showing nose direction control in the first embodiment. FIG. 2 is a conceptual diagram showing a neural network. 1 is a schematic flowchart showing a machine learning method. It is a conceptual diagram showing a planned route and an actual traveling route. FIG. 3 is a conceptual diagram showing a progress mode of the mobile device. It is a schematic flowchart which shows the whole picture of control performed by an autonomous control device. 3 is a schematic flowchart showing control for correcting the nose direction. It is a schematic flowchart which shows control in the vicinity of a bending position and at a bending position. FIG. 7 is a conceptual diagram showing nose direction control in a second embodiment of the present invention. It is a conceptual diagram showing a planned route and an actual traveling route. 3 is a schematic flowchart showing control for correcting the nose direction. FIG. 7 is a conceptual diagram showing nose direction control in a third embodiment of the present invention. It is a conceptual diagram showing a planned route and an actual traveling route. 2 is a schematic flowchart showing control of the nose direction. FIG. 7 is a conceptual diagram showing control according to a fourth embodiment of the present invention. FIG. 3 is a conceptual diagram showing a planned speed and an actual moving speed. 3 is a schematic flowchart showing speed control. It is a schematic flowchart which shows the control of the nose direction in the fifth embodiment of this invention. FIG. 7 is a conceptual diagram showing correction of nose direction control in a sixth embodiment of the present invention. FIG. 3 is a conceptual diagram showing a progress mode of the mobile device. It is a conceptual diagram which shows the relationship between correction|amendment and adjustment amount of the reference amount A, the reference amount B, and the second reference amount. It is a conceptual diagram which shows the relationship between correction|amendment and adjustment amount of the reference amount A, the reference amount B, and the second reference amount. 3 is a schematic flowchart showing correction of nose direction control; FIG. 2 is a conceptual diagram showing a conventional example.

Hereinafter, modes for carrying out the present invention (hereinafter referred to as embodiments) will be described in detail. In the following description, similar configurations are denoted by the same reference numerals, and the description thereof will be omitted or simplified. Note that descriptions of configurations that can be appropriately implemented by those skilled in the art will be omitted, and only the basic configuration of the present invention will be described. The gist of the present invention is to control the traveling direction based on changes in the traveling state during the acquisition of positioning results and Doppler information using positioning radio waves, and finally to reliably control the traveling direction based on the positioning results and Doppler information. This is to control the direction of travel.

<First embodiment>
A tractor 1 (hereinafter referred to as "unmanned aircraft 1") shown in FIGS. 1 and 2 is an example of a moving device and is capable of autonomous movement. The unmanned aerial vehicle 1 is configured to perform operations based on instructions via radio signals from a control device 70 (see FIG. 4, also referred to as a "propo") that controls the unmanned aerial vehicle 1, and to autonomously travel on the ground. . In this specification, autonomous driving means automatically proceeding along a driving route by software without receiving a wireless signal from the control device 70. Autonomous driving is synonymous with automatic driving. Note that, unlike this embodiment, the mobile device is not limited to an unmanned moving object such as an unmanned vehicle or an unmanned flying object (drone), but may also be a vehicle such as a manned heavy machine operated by a human on board. .

Unmanned aircraft 1 is a tractor used in a field. As shown in FIG. 1, the unmanned aircraft 1 performs the work of cultivating a field 100, that is, tilling. After plowing, the soil in the field 100 swells up into waves and becomes a field 100a.

The unmanned aerial vehicle 1 plows the field 100 through planned routes R1 to R6. On the planned routes R1 to R6, the planned position where the nose direction changes is called a turning position. For example, the bending position between the planned route R1 and the planned route R2 is the position P. The position where the bending position starts is position P1.

The unmanned aircraft 1 is configured to receive positioning radio waves from navigation satellites such as navigation satellites 202A to 202D, measure its current position, and proceed along scheduled routes R1 to R6. In this embodiment, the traveling direction is slightly corrected during the output of the positioning result of the positioning performed by receiving the positioning radio waves. Thereby, the difference between the actual travel route and the planned routes R1 to R6 can be made smaller than in the past. The present embodiment relates to control in which the unmanned aircraft 1 travels along the scheduled route in a scheduled direction when the scheduled route R1 or the like of the unmanned aircraft 1 is a straight route.

As shown in FIG. 2, the unmanned aircraft 1 has a vehicle body 10. The vehicle body 10 stores a computer that controls each part of the unmanned aircraft 1, an autonomous control device 32, a wireless communication device, and the like. The autonomous control device 32 includes an inertial sensor and a positioning device. The autonomous control device 32 is an example of an autonomous control device. Inertial sensors include gyro sensors, acceleration sensors, and electronic compasses.

A gyro sensor, also called an angular velocity sensor, is a sensor that measures the amount of change in angle (angular velocity) per time. The gyro sensor in this embodiment is a vibrating gyro sensor using IC-type MEMS (Micro Electro Mechanical System) technology, and is capable of measuring angular velocity in three axes (vertical, horizontal, and longitudinal axes). . A gyro sensor is an example of a direction change measuring means.

The acceleration sensor is an IC type sensor that measures acceleration. The acceleration sensor in this embodiment is a capacitance type sensor, and can measure acceleration in three axes (vertical, horizontal, and longitudinal axes). An acceleration sensor is an example of a speed change measuring means.

An electronic compass is a device that can measure direction by observing the Earth's geomagnetism. The electronic compass in this embodiment is an IC type compass, and is a three-axis electronic compass using Hall element technology. An electronic compass is an example of a relative direction measuring means.

The positioning device is a device for positioning the current position using positioning radio waves from a navigation satellite. Navigation satellites include, for example, satellites used in Michibiki (quasi-zenith satellite system) and satellites used in GPS (Global Positioning System). The positioning device basically measures the position of the unmanned aircraft 1 by receiving positioning radio waves from four or more navigation satellites. Since the positioning method using positioning radio waves is well known, the explanation thereof will be omitted. The positioning device is an example of positioning means.

Furthermore, an antenna 12 for wireless communication and an antenna 14 for receiving positioning radio waves from a navigation satellite are arranged on the vehicle body 10.

