Secure delivery of goods via drones

Abstract

The present invention relates to the secure delivery of cargo by unmanned aerial vehicles (UAVs). A control system in a delivery system performs a method for delivering a payload by an unmanned aerial vehicle (UAV) without the presence of a recipient, even if the payload includes sensitive or fragile cargo. The method comprises the following steps: obtaining (21) a designated location for delivering the payload, obtaining (22) at least one payload vulnerability index for the payload, and obtaining (23) a collection time indicating a point in time when the payload is expected to be retrieved by the recipient. The method further comprises the following steps: selecting (24) a final delivery range at or near the designated location based on the at least one payload vulnerability index and the collection time, and operating (25) the UAV to deliver the payload to the final delivery range.

Description

Technical Field

[0001] The present disclosure relates generally to delivering cargo using drones, and in particular, to improving the security of cargo being delivered. Background Art

[0002] Advances in autonomous aircraft technology have opened up new possibilities in the field of cargo delivery systems. Unmanned aerial vehicles (UAVs), also known as drones, have been configured to carry and deliver various types of cargo. Such cargo can include, for example, packages that would otherwise be delivered in person from land-based delivery vehicles, such as cars and trucks.

[0003] A common problem with both land-based and UAV delivery is that the recipient may not be present when the shipment arrives at the delivery point. Even if the delivery is scheduled to coincide with the recipient's arrival at the delivery point, unforeseen events such as traffic jams may cause delays for the recipient. From a user's perspective, it is often desirable to order shipments without the recipient's presence at the time of delivery.

[0004] This problem is exacerbated when the cargo is perishable, valuable, confidential, or of a sensitive or fragile nature. While land-based delivery vehicles have the option of retaining the cargo on board and rerouting it for another delivery, this option is less practical and / or less attractive for UAVs given their limited energy and payload capabilities.

[0005] Prior art includes WO2019 / 055690, which addresses the challenges that may arise when delivering payloads by UAV to uncontrolled landing sites such as residential buildings. To this end, the UAV is configured to use delivery instructions combined with computer vision technology to select a delivery surface based on images from an onboard camera that meets the delivery instructions and has appropriate characteristics in terms of, for example, surface material, surface area, surface slope, distance to the delivery location, and distance to nearby objects that may hinder delivery. Summary of the Invention

[0006] It is an object to at least partially overcome one or more limitations of the prior art.

[0007] Another purpose is to enable UAV delivery without the recipient being present.

[0008] Another goal is to improve UAV delivery of sensitive or fragile goods.

[0009] Yet another objective is to ensure the integrity of cargo delivered by UAVs.

[0010] One or more of these objects, as well as further objects that may emerge from the description below, are achieved at least in part by a method for delivering a payload by a drone, a control system, a drone, and a computer-readable medium.

[0011] As disclosed herein, some aspects of delivering a payload by a UAV involve selecting a final delivery range at or near a designated location for delivering the payload based on at least one vulnerability index of the payload and based on a collection time indicating a point in time when the payload is expected to be retrieved by a recipient. Thus, the final delivery range can be tailored to one or more characteristics of the payload, such as ensuring the integrity of the payload, which may be relevant when the payload includes sensitive or fragile goods.

[0012] Other objects, features, aspects and advantages will appear from the following detailed description, from the appended claims as well as from the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Hereinafter, embodiments will be described in more detail with reference to the accompanying schematic drawings.

[0014] Figure 1 is a block diagram of an example UAV according to some embodiments.

[0015] Figure 2 is a view of a UAV approaching a designated location to deliver a payload carried by the UAV.

[0016] Figure 3A is a flow chart of an example delivery method for a UAV according to one embodiment, Figure 3B is Figure 3A A flowchart of an example process for selecting a final delivery range in a delivery method of Figure 3C is Figure 3A An example process of predicting environmental conditions in a delivery method.

[0017] Figure 4 is a block diagram of an example system configured to select a final delivery range, according to one embodiment.

[0018] Figure 5 is a flow chart of an example process for generating proof of delivery according to one embodiment.

[0019] Figure 6 is an overview of an example delivery system according to one embodiment.

[0020] Figure 7 is a schematic diagram of an example delivery system according to one embodiment.

[0021] Figure 8is a diagram of an example interactive chart generated according to one embodiment and enabling selection of a pickup or delivery location.

[0022] Figure 9 is a flow chart of an example scheduling method according to one embodiment.

[0023] Figure 10 is a schematic illustration of an example partitioning of a geographic area into sub-areas and locations of UAVs and idle points with respect to potential locations for pickup or delivery.

[0024] Figure 11A is a flowchart of an example process for generating a summarized cost matrix according to one embodiment, and Figure 11B is a flow chart of an example process for calculating an individual cost matrix, according to one embodiment.

[0025] Figure 12A is a step-by-step illustration of an example process for calculating an individual cost matrix according to one embodiment, Figure 12B Instantiates a delivery task with an intermediate destination, and Figure 12C Illustrated for Figure 12B An example of a separate cost matrix calculated for a delivery task in .

[0026] Figure 13 is a conceptual diagram of the alignment of sub-regions in separate cost matrices.

[0027] Figure 14 is a block diagram of an example scheduling module according to one embodiment.

[0028] FIG. 15A to FIG. 15B 2. Side views of an example UAV launch system in a default state and a launch state, respectively, according to one embodiment.

[0029] Figure 16 is a side view of an example UAV launch system according to another embodiment.

[0030] 17A to 17B Side views of an example UAV launch system before and after launch, respectively, according to another embodiment.

[0031] Figures 18 and 19 is a side view of an example UAV launch system according to some other embodiments.

[0032] Figure 20 is a block diagram of an example emission control system according to one embodiment.

[0033] Figure 21 is a flow chart of an example UAV transmission method according to one embodiment.

[0034] FIG. 22A to FIG. 22B Figures 2 and 3 are side and top views of an example UAV flight path from the launch system location to the destination.

[0035] Figure 23A is a diagram of example clusters generated for multiple destination locations of a delivery system, and Figures 23B to 23C It is a flowchart of a method of configuring a delivery system.

[0036] Figure 24 is a schematic diagram of an example delivery system.

[0037] Figure 25A is a schematic illustration of a UAV included in an example delivery system, and Figures 25B to 25C is an overhead view of an example UAV formation for a mission to deliver and / or pick up cargo.

[0038] Figure 26 is a flow chart of an example method of controlling a group of UAVs, according to one embodiment.

[0039] FIG. 27A to FIG. 27B is a top view of an example UAV group on a mission to deliver cargo, illustrating an example arrangement of UAVs based on air resistance.

[0040] Figure 28 is a flow chart of an example process for determining the relative position of a UAV.

[0041] Figures 29A to 29C is a flow chart of an example process for determining the air drag of a UAV.

[0042] Figure 30A is a side view of the forces acting on the UAV in transit, and Figure 30B An example definition of the orientation parameters of a UAV is shown.

[0043] Figures 31A to 31B is a top view of an example UAV group having a master UAV and illustrating example data paths within the UAV group, and Figure 31C is a top view of an example UAV group without a master UAV, and illustrates the data exchange and resulting position transformation.

[0044] Figure 32 is a schematic diagram of a UAV group with respect to UAVs joining and leaving the UAV group.

[0045] Figures 33A to 33C An example delivery system configured in accordance with embodiments is illustrated.

[0046] Figures 34A to 34Dis a flow chart of an example method for configuring a delivery system for delivering cargo using a UAV, according to one embodiment.

[0047] Figures 35A to 35E is an overview of an example configuration method according to one embodiment.

[0048] Figure 36 is a block diagram of an example control system for a delivery system according to one embodiment.

[0049] Figure 37 is a block diagram of a machine that can implement the methods, processes, and functions described herein. DETAILED DESCRIPTION

[0050] Embodiments are described more fully hereinafter with reference to the accompanying drawings, which illustrate some, but not all, embodiments of the invention. Indeed, the subject matter of the present disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present disclosure may satisfy applicable legal requirements.

[0051] Furthermore, it should be understood that, where possible, any of the advantages, features, functions, devices and / or operational aspects of any embodiment described and / or contemplated herein may be included in any other embodiment described and / or contemplated herein, and vice versa. In addition, where possible, any term expressed in the singular herein is intended to include the plural form, and vice versa, unless otherwise expressly provided. As used herein, "at least one" should mean "one or more" and these phrases are interchangeable. Therefore, even if the phrases "one or more" or "at least one" are also used herein, the terms "one" and / or "an" should also mean "at least one" or "one or more". As used herein, unless the context requires otherwise due to clear language or necessary meaning, the word "include" or variations such as "comprising" are used in an inclusive sense, that is, to specify the presence of specified features, but not to exclude the presence or addition of other features in various embodiments. The term "compute" and its derivatives are used in their conventional sense and can be considered to involve, for example, performing a calculation involving one or more mathematical operations using a computer to generate a result.

[0052] As used herein, the terms "plurality," "plurality," and "multiple" are intended to imply the presence of two or more elements, while the term "set" of elements is intended to imply the presence of one or more elements. The term "and / or" includes any and all combinations of one or more of the associated listed elements.

[0053] It should also be understood that, although the terms first, second, etc. may be used to describe various elements in this article, these elements should not be limited to these terms. These terms are only used to distinguish one element from another element. For example, without departing from the scope of this disclosure, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element.

[0054] For the sake of brevity and / or clarity, well-known functions or constructions may not be described in detail. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

[0055] Like numbers refer to like elements throughout.

[0056] Before describing the embodiments in more detail, a few definitions are given.

[0057] As used herein, an "unmanned aerial vehicle" (UAV) refers to an aircraft without a crew on board. The UAV can be controlled by an onboard automated control system, a ground control system, or by a ground crew. Such aircraft are also referred to as unmanned aircraft or "drones." Other synonyms or variations encompassed by the term UAV as used herein include RPAS (remotely piloted aircraft systems), UAS (unmanned aircraft systems), MAV (micro air vehicles), and sUAS (small unmanned aerial vehicle systems). Examples of UAVs include multi-rotor drones (such as tri-rotors, quad-rotors, hexacopters, octocopters, etc.), fixed-wing drones, single-rotor helicopter drones, and fixed-wing hybrid VTOL (vertical take-off and landing) drones. The UAV may, but need not, include a propulsion system.

[0058] As used herein, "payload" refers to any load carried by a vehicle, excluding loads required for the operation of the vehicle. A payload may include any type of item to be delivered by the vehicle, such as to a final destination or an intermediate storage point. For example, a payload may include cargo or merchandise, optionally contained in one or more parcels, packages, containers, bags, and the like.

[0059] Embodiments relate to various aspects of systems and methods that rely, at least in part, on UAVs for cargo delivery. The following description is divided into sections 1 through 5 to discuss the inventive concepts of UAV-based delivery. This division into sections is for clarity and does not imply that these concepts cannot be combined. On the contrary, as those skilled in the art will appreciate, the inventive concepts and embodiments can be combined in various configurations to achieve corresponding synergistic effects.

