Method for electing a master drone within a fleet of drones, method for managing a fleet of drones, computer program product, and assembly comprising a fleet of drones

Abstract

The invention relates to a method (500) for electing a drone (111), referred to as the master drone, to carry out a particular function with respect to the other drones within a fleet (11) of drones. The method makes it possible to select the master drone (111) by: - (521) evaluating a weight w for each drone (112); - (522) comparing the weights w; and - (524) nominating the master drone on the basis of the comparison of the values of the weights w. Figure for the abstract: Figure 1B

Description

[0001] DESCRIPTION

[0002] Title: Method for selecting a master drone within a drone fleet, method for managing a drone fleet, computer program product, and assembly comprising a drone fleet

[0003] TECHNICAL FIELD

[0004] The present invention belongs to the technical field of operations and missions carried out using drones.

[0005] The invention relates more particularly to a method of selecting a master drone to carry out a mission, entirely or in part, in an autonomous and collaborative manner.

[0006] The invention has a direct application in the field of site surveillance and observation, particularly as a complement to human resources.

[0007] STATE OF THE ART

[0008] In the drone systems sector, the effective coordination and management of a fleet to accomplish diverse missions is a central challenge. Historically, drone fleets were operated in a rather rigid manner, with a set of pre-programmed directives, limiting responsiveness to changing circumstances or unforeseen events during missions. The increasing complexity of missions assigned to these unmanned systems, particularly aerial missions, has revealed shortcomings, notably the difficulty of real-time adaptation and the need to redefine mission objectives.

[0009] Previous solutions largely rely on centralized control, where each drone is piloted individually from a command post or according to an operational plan that does not allow for flexibility.

[0010] However, in fluctuating operational circumstances, this configuration results in critical mission delays, or even failures, due to the inability to adapt fleet behavior to the constraints of new missions and the difficulty of managing drone fleet priorities in real time. In the specific context of fire detection, historically, surveillance was conducted by firefighters from lookout towers. More recently, drones and reconnaissance aircraft have been used. These devices allow for the monitoring of an area and the provision of visual information, which is then analyzed. However, these methods require a significant investment in human resources, leading to results that can be redundant or incomplete and suffer from a lack of coordination.

[0011] Various methods have been considered to improve the implementation of drone fleets.

[0012] Centralized drone management systems present challenges related to dependence on the command center, potentially leading to mission interruptions if the connection is lost. Furthermore, drones and aircraft operated by human operators require significant personnel, which is particularly critical in the context of fighting forest fires.

[0013] Document US2018139152 describes a method for facilitating communication between drones in multiple drone networks, particularly to improve the deployment performance of said drones in the field. The drones may also be multi-owned and / or belong to one or more drone networks. These drone groups can be tasked with carrying out various missions such as collecting images of road traffic, capturing meteorological or environmental data, etc.

[0014] This document notes that a fleet of drones can be configured to have autonomous capabilities, and thus be able to coordinate themselves in a relatively autonomous manner.

[0015] However, this document does not describe how the reorganization of the drone fleet is managed when an event such as a drone crash results in the loss of one or more drones, which could lead to the premature abandonment of the mission assigned to the drone fleet. PRESENTATION OF THE INVENTION

[0016] The present invention aims to overcome all or part of the disadvantages presented above, and to propose a method of managing a fleet of drones enabling the coordination of a mission assigned to a fleet of drones, minimizing or at least reducing the need for human intervention for the coordination of the fleet.

[0017] This objective is achieved primarily through the use of 'master drones'. A master drone is defined here as a drone within a fleet of drones tasked with performing a specific function in relation to the other drones in the fleet. However, to ensure the robustness of the system, a procedure must be in place to guarantee the continuity of the function assigned to the master drone even if it is lost (or unable to perform its function). To ensure this continuity of function, a master drone selection procedure is proposed: this procedure allows for the designation of a new master drone in the event of a master drone failure.

[0018] This process is a method for selecting a drone, called the master drone, to perform a specific function in relation to the other drones within a drone fleet. The process is characterized by a master drone selection step comprising the following steps:

[0019] (521) evaluation of a weight w for each drone;

[0020] (522) comparison of weights w; and

[0021] (524) nomination of the master drone, elected on the basis of the comparison of the values ​​of the weights w.

[0022] The master drone (also called the chosen drone) can be, for example, the drone with the highest weight, or conversely, the lowest weight.

[0023] In the preceding definition, weight is preferably a variable considered representative of a drone's ability to perform a particular function. For example, if the drone must be able to perform its specific function for a potentially long period, the weight can be correlated with the remaining battery capacity. Thus, a drone with a fully charged battery, and therefore a long flight time, could have a high weight value, which would favor its selection as the master drone. The execution of this process can be ensured by one or more computing devices. These computing devices can be of any type, but normally possess communication functions to exchange information with the various drones in the fleet. They can be mounted on one or more drones or integrated into a ground-based control and command station.

[0024] Naturally, each drone is equipped with means of communication allowing it to communicate at least with the other drones in the fleet.

[0025] Furthermore, at least some drones (which may be master drones) are equipped with means of communication allowing them to communicate at least with a third party, particularly one located on the ground.