Further, a camera 16 is disposed on the vehicle body 10 for photographing the front and acquiring an image.

Front wheels 18L and 18R and rear wheels 22L and 22R (see FIG. 3) are arranged in the vehicle body 10.

A rotary 30 is attached to the vehicle body 10. The rotary 30 is a member for tilling the field 100. The rotary 30 is configured to be movable in the vertical direction (arrow Z direction in FIG. 2).

In the unmanned aircraft 1, the direction in which the camera 16 is arranged is called the nose direction. The nose direction is the traveling direction of the unmanned aircraft 1, and is the direction shown by the arrow M1 in FIG.

As shown in FIG. 3, the front wheels 18L and 18R are connected to a steering system 20 and are rotatable left and right (in the direction of arrow X in FIG. 3). By controlling the steering system 20, the nose direction of the unmanned aircraft 1 can be controlled.

Further, an engine 26 is arranged inside the vehicle body 10 and near the front wheels 18L and 18R. Engine 26 is, for example, a diesel engine. The engine 26 is connected to the shafts 24 of the rear wheels 22L and 22R via a transmission 28, and rotates the rear wheels 22L and 22R. Note that the engine 26 is not limited to a diesel engine, and may be, for example, a gasoline engine or an electric motor.

The autonomous control device 32 is electrically connected to the steering system 20 and the engine 26, and controls the steering system 20 and the engine 26. The autonomous control device 32 is configured to adjust the traveling direction and traveling speed of the unmanned aircraft 1.

FIG. 4 is a schematic diagram showing an example of the functional configuration of the unmanned aircraft 1. As shown in FIG. As shown in FIG. 4, the unmanned aircraft 1 includes an autonomous control section 40, a wireless communication section 58, an image processing section 60, and a work control section 62. The autonomous control unit 40 includes a CPU (Central Processing Unit) 50, a storage unit 52, a drive control unit 54, and a satellite positioning unit 56. The autonomous control unit 40 corresponds to the autonomous control device 32, and shows the functional configuration of the autonomous control device 32.

The unmanned aircraft 1 is capable of communicating with the outside through a wireless communication unit 58. The unmanned aircraft 1 receives, through the wireless communication unit 58, instructions such as starting from a control device (propo) 70 operated by a user.

The unmanned aerial vehicle 1 uses the image processing unit 60 to operate the camera 16 (see FIG. 2) to acquire an external image, and processes the acquired photographic data. The external image is mainly an image in the direction of movement of the unmanned aircraft 1.

The work control unit 62 of the unmanned aircraft 1 controls the elevation of the rotary 30 and the traveling direction and speed necessary for the work. The work control unit 62 controls the work performed by the unmanned aircraft 1 and transmits a signal regarding the movement of the unmanned aircraft 1 to the CPU 50.

The unmanned aircraft 1 can measure its own position using the satellite positioning unit 56. The satellite positioning unit 56 corresponds to a positioning device, and basically measures the position of the unmanned aircraft 1 by receiving positioning radio waves from four or more navigation satellites. The satellite positioning unit 56 outputs positioning results at predetermined time intervals. The predetermined time interval is, for example, 1 second (s).

The unmanned aircraft 1 controls the steering system 20 and the engine 26 by the drive control section 54 that constitutes the autonomous control section 40 . The drive control unit 54 uses outputs from a gyro sensor, an acceleration sensor, and an electronic compass.

The gyro sensor measures changes in the nose direction of the unmanned aircraft 1 at time intervals shorter than the output interval of positioning results by the satellite positioning unit 56. The time interval at which the gyro sensor measures changes in the nose direction is, for example, one-fourth of a second (0.25 seconds).

The acceleration sensor measures changes in speed (acceleration) of the unmanned aircraft 1 at time intervals shorter than the output interval of positioning results by the satellite positioning unit 56. Since acceleration is a change in speed, it is a change in movement of the unmanned aircraft 1. As described above, the acceleration sensor is a three-axis acceleration sensor and can measure the three-dimensional movement of the unmanned aircraft 1. Therefore, a change in speed having a component perpendicular to the traveling direction of the unmanned aircraft 1 can be measured. For example, when the unmanned aircraft 1 moves in parallel while maintaining its nose direction, the parallel movement can be measured. The time interval at which the acceleration sensor measures changes in movement is, for example, one quarter second (0.25 seconds).

Note that the time interval at which the gyro sensor measures changes in the nose direction and the time interval at which the acceleration sensor measures changes in movement are adjusted depending on the specifications and usage of the unmanned aircraft 1. For example, if the unmanned aircraft 1 is a tractor, the time interval is set relatively long because it is relatively heavier and moves slower than a drone or the like. On the other hand, in the case of vehicles or drones that move quickly, the time interval is set relatively short. For example, the time interval can be adjusted in units of 1/1000 second (0.0001 second), but in this embodiment, for convenience of explanation, it is set to 1/4 second (0.25 second). .

FIG. 5 shows main data used by the autonomous control unit 40 for control. The autonomous control unit 40 uses a gyro sensor output that is a measurement result from a gyro sensor, an acceleration sensor output that is a measurement result from an acceleration sensor, and an electronic compass output that is a measurement result from an electronic compass. The satellite positioning unit 56 uses satellite information such as orbit information of navigation satellites and received positioning radio waves (not shown) to obtain Doppler information indicating the Doppler effect that can be obtained during positioning calculation, and positioning results. Outputs positioning position information indicating the positioning position. Note that the Doppler effect is a change in the frequency of a carrier wave of a positioning radio wave.

As shown in FIG. 6, in the present embodiment, the autonomous control unit 40 refers to the gyro sensor output and the acceleration sensor output to make a slight correction in the nose direction (hereinafter referred to as "first correction process"). ), the position and direction are corrected using the measured position information (hereinafter referred to as "return processing"). The fine correction of the nose direction made with reference to the gyro sensor output is called the "first correction process A," the fine correction of the nose direction made with reference to the acceleration sensor output is called the "first correction process B," and the The first correction process A'' and the first correction process B are collectively referred to as the first correction process.

