WO2023286097A1

WO2023286097A1 - Drone and relative command method

Info

Publication number: WO2023286097A1
Application number: PCT/IT2022/050202
Authority: WO
Inventors: Luca Bigazzi; Michele Basso; Enrico Boni
Original assignee: Universita' Degli Studi Di Firenze
Priority date: 2021-07-13
Filing date: 2022-07-13
Publication date: 2023-01-19
Also published as: EP4370988A1; IT202100018362A1

Abstract

Drone (10), comprising at least one on-board computer (11), a flight controller (12) interfaced with said on-board computer (11) and with a plurality of motors (13) for moving the drone (10), and at least one inertial measurement unit (14) configured to detect the attitude of flight of the drone (10).

Description

“DRONE AND RELATIVE COMMAND METHOD”

FIELD OF APPLICATION

The present invention relates to a drone, in particular to a drone provided with an on-board computer, a flight controller and an interface device between said on-board computer and said flight controller, as well as to the relative control method.

BACKGROUND ART

A remotely piloted aircraft, commonly known as a drone, is a flying aircraft whose flight, being without a pilot on board, is controlled by an on board computer or by means of the remote control of a navigator or pilot, on the ground or in other positions.

Commercial drones, in particular, are becoming more and more popular, their technology is evolving rapidly and the scientific community is dedicating more and more time and resources to them. Each commercial drone, whether self-constructed or pre-assembled, implements a flight controller on board which is capable of stabilizing the attitude and tracking the references which are passed to it by the radio receiver, which communicates with the radio command of the human pilot.

However, the flight controllers used on these remotely piloted aircraft are based on microcontroller technology and do not allow the development of advanced modes, such as autonomous piloting based on artificial vision.

To try to develop applications of this type, dedicated computers are mounted on board, which however cannot dialogue efficiently and completely with the electronics of the drone, in particular with the flight controller, which is normally associated with at least one inertial measurement unit which detects and processes high-frequency data, that is, of the order of one or more kHz.

To such dedicated computers, which can be complex and expensive, it is rather difficult to connect any additional sensors that can be used on board, such as GPS, barometric, magnetometric, optical flow sensors or other. Furthermore, the information provided by such sensors, which is typically at high frequencies, is not correctly read and exploited by such computers. Furthermore, the dialogue between the on-board computer and the drone's flight controller can involve a number of issues, including the complex and burdensome need to intervene on the flight controller's firmware. Such an operation can also be risky, since potential bugs could be introduced in the flight controller's firmware. This may be necessary each time the type of flight controller changes. If the drone also includes a remote manual flight mode, moreover, it would not be possible to send the manual commands to the additional on-board computer in real time and this would preclude the development of Artificial Intelligence algorithms capable of learning from the human pilot.

A known type of drone with related control devices and having the above issues is described in US-A-2019033892. In particular, in such a document a single bus is used to connect all the electronic components on board, making the system inflexible and not universal.

Hence, there is a need to improve a drone which can overcome at least one of the drawbacks of the prior art.

In particular, an object of the present invention is to provide a drone which can be piloted accurately and reliably by means of an on-board computer.

A further object of the present invention is to make a drone which is provided with an interface device capable of creating a dialogue between any on-board computer with any inertial measurement unit, normally provided in the flight controller, without the need for hardware or software changes to the on-board computer or the flight controller.

A further object of the present invention is to make a drone which can include the development of innovative applications for autonomous piloting and in which the on-board computer can use Artificial Intelligence algorithms capable of learning from the commands possibly provided by a human pilot.

A further object of the present invention is to make a drone provided with an on-board computer to which a plurality of sensors provided with the drone can be associated, such as GPS, barometric, magnetometric, optical flow sensors or other. A further object is to devise a method for commanding a drone.

The Applicant has studied, tested and realized the present invention to overcome the drawbacks of the prior art, and to obtain the above as well as further objects and benefits.

DISCLOSURE OF THE INVENTION

The present invention is expressed and characterised in the independent claims. The dependent claims show other features of the present invention or variants of the main solution proposed.

In accordance with the aforesaid objects, a drone according to the present invention comprises at least one on-board computer, a flight controller, interfaced with said on-board computer and with a plurality of motors for moving the drone, and at least one inertial measurement unit configured to detect the attitude of flight of the drone.

According to an aspect of the invention, the drone comprises an interface device comprising electronic circuits configured to receive the commands generated by the on-board computer through a communication bus. Such an interface device is connected to said on-board computer, to said flight controller and to said inertial measurement unit and configured to perform real-time algorithms on data from said inertial measurement unit, send the results of said algorithms to said on-board computer and receive autonomous commands from said on-board computer to be sent to said flight controller. Such a flight controller will provide such motors with the drone attitude and thrust information.