[0026] The drone selection process can be implemented in particular in the following two modes of implementation:

[0027] In the first implementation mode, in addition to the drone fleet, a ground-based computing system is planned, for example, within a control and command post. In this case, the drone selection process (specifically step S520) is primarily performed by this computing system. In the second implementation mode, each drone has an onboard computing system; the drone selection process (specifically step S520) is performed collectively by these computing systems. Advantageously, in this case, the drone selection process is executed autonomously by the drone fleet.

[0028] Unless otherwise indicated or technically impossible, in the processes according to this disclosure, each planned calculation step can in particular be carried out by the calculation device(s) indicated above for the first or second mode of implementation.

[0029] In particular, in certain implementation modes, all or part of the process steps according to this disclosure are carried out in parallel by each of the drones in the fleet.

[0030] In some implementation modes, at least one drone weight w is evaluated by means of a drone weight evaluation function, said function being a function of at least one parameter among representative values ​​of a remaining autonomy of a drone battery, a distance between the drone and a point on the ground, an electrical power consumed by a computing device of the drone, an availability of at least one predetermined sensor of the drone, a quality of connection between the drone and a communication device, carried on another drone or on the ground.

[0031] The drone evaluation function can, for example, be increasing (or decreasing respectively) with respect to its input parameter(s).

[0032] For example, the weight can be calculated by the formula: w = P x Cbat / DC2 with P a coefficient, Cbat the battery level, DC2 the distance of the drone from a point on the ground (a point where the drone can be recharged and / or repaired, for example the location of a control and command post (hereafter referred to as C2)).

[0033] Connection quality can be represented, for example, by a signal-to-noise ratio, a measurement of throughput or latency, the existence of a direct line of sight...

[0034] Taking connection quality into account when calculating weight makes it possible to favour, when selecting the master drone, drones whose communication equipment works well, which is particularly important for smooth management of the drone fleet.

[0035] In some implementation modes, the (520) master drone selection step further includes a step of notifying the elected drone of the result of its election, by each of the remaining drones in the fleet.

[0036] In these implementation modes, preferably, the master drone election step (520) further includes a step (526) of validation of the election by the elected drone on the basis of the meanings received during the step (525) of notifying the elected drone of the result of its election.

[0037] In these latter implementations, preferably the election validation step (526) for the elected drone consists of verifying that the elected drone received, during the notification step (525) for the elected drone, a number of votes (i.e., messages indicating that it is elected) exceeding a predetermined proportion (for example, half) of the number of drones in the fleet that could become the master drone. In some implementations, the method includes an initialization step (510) preceding the master drone selection step (520), the initialization step (510) including a step (512) for identifying at least one drone in the fleet that could become the master drone.

[0038] In some embodiments, the method includes an initialization step (510) preceding the master drone selection step (520), the initialization step (510) including a step (513) of transmitting a status message, either by each drone in the fleet, or by each drone in the fleet capable of becoming a master drone, to at least one entity in charge of executing the weight evaluation step (521), the status message indicating at least one value among values ​​representative respectively of a remaining battery life of the drone (Cbat), a distance between the drone and a point on the ground, an electrical power consumed by a computing device of the drone, an availability of at least one predetermined sensor of the drone, a quality of connection between the drone and a communication device, carried on another drone or on the ground.

[0039] In other implementation modes, each of the drones likely to become a master drone calculates its weight, and transmits the calculated weight value to the computing and communication device.

[0040] More generally, to achieve the objective of the present invention, a method for managing a fleet of drones is also proposed, comprising the election of a master drone by an election method as presented above, and the performance of the particular function by the master drone with respect to the other drones in the drone fleet, the particular function comprising, for example, calculating control commands for other drones in the drone fleet, and / or transmitting messages, including commands, to other drones in the drone fleet.

[0041] In some implementations of this management process, the master drone periodically sends a status message (to at least one entity responsible for executing the master drone election process, for example, to the other drones in the fleet) indicating that the master drone is operational. When at least one expected status message is not received, or when no status message is received for a predetermined period, the drone election process (500) as defined previously is triggered. This process is normally triggered by the recipient(s) of the status messages, that is, by the entity or entities responsible for executing the master drone election process. As previously stated, this entity or these entities include one or more computing devices capable of executing step 520 of the process.In these implementation modes, when at least one expected status message is not received or when no status message is received for a predetermined period, the drone election process as defined above is triggered.

[0042] In some implementation modes, the management process further includes a fleet reorganization step (530); said reorganization step (530) comprising the following steps:

[0043] (531) change of state of the elected drone, from the state of candidate drone to the state of master drone;

[0044] (532) meaning to drones having their state in that of a candidate drone for the election of the new master drone; and

[0045] (533) change of states of drones having their state to that of candidate drone to the state of slave drone.

[0046] In one particular embodiment, the various steps of the processes according to this disclosure are determined by computer program instructions.

[0047] Accordingly, this disclosure also relates to a computer program product characterized in that it comprises a set of program code instructions which, when executed by at least one processor, implement one of the processes defined above. This program or these programs may use any programming language and be in the form of source code, object code, or code intermediate between source and object code, such as in a partially compiled form, or in any other desirable form.