The drive control unit 54 controls the rotation of the engine 26 connected to the rear wheels 22L and 22R, the rotation of the front wheels 18L and 18R by the steering system 20, and controls the running speed and direction.

The storage unit 52 stores various data and programs necessary for autonomous movement, such as data indicating a movement plan for autonomous movement along a scheduled route, as well as the following programs.

The storage unit 52 stores a first modification program, a restoration program, an image recognition program, and an avoidance program. The CPU 50 and the first modification program are an example of first modification means. The CPU 50 and the return program are an example of a return means. The CPU 50 and the image recognition program are examples of recognition means. The CPU 50 and the avoidance program are examples of avoidance means.

The first correction program corrects the nose direction of the unmanned aircraft 1 so as to reduce the change in the nose angle measured by the gyro sensor and the change in speed measured by the acceleration sensor. This correction of the nose direction is the above-mentioned "first correction process." The first correction process is a process for reducing the measured nose angle change and acceleration. For example, if a gyro sensor detects that the nose direction has changed in a certain direction, the nose direction is changed in the opposite direction. Further, for example, when a speed change in a direction perpendicular to the traveling direction of the unmanned aircraft 1 is measured by an acceleration sensor, the nose direction of the unmanned aircraft 1 is changed so that it heads in the opposite direction. This makes it possible to reduce the effects of changes in the measured nose angle and acceleration. It does not matter whether or not the nose direction matches the direction of the planned route as a result of the first correction process. The unmanned aircraft 1 corrects the nose direction by a predetermined reference amount using the first correction program. The reference amount is defined by angle and time, and for example, a 1 degree angle change continues for 0.05 seconds. The reference amount for the first correction process A is called the "reference amount A," the reference amount for the first correction process B is called the "reference amount B," and the reference amounts A and B are collectively called the "reference amount."

The unmanned aerial vehicle 1 is adjusted by the return program so that the unmanned aerial vehicle 1 moves in a scheduled direction on a scheduled route based on the positioning result by the satellite positioning unit 56. This adjustment to match the nose direction with the planned route is the above-mentioned "return process." Therefore, even with the above-described first correction process, even if the unmanned aircraft 1 is not traveling along the scheduled route in the scheduled direction, the unmanned aircraft 1 can move on the scheduled route in the scheduled direction by the return program.

The unmanned aerial vehicle 1 recognizes the object to be avoided in the image captured by the camera 16 using the image recognition program. The unmanned aerial vehicle 1 stores, for example, a neural network in the storage unit 52, and recognizes the object to be avoided using the neural network.

An outline of the learning method (machine learning process) will be explained with reference to FIGS. 7 and 8. When a large amount of image data is input as learning data into a computer neural network model (see FIG. 7) (step S1 in FIG. 8), learning is performed by the neural network algorithm (step S2). The learning data is data showing humans in various forms.

The neural network model is, for example, the neural network conceptually shown in FIG. Note that neural networks and deep learning are well known, so they will only be outlined in this specification. A neural network is composed of an input layer, one or more hidden layers, and an output layer. Although only one hidden layer is shown in FIG. 7, there are actually multiple hidden layers. That is, the learning in this embodiment is machine learning (deep learning) using a multilayer neural network (deep neural network).

In machine learning, by adjusting the weight w1a of the input value input to the hidden layer and the output layer, and the bias (threshold value in each layer) B1 etc. in the input layer, hidden layer, and output layer, the output from the output layer is correct. I will try to get closer to . As a result of learning, a neural network with adjusted weights and biases is generated as learning result data (step S3 in FIG. 8). Through this learning, for example, a neural network is generated that can accurately recognize that an object in an image is a human. The storage unit 52 stores the neural network generated in this way.

The unmanned aerial vehicle 1 recognizes objects in the acquired images using an image recognition program.

The unmanned aerial vehicle 1 performs an avoidance operation based on the recognition result by the image recognition program using the avoidance program. The unmanned aerial vehicle 1 is configured to perform an avoidance operation when the avoidance target recognized by the recognition program satisfies a predetermined condition.

For example, the unmanned aerial vehicle 1 performs the avoidance operation when the distance between the unmanned aerial vehicle 1 and the avoidance target satisfies a predetermined condition that the distance is within a predetermined distance range. The distance of the predetermined distance range is, for example, 5 meters (m).

Hereinafter, with reference to FIGS. 9 and 10, control of the traveling direction of the unmanned aircraft 1 will be described. It is assumed that the planned route of the unmanned aircraft 1 is a straight route A1 shown in FIG. 9(a). The autonomous control device 32 uses the satellite positioning unit 56 to output positioning results at predetermined time intervals such as times t0, t1, t2, and t3. The predetermined time interval is, for example, one second. In addition, the unmanned aircraft 1 detects changes in the nose direction and changes in speed (acceleration ) to measure. The predetermined time interval for measuring changes in nose direction and changes in speed (acceleration) is, for example, four times per second. At time t1 and the like at which the positioning result is acquired, the unmanned aerial vehicle 1 performs a return process using a return program. Then, while acquiring the positioning results, the first correction process is performed, for example, at times t0.25, t0.5, and t0.75. If there is a change in both the nose direction and the speed, the conflict in control is resolved by, for example, giving priority to the change in the nose direction or to giving priority to the change in speed. In this embodiment, if there are changes in both the nose direction and the speed, the first correction process A is performed with priority given to the change in the nose direction.

As described above, the autonomous control device 32 performs the first correction process while acquiring the positioning result, and performs the return process after acquiring the positioning result. That is, during the return process, since the fine correction is performed by the first correction process, the deviation between the planned route A1 and the actual route A1b is relatively small and has a width W2. The width W2 is smaller than the width W1 of the conventional example (see FIG. 29).