[0060] Generally, a UAV for use in a delivery system can be considered to include a body or structure, a control system, and a mechanism for holding a payload. The control system includes logic configured to control the operation of the UAV. This logic can be implemented in hardware, software, or a combination of hardware and software.

[0061] Figure 1 Detailed example of a UAV is depicted. As shown, the UAV may include: a propulsion system 1002, control surfaces 1003, a delivery mechanism 1004, a flight controller 1005, a peripheral interface 1006, one or more processors 1007, a memory controller 1008, memory 1009, a power system 1010, an external port 1011, a GNSS module 1012, a communication interface 1013, an audio circuit 1014, a proximity sensor 1015, an inertial measurement unit (IMU) 1016, an imaging device controller 1017 and one or more associated imaging devices 1018, and any other input controllers 1019 and input devices 1020. One or more of these components may communicate via one or more communication buses or signal lines, such as Figure 1 As shown by the arrow in .

[0062] Figure 1 is only one example of a UAV system. Thus, a UAV may include Figure 1 More or fewer components may be shown, two or more components may be combined into a functional unit, or the components may be arranged differently. The present invention may be implemented in hardware, software, or a combination of hardware and software (including one or more signal processing and / or application specific integrated circuits). Figure 1 Some of the various components shown.

[0063] In some embodiments, the propulsion system 1002 may include one or more fixed-pitch rotors, variable-pitch rotors, variable-pitch jet engines, or any other propulsion mode that provides thrust. The propulsion system 1002 may be operable to vary the applied thrust, for example, via an electronic speed controller that varies the speed of each rotor. Furthermore, according to some embodiments, the propulsion system 1002 may include a propulsion subsystem dedicated to the delivery mechanism 1104 to enable the payload to be maneuvered independently of the main body of the UAV.

[0064] In some embodiments, the UAV may include control surfaces 1003, such as flaps, rudders, ailerons, etc. The control surfaces 1003 may include regulating components such as electric motors, cable pulley systems, or hydraulic systems.

[0065] The delivery mechanism 1004 may include the payload holder described above, which may include a coupler operable to secure and release the payload.

[0066] The flight controller 1005 (sometimes referred to as a "navigation system" or "autopilot") may include a combination of hardware and / or software configured to receive input data (e.g., sensor data from the imaging device 1018 and / or the proximity sensor 1015), interpret the input data, and output control signals to the propulsion system 1002, the control surfaces 1003, and / or the delivery mechanism 1004. Alternatively or in addition, the flight controller 1005 may be configured to receive control commands generated by another component or device (e.g., the processor 1007 and / or a remote computing device), interpret those control commands, and output control signals for the propulsion system 1002, the control surfaces 1003, and / or the delivery mechanism 1004.

[0067] A peripheral device interface (I / F) 1006 can connect input and output peripheral devices to the processor 1007 and memory 1009. The one or more processors 1007 can run or execute various software programs and / or instruction sets stored in the memory 1009 to perform various functions of the UAV and process data. In some embodiments, the processor 1007 can include a general-purpose central processing unit (CPU), a special-purpose processing unit (e.g., a graphics processing unit (GPU) suitable for parallel processing applications), or any combination thereof. The memory 1009 can include one or more computer-readable storage media (such as high-speed random access memory) and / or non-volatile memory (such as one or more disk storage devices, flash memory devices, or other non-volatile solid-state memory devices). Access to the memory 1009 by other components such as the processor 1007 and the peripheral device I / F 1006 can be controlled by a memory controller 1008. In some embodiments, the peripheral device I / F 1006, the processor 1007, and the memory controller 1008 can be implemented on a single integrated chip (represented by a solid rectangle). In other implementations, one or more of the peripheral device I / F 1006, the memory controller 1008, and the processor 1009 may be implemented on a separate chip.

[0068] In some embodiments, the UAV includes a power system 1010 for powering various components of the UAV. The power system 1310 may include a power management system, one or more local power sources (such as batteries, fuel cells, etc.), a recharging system, power failure detection circuitry, a power converter or inverter, a power status indicator, and any other components associated with the generation, management, and distribution of power in a computerized device.

[0069] In some embodiments, a communication module (not shown) can facilitate wired communication with other devices through one or more external ports 1011. The external port 1011 (e.g., USB, FIREWIRE, etc.) can be suitable for direct connection to other devices, or indirectly connected to other devices through a network (e.g., the Internet, wireless LAN, etc.).

[0070] In some embodiments, the UAV may include a receiver such as a Global Navigation Satellite System (GNSS) receiver 1012 for obtaining the position of the UAV. The GNSS receiver 1012 may be configured to communicate with any satellite navigation system such as GPS, GLONASS, Galileo, or BeiDou. Figure 1 A GNSS receiver 1012 is shown coupled to the peripherals I / F 1006. Alternatively, the GNSS receiver 1012 may be coupled to the input controller 1019. The GNSS receiver 1012 may determine the current global position of the UAV based on satellite signals.

[0071] The communication interface (I / F) 1013 can facilitate the transmission and reception of communication signals, for example, in the form of electromagnetic signals. The transmission and reception of communication signals can be performed wirelessly, for example, via a radio frequency (RF) transceiver. In some embodiments, the communication I / F 1013 may include RF circuitry. In such embodiments, the RF circuitry can convert electrical signals into / from electromagnetic signals and communicate with a communication network and other communication devices via the electromagnetic signals. The RF circuitry can include well-known circuitry for transmitting and receiving communication signals, including, but not limited to, an antenna system, an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, a codec chipset, a subscriber identity module (SIM) card, memory, and the like. The RF circuitry can facilitate the transmission and reception of data over a communication network, including public networks, private networks, local networks, and wide area networks. For example, wireless communication can be performed over a wide area network (WAN), a cellular telephone network, a local area network (LAN), or any other wireless communication mode. Wireless communications may use any of a variety of communication standards, protocols, and technologies, including but not limited to GSM, EDGE, HSDPA, W-CDMA, CDMA, TDMA, LTE, 5G, Bluetooth, VoIP, Wi-MAX, or any other suitable communication protocol.

[0072] The audio circuit 1014 may include one or more of a speaker and a microphone and may provide an audio interface between the surrounding environment and the UAV. The audio circuit 1014 may receive audio data from the peripheral device I / F 1006, convert the audio data into electrical signals, and send the electrical signals to the speaker. The speaker may convert the electrical signals into sound waves audible to humans. The audio circuit 1014 may also receive electrical signals converted from sound waves by the microphone. The audio circuit 1014 may convert the electrical signals into audio data and send the audio data to the peripheral device I / F 1006 for processing. The audio data may be retrieved by the peripheral device I / F 1006 from the memory 1009 and / or the communication I / F 1013 and / or sent to the memory 1009 and / or the communication I / F 1013.

[0073] The UAV may include one or more proximity sensors 1015 . Figure 1 A proximity sensor 1012 is shown coupled to the peripheral device I / F 1006. Alternatively, a proximity sensor 1015 may be coupled to the input controller 1019. The proximity sensor 1015 may generally include long-range sensing technologies for proximity detection, distance measurement, target identification, and the like. For example, the proximity sensor 1015 may include a radio-based sensor (e.g., radar), an audio-based sensor (e.g., sonar), an optical-based sensor (e.g., LIDAR), and the like.

[0074] The UAV may include one or more IMUs 1016. The IMU 1016 may measure and report one or more of velocity, acceleration, orientation, and gravity. The IMU 1016 may include one or more subcomponents such as a gyroscope, an accelerometer, a magnetometer, etc. Figure 1 The IMU 1016 is shown coupled to the peripherals I / F 1006, but may alternatively be coupled to the input controller 1019. Furthermore, the IMU 1016 may be supplemented or replaced by a separate gyroscope, accelerometer, magnetometer, or the like.

[0075] The UAV may include at least one imaging device 1018. Figure 1In the example of , the imaging device 1018 is coupled to the imaging device controller 1017 in the I / O subsystem (represented by a solid rectangle). The corresponding imaging device 1018 may include imaging optics and an optical sensor responsive to electromagnetic radiation. The corresponding imaging device 1018 may be responsive to electromagnetic radiation within any wavelength range, including but not limited to ultraviolet, visible, or infrared radiation, or any portion or combination of these. The optical sensor may include a digital image sensor, such as a CCD or CMOS sensor. As used herein, the combination of the optical sensor and the imaging optics may be referred to as a "camera." In conjunction with the imaging module located in the memory 1009, the imaging device 1018 may capture digital images (including still images and / or video), which may be monochrome or multi-color. In some embodiments, the imaging device 1018 may include a single fixed camera. In some embodiments, the imaging device 1018 may include a camera with an adjustable orientation, for example, by using a gimbal mechanism with one or more axes of motion. In some embodiments, the imaging device 1018 may include a camera with a wide-angle lens that provides a wider field of view. In some embodiments, the imaging device 1018 may include an array of multiple cameras that provide up to a full 360-degree view in all directions. In some embodiments, the imaging device 1018 may include two or more cameras placed in close proximity to each other to provide stereoscopic vision. In some embodiments, the imaging device 1018 may include multiple cameras in any combination as described above. In some embodiments, the UAV may include one or more cameras dedicated to image capture for navigation, such as using visual inertial odometry (VIO).

[0076] In some embodiments, the input controller 1019 can receive / send electrical signals from / to one or more other input or control devices 1020, such as physical buttons (e.g., push buttons, rocker buttons, etc.), dials, touch screen displays, slider switches, joysticks, click wheels, etc. The touch screen display can be used to implement virtual or soft buttons and one or more soft keyboards, and can provide an input interface and an output interface between the UAV and the user. In some embodiments, the other control devices 1020 include a code reader that detects a machine-readable code from a payload carried by the UAV. The machine-readable code can include a series of characters, a barcode, a 2D code, etc.

[0077] In some embodiments, the software components stored in memory 1009 may include an operating system, and a collection of modules or applications (such as a communication module, a flight control module, a navigation module, a computer vision (CV) module, or a graphics module). For clarity, Figure 1Such modules are not shown. An operating system (e.g., Darwin, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks) may include various software components and / or drivers for controlling and managing routine system tasks (e.g., memory management, storage device control, power management, etc.), and facilitate communication between various hardware and software components. A graphics module may include various software components for processing, rendering, and displaying graphical data. As used herein, the term "graphics" may include any object that can be displayed to a user, including but not limited to text, static images, videos, animations, icons, etc. The graphics module, in conjunction with a GPU, may process graphical data captured by the imaging device 1018 and / or the proximity sensor 1015 in real time or near real time. A CV module, which may be a component of the graphics module, may provide analysis and recognition of graphical data. For example, when the UAV is in flight, the CV module, together with the graphics module (if separate), the processor 1007, and the imaging device 1018 and / or the proximity sensor 1015, may identify and track captured images of one or more objects in a natural environment. The CV module can also communicate with the navigation module and / or the flight control module to update the relative position between the UAV and a reference point (for example, a target object such as a payload, a delivery surface, a landing point, etc.). The navigation module can determine the position and / or orientation of the UAV relative to other objects and provide this information for use by various modules and applications, for example, the flight control module generates commands for use by the flight controller 1005. Each of the modules and applications identified above can correspond to an instruction set that performs one or more functions described above. The described modules need not be implemented as separate software programs, processes, or modules, and thus, in various embodiments, various subsets of the depicted modules can be combined or otherwise rearranged. In some embodiments, the memory 1009 can store a subset of the modules or data structures identified above. Furthermore, the memory 1009 can store additional modules and data structures not described above.