[0048] This disclosure also relates to a set of computer-readable non-transient recording media on which is recorded a set of program code instructions which, when executed by at least one processor, implement one of the processes as defined above.

[0049] The information storage medium can be any entity or device capable of storing the program(s). For example, the media may include a storage medium, such as a ROM (e.g., a CD-ROM or a microelectronic circuit ROM), or a magnetic recording medium, such as a floppy disk or a hard disk drive. Alternatively, the storage media may be integrated circuits in which the program(s) are embedded, the circuits being adapted to perform, or to be used in performing, the processes described in this disclosure.

[0050] By extension, this disclosure also covers an assembly comprising a fleet of drones, the assembly comprising at least one computing device, said at least one computing device having at least one processor and a set of recording media on which a set of program code instructions is or are recorded; said at least one computing device being configured so that, when said program code instructions are executed by said at least one processor, said at least one computing device implements one of the methods as defined above. The fundamental concepts of the invention having been set forth above in their most elementary form, further details and features will become clearer upon reading the following description and with reference to the accompanying drawings.

[0051] BRIEF DESCRIPTION OF THE FIGURES

[0052] The figures are provided for illustrative purposes only to aid understanding of the invention without limiting its scope. The various elements may be represented schematically and are not necessarily to the same scale. Throughout the figures, identical or equivalent elements are identified by the same numerical reference.

[0053] This is illustrated as follows:

[0054] - Figure 1A: A diagram of a first system for carrying out missions using a fleet of drones, according to one embodiment of the invention; - Figure 1B: A diagram of a second system for carrying out missions using a fleet of drones, according to another embodiment

[0055] - Figure 2: a diagram of the main functions embedded in a drone of the system, according to one embodiment of the invention;

[0056] - Figure 3: a diagram of the main internal functions of a control and command station of the system, according to one embodiment of the invention;

[0057] - Figure 4: a diagram of the main steps of a mission coordination process involving at least one fleet of drones as well as the control and command post.

[0058] - Figure 5: a diagram of the main steps in the process of electing the master drone of the drone fleet.

[0059] DETAILED DESCRIPTION OF IMPLEMENTATION METHODS

[0060] The embodiment described below refers to a method for selecting a master drone from a fleet of drones to autonomously carry out one or more missions, in whole or in part. An example of an application of the invention is the performance of a support mission for surveillance and firefighting. This example is not limiting; the invention can be applied in all fields where aerial, terrestrial, marine, and underwater drones are used for surveillance, exploration, rescue, observation, and other missions.

[0061] Figure 1A and Figure 1B respectively represent a system 100a and a system 100b for carrying out missions using a set of drones comprising a first fleet of drones 11 and a second fleet of drones 12. Each of the systems 100a and 100b includes a control and command station, C2, 10, and two fleets of drones, referenced 11 and 12. Each of these fleets includes a master drone 111 and at least one slave drone 112. Each drone can be piloted using a dedicated remote control 113. The drones can be located at varying distances from each other in order to cover, in particular, the widest possible area, depending on the missions assigned to the fleets 11 and 12.The C2 station (referenced 10) can be a building or a moving means of transport such as a land vehicle to coordinate support missions to firefighters, and includes at least one operator 101 as well as a computing device 102 to perform certain calculation steps and exchange data with the drone fleets 11 and 12.

[0062] In the first implementation mode illustrated by Fig.1 A, the master drone election process is implemented mainly by the computing device 102 of station C2, schematically represented in Figure 1A.

[0063] Thus, in this example, the master drone selection process is primarily executed by the entity constituted by the computing device 102 at station C2. This computing device 102 has the hardware architecture of a computer, as schematically illustrated in Figure IA. Generally, any data processing device comprising at least one memory capable of storing data and the program that will be presented later, and one or more processors capable of executing this program, can be used.

[0064] In this embodiment, the computing device 102 includes in particular a screen 103, a processor 104, a non-volatile flash memory 105, as well as communication means 106 for communicating with the drones of fleets 11 and 12.

[0065] The non-volatile memory 105 of the computing device 102 constitutes a recording medium conforming to this disclosure, readable by the processor 104, on which is stored a computer program conforming to this disclosure, containing instructions for executing the steps of a fleet management process conforming to this disclosure. This program includes, in particular, a program containing instructions for executing the steps of a master drone selection process conforming to this disclosure.

[0066] Although in this implementation mode, the drone selection process (and more generally the fleet management process) is implemented essentially in the computing device 102 integrated into the control station C2, in other implementation modes, such as that presented by Fig.1 B, the drone selection process (and more generally the fleet management process) is implemented essentially by the drones themselves, more precisely by computing devices, referenced 200, integrated respectively into each of the different drones in the fleet considered.

[0067] The hardware architecture of these computing devices 200 is schematically the same as that of device 102 described previously as shown in Fig.1 A.

[0068] In this second mode of implementation, the computing devices 200 thus include recording media on which are recorded instructions enabling the implementation of the processes presented in this disclosure for the election of a master drone and more generally for the management of the drone fleet.