As shown in FIG. 10(a), for example, when the autonomous control device 32 measures a change in the nose direction indicated by the arrow B1 at time t0.25, the autonomous control device 32 measures the change in the nose direction as indicated by the arrow B1. The first correction process is performed as a change in the direction indicated by arrow B2. In the example of FIG. 10(a), it is assumed that the first correction process causes the unmanned aircraft 1 to travel along a route parallel to, but not coincident with, the planned route A1. At time t1, the autonomous control device 32 performs a return process based on the positioning result so that the current position is on the scheduled route and the traveling direction matches the direction of the scheduled route A1. As a return process, the autonomous control device 32 corrects the nose direction, for example, as shown by arrow D1, and performs control so that the current position is on the planned route and the traveling direction matches the direction of the planned route A1. do. FIG. 10A shows an example in which the positioning result is output at time t1, the return process is started in an extremely short time, and the return process is completed at time t1.05.

As shown in FIG. 10(b), for example, at time t0.25, the autonomous control device 32 does not measure (detect) a change in the nose direction, but measures acceleration in a direction that deviates from the planned route A1. , the first correction process is performed as a change in the direction shown by arrow C2, which is the opposite direction to the change in position, so as to reduce the change in position due to the change in acceleration. In the example of FIG. 10(b), it is assumed that the first correction process causes the unmanned aircraft 1 to travel along a route parallel to the planned route A1. At time t1, the autonomous control device 32 performs a return process based on the positioning result so that the current position is on the scheduled route and the traveling direction matches the direction of the scheduled route A1. The return process is started by, for example, changing the nose direction by a predetermined angle, and when the current position matches the planned route A1, it is completed by making the traveling direction match the direction of the planned route A1.
In this embodiment, the magnitude of the return process is defined as an "adjustment amount." The size of the return process is defined by angle and time. In this embodiment, for convenience of explanation, the time from the start to the completion of the return process is referred to as an "adjustment amount."

The outline of the control by the autonomous control device 32 will be described below with reference to the flowcharts of FIGS. 11 to 13.

When the autonomous control device 32 launches the unmanned aircraft 1 (step ST1 in FIG. 11), it acquires an external image (step ST2), performs image recognition of the entire image using a neural network (step ST4), and detects the avoidance in the image. When it recognizes the target (step ST5) and determines that it is within the avoidance distance (step ST6), it stops advancing. If there is no object to be avoided in the image or if it is not within the avoidance distance, it is determined that the mission has been completed (step ST7), and the control is terminated. In parallel with steps ST2 to ST6 described above, the unmanned aerial vehicle 1 moves along a predetermined route while measuring its current position and executes the work (step ST3).

Next, step ST3 will be explained with reference to FIG. When the autonomous control device 32 determines that the vehicle is near the bending position (step ST31 in FIG. 12), it performs control near the bending position (step ST40). The bending position is, for example, position P in FIG. 1, and is a position at which the direction of travel should be changed in order to transition from route R1 to route R2. Step ST40 will be described later.

The autonomous control device 32 determines that the turning position is not near (step ST31), and when a change in direction or speed is measured (step ST32), performs a first correction process (step ST33). Further, the autonomous control device 32 acquires the positioning result (step ST34), and when determining that the current traveling route is not on the planned route or in the planned direction (step ST35), performs a return process (step ST36). Steps ST32 and ST33 and steps ST34 to ST36 are executed in parallel, and when the timings of the first correction process and the return process coincide, the return process is given priority.

Next, step ST40 will be explained with reference to FIG. 13. When the autonomous control device 32 determines that it is near the turning position, it reduces the speed (step ST41 in FIG. 13), and when it determines that it is at the turning start position (step ST42), it starts turning (step ST43) and follows the schedule. (step ST44), it is determined whether the vehicle is located on the planned route (step ST45), and if it is not on the planned route, the traveling direction is adjusted to correct the position (step ST46). ). The vicinity of the bending position is, for example, a position 10 meters before the position where the route should be corrected. The bending start position is position P1 in FIG.

<Second embodiment>
Next, a second embodiment will be described with reference to FIGS. 14 to 16. Note that descriptions of items common to the first embodiment will be omitted, and the description will focus on parts that are different from the first embodiment.

As shown in FIG. 14, in the second embodiment, in addition to the first correction process and the return process, the autonomous control unit 40 uses the electronic compass output to perform a fine direction correction (hereinafter referred to as "first correction process"). 2).

The electronic compass measures the heading direction of the unmanned aircraft 1 at time intervals shorter than the output interval of positioning results by the satellite positioning unit 56. The time interval at which the electronic compass measures changes in the nose direction is, for example, one-half second (0.5 seconds). An electronic compass measures the heading direction relative to the direction of the earth's magnetic field. The planned direction, which is the planned direction in which the unmanned aircraft 1 travels along the planned route A1, is also defined in relation to the earth's magnetism. Therefore, the electronic compass can measure the difference between the planned direction and the actual direction of travel.

The unmanned aircraft 1 stores the second correction program in the storage unit 52 (see FIG. 4). The CPU 50 and the second correction program are an example of the second correction means. The autonomous control device 32 corrects the nose direction of the unmanned aircraft 1 so as to reduce the difference between the planned nose direction and the nose direction measured by the electronic compass (hereinafter referred to as "nose direction deviation"). (This is the above-mentioned "second correction process.") Note that when the timings of the first correction process and the second correction process match, priority is given to the second correction process. If the timings of the second correction process and the return process match, the return process is given priority.

As shown in FIG. 15(a), for example, when the autonomous control device 32 measures a change in the nose direction indicated by the arrow B1 at time t0.25, the autonomous control device 32 measures the change in the nose direction as indicated by the arrow B1. The first correction process is performed as a change in the direction indicated by arrow B2. In the example of FIG. 15A, it is assumed that the first correction process causes the unmanned aircraft 1 to not return to the route parallel to the planned route A1, but to move along the route indicated by the arrow A2. The autonomous control device 32 measures the deviation in the nose direction using the electronic compass at time t0.5. At time t0.5, the nose direction is the direction shown by arrow A2, and the deviation in the nose direction is measured. In this case, the autonomous control device 32 corrects the nose direction by a second predetermined reference amount (hereinafter referred to as "second reference amount"). For example, the second reference amount is that the angle change of 1.5 degrees continues for 0.05 seconds. In the example of FIG. 15A, as a result of the second correction process, the traveling direction of the unmanned aircraft 1 matches the direction of the planned route A1, but does not match the planned route A1 and is deviated in parallel. At time t1.0, the autonomous control device 32 acquires the positioning result and corrects the nose direction as shown by arrow D1, thereby determining that the current position is on the planned route and that the traveling direction matches the planned direction. Execute the recovery process as follows.