[0078] 1. Secure delivery of goods via UAVs

[0079] This portion of the disclosure relates to enabling the delivery of items via UAVs without requiring the consignee (e.g., a user who ordered the goods to be delivered) to be present when the item arrives at a designated delivery point. This portion of the disclosure relates particularly, but not exclusively, to the delivery of items that are sensitive in one or more respects. In one example, the item may have a high monetary value and / or sentimental value. In another example, the item may be confidential. In yet another example, the item may be perishable and, for example, need to be stored at a specific ambient temperature, at least for a period of time. Examples of such perishable items include various foods (which may be raw or processed) and medications.

[0080] While drone deliveries can often be pre-scheduled to ensure the consignee is present to receive the item, this may be found to be limited to consignees who need to adhere to an agreed-upon schedule. Furthermore, unforeseen events such as traffic jams may prevent the consignee from appearing in time for the delivery. In conventional road deliveries (e.g., by van or truck), one may choose to keep the item on board and reroute it for another delivery. For UAV delivery systems, this option is less practical given that UAVs typically have limited carrying capacity and are propelled by local power sources of limited capacity.

[0081] Figure 2 The UAV 1 is shown in a transport for delivering a payload 4. The UAV 1 includes a control system 2, a delivery mechanism 3 (see Figure 1 1004) and imaging device 5 (refer to Figure 1 1018). The delivery mechanism 3 is releasably attached to the payload 4 to be delivered to Figure 2 , a destination indicated by an X. In the example shown, the destination comprises a residential building 10 having a backyard enclosed by a fence 11. A car 12 is parked in the driveway of the building 10.

[0082] Figure 3A An example delivery method 20 according to some embodiments is illustrated. The method 20 may be performed by a control system of a UAV and will be referred to as Figure 2. Step 21 obtains a designated delivery location for the payload 4. The designated delivery location DDL may be given in any suitable format (e.g., GNSS coordinates, street address, proprietary map coordinates, etc.). Step 22 obtains at least one payload vulnerability index VID. The VID may be predefined for the payload 4 and indicate the sensitivity or vulnerability of the payload 4 to general or specific environmental conditions. In some embodiments, the VID may represent the monetary value or sentimental value or confidentiality of the payload 4 and thereby indicate susceptibility to theft. In some embodiments, the VID may represent temperature sensitivity, for example, indicating the vulnerability of the payload 4 to temperatures that are too high or too low relative to the nominal temperature. In some embodiments, the VID may represent an acceptable range of storage temperatures. In some embodiments, the VID may represent the moisture sensitivity of the packaging surrounding the item to be delivered or the item itself or both. In some embodiments, the VID may represent wind sensitivity, for example, indicating the risk that the payload 4 will be blown away or damaged by the wind.

[0083] Step 23 obtains a consignee pickup time RPT, at which the consignee ("consignee") expects to retrieve the payload 4 from the DDL. The RPT may be different from, and typically is different from, the point in time when the UAV is scheduled / estimated to deliver the payload at the DDL ("delivery time"). The UAV 1 may perform one or more of steps 21 to 23 in communication with one or more external computing resources (not shown), which may provide the UAV 1 with one or more of the DDL, VID, or RPT. Alternatively or in addition, the UAV 1 may obtain one or more of the DDL, VID, or RPT by detecting and interpreting one or more machine-readable codes on the payload 4. In step 24, a final delivery range FDA is selected at or near the DDL based on the VID and RPT. In step 25, the method 20 may be inferred by operating the UAV 1 to deliver the payload 4 to the FDA (e.g., by releasing the delivery mechanism 3).

[0084] In some embodiments, the FDA is selected to mitigate the vulnerability of the payload 4 while it is at or near the DDL during the period from delivery to the RPT. Step 24 generally facilitates the delivery of the payload by the UAV and opens up the possibility that the FDA will be customized to the respective payload to be delivered and the integrity of the respective payload will be ensured until the payload is expected to be collected by the consignee. It will be understood that while the DDL can be a specific location, the FDA is generally spaced apart from the DDL. In some embodiments, the UAV can store a range definition that specifies the boundaries of a search range within which the FDA can be selected. For example, the search range can be specified as a predefined distance from the DDL or a defined boundary relative to the DDL. Figure 2 In the example, you can limit the search to real estate that includes DDL, for example, Figure 2 In some embodiments, the UAV 1 autonomously determines the limits of the search range, for example, by performing computer vision analysis on images taken by the imaging device 5 around the DDL.

[0085] In some embodiments, the FDA can be selected from a set of candidate ranges, which can be predefined for the DDL or dynamically identified by the UAV during delivery at the DDL. The predefined candidate ranges can be determined by computer analysis of local terrain, imagery, etc., by or for the external computing resources mentioned above, and sent to the UAV 1. The imagery can be captured, for example, by a satellite and / or by one or more other UAVs, which may or may not be part of the delivery system. Figure 2 Examples of candidate ranges are shown in FIG and designated by 13A through 13H. Note that candidate range 13G is within a dedicated secure container 14, which is located on the roof of building 10 and may or may not be closed or even locked. In some embodiments, UAV 1 is operable to remotely open secure container 14, for example, by transmitting a security code to the control system of secure container 14. Candidate range 13H is a secure space associated with car 12 on the roadway, such as the trunk of the car or an attached secure container, which UAV 1 can similarly approach.

[0086] like Figure 3AAs shown, step 24 may include step 24A of predicting the environmental conditions of the candidate delivery range set. In some embodiments, the environmental conditions are predicted for the dwell period extending from the delivery time to the RPT, and the FDA is selected from the candidate delivery range set based on the predicted environmental conditions and the VID. The predicted environmental conditions may be related to the VID, and vice versa. For example, the VID may indicate the vulnerability of the payload 4 under the environmental conditions predicted by step 24A. In some embodiments, the environmental conditions may include sunlight. Such environmental conditions may be related to a VID indicating temperature sensitivity. In some embodiments, the environmental conditions may include rainfall, dew or other moisture. Such environmental conditions may be related to a VID indicating moisture sensitivity. In some embodiments, the environmental conditions may include wind, drafts, etc. Such environmental conditions may be related to a VID indicating wind sensitivity.

[0087] Some non-limiting examples of what to consider when selecting an FDA for delivering an item include: finding an area of ​​shade that can be maintained for a certain period of time, taking into account the RPT, to keep the item cool; finding an area of ​​direct sun that can be maintained for a certain period of time, taking into account the RPT, to keep the item warm; finding an interfering surface; finding an area that is specified based on a photo of a part of the recipient's residence (e.g., a part of a garden) or text describing the location of the recipient's residence; finding an area that is immediately behind a fence or wall, between two large structures, or on an elevated structure (such as a rooftop or balcony) to keep the item concealed and / or inaccessible to non-recipients; or identifying in-vehicle options, such as the trunk or special compartment of a car owned by the recipient.

[0088] Figure 3B is a flow chart of an example process for selecting an FDA. As shown, the example process may be Figure 3A . This process assumes that a candidate range, denoted below by CAD0, is nominally selected and may include all available candidate ranges. In step 30, the VID is evaluated to determine whether the payload 4 is of high value, for example in terms of monetary, sentimental or confidentiality. If so, step 31 determines a first set of candidate ranges CAD1 that match the VID by reducing the risk of the payload 4 being stolen, for example, out of sight ranges ("overlooked ranges"). Figure 2In the example, CAD1 may include all candidate ranges except 13A. In some embodiments, the VID may indicate different categories of value, and CAD1 may be different depending on the category. For example, for the most valuable payload, CAD1 may consist of 13G and 13H. With respect to 13H, step 31 may include verifying that the car 12 belongs to the consignee based on, for example, the license plate of the car 12. The process proceeds to step 32, which evaluates the VID to determine if the payload 4 is moisture sensitive. If so, step 33 predicts the risk of rainfall at the DDL during the dwell period. Thus, step 33 may be Figure 3A Part of step 24A in . Step 33 may comprise: predicting the risk of rainfall for the available candidate ranges 13A to 13H or a subset thereof. The prediction in step 33 may be based on weather data of the DDL, for example a weather forecast for the dwell period or a portion of the dwell period. Step 33 may take into account not only rainfall but also the local topography (see 3D model below) and / or the estimated wind direction for the candidate ranges. For example, candidate range 13E may be considered to be rainproof given that it is located under a tree. Based on step 33, step 34 determines a second selected candidate range CAD2 which is considered to reduce the risk of the payload 4 getting wet. In Figure 2 In the example shown, CAD2 may include 13E, 13G, 13H, and, depending on wind direction, one or more of 13B through 13D. The process proceeds to step 35, where the VID is evaluated to determine whether payload 4 is temperature sensitive. If so, step 36 predicts the sunlight conditions for the available candidate range 13A through 13H, or a subset thereof.

[0089] Figure 3C An example process that can be performed by step 36 is shown. Step 40 obtains a three-dimensional (3D) representation of the DDL. The 3D representation can be a 3D model of the area that includes the DDL. The scope of the area can be defined to include any structure that casts a shadow on the available candidate range at the DDL. Such structures can include surrounding buildings and naturally formed terrain. Step 41 obtains weather data for the DDL or area, such as the weather forecast mentioned above. Then, step 42 predicts the sunshine conditions in the 3D model (e.g., in the available candidate area or a subset thereof) based on the weather data and the movement of the sun during the dwell period. The weather data can be used to estimate cloud cover and thus estimate the possibility of sunshine.

[0090] Back to Figure 3BStep 37 determines a third selection of candidate ranges, CAD3, that are considered compatible with the temperature requirements of payload 4 as indicated by VID. If the payload is sensitive to high temperatures, CAD3 may include candidate ranges with little or no sunlight. If the payload is sensitive to low temperatures, CAD3 may include candidate ranges with very long sunlight exposure. The process then proceeds to step 38, which selects FDA from among the selections CAD0 to CAD3 for the ranges already determined. For example, step 38 may select FDA from the intersection of the determined selections CAD0 to CAD3, i.e., the candidate range included in all the determined selections.

[0091] like Figure 3B As shown, the FDA can also be selected based on input data 219, which can be provided by the consignee, for example, when placing a delivery order. The input data 219 can indicate a nominal selection of one or more candidate ranges in the CAD0. For example, the input data 219 can include one or more photos of the suggested candidate ranges and / or a textual description of the candidate ranges.