[0069] In the embodiments described here, drones are unmanned aerial vehicles (UAVs), the present invention being naturally applicable to any set of connected devices without a human operator, in particular to any fleet of vehicles.

[0070] Preferably, each of these drones includes a GNSS (Geolocation and Navigation Satellite System) receiver, at least one camera, and a plurality of sensors, depending on the type of missions for which these drones are used.

[0071] In one particular embodiment, all drones are equipped identically. In other words, in this embodiment, fleets 11a and 11b are said to be "homogeneous." However, in other embodiments, the drones are equipped differently, for example in terms of sensors and / or computing resources (RAM, processor, etc.), in order to perform specific tasks. In this embodiment, fleets 11 and 12 are said to be "heterogeneous."

[0072] In addition to telecommunications via remote controls 113, drones can exchange data with each other, and at least some of the drones are able to exchange data with the C2 10 station.

[0073] Figure 1A represents the first implementation mode, in which communication is centralized. In this case, the master drone 111 centralizes all data from the slave drones 112 and transmits it to station C2 10. In this embodiment, station C2 10 coordinates the mission, and the computing and communication device 102 of station C2 executes the steps of the master drone selection process according to this disclosure.

[0074] For this implementation method, the distance between the master drone 111 and the C2 station must remain sufficiently small at all times to allow communication to be maintained between the C2 station 10 and the master drone 111. On the other hand, the slave drones 112 can be out of communication range of the C2 station 10, as long as the master drone 111 maintains communication with the C2 station.

[0075] In an alternative implementation with centralized communication (not shown in the figures), station C2 10 exchanges data directly with each of the drones 112 in fleets 11 and 12, without using a master drone 111 to relay the information. This implementation can be used while the master drone selection process is running.

[0076] In the embodiment shown in Figure 1A, all decisions are managed by station C2 10, and drones 111 and 112 merely execute commands from this station. The specific function performed by the master drone is the transmission of information from the slave drones to station C2 and vice versa.

[0077] This implementation method has one main constraint during mission execution: the need to maintain communication between the C2 10 station and the drone fleet(s). Therefore, if communication is interrupted between the C2 10 station and one of the drone fleets, that fleet is unable to continue its mission.

[0078] Figure 1B illustrates the second implementation mode, in which communication is distributed. In this case, all the drones (slave drones 112 and master drone 111) exchange data so that it is received by the fleet's computing and communication system. In the example shown, this system is collectively comprised of all the computing systems 200 of the drones in the fleet 11. Furthermore, the master drone 111 preferably remains the sole interface between the drone fleet 11b and the C2 station 10. This master drone may choose not to maintain communication with the station and continue the mission autonomously. Thus, the drone fleet 111 is capable of carrying out the mission autonomously, but also collaboratively.

[0079] Advantageously, since the 100b system is based on distributed communication, the problems of processing a large data stream by a single drone (111) causing a bottleneck that impacts the performance of the 100b system (processing and response times), as well as problems arising from potential hardware failures of the drone (111) centralizing the data, are resolved within the 100b system. Furthermore, this distributed approach significantly increases the operational resilience, adaptability, and overall efficiency of the 11b fleet, while minimizing points of failure caused by overly centralized control.

[0080] Beyond communication via a distributed system, thus ensuring inter-drone collaboration within Fleet 11, for Fleet 11 to operate autonomously throughout the mission, it is first necessary that the master drone 111 coordinate the fleet during its mission. It is also necessary that a master drone selection process, described later, be implemented by the 100b system. This master drone selection process is ensured by various functionalities integrated into drones 111 and 112, as well as into the C2 station.

[0081] The 100b system is capable of carrying out the mission autonomously, collaboratively by self-organizing tasks and objectives according to the state of the fleet 1 1 and through the election of the master drone 111 which distributes the different commands or instructions to the slave drones 1 12.

[0082] Advantageously, the 100b system is robust to the loss or reassignment of the master drone 111 or one of the slave drones 112.

[0083] Thus, systems 100a and 100b are used to carry out separate operations in order to achieve one or more objectives collaboratively; said objectives may be wholly or partly common to fleets 11a or 11b of drones.

[0084] Figure 2 shows a diagram of the main functions of the different drones (drones 111 and 112). These functions are performed by a computing and communication device 200 onboard each of the drones.

[0085] Thanks to device 200, first of all, the drone has a set of basic functions 210, called high-level functions which convert an intention into a movement objective. Among the basic functions 210 are functions to control takeoff 211, landing 212, stabilization 213 of the drone 111 or 112 but also to reach a specific position 214, to follow a given trajectory 215 or to perform a movement 216, such as a longitudinal, lateral, vertical or yaw rotation for example. Next, the drone also has an avoidance function 220. The avoidance function 220 analyzes the movement objective defined by one of the basic functions 210 and, once the intra-drone distance is determined, for example from the GNSS position of other drones or proximity sensors, or by other methods, adjusts if necessary the movement objective of drone 111 or 112 to avoid a collision.Once the avoidance function 220 has been performed, the movement objective is sent to a low-level control function 230 which acts directly on drone motors 111 or 112 to obtain, for example, a thrust or a rotation speed of the propellers of said drone.