FIG. 15(b) shows a state in which the deviation in the nose direction is not resolved by the second correction process. At time t0.5, the nose direction is the direction shown by arrow A2, and the autonomous control device 32 measures the deviation in the nose direction. In this case, the autonomous control device 32 corrects the nose direction by a second predetermined reference amount. In the example of FIG. 15(b), as a result of the second correction process, the traveling direction of the unmanned aircraft 1 does not match the direction of the planned route A1, and becomes an arrow A3. At time t1.0, the autonomous control device 32 acquires the positioning result and corrects the nose direction as shown by arrow D2, thereby determining that the current position is on the planned route and that the traveling direction matches the planned direction. Execute the recovery process as follows.

Next, with reference to FIG. 16, the first correction process, second correction process, and return process will be described. The autonomous control device 32 determines that the turning position is not near (step ST31 in FIG. 16), and when a change in direction or speed is measured (step ST32), performs a first correction process (step ST33). Furthermore, when the autonomous control device 32 measures a deviation between the nose direction and the planned route direction (deviation in the nose direction) (step ST101), it performs a second correction process (step ST102). Further, the autonomous control device 32 acquires the positioning result (step ST34), and when determining that the current traveling route is not on the planned route or in the planned direction (step ST35), performs a return process (step ST36). Steps ST32 and ST33, steps ST101 and ST102, and steps ST34 to ST36 are performed in parallel, and if the timings of the first correction process, second correction process, and return process coincide, the return process is given top priority. , followed by the second modification process.

<Third embodiment>
Next, a third embodiment will be described with reference to FIGS. 17 to 19. Note that descriptions of matters common to the first embodiment or the second embodiment will be omitted, and the description will focus on parts that are different from the first embodiment and the second embodiment.

As shown in FIG. 17, in the present embodiment, the autonomous control unit 40 uses Doppler information to perform slight corrections in direction (hereinafter referred to as (referred to as the "Third Amendment Process").

In the third embodiment, in addition to the configuration in the second embodiment, the storage unit 52 (see FIG. 4) further stores a nose direction calculation program, a direction deviation calculation program, and a third correction program. are doing. The CPU 50 and the nose direction calculation program are an example of a nose direction calculation means. The CPU 50 and the directional deviation calculation program are an example of directional deviation calculation means. The CPU 50 and the third modification program are an example of third modification means.

The unmanned aircraft 1 calculates Doppler information indicating the Doppler effect for each positioning satellite 202A and the like using the nose direction calculation program, and calculates the nose direction based on the Doppler information. For example, the Doppler information of the positioning satellites 202A and 202B in FIG. 1 indicates that the unmanned aerial vehicle 1 and the positioning satellites 202A and 202B are approaching, and the approaching speeds of the positioning satellites 202A and 202B are different. This indicates that the Doppler information of the positioning satellites 202C and 202D indicates that the unmanned aircraft 1 and the positioning satellites 202C and 202D are moving away from each other, and the degree of distance is different between the positioning satellites 202C and 202D. If so, the traveling direction of the unmanned aircraft 1 can be calculated based on that information. Since azimuth measurement using Doppler information of positioning radio waves from a plurality of positioning satellites is well known, detailed explanation will be omitted.

The unmanned aerial vehicle 1 calculates the deviation between the above-mentioned direction of travel and the direction of the planned route using a direction deviation calculation program. Since the planned route is stored in the storage unit 52, the unmanned aircraft 1 can calculate the deviation between the above-mentioned traveling direction and the direction of the planned route.

The unmanned aerial vehicle 1 corrects the nose direction of the unmanned aerial vehicle 1 by the third modification program so as to reduce the above deviation. This correction of the nose direction is the above-mentioned "third correction process." The autonomous control device 32 corrects the nose direction by a third predetermined reference amount (hereinafter referred to as "third reference amount"). For example, the third reference amount is that the angle change of 1.5 degrees continues for 0.05 seconds.

Since the Doppler information can be acquired before the calculation for acquiring the positioning result converges, the unmanned aerial vehicle 1 can perform the third correction process during the return process using the positioning result. . If the timings of the first correction process or the second correction process and the third correction process match, the third correction process is given priority.

Hereinafter, control of the traveling direction of the unmanned aircraft 1 will be described with reference to FIG. 18. It is assumed that the planned route of the unmanned aircraft 1 is a straight route A1 shown in FIG. 18(a). The unmanned aerial vehicle 1 outputs positioning results by the satellite positioning unit 56 at predetermined time intervals such as times t0, t1, t2, and t3 (for example, once every second). Additionally, the unmanned aircraft 1 measures changes in the nose direction and speed at predetermined time intervals (for example, 4 times per second) such as times t0, t0.25, t0.5, and t0.75. . At time t1 and the like at which the positioning result is acquired, the unmanned aerial vehicle 1 performs a return process using a return program. While acquiring the positioning results, for example, a first correction process is performed at time t0.25, a second correction process is performed at t0.5, and a third correction process is performed at t0.75.

As mentioned above, while acquiring the positioning results, the unmanned aerial vehicle 1 performs fine corrections through the first correction process, the second correction process, and the third correction process, so the difference between the planned route A1 and the actual route A1c is The deviation is relatively small and has a width W3. The width W3 is smaller than the width W2 of the first embodiment (see FIG. 9).