[0092] In some embodiments, step 24 may include obtaining a signal from a sensor device onboard the UAV or one or more other UAVs (see Figure 1 1018, 1020 and Figure 2 5) obtain sensor data, and step 24A can predict the environmental conditions at the DDL based on the sensor data. Such sensor data may include ambient humidity from a humidity sensor, wind data (speed and / or direction) from a wind sensor, images from two or more cameras or stereo cameras for determining 3D terrain, images from a thermal imaging camera for determining temperature or sunshine, and images from any type of camera for identifying shadows and / or ground moisture. Such images are non-limiting examples of 2D representations that can be obtained near the DDL for use in predicting environmental conditions. In some embodiments, the sensor data can supplement or even replace the weather data mentioned above. In some embodiments, the sensor data can provide or supplement the 3D representation mentioned above. In some embodiments, the sensor data can enable prediction of sunshine during the stay period without the need for further weather data.

[0093] Figure 4is a block diagram of an example system for determining FDA. The system includes a computing device 50 that is configured to obtain RPT, DDL, and VID from one or more first modules 51 (one first module is shown) and calculate FDA according to any of the methods and processes described herein. In the example shown, the computing device 50 is also configured to obtain a 3D representation of the DDL from one or more second modules 52 (see step 40) and weather data from one or more third modules 53 (see step 41). As understood above, the computing device 50 can be included in the UAV 1, and the first module 51 can be included in one or more external computing resources. Alternatively, the first module 52 can be positioned on the UAV 1 and connected to a code reader. Similarly, the second module 52 and the third module 53 can be included in one or more external computing resources and / or included in the UAV 1 and connected to one or more sensor devices.

[0094] It is also contemplated that the computing device 50 may be included in an external computing resource that is configured to transmit the FDA to the UAV. For example, such an external computing resource may execute Figure 3A or performing at least steps 21 to 24.

[0095] Figure 3A The method 20 in further includes the following optional step 26: generating proof of delivery when the payload has been delivered to the FDA. Step 26 may include: operating the imaging device 5 on the UAV 1 to capture one or more images (e.g., in the form of still images or videos) that visually include the payload 4 at the FDA and providing the image as proof of delivery. In some embodiments, the proof of delivery may also include the sensor data and / or prediction data related to the payload 4 mentioned above. In a specific example, the proof of delivery may state "The package has been delivered. The current temperature is 7°C and is expected to be 9°C upon collection with a confidence level of 88%."

[0096] In some embodiments, an image of the evidence of delivery can be taken to meet at least one predefined image standard. Such an image standard can specify one or more objectively identifiable characteristics or visual features to be included in the image. The visual feature can broadly represent the DDL or the recipient. One visual feature can be an identifier associated with the recipient's address, such as a trademark, a building or apartment sign, a number symbol, etc. Another visual feature can be an identifier associated with an object owned by the recipient, such as the license plate of the recipient's car 12. Still another visual feature can be provided and requested by the recipient, such as a specific window, balcony, door or other feature of the recipient's residence. In one example, the image standard includes a photograph provided by the recipient that roughly matches the image provided with the evidence of delivery. In some embodiments, the UAV 1 can be configured to automatically detect the presence of visual features in the image, for example by using vision-based machine learning for object detection.

[0097] Figure 5 is a flow chart of an example process for providing proof of delivery by operating UAV 1. As shown, the example process may be Figure 3A Part of step 26 in the reference. Figure 2In an example, step 60 may include operating the imaging device 5 on the UAV 1 to generate an image of the FDA. Step 61 evaluates whether the at least one image criterion is met. If not, step 62 moves the UAV 1 relative to the FDA and / or moves the imaging device 5 relative to the UAV, whereupon step 60 is repeated. In one example, steps 60 to 62 may cause the UAV 1 to hover and / or move in / out relative to the payload 4 to adjust its position until the at least one image criterion is met. When the criterion is met, step 63 may transmit the image from the UAV 1 in real time or substantially real time for display to the recipient. At this time, the recipient may be given the opportunity to accept or reject the proposed delivery, as detected by step 64. If the delivery is accepted, step 64 proceeds to step 65, which operates the UAV 1 to continue its mission, e.g., to deliver another payload. Otherwise, step 64 may proceed to step 66, which may grant the recipient access to the UAV 1's imaging device 5 and / or flight controls (see 1005), for example, through a dedicated control interface, while the UAV 1 continues to provide real-time imagery for display to the recipient. This control interface may also enable other interactions with or by the recipient, as described herein. For example, this control interface may be provided on a mobile device such as a smartphone. Step 66 may restrict the recipient's access, for example, by limiting the movement of the UAV 1 to a predefined airspace around the DDL. Step 66 may give the recipient the option of selecting a new FDA from a range of candidates and / or another pre-calculated location, whereupon the UAV 1 places the payload 4 at the new FDA. If the recipient then accepts the placement of the payload 4 at the new FDA, as detected by step 67, the UAV may continue its mission (step 65). Otherwise, step 68 may operate the UAV 1 to retrieve the payload 4 and move it to a storage location.

[0098] It is conceivable that, before step 25, instead of executing Figure 5 The example process in FIG. 5 is to give the recipient an opportunity to accept or reject the FDA before placing the payload 4 thereon. In the event of rejection in steps 64 or 67, this would eliminate the need for the UAV 1 to retrieve the payload 4.

[0099] Figure 6is a schematic overview of an example delivery system according to an embodiment. The delivery system includes: a fleet of UAVs 1 (one shown); and a server 70 configured to communicate with the UAVs 1 in the fleet. It should be understood that the external computing resources mentioned above may include the server 70. In the illustrated example, the server 70 is also configured to communicate with an electronic user device 71 (one shown), which can be operated by the recipient mentioned above to place a delivery order using the delivery system (optionally, via a merchant or any other goods provider). For example, the user device 71 can be operated to enter the delivery data and route prediction (RPT) for the delivery order. The user device 71 can also be operated to receive and display the FDA image provided to the server 70 from the corresponding UAV 1 in step 63. The user device 71 can also be operated to control the corresponding UAV 1 via the server 70 in accordance with step 66, for example, to change the direction of the field of view 5' of the imaging device 5 and / or to change the position and / or orientation of the UAV 1.

[0100] Below, several clauses are cited to summarize some aspects and embodiments disclosed above.

[0101] Clause 1: A method for delivering a payload (4) by an unmanned aerial vehicle (UAV) (1), the method comprising the following steps:

[0102] obtaining (21) a designated location for delivering said payload (4),

[0103] obtaining (22) at least one payload vulnerability index of said payload (4),

[0104] obtaining (23) a collection time indicating the point in time when the payload (4) is expected to be retrieved by the recipient,

[0105] selecting (24) a final delivery range at or near the designated location based on the at least one payload vulnerability index and the pickup time, and

[0106] The UAV (1) is operated (25) to deliver the payload (4) to the final delivery range.

[0107] Clause 2: The method of clause 1, wherein the at least one payload vulnerability index represents one or more of: monetary value, temperature sensitivity, an acceptable range of storage temperature, moisture sensitivity, or wind sensitivity.

[0108] Clause 3: A method according to clause 1 or 2, wherein the step of selecting (24) includes: predicting (24A) environmental conditions in a set of candidate delivery ranges located near the specified location, the environmental conditions being predicted for a dwell period extending from the estimated delivery time to the collection time, and selecting (204) the final delivery range from the set of candidate delivery ranges based on the predicted environmental conditions and the at least one payload vulnerability index.

[0109] Clause 4: The method of clause 3, wherein the environmental conditions include at least one of the following: sunlight, moisture, and wind.

[0110] Clause 5: The method according to clause 3 or 4, wherein the environmental conditions are predicted based on a three-dimensional model of the area surrounding the delivery location and weather data of the area during the dwell period.

[0111] Clause 6: A method according to any one of clauses 3 to 5, wherein the environmental condition is predicted based on sensor data from a sensor device (5; 1018; 1020) on the UAV (1) or on another UAV.

[0112] Clause 7: The method of clause 6, wherein the sensor data comprises at least one two-dimensional representation taken in the vicinity of the delivery location.

[0113] Clause 8: A method according to any preceding clause, wherein the final delivery range is selected based on the following input data (39): the input data is provided by the recipient and indicates at least one candidate delivery range in the set of candidate delivery ranges.

[0114] Clause 9: The method of clause 8, wherein the input data (39) comprises one or more of a photographic reproduction and a textual description.

[0115] Clause 10: A method according to any preceding clause, further comprising: operating (26; 60) an imaging device (5; 1018) on the UAV (1) to capture at least one image of the final delivery range according to at least one image standard.

[0116] Clause 11: The method of clause 10, wherein the at least one image criterion indicates one or more objectively identifiable features to be included in the at least one image.

[0117] Clause 12: The method of clause 10 or 11, further comprising the step of providing (63) the at least one image to the recipient for display in real time or substantially real time.

[0118] Clause 13: The method according to any one of clauses 10 to 12, further comprising the step of enabling the recipient to (66) externally control at least one of the imaging device (5; 1018) and flight controller (1005) of the UAV (1) while providing the at least one image for display in real time.

[0119] Clause 14: The method according to any one of clauses 10 to 13, further comprising the step of operating (60-63) the UAV (1) to capture an image of the final delivery range while moving at least one of the UAV (1) and the imaging device (5; 1018) until the at least one image is deemed to match a predefined image.

[0120] Clause 15: The method of any one of clauses 10 to 14, wherein the at least one image is taken after delivering the payload (4) to the final delivery range to include the payload (4).

[0121] Clause 16: A control system comprising logic (1101, 1102) configured to perform the method of any one of clauses 1 to 15.

[0122] Clause 17: A drone comprising a control system according to clause 16.

[0123] Clause 18: A computer readable medium comprising computer instructions (1102A) which, when executed by a processing system (1101), cause the processing system (1101) to perform the method of any one of clauses 1 to 15.

[0124] Clause 19: A drone configured to: obtain a designated location for delivering a payload (4); obtain at least one payload vulnerability index for the payload (4); obtain a collection time indicating a point in time when the payload (4) is expected to be retrieved by a recipient; select a final delivery range at or near the designated location based on the at least one payload vulnerability index and the collection time; and deliver the payload (4) to the final delivery range.

[0125] Clause 20: The drone according to Clause 19 is further configured to: predict environmental conditions in a set of candidate delivery ranges located near the specified location, wherein the environmental conditions are predicted for a stay period extending from the estimated delivery time to the collection time; and select the final delivery range from the set of candidate delivery ranges based on the predicted environmental conditions and the at least one payload vulnerability index.

[0126] Clause 21: The drone of clause 20, further configured to predict the environmental condition based on sensor data from a sensor device (5; 1018; 1020) on the drone or on another drone.

[0127] Clause 22: A drone according to any one of clauses 19 to 21, wherein the drone is further configured to: operate an imaging device (5; 1018) on the drone to capture at least one image of the final delivery range according to at least one image standard.

[0128] Clause 23: The drone according to clause 22, further configured to enable the recipient to manipulate at least one of the imaging device (5; 1018) and flight controller (1005) of the drone while the drone (1) provides the at least one image for display in real time.