[0086] The control function 230 includes a control function for the drone's motors 231 to physically control said motors, for example by adjusting their supply voltage to enable the requested command to be carried out.

[0087] In addition to these flight functions, the drone also has:

[0088] - a localization function 221;

[0089] - an image acquisition function 222;

[0090] - a function indicating a state 223;

[0091] - a function for electing a master drone 224; and

[0092] - a coordination function 225 for the drone fleet, in the event that said drone is designated as the master drone. The localization function 221 determines the geographic coordinates of the drone 111, 112 in the terrestrial reference frame (latitude, longitude, altitude), from the onboard GNSS receiver.

[0093] The 222 image acquisition function allows the 111 or 112 drone to capture and record images using onboard cameras.

[0094] The status indication function 223 consists of transmitting, at regular intervals, to station C2 and / or to drones 111 or 112 of the drone fleet 11a or 11b, a status message indicating, in particular, that drone 111 or 112 is still operational and dedicated to the current mission. The status indication function 223 also transmits the drone's status from among the following: master drone, slave drone, or candidate drone.

[0095] Using the status indication function 223, the master drone is configured to periodically send a status message to the computing and communication device indicating that the master drone is operational.

[0096] The function of electing a master drone 224 consists of electing the master drone 111, based on predefined criteria.

[0097] The master drone 111, which was previously elected, then coordinates the drone fleet to which it belongs (fleet 11 or 12 respectively) via the drone fleet coordination function 225. This drone fleet coordination function 225 is only activated for the master drone 111.

[0098] The master drone selection function 224 uses an evaluation function 299 for a weight w (the drone's weight). This evaluation function allows the weight to be calculated based on one or more parameters. These parameters are chosen appropriately according to the mission assigned to the drone fleet. They may include the drone's remaining battery capacity, its position (its geographical coordinates), etc.

[0099] Finally, the drone includes a set of 240 communication functions:

[0100] - a communication function 241 with station C2 10;

[0101] - a communication function 242 with the remote control radio 113 of said drone; and

[0102] - a 243 inter-drone communication function.

[0103] Figure 3 illustrates the main internal functions of the C2 10 station. The C2 10 station controls the fleet of drones (11a or 11b) when coordinating the mission, and also launches and terminates the mission. The C2 10 station has the following basic functions:

[0104] - send a takeoff order 31 1;

[0105] - send a landing order 312;

[0106] - send a stabilization order 313.

[0107] Following the execution of these functions, the commands which are calculated are sent to all drones 111 and 112 of the fleet in question (11 or 12), in particular at the time of the launch of the mission.

[0108] Next, station C2 10 includes a mission coordination definition function 315. The operator 101 of station C2 10 selects, using the mission coordination definition function 315, whether the mission is coordinated by station C2 10 or by the master drone 111.

[0109] In the event that operator 101 chooses that the mission be coordinated by post C2 10, the fleet in question is coordinated by the post and drones 111 and 112 then follow the orders coming from this post.

[0110] Conversely, if operator 101 chooses that the mission be coordinated by master drone 111, it is master drone 111 that coordinates the tasks and actions carried out by slave drones 1 12. Thus, station C2 10 does not intervene directly in the coordination of the mission.

[0111] At any time during the mission, the operator 101 can modify the state of the coordination using the mission's coordination definition function 315.

[0112] Station C2 10 also includes a 314 display function showing the position of drones 11 1 and 1 12 on a map displayed on a control screen. By extension, this drone position display function can also show other information such as the planned trajectories for drones 11 1 and 1 12, mission objectives, and the position of station C2 10 itself.

[0113] Whether the mission is coordinated by station C2 10 or master drone 111, station C2 10 also includes a function for viewing images and videos 333 acquired by the various drones. When the mission is coordinated by station C2 10, it is the device 102 integrated into station C2, as a computing and communication device, that executes the steps of the master drone selection process. In particular, by executing the selection function 324, it evaluates the weight w of each drone using a weight evaluation function 399. In this case, the master drone 111 has the specific function of relaying communications (in both directions) between station C2 and the drones of the fleet under consideration.

[0114] As with the master drone 111, the C2 station also includes a 325 drone fleet coordination function.

[0115] The image and video viewing function 333 provides operator 101 at station C2 10 with live visual feedback of the mission environment from drones 111 and 112 present in the area.

[0116] The images and videos transmitted to station C2 10 are analyzed using various image processing algorithms, these algorithms being used in functions such as a verification function 331 of a state or a site for example, an object detection function 332, a fire start detection function 335 or a threat detection function 334.

[0117] The results of functions 331, 332, 334 and 335 are displayed and superimposed on the images and videos obtained by the image and video viewing function 333. Advantageously, the operator 101 is thus assisted in the management of the mission and can inform people intervening in the area and coordinate them if necessary.

[0118] It should be noted that functions 331, 332, 333, 334 and 335 can only be executed when communication between station C2 10 and fleet 11 a or 11 b of drones is available, for example by means of a connection made on a Wi-Fi ®, 4G, 5G or satellite network.