Next, with reference to FIG. 19, the first to third correction processing and return processing will be described. The autonomous control device 32 determines that the turning position is not near (step ST31 in FIG. 19), and when a change in direction or speed is measured (step ST32), performs a first correction process (step ST33). Furthermore, upon measuring the deviation between the nose direction and the planned route direction (step ST101), the autonomous control device 32 performs a second correction process (step ST102). Further, the autonomous control device 32 acquires Doppler information (step ST201), and when determining that the current traveling route is not in the direction of the planned route (step ST202), performs a third correction process (step ST203). Further, the autonomous control device 32 acquires the positioning result (step ST34), and performs the return process by the control described above. If the timings of the above-mentioned first to third modification processing and restoration processing match, the restoration processing is given the highest priority, followed by the third modification processing, second modification processing, and first modification processing. be done.

<Fourth embodiment>
Next, a fourth embodiment will be described with reference to FIGS. 20 to 22. Note that descriptions of matters common to the first to third embodiments will be omitted, and the description will focus on parts that are different from the first to third embodiments. In the fourth embodiment, the moving speed is controlled.

As shown in FIG. 20, in this embodiment, the autonomous control device 32 performs a speed correction process to correct the speed by referring to the acceleration sensor output, and uses Doppler information to restore the speed to the scheduled speed. A speed recovery process is performed.

In the fourth embodiment, in addition to the configurations in the first to third embodiments, the storage unit 52 (see FIG. 4) further stores a speed correction program and a speed recovery program. There is. The CPU 50 and the speed correction program are an example of speed correction means. The CPU 50 and the speed return program are an example of speed return means. The acceleration sensor forming the autonomous control device 32 also measures changes in speed in the nose direction. The acceleration sensor measures changes in velocity at time intervals shorter than the Doppler information acquisition interval.

The unmanned aircraft 1 corrects its speed using the speed correction program so as to reduce the above-mentioned change in speed. This speed correction is the above-mentioned "speed correction process."

The unmanned aircraft 1 calculates the moving speed of the unmanned aircraft 1 based on the Doppler information using the speed recovery program, and adjusts the speed so that the moving speed matches the scheduled speed. This speed adjustment is the above-mentioned "speed recovery process." Since the orbits of the navigation satellites 202A and the like are known, the traveling speed of the unmanned aircraft 1 can be calculated using Doppler information. If the timings of the speed correction process and speed recovery process match, priority is given to the speed recovery process.

Hereinafter, with reference to FIG. 21, control of the traveling speed of the unmanned aircraft 1 will be described. It is assumed that the scheduled speed of the unmanned aircraft 1 is speed V1. The unmanned aerial vehicle 1 outputs positioning results by the satellite positioning unit 56 at predetermined time intervals (for example, once every second) such as times t0, t1, t2, and t3. Additionally, the unmanned aircraft 1 measures changes in the nose direction and speed at predetermined time intervals (for example, 4 times per second) such as times t0, t0.25, t0.5, and t0.75. . The unmanned aircraft 1 measures changes in speed in the nose direction, and performs speed correction processing if there is a deviation from the scheduled speed. The speed correction process is performed only at a predetermined reference speed. The reference speed is, for example, an increase/decrease of 1 meter/second.

Then, the unmanned aircraft 1 performs speed recovery processing based on the Doppler information at predetermined time intervals (for example, every 0.75 seconds) such as times t0.75, t1.5, and t2.25. Even if the speed correction process is performed, if a discrepancy exists between the planned speed and the actual speed, the discrepancy is eliminated by the speed restoration process.

Next, adjustment of the speed of the unmanned aircraft 1 will be described with reference to the schematic flowchart of FIG. 22. The speed adjustment is performed in parallel with the direction of travel adjustment. When the unmanned aircraft 1 measures the speed change (step ST301 in FIG. 22), it performs speed correction processing (step ST302). The unmanned aircraft 1 acquires the Doppler information (step ST303), and when determining that the speed deviates from the planned speed (step ST304), performs speed recovery processing (step ST305). If the timings of the speed correction process and speed recovery process match, priority is given to the speed recovery process. Furthermore, if the speed correction processing or speed restoration processing and the timing of the adjustment of the traveling direction coincide, priority is given to the adjustment of the traveling direction.

<Fifth embodiment>
Next, a fifth embodiment will be described with reference to FIG. 23. Note that descriptions of matters common to the first to fourth embodiments will be omitted, and the description will focus on parts that are different from the first to fourth embodiments. In the fifth embodiment, work such as plowing a field is prioritized in terms of direction and speed.

In the fifth embodiment, in addition to the configurations in the first to fourth embodiments, a priority program is further stored in the storage unit 52 (see FIG. 4). The CPU 50 and the priority program are examples of priority means.

When the unmanned aerial vehicle 1 receives a signal related to the movement of the unmanned aerial vehicle 1 from the work control unit 62 that controls the work carried out by the unmanned aerial vehicle 1 according to the priority program, the unmanned aerial vehicle 1 reduces the deviation between the planned route and the actual traveling route. The configuration is such that the signal from the work control unit 62 is given priority over the control to perform the work. The work control unit 62 is an example of work control means. The work control unit 62 is also called a work module. That is, even if a change in direction or movement is measured, if the change is due to a signal from the work module, the direction of movement is not adjusted by the first correction process or the like.

Next, control of the unmanned aircraft 1 will be described with reference to the schematic flowchart of FIG. 23. The flowchart in FIG. 23 is an example in which the priority program is applied to the second embodiment. If the autonomous control device 32 has acquired the signal from the work module (step ST401), it gives priority to the signal from the work module (step ST402). On the other hand, if the autonomous control device 32 has not acquired the signal from the work module (step ST401), it executes the processes from step ST31 onwards.

Note that, unlike this embodiment, the first modification process, the second modification process, or the third modification process may not be performed, and the configuration may be configured by combining the restoration process and the priority process.

<Sixth embodiment>
Next, a sixth embodiment will be described with reference to FIGS. 24 to 26. Note that descriptions of matters common to the first to fifth embodiments will be omitted, and the description will focus on parts that are different from the first to fifth embodiments.