Advantageously, the present drone, by virtue of the interface device connected to the on-board computer, the flight controller and the inertial measurement unit, can be piloted in a precise and reliable manner by means of such an on-board computer.

The present interface device is capable of creating a dialogue between any on-board computer with any commercial flight controller, without requiring hardware or software changes to the on-board computer or flight controller.

According to a further aspect of the invention, said inertial measurement unit has a high frequency information contribution, of the order of one or more kHz, and said interface device can be configured to perform a frequency intermediation and lower said frequency to about 100-200 Hz. Such an interface device is then configured to provide an estimate of attitude and position which is transmitted at lower frequencies while maintaining the quality and integrity of the information contribution. Such an interface device can perform high-frequency, real-time algorithms on data from the inertial measurement unit and any other sensors. For example, such algorithms for the inertial sensor can relate to the reconstruction of the attitude estimate, which once reconstructed can be sent to the on-board computer at any frequency compatible with the capabilities of the latter, without loss of information, as the estimate was obtained by exploiting all the available information content.

The interface device is able to execute the algorithm for estimating the attitude and position at frequencies of the order of KHz. The more the frequency increases, the more the estimation accuracy increases. Since it is not possible to reach such frequencies if such operations are performed on the on-board computer, the presence of the interface device produces an advantage in terms of performance.

According to a still further aspect of the invention, said interface device can be connected to a receiver installed on board the drone and configured to receive drone command signals from a radio command for remote piloting.

Such an interface device can further be configured to perform a switching mode by means of which it can send automatic commands to the flight controller from the on-board computer, manual commands from the radio command, or manual and automatic commands, thereby obtaining a hybrid mode that assists the human pilot during piloting.

A neural network for Artificial Intelligence applications can also be installed on such an on-board computer.

The present drone can therefore support the development of innovative applications for autonomous piloting and the on-board computer can use Artificial Intelligence algorithms capable of learning from the commands provided by a human pilot. Such an interface device can communicate with such a flight controller by at least one of the following protocols: PWM (Pulse-Width Modulation), PPM (Pulse Position Modulation), S-BUS, or others.

According to a further aspect of the invention, further sensors, such as inertial sensors, magnetometers, GPS, barometers, optical flow sensors or other, can be connected to such an interface device.

The interface device therefore functions as a sensor HUB or concentrator, therefore leaving the Input/Output contacts of the on-board computer free.

Such an interface device can also be provided with a non-volatile memory for storing the processed data and can comprise electronic circuits configured to receive the commands generated by said on-board computer through a communication bus.

A further aspect of the present invention is a method for commanding a drone, comprising detecting data on the attitude and flight position of the drone by means of such an inertial measurement unit, performing algorithms on such data by means of the interface device, sending the results of such algorithms to such an on-board computer, receiving from such an interface device autonomous commands processed by such an on board computer, providing such motors with such attitude and thrust information through such a flight controller. The processed autonomous commands can include, for example, attitude, drone height control, or other commands.

ILLUSTRATION OF THE DRAWINGS

These and other aspects, features and advantages of the present invention will become clear from the following embodiment disclosure, given by way of example only, with reference to the attached drawings in which:

- fig. 1 is a schematic drawing of the main components of a drone according to the present invention.

To facilitate understanding, identical reference numbers have been used, where possible, to identify identical common elements in the figures. It should be noted that elements and features of an embodiment can be conveniently combined or incorporated into other embodiments without further clarification. DESCRIPTION OF EMBODIMENTS

Detailed reference will now be made to a preferred embodiment of the invention, of which an example is illustrated in the attached drawing by way of non-limiting example. The phraseology and terminology used herein is also exclusively for illustrative purposes.

With reference to figure 1, a drone 10 according to the present invention comprises at least one on-board computer 11, a flight controller 12 interfaced with such an on-board computer 11 and with a plurality of motors 13 for moving the drone 10, and at least one inertial measurement unit 14 configured to detect the attitude of flight of the drone 10.

The drone 10 comprises an interface device 15 connected to such an on board computer 11, such a flight controller 12 and such an inertial measurement unit 14 and configured to perform algorithms on data from such an inertial measurement unit 14, send the results of such algorithms to such an on-board computer 11 and receive from such an on-board computer 11 autonomous commands to be sent to such a flight controller 12. Thereby, the flight controller 12 can command the motors 13 of the drone 10 based on data received from such an interface device 15. Thus substantially, such a flight controller 12 provides such motors with the attitude and thrust information of the drone 10.

The execution of algorithms on this data is performed directly on board the interface device 15, ensuring the performance on the data estimation accuracy which is significantly higher with respect to the estimates performed to date on the on-board computers of drones. The data pre- processed by the interface device 15 are sent in aggregate form to the on board computer 11.