[0119] To implement this communication, the C2 station 10 includes a communication function 341 with the drone fleet. In an alternative embodiment, the C2 station 10 is capable of receiving communications from third parties, such as at least one other C2 station acting as a relay, particularly for the purpose of coordinating a mission. Figure 4 shows a diagram of a method 400 for coordinating a mission assigned to the fleet 11 (as an example of a drone fleet).

[0120] The description is given in the case where the process is applied to the coordination of fleet 11, but it naturally applies to the coordination of any fleet of drones.

[0121] The 400 mission coordination process mainly comprises the following steps:

[0122] - 410 takeoff command for fleet 11;

[0123] - 420 for hovering positioning of drones in this fleet;

[0124] - 430 for input and transmission of mission parameters (objectives, area to monitor, etc.) by station C2;

[0125] - 440 mission coordination, via station C2 or 445 coordination via master drone 111;

[0126] - 450 manual resumption of drone control for drones 111 and 112;

[0127] - 455 drone stabilization control for drones 1, 11 and 112; and

[0128] - 460 drone landings at the end of the mission.

[0129] The 400 method for coordinating a mission also includes a 500 method for selecting the master drone 111, the steps of which are shown in Figure 5.

[0130] The steps of process 500 for the election of the master drone 111 are executed during process 400 for the coordination of a mission, whether the C2 10 station or the master drone 111 is responsible for coordinating the mission.

[0131] Step 410, the takeoff command for fleet 11 or 12, is executed when operator 101 at station C2 orders the mission to begin. During this step, the drones 112 in the fleet are informed via communication function 341 at station C2 that they are authorized to take off.

[0132] Step 420, positioning the drones of fleet 11a or 11b into a hover, is then carried out in order to stabilize said drones before the mission parameters or other orders from station C2 10 are transmitted to said drones.

[0133] Step 430, the entry and transmission of mission parameters (objectives, positioning of the area to be monitored, coverage of a specific area, object search, etc.) by station C2 10, involves operator 101 entering the parameters necessary for the mission to run smoothly. Once entered, these parameters are transmitted to fleet 11a or 11b. During step 430, operator 101 selects whether the mission is coordinated by station C2 10 or autonomously by the master drone 111. Following step 430, the selection of the master drone 111 is executed. Then, depending on the case, either step 440 of mission coordination by station C2, or step 445 of mission coordination by master drone 1 1 1 is carried out.

[0134] When the mission objectives (defined in step 430) have been achieved, step 460 of landing the drones at the end of the mission is executed, either under the control of operator 101 from station C2 or autonomously by fleet 1 1 under the control of master drone 1 1 1.

[0135] In an unrepresented embodiment, it is possible to execute step 430 of entering and transmitting the parameters of a new mission again if the first mission ended successfully.

[0136] Furthermore, in the event of a technical problem or unexpected occurrence, and in order to secure the fleet, a step 450 involving manual takeover of drone control by operator 101 and / or a step 455 involving drone stabilization by the master drone are executed. The manual takeover of drones 111 and 112 requires the intervention of at least one operator 101, who then uses the remote controls 113 to pilot each drone. Following the manual takeover of drones 111 and 112 and / or the stabilization of said drones, the mission parameters are redefined, or the mission is ordered to terminate.

[0137] Figure 5 shows a diagram of the main steps of the 500 method for selecting the master drone 111, which is implemented by the selection function 224 onboard each drone in the fleet 11 or by the selection function 324 of station C2. The 500 method for selecting the master drone 111 mainly comprises the following steps:

[0138] - 510 initialization of the master drone election 111;

[0139] - 520 master drone selection 111; and

[0140] - 530 reorganization of the 11a or 11b drone fleet by reassigning roles within said fleet.

[0141] Step 510 of the master drone election initialization 111 includes the following steps:

[0142] - 511 change of states 223 of the master drone 111 and the slave drones 112 of the fleet 11 a or 11 b into a state 223 called candidate drone.

[0143] - 512 identification of drones 111 and 112 of said fleet having their status 223 to that of candidate drone; and

[0144] - 513 for transmitting a status message.

[0145] The method 500 for electing the master drone 111 is triggered when an event 599 occurs, such as the cessation for a predetermined duration D of the sending of the status message 223 from the master drone 111, either to station C2, or to at least one slave drone 112. The cessation of sending status messages may follow a technical problem or an accident of the master drone.

[0146] For example, if Fleet 11 is following a vehicle on the ground, the duration D is approximately 3 seconds, whereas if Fleet 11 is statically observing a given site, the duration D is approximately 30 seconds. The duration D can vary. For example, it can depend on the distance Dc2 between Fleet 11 and station C2.

[0147] Following the occurrence of event 599, at stage 511, each drone in the fleet is placed in candidate drone status.

[0148] For security reasons, to ensure that the master drone selection process 500 is carried out only by drones in fleet 11 and to prevent intruder drones from participating, in identification step 512, drones in fleet 11 whose status is that of a candidate drone are identified. Indeed, in order to implement the master drone selection process 500, station C2 10, or each drone in fleet 11, must have knowledge of the exact list of drones in fleet 11. Drone identification step 512 is performed either by means of the drones' communication function 241 with station C2, or by the inter-drone communication function 243, implemented by the computing and communication device 200 onboard each drone. Each drone 112 is identified by means of a unique identifier within the fleet 11 such as a serial number or equivalent.