[0129] Clause 24: The drone of clause 22 or 23, further configured to compare the at least one image with a predefined image and capture images of the final delivery range until the at least one image is deemed to match the predefined image.

[0130] 2. Improve the utilization of a fleet of UAVs delivering cargo

[0131] This section of the disclosure relates to improvements that utilize a fleet of drones in a delivery system and takes a different approach than conventional centralized planning. This centralized planning optimizes for maximum utilization of the delivery system and requires that all tasks and parameters within a specific time span be given, a plan be constructed, and the plan be executed. This time span causes delays in traditional deliveries. The time span must be long enough because if it is too short, then optimization becomes more of a matter of chance. During the execution of the plan, the plan cannot adapt to changes, such as new deliveries, drone failures, and unexpected weather conditions, without sacrificing the level of optimization. In order to perform proper optimization, centralized planning relies on access to all relevant parameters and therefore typically relies on multiple real-time sensors, which can be difficult and costly to implement.

[0132] This section of the disclosure does not address specific techniques for scheduling individual UAVs in a fleet to perform delivery tasks, but rather focuses on enabling features for different types of logistics systems within the fleet, particularly those that apply swarm intelligence to optimize utilization. Swarm intelligence (SI) is a well-known field of technology (also known as collective intelligence or symbiotic intelligence) that involves the behavior of natural and artificial systems consisting of multiple agents that interact locally with each other and their environment. Specifically, SI systems do not have any centralized control structure. Instead, local interactions between agents, and to some extent random interactions, lead to the emergence of global behaviors that are "intelligent" and unknown to the individual agents. The individual behaviors of agents can often be described in probabilistic terms, and each agent can take actions based solely on its local perception of its neighborhood. This mechanism of indirect coordination or self-organization is also known in the art as stigmergy. The principle of stigmergy is that the traces left by individual actions in the environment stimulate the execution of subsequent actions, which may be performed by the same agent, but in many scenarios by different agents. As such, SI systems are inherently flexible and robust, as well as decentralized and unsupervised.

[0133] The principles of SI can be applied to drone delivery by providing an interactive ordering system that allows users to evaluate different locations for drone delivery or pickup based on location-specific shipping prices calculated based on previously placed orders, and allows users to place delivery / pickup orders at selected locations based on these prices. By providing an interactive ordering system that presents shipping prices that reflect the impact of user decisions regarding delivery / pickup locations, a dynamically adaptable solver can be configured for the delivery system to utilize the fleet of UAVs based on swarm intelligence, with users acting as agents in such SI logistics systems. By assigning location selection to individual users and basing their choices largely on the impact of their choices on UAV utilization, the logistics system effectively leverages the neural computing power of each user, acting as a search function, to create optimal utilization at any given moment. For example, users can, through the interactive ordering system, predict changes based on a plethora of external factors over time and, for example, wait to order a delivery / pickup until a certain point in time, or decide to select a location that is more distant but associated with a lower shipping price. By allowing individual users to choose locations in real time based on a live profile of their impact on utilization, the collective of users will seamlessly adapt to changes in the delivery system with the goal of maximizing utilization. In other words, when individual users make their individual choices in their own best interest (e.g., keeping costs down), the logistics system will self-converge.

[0134] Figure 7 An example delivery system including a fleet of UAVs 1 is illustrated. The delivery system includes a scheduling device 100 configured to communicate with the UAVs 1 via a wireless network 101. The scheduling device 100 is also configured to execute the method described herein to determine the utilization schedule of the UAVs 1 and transmit the time points and locations for delivery / collection of goods to each UAV 1. The scheduling device 100 is also configured to communicate with a user device 102 (one user device is shown) including a display 103 (the display is operable to display an interactive chart 104 generated by the scheduling device 100), and one or more input devices (not shown) for data input. The user device 102 may include any type of electronic device, including but not limited to a personal computer, a laptop, a mobile communication device, a tablet computer, etc.

[0135] Figure 8 A detailed example of an interactive chart 104 is shown that corresponds to a geographic area and that may include a background image of the area, such as a satellite photo, a map, etc. The interactive chart 104 also depicts shipping prices for various locations. In the example shown, different shipping prices are represented by curves 108 drawn through points of equal shipping prices. In an alternative, the chart 104 may include a heat map that represents shipping prices in different colors or shades. Alternatively or in addition, the chart 104 may display the shipping prices in plain text when a location is selected (e.g., by the user positioning a cursor at the location). Figure 8 In the example, two locations are indicated by the mouse cursor, which are associated with shipping prices of 10 USD and 45 USD respectively. Figure 8 The diagram in FIG10 is also personalized for the user because it includes the route that the user is to take from the starting point 105 to the end point 106 along the movement path 107. The movement path 107 may be manually input by the user or may be automatically derived from an application associated with the user on the user device 102 or another device. Such an application may be a digital personal assistant that has learned the user's behavior, a navigation service, etc. Figure 8The black dots in the diagram represent proposed delivery / collection locations that are on or near the moving path 107. The proposed locations can be determined by the scheduling device 100 to be on or near the moving path. The user can select any location on the chart 104 for delivery / collection by any suitable selection action. As previously understood, the scheduling device 100 can update the chart 104 upon receiving the user selection, which, since it is displayed to both the user and other users, affects all future selections by all users. As an alternative to allowing the user to freely select a location by pointing to the chart 104, a fixed set of locations can be presented to the user (e.g., as a list or geographic chart) for selection based on the associated transportation costs.

[0136] Figure 9 An example method 110 for scheduling the delivery and / or collection of goods according to one embodiment is illustrated. Figure 7 The scheduling device 100 in the embodiment of the present invention performs the method 110. Figure 10 Describing method 110, the Figure 10 A geographic area 120 is shown, which may correspond to Figure 8 The area of ​​the chart 104 in FIG. 1 is divided into sub-areas (“cells”) 122 . Figure 10 Also shown is the current position of the UAV 1 within the area 120, as well as the current or future idle points of the UAV 1 (see below).

[0137] In method 110, step 111 calculates aggregate cost data ACM for the fleet of UAVs based on the current and / or future positions of the UAVs within a geographic area 120 (e.g., given by a utilization schedule 110' (see below)). The ACM associates a "utilization cost" with a corresponding cell 122, wherein the utilization cost of a cell 122 represents the estimated cost value of directing at least one of the UAVs 1 to that cell 122. As used herein, "cost" or "cost value" does not have to be a monetary value, but can be given in any type of unit. In some embodiments, the utilization cost corresponds to a distance, which can be a physical distance in two or three spatial dimensions. It is also conceivable that the distance includes other non-spatial dimensions, such as taking into account wind direction, wind speed, etc. In some embodiments, the utilization cost corresponds to energy consumption. The ACM can be represented as a matrix. As used herein, a matrix generally refers to any form of addressable data structure including values ​​in any format, such as an array of arbitrary dimensions.

[0138] Step 112 receives a query indicating one or more potential locations for delivery / collection of a payload. The query may be generated based on a user selection, such as by a user submitting a request to place an order for delivery or collection of goods. Alternatively, the query may be generated by the system, such as when the user device 102 is connected to the scheduling device 100. The query may indicate a discrete set of potential locations at a given spatial resolution, one or more coherent areas of potential locations within the geographic area 120, or all locations within the geographic area 120. In one example, the query may identify a user and cause step 112 to retrieve potential locations specifically for that user, such as based on a user profile and / or historical data. For example, Figure 8 and Figure 10 The locations indicated by the black dots in FIG are pre-stored as potential locations of the user. In another example, the query may include the moving path 107 ( Figure 8 and Figure 10 ), and causes step 112 to determine a potential position relative to the user's movement path 107, e.g., resulting in Figure 8 and Figure 10 In one example, step 112 may determine a potential location along the movement path 107 that has the lowest utilization cost.

[0139] Step 113 determines a transportation price of at least one potential location among the potential locations based on the ACM, and step 114 presents the transportation price of the at least one potential location to the user. The transportation price may be given as a monetary value, so step 113 may involve searching the ACM to obtain a utilization cost of the unit 122 including the potential location, and converting the utilization cost into a transportation price.

[0140] The delivery system can be configured to allow the respective user to change the previously selected location and / or delivery / collection time. It is contemplated that, in the event of such a change, step 113 determines an environmental utilization cost representing the change in the overall delivery cost of the delivery system resulting from the change, and adds the environmental utilization cost to the utilization cost provided by the ACM when determining the transport price. It should be noted that the environmental utilization cost may be negative.

[0141] Step 114 may present a list of shipping prices for different potential locations. Alternatively or in addition, step 114 may present Figure 8 or any other graphical representation 104 of the distribution of transportation prices within a geographic area 120 or a portion of that geographic area. Figure 9 As shown, step 113 may include a sub-step 113A of generating the graphical representation 104. Step 114 may also include displaying the transportation price for a location given by the user pointing to a specific location in the graphical representation (see Figure 8(the mouse pointer in the ).

[0142] In another variation, the query received in step 112 may be generated in response to a user inputting a potential location, such as by the user selecting a location / area on a displayed map of the geographic area 120 or a portion thereof using a mouse, stylus, finger, etc. Alternatively, the user may manually input the coordinates of the potential location using a keyboard, thereby generating the query of step 112 .

[0143] It will be appreciated that the combination of steps 111 to 114 provides an interactive ordering system as discussed above that allows the delivery system to optimize utilization of the delivery system based on swarm intelligence (SI).

[0144] exist Figure 9 In the example of FIG, method 110 further includes step 115 of receiving a user selection of a shipping price and a corresponding potential location. The user may indicate the selection in any conventional manner, such as by using a mouse, keyboard, touch screen, etc. Optionally, in response to the proposed time point / window set being presented to the user, step 115 may also include receiving a selected time point or time window for delivery / collection, the selected time point or time window being input by the user.

[0145] Following the selection in step 115, step 116 updates the utilization schedule 110' for the delivery system. The utilization schedule 110' defines the current and future locations of the UAVs 1 in the fleet within the geographic area 120 and therefore serves as a record of the current plan for utilizing the UAVs in the fleet. For example, the utilization schedule 110' may contain a list of delivery and pickup locations and associated time points selected by a user of the delivery system, and the utilization schedule may also include assigning specific UAVs to tasks involving the locations / time points. Step 116 may update the utilization schedule 110' by adding the locations selected by the user in step 115 to the utilization schedule (optionally along with the associated time points entered by the user). The details of the utilization schedule 110' and its updating are implementation-specific and are therefore not described further.

[0146] In the example shown, since the ACM is calculated by step 111 based on the utilization schedule 110', subsequent execution of steps 111 to 116 will take into account the updated utilization schedule 110' and thus automatically adapt to the choices made by the users. Thus, over time, the transportation price is adapted to the choices made by all users in the system, resulting in inherent utilization convergence as the respective users make their own choices in their own best interests (e.g., keeping the transportation price down).