As can be seen, it is possible to include a plurality of inertial measurement units 14 associated with the interface device 15.

The inertial measurement unit 14 is provided with inertial sensors which have an information contribution focused on high frequencies, such a contribution cannot be exploited by a normal on-board computer 11 based on a non-real-time operating system. Such an information contribution can instead be fully exploited by the interface device 15, which uses microcontroller technology and is capable of processing such high- frequency data.

Substantially, the interface device 15 pre-processes the data related to the attitude and provided by the inertial measurement unit 14 at the same frequencies as the inertial measurement unit 14, that is, of the order of one or more kHz, and provides them in aggregate form to the on-board computer 11 at a frequency of the order of 100-200 Hz, so that they can be correctly processed by such an on-board computer 11 , therefore performing a frequency intermediation.

Such a frequency intermediation allows the on-board computer 11 to process all the data received from the interface device 15 and thus provide a result containing the autonomous commands to be provided to the flight controller 12 which drives the motors 13 accurately and reliably. The frequency intermediation is performed by means of an appropriate algorithm.

Such an interface device 15 comprises electronic circuits configured to receive the commands generated by the on-board computer 11 through a communication bus. Such electronic circuits are in fact installed in a further electronic board included in the present system.

Such electronic circuits can comprise high-performance microcontrollers normally present on the market.

The interface device 15 can also be connected to a receiver 18 installed on board the drone 10 and configured to receive command signals of the drone 10 from a radio command 16 for remote piloting.

The interface device 15 is configured to perform a switching mode by which it can send automatic commands to the flight controller 12 from the on-board computer 11 , manual commands from the radio command 16 or manual and automatic commands, thereby obtaining a hybrid mode that assists the human pilot during piloting.

For example, it is possible to include that the human pilot, by means of the radio command 16, can choose a piloting mode of the drone 10 with manual piloting, an automatic piloting mode by means of the on-board computer 11, or a hybrid mode, with manual piloting assisted by the on- board computer 11. Preferably, moreover, the interface device 15 can allow remote piloting as preferential piloting as soon as the human pilot operates the radio command 16.

The interface device 15 can then send to the on-board computer 11 the commands received from the radio command 16 and then from the receiver 18. This feature can be further useful if a neural network for Artificial Intelligence applications is installed on the on-board computer 11 and which must learn to pilot the drone 10. In this case it is possible to pilot the drone 10 manually by having the on-board computer 11 record all the commands that the human pilot sends to the flight controller 12 of the drone 10 through the radio command 16.

If the interface device 15 is also connected to the radio command 16 and by virtue of the aforesaid switching mode between the manual controls and the autonomous ones, it is possible to greatly improve the safety of the drone 10, since there is the possibility of excluding the commands generated by the on-board computer 11 at any time.

The interface device 15 can be capable of communicating with the flight controller 12 of the drone by regenerating the input signals it receives in the protocols commonly used on these systems. In particular, it is possible to choose the protocol used by the interface device 15 to communicate with the flight controller 12 among at least one of the following protocols: PWM, PPM, S-BUS. These protocols would not be replicable directly by the on board computer 11 without the use of the interface device 15, since to generate such signals a system capable of managing the interrupts, in real time, is necessary, such as a microcontroller.

The interface device 15 is configured to also function as a concentrator or HUB for additional sensors 17, such as inertial sensors, magnetometers, GPS, barometers, optical flow sensors, or other. Also for these sensors 17, the interface device 15 is capable of executing high-frequency algorithms on the information and sending the results in aggregate form to the on-board computer 11, which can use them for high-level algorithms.

The interface device 15 can be associated with the flight controller 12 of any commercial drone 10 without any modification to the other components of the drone 10, in fact it can generate an output command signal encoded in the same standard used by the receiver 18.

Furthermore, the fact that other sensors 17 can be connected to the interface device 15, which functions as a concentrator or HUB, also allows the Input/Output contacts of the on-board computer 11, which otherwise should be occupied by such sensors 17, to be left free. This allows the on board computer 11 to perform other tasks and use the pre-processed and reconstructed data from the interface device 15, by means of for example serial bus communication.

The interface device 15 can further be provided with a non-volatile memory 19 for storing the processed data, thus a sort of "black box" of the drone 10. This aspect can be useful since the stored data can be used in neural networks that can be included in the on-board computer 11 of the drone 10.

The method for commanding the drone 10 according to the present invention substantially includes the detection of data on the attitude of flight of the drone 10 by means of such an inertial measurement unit 14 and/or such a plurality of sensors 17, the execution of algorithms on such data by means of the interface device 15, the sending of the results to such an on board computer 11, the receipt by such an interface device 15 of the autonomous commands generated by such an on-board computer 11, the provision to such motors 13 of said attitude information through such a flight controller 12.