[0149] Once the drones 112 have been identified and recognized as part of the fleet 11, each drone transmits, during step 513 (status message transmission), either to station C2 10 or to the other drones in the fleet, one or more pieces of information that will be used to calculate the drone's weight during step 521 (drone weight calculation). This information may include the remaining battery capacity (expressed, for example, in kWh), the drone's position (latitude, longitude, altitude), etc. This information will then be used during step 520 (selection of the master drone 111).

[0150] Step 520, the selection of the master drone 111, is then completed. It mainly comprises the following steps:

[0151] - 521 evaluation of the weight of each drone, provided that it is in the state 223 of candidate drone;

[0152] - 522 comparison of weights w;

[0153] - 524 nomination of the elected drone;

[0154] - 525 meaning to the elected drone of the result of his election; and

[0155] - 526 validation of the election by the elected drone.

[0156] Step 521, which evaluates the weight w of each drone, is performed either by a drone using evaluation function 299, or by station C2 (via the computing and communication device 102) using evaluation function 399. When calculated by a drone, the weight w of a drone can be calculated by the drone itself. The drone then directly transmits its calculated weight to the fleet's computing and communication device for the execution of steps 522-526 of the process. In another implementation, the information used to calculate the drone's weight w is first transmitted either to station C2 or to the other drones in the fleet. In this case, the weight calculations for the individual drones are performed either by station C2's computing and communication device or by all the remaining drones in the fleet.

[0157] In one embodiment, the weight w of each drone 112 corresponds to the remaining capacity Cbat of its battery. In this case, the weight evaluation function 299 or 399 is as follows:

[0158] [Math. 1] w = C bat

[0159] In another embodiment, the weight w of each drone 112 is evaluated from the remaining capacity Cbat of its battery as well as the latitude / , longitude L and altitude z of the drone to calculate the distance Dc2 away from the post C2.

[0160] The weight evaluation function 299 or 399 can then be defined, for example, as follows:

[0161] [Math. 2] where P is a multiplier applied to the ratio between the remaining battery capacity Cbat and the distance Dc between each of the drones 112 and station C2. In other words, in this embodiment, we will seek to allocate a weight w to a drone that is higher the closer it is to station C2. The coefficient P for the fleet 11 can be adjusted according to the type of mission, thus allowing the values ​​of the parameters taken into account in the evaluation function 299 or 399 of the weight w to be weighted.

[0162] Preferably, as indicated, the weight evaluation function 299 or 399 takes as input parameters one or more technical characteristics of the drone at the given time, such as its computing power, one or more operating states of the drone's sensors, and / or one or more indicators of the communication quality between said drones and station C2 10, such as the signal-to-noise ratio of the communications, latency, etc. Once step 521, which evaluates the weight w of each drone 112, has been performed by each drone, the values ​​of the different calculated weights are compared. Step 522, which compares the weights, is performed either by each drone 112 implementing the master drone 111 selection process 500, or by station C2 if it is coordinating the mission.

[0163] Thus, depending on the situation, either station C2 or each drone 112 selects (i.e., determines) the drone with the highest weight among the drones 112 in the fleet. This determination constitutes step 524, the nomination of the selected drone. Then, in step 525, if station C2 10 is coordinating the mission, it directly informs the selected drone that it is the new master drone 111 of the fleet for the current mission.

[0164] Conversely, in the case where the mission is coordinated by the master drone or successive master drones, each of the 112 drones will then signal to the drone concerned that it obtained the highest weight during the 522 wet weight comparison step and therefore becomes the new master drone.

[0165] During all or part of the steps of the 500 process, the information sent is preferably encrypted in order to avoid any disruption of the drone fleet by third parties.

[0166] In the case where, during step 522 of comparing weights w, at least two drones have weights w of equal values, in certain embodiments, the method 500 of selecting the master drone 111 further includes a step 523 during which other representative values ​​of the different drones are compared, in order to be able to select the master drone.

[0167] For example, during comparison step 523, the battery charge levels (Cbat) of the tied drones are compared, and the drone with the highest Cbat charge level (112) is selected. This step 523 allows the master drone to be selected as the one with the greatest autonomy to complete its mission due to its high battery charge level, which, in principle, minimizes the number of times the master drone selection process (500) is implemented.

[0168] Finally, if the mission is carried out autonomously by the fleet, step 520, the selection of the master drone 111, concludes with step 526, the validation of the election by the chosen drone. During this step 526, a drone that is expected to become the chosen drone (insofar as it has itself determined that it should become the master drone) verifies that, during step 525, the notification to the chosen drone, it received a number of messages indicating that it should be the new master drone that is strictly greater than a predetermined fraction (for example, half) of the number of drones in the fleet that are in the candidate drone state.

[0169] If the election validation step 526 is not completed within a predetermined time (for example, five seconds), step 525, which notifies the elected drone of the election result, is performed again; and this is repeated up to a maximum number of iterations n. If, after n iterations, the election has still not been validated, the master drone election process 500 is executed again from step 512, which identifies the drones in the fleet whose status 223 is that of a candidate drone.