As shown in FIG. 24, in this embodiment, the autonomous control device 32 determines the magnitude of the corrections made by the first correction process A, the first correction process B, and the second correction process based on the size of the adjustment made by the return process. Correct the The reference amount A, the reference amount B, and the second reference amount are increased or decreased so that the magnitude of adjustment (adjustment amount) due to the return process is minimized. The initial settings of the reference amount A, the reference amount B, and the second reference amount are defined as amounts that are assumed not to recover the deviation from the planned route. The initial setting is, for example, an amount smaller than the adjustment amount when only the return process is performed regarding the correction of the traveling direction of the unmanned aircraft 1. Therefore, basically, by increasing the reference amount A, the reference amount B, and the second reference amount, the adjustment amount becomes smaller. The smaller the amount of adjustment, the greater the extent to which the actual traveling route of the unmanned aircraft 1 matches the planned route A1.

In the sixth embodiment, in addition to the configurations in the first to fifth embodiments, a correction program is stored in the storage unit 52 (see FIG. 4). The CPU 50 and the correction program are an example of correction means.

The unmanned aircraft 1 corrects the magnitude of the correction in the nose direction by the first correction means and the second correction means based on the magnitude of adjustment (adjustment amount) in the return process using the correction program. As described above, the first correction process corrects the nose direction by a predetermined reference amount. The second correction process corrects the nose direction by a predetermined second reference amount. Here, the first reference amount etc. is not necessarily appropriate. In this regard, the unmanned aircraft 1 performs correction to increase or decrease the first reference amount and the second reference amount using the correction program, thereby minimizing the adjustment amount due to the return process. Regarding the first correction process, the unmanned aircraft 1 corrects the reference amount separately for the first correction process A and the second correction process B.

With reference to FIG. 25, the correction in the first correction process A will be described. For example, as shown in FIG. 25, assume that the gyro sensor measures a change in the nose direction indicated by arrow B1 at time t0.25. If the reference amount A of the first correction process A is too small, the correction in the nose direction will be too small, as shown by arrow B2A. As a result, the amount of adjustment due to the return process increases. This also applies to the first correction process B and the second correction process.

In this regard, the autonomous control device 32 increases or decreases each of the reference amount A, the reference amount B, and the second reference amount, and makes corrections so that the adjustment amount is the smallest. In the correction of the reference amount A, the reference amount B, and the second reference amount, the reference amount A, the reference amount B, and the second reference amount when the adjustment amounts are the smallest are called "optimal values." The optimal value is the reference amount A, the reference amount B, and the second reference amount when the adjustment amount shows the minimum value when the reference amount A, the reference amount B, and the second reference amount are graphed in correspondence with the adjustment amount. It is.

For example, as shown in FIG. 26, the reference amount A is corrected from time t2 to time t5. Time t1 to time t5 are the timings at which the return process is executed. Since the adjustment amount is determined by the return process, the corrected reference amount A is applied in the next return process.

In the example of FIG. 26, the adjustment amount at time t1 is 0.3 seconds. At time t2, by increasing the reference amount A, the adjustment amount is decreased. If the reference amount A is such that a 1 degree angle change continues for 0.05 seconds, the reference amount A is increased and corrected to, for example, a 1 degree angle change continues for 0.06 seconds. Assume that the adjustment amount decreases as a result of increasing the reference amount A at the same rate at times t3 and t4. However, at time t5, as a result of increasing the reference amount A, the adjustment amount actually increases. This means that the reference amount A at time t4 was the optimal value. Therefore, the autonomous control device 32 sets the reference amount A to the condition at time t4.

Subsequently, if the reference amount B is such that the angle change of 1 degree continues for 0.05 seconds, the reference amount B is increased at time t6, and for example, the angle change of 1 degree continues for 0.06 seconds. Correct it so that it does. Assume that the adjustment amount decreases as a result of increasing the reference amount B at the same rate from time t7 to time t9. However, at time t10, as a result of increasing the reference amount B, the adjustment amount actually increases. This means that the reference amount B at time t9 was the optimal value. Therefore, the autonomous control device 32 sets the reference amount B to the condition at time t9.

Subsequently, if the second reference amount is such that the angle change of 1.5 degrees continues for 0.05 seconds, the second reference amount is increased at time t11, and for example, the angle change of 1.5 degrees is increased. is corrected to last for 0.06 seconds. Assume that the adjustment amount decreases as a result of increasing the second reference amount at the same rate from time t12 to t14. However, at time t15, as a result of increasing the second reference amount, the adjustment amount actually increases. This means that the second reference amount at time t14 was the optimal value. Therefore, the autonomous control device 32 sets the second reference amount to the condition at time t14.

As described above, the adjustment amount can be minimized by setting each of the plurality of reference quantities, ie, the reference quantity A, the reference quantity B, and the second reference quantity, to optimal values. This is because the first correction process A, the first correction process B, and the second correction process are correction processes based on outputs from different sensors such as a gyro sensor, an acceleration sensor, and an electronic compass, respectively.

Note that in the present embodiment, the reference amount A, the reference amount B, and the second reference amount are corrected by increasing the time, but different from this, they may be corrected by increasing the angle. Alternatively, both an increase in time and an increase in angle may be implemented.

As mentioned above, basically, by increasing the reference amount A, the reference amount B, and the second reference amount, the adjustment amount becomes smaller, but depending on the actual situation of the place where the unmanned aircraft 1 travels, Increasing the reference amount A or the like may actually increase the adjustment amount. In this regard, the autonomous control device 32 determines whether increasing or decreasing the reference amount is appropriate. For example, as shown in FIG. 28, assume that the adjustment amount is increased by increasing the reference amount A at time t2. This means that increasing the reference amount A is inappropriate as a correction for the reference amount A. In this case, the autonomous control device 32 decreases the reference amount A at time t3, and corrects it so that, for example, a 1 degree angle change continues for 0.04 seconds. Assume that at t4, as a result of reducing the reference amount A at the same rate, the adjustment amount is reduced. However, at time t5, as a result of decreasing the reference amount A, the adjustment amount actually increases. This means that the reference amount A at time t4 was the optimal value. Therefore, the autonomous control device 32 sets the reference amount A to the condition at time t4. Note that in both cases where the reference amount A is increased and decreased, if the adjustment amount increases, the current reference amount A is the optimal value, so the autonomous control device 32 maintains the current reference amount A. . The same applies to the reference amount A and the second reference amount.