[0147] although Figure 9Not shown, but method 110 may also include, for example in at least some iterations, after step 116, the step of controlling the fleet of UAVs by directing one or more of UAVs 1 to the locations selected in step 115 based on the utilization schedule 110'.

[0148] Figure 11A is a flow chart of an example process for calculating ACM. As shown, the example process can be Figure 9 In step 131, an individualized cost matrix of the available UAVs is calculated. The respective individualized cost matrix ICM defines the cost matrix for directing one of the available UAVs 1 to the different sub-areas 122 ( Figure 10 ) cost value. Therefore, in Figure 10 In the example shown, each drone 1 has an ICM, and each ICM includes a cost value for each sub-region 122. Each ICM can be represented by a matrix or corresponding data elements. After step 131, process 111 forms the ACM as a (logical) combination of the ICMs. This is a highly efficient way to calculate the ACM. In the example shown, the ACM is formed via steps 132 through 135.

[0149] In optional step 132, a cost threshold is applied to the ICM, and cost values ​​above the cost threshold are replaced with a predefined value, such as a null value. In the absence of step 132, an ACM can be generated to show the overutilization costs of some units. It may be undesirable to allow users to select locations in such units because this may force the delivery system to perform deliveries / collections that are disadvantageous in terms of resource efficiency. Step 132 may also enable step 134 (see below).

[0150] In step 133 , the utilization cost of the corresponding unit 122 is determined based on the cost value associated with the corresponding unit 122 in the ICM. Figure 13 is a conceptual diagram of step 133, wherein cells 122 in different ICMs are aligned and processed to generate an ACM. Figure 13 In , there are N different ICMs specified by ICMi, where i = 1,…,N. Figure 13 Also shown is how cells 122 are defined relative to geographic area 120 .

[0151] In some embodiments, the utilization cost of a cell 122 is determined by finding the minimum (lowest) cost value among all the cost values ​​of that cell 122 in the ICMs. Depending on the definition of cost, this may correspond to identifying the UAV with the shortest or most energy-efficient path to the cell 122 given by the ICM containing the minimum cost value for the cell 122. In some embodiments, the utilization cost of a cell 122 may be set to the minimum cost value for the cell. In other embodiments, the utilization cost of a cell 122 may be set to the minimum cost value for the cell. Figure 11A As represented by steps 134 to 135 in , the utilization cost of a cell 122 can be given by the minimum cost value of the cell, which is reduced according to the number of available UAVs for the cell 122. In this context, "available UAVs" can be given by all ICMs whose cost value is lower than the threshold mentioned above. If step 132 has been executed, the number of available UAVs is the number of ICMs whose cost value for the cell 122 is not a predefined value (e.g., a null value). The basic principle of reducing the minimum cost value is to encourage the user to select (in step 115) a location with multiple available UAVs rather than a location with only one or a few UAVs, because selecting a location with only one or a few UAVs may result in a task that is disadvantageous in terms of resource efficiency. Therefore, in step 11A, step 134 determines the count of available UAVs for each cell 122, and step 135 fills the corresponding cell of the ACM according to the minimum cost value reduced by the count.

[0152] In some embodiments, step 111 may include an interpolation step to increase the granularity of the ACM. Such an interpolation step may operate on the cost values ​​of cells 122 determined in step 135 to generate interpolated cost values ​​within or between these cells 122. This is an efficient process for increasing the resolution of the ACM. Any interpolation technique may be used, such as linear interpolation, polynomial interpolation, spline interpolation, etc.

[0153] In some embodiments, step 111 may include a contrast enhancement step to increase the accuracy of the ACM, for example, after or during the interpolation step mentioned above. Such a contrast enhancement step may include determining a local derivative of the cost value of the cell 122 within the ACM, and updating the cost value in the ACM based on the local derivative to enhance the contrast between adjacent cells 122.

[0154] Figure 11B is a flow chart of an example process for calculating ICM. As shown, the example process can be Figure 11AAs mentioned above, the ICM is calculated for a particular UAV in the fleet. Step 136 determines the starting point of the UAV in the geographic area 120. In some embodiments, the starting point is set to the current location of the UAV. By setting the starting point to the location where the UAV 1 is scheduled to be idle, a simple calculation and sufficiently accurate results can be obtained. In one example, whenever the UAV 1 arrives at an infrastructure component of the delivery system, such as a payload storage facility (warehouse, intermediate storage point, etc.) or a charging station, the UAV 1 is scheduled to be idle. The locations of these infrastructure components ("idle points") are known, and the time points of arrival and departure of the corresponding UAVs can be included in the utilization schedule ( Figure 9 110'). Back Figure 10 , the idle points 123 are marked by crosses, and the scheduled paths of the UAVs 1 from their current positions to their next idle points 123 are schematically represented by solid arrows. Figure 10 The dashed arrow in illustratively represents a path from an idle point to one of the cells 122 .

[0155] Step 137 determines the distance value from the starting point to the cell 122 for a particular UAV, and step 138 determines the cost value of the cell based on the distance value. Steps 137 to 138 are performed for all cells 122 in the area 120 or a subset thereof. As mentioned above, the distance value can be a physical distance, but can also include non-spatial dimensions. Step 138 can set the cost value to the distance value or convert the distance value in any suitable manner. For example, the cost value can be calculated to represent the energy consumption of the UAV. Step 138 can also take into account other UAV parameters, such as the scheduled stay time ("idle time") of the UAV at its idle point 123. For example, in order to promote the utilization of the UAV and reduce the scheduled stay time, the cost value can be reduced according to the scheduled idle time.

[0156] Step 137 may resolve obstacles that prevent the UAV from taking a straight path from the starting point to the corresponding cell. In some embodiments, a flood-fill algorithm may be applied to calculate the distance value. Figure 12A One such implementation is illustrated in FIG. Figure 12AThe step-by-step calculation of the ICM matrix for cells 122 is illustrated. Some of the cells are occupied by obstacles 124 (filled squares), obstacles that the UAV is deemed unable to overcome, for example, by flying over spatial obstacles. Initially, the matrix is ​​empty, except for the cell containing an idle point 123, which is assigned a distance value of 0. This calculation is performed over a series of iterations to determine the distance values ​​for all cells. Each iteration determines the distance values ​​of cells adjacent to the cell to which a distance value has been assigned so far ("non-empty cells"). In the example shown, the cell width is assumed to be 1 and the elevation difference between cells is assumed to be 0. To determine the distance value for a particular cell in an iteration, step 137 can calculate, for each non-empty adjacent cell, the sum of the center-to-center distance between these cells and the distance values ​​of the non-empty cells, and set the distance value for the particular cell to the minimum of these sums. This iteration terminates when all relevant cells have been assigned distance values, thus estimating the minimum travel distance from the corresponding cell 122 to the idle point 123. In the example shown, four repetitions are required to fill the ICM.

[0157] Step 137 may also take into account the current position of the UAV relative to the idle point 123 . Figure 12B An example is shown, where 123' specifies the current location of the UAV and 123 specifies an idle point, such as a storage facility where the UAV will receive a payload to be delivered and / or a parking place for charging. In this example, step 137 can be performed using the current location 123' as a starting point as for Figure 12A The first iterative sequence described is to determine the distance value of the cell containing the idle point 123. As in Figure 12A Then, step 137 uses the idle point 123 as the starting point and the distance value determined by the first iterative sequence to perform a second iterative sequence, such as for Figure 12A described. Figure 12C The resulting ICM is shown.

[0158] Step 137 may also include an interpolation step and / or a contrast enhancement step operating on the ICM, for example, as described above for the ACM.

[0159] Any available algorithm that finds the shortest path in a dataset can be implemented to fill the ICM. Examples include: flood fill, Floyd-Warshall, Johnson, or any wave propagation algorithm such as the adaptive light propagation volume algorithm.

[0160] Back to Figure 9In the method 110 of FIG. 1 , it can be noted that at each single point in time, the user is presented with a shipping price based on the utilization schedule 110 ′. It is contemplated that step 116 also updates the utilization schedule 100 ′ based on manual or automatic forecasts (e.g., statistically based extrapolation to introduce virtual deliveries) or based on relevant events (such as changing weather conditions).

[0161] In some embodiments, method 110 is performed for the current time point, without regard to the user-selected time point at which the delivery / collection is to be performed. Thus, the ACM may be calculated based on the UAV's current location, the UAV's upcoming or current idle point, etc. In such embodiments, at least the user-selected delivery / collection time point may be considered when updating the utilization schedule 110'. As used herein, a "user-selected time point" may be freely selected by the user, or a time point suggested by the system and accepted by the user.

[0162] In other embodiments, method 110 is performed at at least one future time point, which may or may not be user-selectable. Given the ever-changing availability of users to interface with UAVs for pickup or delivery, utilization costs are likely to depend on the time of day. In these embodiments, the ACM can be calculated based on predictions of future UAV locations, predictions of idle spots for the UAV, and the like. These predictions can be based on utilization schedules, as well as historical data, weather forecasts, forecasts from merchants, and similar sources.

[0163] In some embodiments, the method 110 can be configured to present shipping prices at potential locations for a series of time points (e.g., the current time point and one or more future time points) in step 114, and allow the user to select a location and time point for delivery / collection based on the presented data. The intervals between time points can be one or more hours or one or more days. For example, Figure 8 The interactive chart 104 in FIG. 1 can be supplemented with a scrolling function (e.g., a time bar) that allows the user to browse the interactive chart 104 at different time points. In such an embodiment, step 111 can calculate the ACM for multiple different time points to generate a time series of the ACM. Step 113A can generate the interactive chart 104 for the corresponding ACM according to the time series, and step 114 can provide the interactive chart 104 for display.

[0164] Figure 14 is a block diagram of an example scheduling module 140, which may be included in Figure 7In the scheduling device 100 in . Module 140 includes sub-modules 141 to 147. The UAV cost estimator 141 is configured to calculate the ICM of different UAVs specified by [ICMi], for example, according to step 131. The thresholder 142 is configured to perform a thresholding operation on the ICM, for example, according to step 132, thereby obtaining a pre-processed ICM specified by [ICMi']. The UAV counter 143 is configured to count the number of available UAVs per unit, for example, according to step 134, thereby obtaining a count specified by [Cj]. The aggregator 144 is configured to generate an ACM, for example, according to steps 133 and 135. The price calculator 145 is configured to generate a transportation price TP, for example, according to step 113. The transportation price TP can be generated for a first potential location set and / or a second potential location set, the first potential location set can be generated by the system and specified by [LOC1], and the second potential location set can be generated by the user and specified by [LOC2]. In the example shown, [LOC2] is received from the presentation controller 146. Some further examples of sources for [LOC1] and [LOC2] are given above with reference to step 112. The presentation controller 146 is configured to, for example, display the [LOC1] and [LOC2] according to step 114. Figure 7 The transport price TP is presented to the user at step 103 and a user selection of a location designated as SL is received, e.g., according to step 115. The user selection SL is detected and output by the presentation controller 146 and the scheduler 147 updates the utilization schedule 110′ based on the SL, e.g., according to step 116.

[0165] Below, several clauses are cited to summarize some aspects and embodiments disclosed above.