According to the present invention, to connect the on-board computer 11, in particular of high level, to the flight controller 12, said interface device 15, i.e., an electronic board, is substantially interposed. This board can be connected with various sensors and be capable of making various types of filters, such as reconstruction of attitude and position, then sending all this data to said on-board computer 11.

Another advantage that can be obtained with the present invention is to increase the degree of safety of the system, since if the high level on-board computer 11 which generates the autonomous commands fails, it is always possible to switch to the manual commands, since it is the interface device 15 which operates this decision and being a separate board it works independently of the correct operation of said on-board computer 11.

With respect to, for example, what is described in the aforementioned document US-A-2019/033892, furthermore, the present assembly or system consisting of the on-board computer 11 and the interface device 15 is a universal system. In fact, by virtue of the addition of said interface device 15 it is possible to abstract the system and use it on any type of commercial drone/rover, whether closed-source, open-source or built car. This is possible because said interface device 15 can be interconnected between the receiver receiving the manual commands of the human operator and the flight controller 12.

The interface device 15 is thus capable of replicating all the protocols normally used on these systems, such as: SBUS, PPM, PWM etc... In fact, the flight controller 12 is "unaware" of who is generating the input references, since both the manual and synthetic commands generated by the on-board computer 11 share the same protocol and the same input port towards the flight controller 12.

It is clear that modifications and/or additions of parts or steps can be made to the drone 10 or to the relative command method described so far, without departing from the scope of the present invention as defined by the claims. In the following claims, the references in parentheses have the sole purpose of facilitating reading and must not be considered as limiting factors as regards the scope of protection underlying the specific claims.

Claims

1. Drone (10) comprising at least one on-board computer (11) to which a flight controller (12) is interfaced, a plurality of motors (13) for its movement and at least one inertial measurement unit (14) configured to detect its attitude of flight, characterized in that it comprises an interface device (15) comprising electronic circuits configured to receive the commands generated by said on-board computer (11) through a communication bus, said interface device (15) being connected to said on board computer (11), to said flight controller (12) and to said inertial measurement unit (14), configured to perform algorithms on the data coming from said inertial measurement unit (14), send the results of said algorithms to said on-board computer (11) and to receive from the latter the autonomous commands to be sent to said flight controller (12), so that said flight controller (12) can provide said motors (13) with the information about the attitude and thrust of the drone (10).

2. Drone (10) according to claim 1, characterized in that said inertial measurement unit (14) has a high frequency information contribution, of the order of one or more kHz, and said interface device (15) is configured to perform a frequency intermediation and lower said frequency to about 100- 200 Hz, that is, to provide an estimate of attitude and position which is transmitted at lower frequencies while maintaining the quality and integrity of the information contribution.

3. Drone (10) according to claim 1 or 2, characterized in that said interface device (15) is connected to a receiver (18) installed on board of said drone (10) and configured to receive the command signals from a radio command (16) for remote piloting.

4. Drone (10) according to claim 3, characterized in that said interface device (15) is configured to perform a switching mode by means of which it can send automatic commands to the flight controller (12) from the on board computer (11), manual commands from the radio command (16), or manual and automatic commands, obtaining a hybrid mode that assists the human pilot during piloting.

5. Drone (10) according to any one of the preceding claims, characterized in that a neural network for Artificial Intelligence applications is installed on said on-board computer (10).

6. Drone (10) according to any one of the preceding claims, characterized in that said interface device (15) communicates with said flight controller (12) by means of at least one of the following protocols: PWM, PPM, S- BUS, or other standard protocols compatible with commercial flight controllers.

7. Drone (10) according to any one of the preceding claims, characterized in that further sensors (17) are connected to said interface device (15), such as inertial sensors, magnetometers, GPS, barometers, optical flow sensors or other.

8. Drone (10) according to any one of the preceding claims, characterized in that said interface device (15) is provided with a non-volatile memory (19) for storing the processed data.

9. Drone (10) according to any one of the preceding claims, characterized in that said interface device (15) comprises an electronic board configured to receive the commands generated by said on-board computer (11) through said communication bus.

10. Method for commanding a drone (10), comprising at least one on-board computer

(11) to which a flight controller

(12) is interfaced, with a plurality of motors

(13) for moving the drone (10) and at least one inertial measurement unit (14) configured to detect the attitude and position of flight of the drone (10), wherein said method comprises detecting data on the attitude of flight of the drone by means of said inertial measurement unit

(14), executing algorithms on said data by means of an interface device

(15), comprising electronic circuits configured to receive the commands generated by the on-board computer (11) through a communication bus, sending said algorithm results to said on-board computer (11), receiving by said interface device (15) the autonomous commands processed by said on board computer (11), providing said motors (13) with said attitude and thrust information through said flight controller (12).