Hereinafter, reference amount correction processing will be described with reference to the flowchart of FIG. 28. Note that the reference amount correction processing is performed at an appropriate timing in parallel with the correction processing of the nose direction of the unmanned aircraft 1. The autonomous control device 32 increases or decreases the reference amount A of the first correction process A (step ST501 in FIG. 28), and when determining that the reference amount A is the optimal value (step ST502), maintains the reference amount A. Subsequently, the autonomous control device 32 increases or decreases the reference amount B of the first correction process B (step ST503), and when determining that the reference amount B is the optimal value (step ST504), maintains the reference amount B. Subsequently, the autonomous control device 32 increases or decreases the second reference amount in the second correction process (step ST505), and when determining that the second reference amount is the optimal value (step ST506), maintains the second reference amount. do. When the autonomous control device 32 determines that the termination condition is satisfied (step ST507), the autonomous control device 32 terminates. When the autonomous control device 32 determines that the termination condition is not satisfied (step ST507), it repeats the processing from step ST501 to step ST506.

Note that the present invention is not limited to the above-described embodiments, and the present invention includes modifications, improvements, etc. within the range that can achieve the object of the present invention.

1 Tractor (drone)
10 Vehicle body 12 Antenna 14 Antenna 16 Cameras 18L, 18R Front wheels 20 Steering systems 22L, 22R Rear wheels 24 Shaft 26 Engine 28 Transmission 30 Rotary 32 Autonomous control device 202A, 202B, 202C, 202D Navigation satellite

Claims (8)

An autonomous control device for controlling a mobile device capable of autonomous movement,
a positioning means for positioning the current position using positioning radio waves from a navigation satellite;
Direction change measuring means for measuring a change in the nose direction of the mobile device at a time interval shorter than an output interval of positioning results by the positioning means;
first correction means for correcting the nose direction of the moving device so as to reduce the change in the nose direction measured by the direction change measurement means;
a return means that adjusts the moving device to move in a planned direction on a planned route based on the positioning result by the positioning device;
Speed change measuring means for measuring a change in the moving speed of the mobile device in a direction perpendicular to the traveling direction of the mobile device at a time interval shorter than an output interval of the positioning result by the positioning means,
The first correction means adjusts the movement so as to reduce the change in the moving position due to the change in the nose direction measured by the direction change measuring means and the change in the moving speed measured by the speed change measuring means. configured to correct the nose heading of the device;
Autonomous control device. Relative direction measuring means for measuring the nose direction of the forward moving device in relation to the direction of geomagnetism at time intervals shorter than output intervals of positioning results by the positioning means;
correcting the nose direction of the moving device so as to reduce the difference between the planned direction defined in a relative relationship with the direction of the earth's magnetic field and the nose direction measured by the relative direction measuring means; two corrective means;
The autonomous control device according to claim 1, having: nose direction calculation means for calculating the nose direction of the mobile device based on Doppler information indicating a Doppler effect of the positioning radio waves caused by relative movement between the navigation satellite and the mobile device;
directional deviation calculation means for calculating a deviation between the nose direction calculated by the nose direction calculation means and the planned direction;
third correction means for correcting the nose direction of the moving device so as to reduce the deviation calculated by the direction deviation calculation means;
The autonomous control device according to claim 2, having: The speed change measuring means is further configured to measure a change in speed in the scheduled direction at a time interval shorter than an acquisition interval of Doppler information, and further,
speed correction means for correcting the speed of the moving device so as to reduce the change in speed measured by the speed change measurement means;
a speed return unit that calculates a moving speed of the moving device based on the Doppler information and adjusts the speed so that the moving speed of the moving device matches a scheduled speed;
The autonomous control device according to claim 1, having: comprising priority means for prioritizing the signal from the work control means when a signal regarding movement of the mobile device is received from a work control means for controlling work performed by the mobile device;
An autonomous control device according to any one of claims 1 to 4. comprising a correction means for correcting the magnitude of the correction of the nose direction by the first correction means and/or the second correction means, based on the magnitude of the adjustment by the return means;
The autonomous control device according to claim 2. It has a positioning means that uses positioning radio waves from a navigation satellite to determine the current position,
A control method using an autonomous control device for controlling a mobile device capable of autonomous movement, the method comprising:
a direction change measuring step of measuring a change in the nose direction of the mobile device at a time interval shorter than an output interval of positioning results by the positioning means;
a first correction step of correcting the nose direction of the moving device so as to reduce the change in the nose direction measured in the direction change measuring step;
a return step of adjusting the mobile device to move in a scheduled direction on a scheduled route based on the positioning result by the positioning means;
has, and furthermore,
implementing a speed change measuring step of measuring a change in the moving speed of the mobile device in a direction orthogonal to the traveling direction of the mobile device at a time interval shorter than an output interval of positioning results by the positioning means;
In the first correction step, the movement is performed so as to reduce the change in the moving position due to the change in the nose direction measured in the direction change measuring step and the change in the moving speed measured in the speed change measuring step. A control method for implementing processing for correcting the nose direction of a device . A computer that controls an autonomous control device for controlling an autonomously movable mobile device,
a positioning means that uses positioning radio waves from a navigation satellite to determine the current position;
direction change measuring means for measuring a change in the nose direction of the mobile device at a time interval shorter than an output interval of positioning results by the positioning means;
first correction means for correcting the nose direction of the moving device so as to reduce the change in the nose direction measured by the direction change measurement means; and
return means for adjusting the mobile device to move in a planned direction on a planned route based on the positioning result by the positioning device;
function as , and furthermore,
The computer,
functioning as a speed change measuring means for measuring a change in the moving speed of the mobile device in a direction perpendicular to the traveling direction of the mobile device at a time interval shorter than an output interval of positioning results by the positioning means;
The first correction means adjusts the movement so as to reduce the change in the moving position due to the change in the nose direction measured by the direction change measuring means and the change in the moving speed measured by the speed change measuring means. correcting the nose direction of the device;
Autonomous control program.