[0166] Clause 1: A method of dispatching transport by a fleet of UAVs (1), the method comprising the following steps:

[0167] calculating (111) aggregated cost data (ACM) for the fleet of UAVs (1) based on current and / or future positions of the UAVs (1) within the geographic area (120), the aggregated cost data (ACM) associating utilization costs with distributed sub-areas (122) within the geographic area (120), the utilization cost of a respective sub-area (122) representing an estimated cost of directing at least one of the UAVs (1) to the respective sub-area (122);

[0168] receiving (112) a query indicating one or more potential locations for pickup / delivery of a payload (4);

[0169] determining (113) a transportation price for at least one of the one or more potential locations based on the aggregated cost data (ACM); and

[0170] Transportation prices for the at least one potential location are presented (114).

[0171] Clause 2: A method according to clause 1, wherein the step of calculating (111) the aggregated cost data comprises: calculating (131) an individualized cost matrix (ICMi) of the UAV (1), which defines the cost value of directing the corresponding UAV (1) to the distributed sub-area (122); and forming (132 to 135) the aggregated cost data (ACM) as a combination of the individualized cost matrices (ICMi).

[0172] Clause 3: A method according to clause 2, wherein the step of forming (132 to 135) the aggregated cost data includes: for the corresponding sub-area (122), determining (133) the utilization cost based on the cost value associated with the corresponding sub-area (122) in the individualized cost matrix (ICMi).

[0173] Clause 4: The method according to clause 3, wherein the utilization cost of the corresponding sub-region (122) is determined based on the minimum cost value among the cost values ​​associated with the corresponding sub-region (122) in the individualized cost matrix (ICMi).

[0174] Clause 5: A method according to clause 4, wherein the step of forming (132 to 135) the aggregated cost data comprises determining (132, 134) the count of available UAVs (1) for the corresponding sub-area (122), wherein the utilization cost of the corresponding sub-area (122) can be given by a minimum cost value that is reduced according to the count of available UAVs (1) for the corresponding sub-area (122).

[0175] Clause 6: A method according to clause 5, wherein the step of determining (132, 134) the count of available UAVs in the corresponding sub-area includes: counting cost values ​​associated with the corresponding sub-area (122) in the individualized cost matrix (ICMi) that are smaller than a threshold cost.

[0176] Clause 7: A method according to any one of clauses 2 to 6, wherein the step of calculating (131) an individualized cost matrix for a UAV comprises determining (136) a starting point (123) of the corresponding UAV (1) in the geographic area (120), determining (137) a distance value from the starting point (123) to the corresponding sub-area (122), and determining (138) a cost value of the corresponding sub-area (122) based on the distance value of the corresponding sub-area (122).

[0177] Clause 8: The method of clause 7, wherein the step of determining (136) the starting point of the respective UAV comprises determining a location at which the respective UAV (1) is scheduled to be idle.

[0178] Clause 9: The method according to clause 7 or 8, wherein the cost value of the respective sub-area (122) is further determined based on the stay time of the respective UAV (1) at the starting point (123).

[0179] Clause 10: A method according to any preceding clause, wherein the step of calculating (111) aggregated cost data comprises determining the utilization costs within or between sub-regions (122) by interpolating the utilization costs associated with the sub-regions (122).

[0180] Clause 11: A method according to any preceding clause, wherein the step of calculating (111) the aggregated cost data comprises determining a local derivative of the utilization cost within the aggregated cost data (ACM), and updating the utilization cost within the aggregated cost data (ACM) based on the local derivative to enhance the contrast of the utilization costs between adjacent sub-regions (122).

[0181] Clause 12: A method according to any preceding clause, further comprising: generating (113A) a graphical representation (104) of the distribution of transport prices within a geographical area (120), wherein the step of presenting (114) comprises: presenting the graphical representation (104).

[0182] Clause 13: A method according to any preceding clause, wherein the step of calculating (111) the aggregated cost data (ACM) comprises generating a time series of the aggregated cost data by calculating the aggregated cost data (ACM) for a plurality of different points in time, wherein the step of determining (113) the transport price comprises determining a time series of the transport price by determining the transport price for the at least one potential location at the plurality of different points in time, wherein the presenting step (114) comprises presenting the time series of the transport price.

[0183] Clause 14: The method of any preceding clause, wherein the query is received in response to a user selection of the one or more potential locations in a geographic area (120).

[0184] Clause 15: A method according to any preceding clause, wherein the query comprises a user movement path (107) in a geographic area (120), and wherein the method further comprises the step of determining the one or more potential locations related to the user movement path (107).

[0185] Clause 16: The method of any preceding clause, further comprising: after said step of presenting (114), receiving (115) a user selection of a transportation price and a corresponding potential location.

[0186] Clause 17: The method of clause 16, further comprising the step of including (116) user selections in a utilization schedule (110') defining current and future locations of the UAV within a geographic area, wherein the step of calculating (111) aggregated cost data (ACM) is based on the utilization schedule (110').

[0187] Clause 18: The method of clause 17, further comprising the step of directing the at least one of the UAVs (1) to a corresponding potential location based on the utilization schedule (110').

[0188] Clause 19: A method as described in any preceding clause, wherein the aggregated cost data (ACM) comprises a matrix relating utilization costs to sub-regions.

[0189] Clause 20: An apparatus for scheduling transport by a fleet of drones (1), the apparatus comprising: logic (1101, 1102) configured to perform the method according to any one of clauses 1 to 19.

[0190] Clause 21: A delivery system comprising: a fleet of drones (1); and an apparatus according to clause 20, wherein the apparatus is configured to communicate with the drones (1).

[0191] Clause 22: A computer readable medium comprising computer instructions (1102A) which, when executed by a processing system (1101), cause the processing system (1101) to perform the method of any one of clauses 1 to 19.

[0192] 3. Improvements in launching UAVs

[0193] This part of the disclosure relates to the launch of a UAV and is based on the insight that by using a power source external to the UAV and raising the UAV before launching the UAV, the performance of the UAV can be significantly improved. By raising the UAV, energy from the power source is converted into increased potential energy of the UAV. The increased potential energy can be used to improve the performance of the UAV, for example in terms of range, energy consumption, ability to carry a payload, transportation time to the destination, etc., and can also enable the use of UAVs with simpler structures, such as UAVs without a local power source. In addition, the UAV can be raised to avoid obstacles in the flight path. Furthermore, the UAV can be raised so that the UAV can use gravity to collect energy, for example to charge a battery.

[0194] Some embodiments relate to a system for launching a UAV. The launch system includes a lifting device that is physically separate from the UAV and powered by a power source. The lifting device is operable to vertically raise the UAV prior to launching the UAV. The launch system is useful for all UAV missions, but has significant advantages for UAVs carrying payloads. The power consumption of the UAV's vertical movement depends on the total weight of the UAV, and if the payload is heavy, the power consumption can be significant. Therefore, the launch system can significantly save the UAV's power consumption. The launch system can also allow the UAV to carry a weight that exceeds the weight that the UAV can lift because the UAV can be launched at an altitude above the destination.

[0195] In the following examples, the launch system is located in a mobile station, such as a vehicle, a ship, or a trailer attached to a truck. Thus, the launch system can be moved to a desired geographic location, from which one or more UAVs can be launched for a corresponding mission. However, it is also conceivable that the station is stationary.

[0196] FIG. 15A to FIG. 15BAn example launch system 200 according to an embodiment is illustrated. The launch system includes a control system 203, which is located in a station 200' and is configured to operate a lifting device to vertically raise the UAV 1 to one or more selected heights. The control system 203 may include control electronics that generate control signals for one or more actuators (not shown) of the lifting device. Such actuators may be electrically, hydraulically, or pneumatically operated. The system also includes a system power supply 204, which is ground-based and supplies electricity for operating the control system 203 and thereby the lifting device. The system power supply may include batteries, fuel cells, internal combustion engines, solar panels, wind turbines, etc., and / or be connected to a power distribution network. In the illustrated example, the lifting device includes an extendable mechanical structure 202A, such as a telescopic device or mast, a pivotable device, a bay lift, a four-bar linkage, etc. The extendable structure 202A is mechanically connected to a platform 201, which carries one or more UAVs 1 to be launched. Figure 15A The system 200 is shown in a default state, wherein the extendable structure 202A is retracted with the platform 201 and the UAV 1 is at a default or reference level ( Figure 15B ), in this example, the default or reference horizontal plane is approximately flush with the top of the station 200'. Figure 15B System 200 is shown in a launch configuration, wherein platform 201 and UAV 1 have been vertically raised to a selected altitude ("altitude") by operating extendable structure 202A via control system 203. In the illustrated example, platform 201 has been sequentially operated to altitudes H1, H2, and H3 for launching respective UAVs 1. As will be described in further detail below, altitudes H1 through H3 may be selectively determined for respective UAVs 1.

[0197] When one or more UAVs 1 have been launched, the platform 201 can be operated to a predefined altitude, such as a reference level H0 below altitudes H1 to H3. This can reduce the power consumption of any UAVs that may need to land on the platform 201, such as a UAV 1 returning after completing its mission.

[0198] Figure 16An example launch system 200 according to another embodiment is illustrated. The lifting device includes an aircraft 202B, such as a dedicated lifting UAV, which can be configured to carry a large payload. In one example, aircraft 202B is a so-called "cargo drone." Other examples of aircraft include aerostats, airships, and hot air balloons. The platform 201 of the UAV 1 is positioned on the aircraft 202B and is thus vertically elevated along with the aircraft 202B. If the aircraft 202B has a propulsion system, the platform 201 is preferably positioned to shield against air motion caused by the propulsion system, for example, to avoid downward forces on the UAV 1 caused by the air motion.

[0199] In the example shown, the control system 203 is located on the aircraft 202B. The aircraft 202B may include a local power source (not shown) that is charged by the system power source 204, for example, when the aircraft 202B has landed on the docking station 205 on the station 200'. Alternatively or in addition, the aircraft 202B may be connected to the system power source 204 via a cable.

[0200] 17A to 17B An example launch system 200 according to another embodiment is illustrated. The lifting device includes a cable 202C that is releasably connected to the UAV 1 to supply energy from a system power supply 204 to the UAV 1 when the UAV 1 is raised to a selected altitude. In this embodiment, a control system 203 is provided on the UAV 1 and operates the propulsion system of the UAV 1 (see FIG. Figure 1 1002 in ) to vertically raise the UAV. The platform 201 (if provided) is fixed in the system, for example, fixed on the top of the station 200 ', as shown. Figure 17A illustrates the system 200 prior to launch when transporting a UAV to a selected altitude, and Figure 17B The system 200 is illustrated after launch when the control system 203 has operated a retaining mechanism (not shown) on the UAV 1 to detach the cable 202C. In the example shown, the cable includes a parachute 206 that can be automatically released upon detachment of the cable 202C to facilitate a controlled return path for the cable 202C. The system 200 may also include a traction mechanism (e.g., a reel) for retracting the cable 202B, and an attachment mechanism for reattaching the cable 202B to another UAV 1.