[0170] Advantageously, steps 525 of notifying the elected drone of its election and 526 of validating the election by the elected drone give the method 500 of electing the master drone 111 a high robustness.

[0171] Furthermore, by means of step 520 of master drone selection, the master drone is elected dynamically.

[0172] Once step 520, the master drone selection step, is completed, step 530, the fleet drone role update step, is executed. This step 530 includes the following steps:

[0173] - 531 change of state of the elected drone, which goes from the state of candidate drone to the state of master drone;

[0174] - 532 meaning of the identifier of the new master drone to the 112 drones that are in the candidate drone state; and

[0175] - 533 change of state of drones which are in the state of candidate drone, which pass into the state of slave drone.

[0176] Following step 530, which updates the drone roles, and then periodically in step 223 throughout the mission execution, the new master drone 111 transmits a status message to the slave drones 112, or, if applicable, to station C2, indicating that it is operational as the master drone. Advantageously, following steps 521 (evaluating the wet weight) and 522 (comparing the w weights), step 523 (comparing with one or more other criteria) allows for handling cases where at least two drones 112 have obtained equal w weights. Furthermore, step 532 (notifying the drones 112 that are in the candidate drone state) optimizes the computation time required for the master drone selection process 500, as this process does not require significant computing resources, especially compared to algorithms requiring logging systems, which consume a large amount of RAM.

Claims

DEMANDS 1. Method (500) for selecting a drone (111), called the master drone, to perform a particular function with respect to other drones within a fleet (11) of drones, the method being characterized in that it comprises a step (520) for selecting the master drone (111) comprising the following steps: - (521) evaluation of a weight w for each drone (1 12), - (522) comparison of weights w; and - (524) nomination of the master drone, elected on the basis of the comparison of the values ​​of the weights w.

2. Method according to claim 1, wherein at least one drone weight (112) is evaluated by means of a drone weight evaluation function (299, 399) (112), said function being a function of at least one parameter among representative values ​​of a remaining autonomy of a drone battery (Cbat), a distance between the drone and a point on the ground, an electrical power consumed by a computing device of the drone, an availability of at least one predetermined sensor of the drone, a quality of connection between the drone and a communication device, mounted on another drone or on the ground.

3. Method according to claim 1 or 2, wherein the step (520) of selecting the master drone (111) further includes a step (525) of notifying the elected drone of the result of its election, by each of the remaining drones in the fleet.

4. Method according to claim 3, wherein the step (520) of election of the master drone (11 1) further comprises a step (526) of validation of the election by the elected drone on the basis of the meanings received during the step (525) of notifying the elected drone of the result of its election.

5. Method according to claim 4, wherein the step (526) of validation of the election by the elected drone consists of verifying that the elected drone has received during the step (525) of notification to the elected drone, a number of votes greater than a predetermined proportion of the number of drones (112) in the fleet (11) likely to become master drone.

6. A method according to any one of claims 1 to 5, comprising an initialization step (510) preceding the master drone selection step (520), the initialization step (510) comprising a step (512) for identifying at least one drone (112) from the fleet (11) capable of becoming the master drone (111).

7. A method according to any one of claims 1 to 5 comprising an initialization step (510) preceding the master drone selection step (520), or a method according to claim 6, the initialization step (510) comprising a step (513) of transmitting a status message, either by each drone (112) in the fleet (11), or by each drone (112) in the fleet (11) capable of becoming a master drone, to at least one entity responsible for executing the weight evaluation step 521, the status message indicating at least one value among values ​​representative respectively of a remaining battery life of the drone (Cbat), a distance between the drone and a point on the ground, an electrical power consumed by a computing device of the drone, an availability of at least one predetermined sensor of the drone, a quality of connection between the drone and a communication device, onboard on another drone or on the ground.

8. A method (400) for managing a fleet (11a, 11b) of drones (111, 112), comprising the election of a master drone by an election method (500) according to any one of claims 1 to 7, and the performance of the particular function by the master drone in relation to the other drones in the fleet (11) of drones, the particular function including for example a calculation of control commands of other drones in the fleet of drones, and / or a transmission of messages, including commands, to other drones in the fleet of drones.

9. Management method according to claim 8, wherein the master drone periodically sends a status message indicating that the master drone is operational; and when at least one expected status message is not received, or when no status message is received for a predetermined period, the drone election method (500) according to any one of claims 1 to 7 is triggered.

10. Computer program product characterized in that it comprises a set of program code instructions which, when executed by at least one processor (102;200), implement a method (400;500) according to any one of claims 1 to 9.

11. Set of non-transient recording medium(s) readable by a computer on which is recorded a set of program code instructions which, when executed by at least one processor (102;200), implement a method (400;500) according to any one of claims 1 to 9.

12. An assembly comprising a fleet of drones (11a, 11b), the assembly comprising at least one computing device, said at least one computing device having at least one processor and a set of recording medium(s) on which is or are recorded a set of program code instructions; said at least one computing device being configured so that, when said program code instructions are executed by said at least one processor, said at less a computing device implements a method according to any one of claims 1 to 9.

Images