[0201] Figure 18An example launch system 200 according to another embodiment is illustrated. The lifting device includes a flow generator 202D that is operable to generate a vertical airflow toward the UAV 1. The flow generator 202D is powered by a system power supply 204. In some embodiments, under the control of a control system 203 in a station 200', the flow generator 202D is operable to lift the UAV 1 to a selected altitude. In other embodiments, when the control system 203 is located in the UAV 1 to operate the propulsion system ( Figure 1 1002 in ) so that when the selected altitude is reached, the flow generator 202D is operated to maintain or assist in the vertical ascent of the UAV 1. As shown, the launch system 200 may include a platform 201 carrying the UAV, which is fixed to the station 200' and is configured to allow airflow to pass through.

[0202] Figure 19 An example launch system 200 according to another embodiment is illustrated. The lifting device includes a conveyor device 207 configured to lift one or more UAVs 1 onto a platform 201. The platform 201 is attached to an extendable mechanical structure 202A. As in Figure 15A and Figure 15B As in the example, the control system 203 in the station 200' can operate the extendable structure 202A to vertically raise the platform 201 to a selected height, and operate the conveyor device 207 to raise the UAV 1 to the level of the platform 201, e.g. Figure 19 shown.

[0203] It should be noted that Figures 15A to 19 Any combination of the examples in is conceivable. For example, Figure 18 The flow generator 202D in Figures 16 to 17B In addition, Figures 15A to 19 Non-limiting examples of transmission systems are depicted, and the skilled artisan will be able to envision variations and alternatives based on the foregoing disclosure.

[0204] Figure 20 is a schematic block diagram of an example control system 203 for controlling a lifting device in a launch system 200. The control system 203 includes a calculation module 210 configured to calculate a selected altitude for launching one or more UAVs. Figure 20, the selected altitude is specified by HL. The control system 203 also includes a controller 211 that generates a control signal CS corresponding to HL and provides CS to an actuator mechanism 212 of the lifting device. The controller 211 can have any suitable configuration, such as a feedback and / or feedforward controller. Thus, the controller 211 can operate in response to signals from one or more sensors that measure the current altitude of the lifting device, such as a barometer or a rangefinder (radar, lidar, time of flight, etc.). The actuator mechanism 212 will depend on the type of lifting device and will not be described further.

[0205] Figure 21 2 is a flow chart of an example method 220 for launching a UAV according to one embodiment. The method 220 may be performed by the control system 203. Step 221 calculates a selected altitude (sea height) HL of the launching UAV relative to a reference altitude or reference data (e.g., mean sea level (MSL) or a position on the station 200', such as Figure 15B The selected height HL is calculated from the reference horizontal plane H0 in step 222. In step 223, the UAV is launched from the selected height, thereby starting its mission. Figure 20 In the example of FIG, step 221 may be performed by the computing module 210, and step 222 may be performed by the controller 211. Step 223 may include providing a transmit signal to the UAV when the lifting device has raised the UAV to the selected height, thereby causing the UAV to transmit. The transmit signal may be generated by the control system 203 and sent to the local controller of the UAV. If the control system 203 is located in the station 200' ( Figure 15A and Figure 15B 、 Figure 18 、 Figure 19 ), the transmission signal can be transmitted via wireless communication or via a cable integrated into the lifting mechanism ( Figure 15A and Figure 15B 、 Figure 19 If the control system 203 is located in the UAV ( Figure 16 、 Figure 17A and Figure 17B ), the transmission signal can be delivered to the local controller of the UAV by wire or wirelessly. It is also conceivable to integrate the control system 203 into the local controller.

[0206] Step 221 may calculate the selected altitude HL based on parameter data representing the surrounding conditions of the UAV to be launched, the lifting device, or between the lifting device and the destination of the UAV. Figure 20Examples of such parameter data are indicated in . In one example, the parameter data include the position of the lifting device (SP, "starting point") and the destination of the UAV to be launched (DP, "destination point"). The starting point SP and the destination point DP may be given as two-dimensional (2D) or three-dimensional (3D) coordinates in any global coordinate system (e.g. a geographic coordinate system). In one example, the parameter data of the UAV may indicate the type of the UAV and / or the current state of the UAV. The current state may include one or more of the following: current weight data (WD) of the UAV, current size data (SD) of the UAV, or current energy content (ED) of the local power source in the UAV (cf. Figure 1 1010 in ). If the UAV carries a payload to be delivered at the destination, then WD can represent the weight of the payload or the total weight of the UAV and the payload. Similarly, SD can represent the size of the payload or the total size of the UAV and the payload. Here, the size can be expressed as area, shape, one or more scales, or any equivalent measurement. In one example, the parameter data of the surrounding conditions may include one or more of the following items: terrain data (TD) including the SP and DP for the area, or current environmental data (AD) for the area. Here, TD can indicate the 3D shape of the area and include the location, range and height of obstacles between the SP and DP, while AD can indicate current weather conditions, air pressure, humidity, etc. Obstacles may include buildings, trees, power lines, etc. In one example, the parameter data of the lifting device may include one or more of the following items: energy content data (ES) of the system power supply 204, and lifting speed data (LS) of the lifting device. Here, LS can indicate the nominal or estimated speed of lifting by the lifting device.

[0207] Step 221 can calculate the selected height HL by evaluating parameter data related to at least one predefined consumption parameter. In one example, step 221 evaluates the parameter data by optimizing a cost function representing the predefined consumption parameters. Such a cost function can use the parameter data as known variables and the height parameter as an unknown variable. The cost function can be optimized by finding the value of the height parameter that generates the maximum or minimum value of the cost function, which value represents the selected height. Any suitable optimization algorithm can be used to optimize the cost function, including but not limited to gradient methods (such as gradient descent, coordinated descent, conjugate descent, etc.) or evolutionary algorithms (such as genetic algorithms, genetic programming, etc.).

[0208] In some embodiments, the at least one predefined consumption parameter represents the consumption cost of performing the delivery at the destination. In one example, the selected altitude HL is calculated to minimize the energy consumption of the UAV, and the consumption cost may represent the energy consumption required by the UAV to deliver the payload at the destination. In one example, the selected altitude HL is calculated to minimize the total energy consumption required to deliver the payload at the destination, and the consumption cost may represent the energy consumption required by the UAV and the energy consumption required by the lifting device to vertically raise the UAV to the selected altitude. In one example, the selected altitude HL is calculated to minimize the delivery time, and the consumption cost may represent the time required for the UAV to reach the destination from the time of launch or from the time the lifting device is activated to raise the UAV. In one example, the selected altitude HL is calculated to minimize the monetary cost of delivering the payload at the destination.

[0209] It may be noted that step 221 may simultaneously calculate the selected altitudes of a plurality of UAVs 1, for example by jointly optimizing the cost functions of the different UAVs.

[0210] The basic principle will be further described for a specific example, where the consumption parameter is the total energy consumption and the cost function contains the following parameters: the type of UAV (TYPE), the weight of the payload (WD), the obstacle height (OH) and the height parameter to be optimized (HP). In this example, the cost function C can be defined as:

[0211] C=C1(TYPE,WD)·(OH-HP)+C2(TYPE,WD)·HP

[0212] Here, the first function C1 provides the specific energy consumption of the UAV when the UAV moves upward with the payload, and the second function C2 provides the specific energy consumption of driving the lifting device upward when the lifting device carries the UAV and the payload of the UAV. It is recognized that the selected altitude can be determined by minimizing the cost function C.

[0213] FIG. 22A to FIG. 22B Another simplified example is shown. Figure 15A and Figure 15B A launch system 200 of the type shown in FIG. 1 is configured to launch a UAV 1 to a destination 231 (e.g., a Figure 20 Position the launch system 200 at the launch position 232 (given by DP in Figure 20 The obstacle 233 is located between the launch location 232 and the destination 231. The characteristics of the obstacle 233 can be given by the terrain number TD mentioned above. Figure 22BAs shown by arrows 236, 246 in the top view of FIG, the wind pattern (e.g., direction and speed) between the launch location 232 and the destination 231 is known, for example, given by the weather data DW mentioned above. Based on this input data, step 221 can calculate a selected altitude to, for example, optimize total energy consumption, UAV energy consumption, or transportation time. In this calculation, step 221 can evaluate the following options: raising the UAV 1 to a level above the obstacle 233 to provide the UAV 1 with a straight route 234 to the destination location 231, or raising the UAV 1 to a lower level and causing the UAV 1 to take a route 235 around the obstacle 233.

[0214] As mentioned above, the launch system 200 can be mobile. Consider a delivery system that includes a fleet of UAVs 1 and multiple launch systems 200, for example, Figure 24 For example, the delivery system includes a scheduling device 270. The launch system 200 is located on a mobile station 200' that holds a corresponding set of UAVs 1. In some embodiments, the launch system is optimized to deliver payloads to multiple destinations by actively selecting the launch position of one or more of the launch systems 200 relative to the multiple destinations.

[0215] Figure 23B 2 is a flow chart of an example method 250 for configuring such a delivery system. The method 250 may be performed by a scheduling device 270. Step 251 obtains the plurality of destinations ("destination points," [DP]) to which payloads are to be delivered by the delivery system. Step 252 obtains the number of launch locations to be determined by the method 250. Step 253 determines the number of locations based on the plurality of destinations. Step 254 directs a group of launch systems 200 to the launch locations (at Figure 24 (denoted by [SP] in the figure). Step 254 may be performed based on electronic communication from the dispatch device 270 to the station 200' and / or the driver arriving at the station. Alternatively, the launch location may be manually communicated to the driver / station. It should be understood that step 253 may assign a subset of destinations to respective launch locations, and step 254 may communicate the subset of destinations to the UAV 1 to be launched from the respective launch location. Furthermore, step 254 may cause the cargo to be delivered to be loaded onto the respective station 200'.

[0216] The delivery system may cover a geographic area that is divided into zones with corresponding distribution centers and stations 200'. A scheduling device 270 may assign shipments to different distribution centers based on the destinations of the shipments, grouping the destinations into groups based on expected delivery times, and then executing method 250 for each group to determine a launch location 232.

[0217] Figure 23A A plurality of destinations (solid dots) 231 of freight to be delivered by the delivery system and launch locations (open dots) 232 determined by the method 250 for the destinations 231 are illustrated. The method 250 may also assign corresponding subsets of the destinations 231 to the launch locations 232. Figure 23A , the corresponding subset is designated as cluster 240. In some embodiments, step 253 of method 250 includes: running a clustering algorithm on destination 231 to determine a number of clusters 240 equal to the number given in step 252; calculating the geometric center of cluster 240; and determining launch location 232 relative to the geometric center. For example, launch location 232 may be set at or near the geometric center to take into account physical circumstances such as road availability, parking, etc. Launch locations 232 may be assigned to different stations 200', but it is also contemplated that a single station 200' may be assigned to a parking spot at two or more of launch locations 232.

[0218] The clustering algorithm can be configured to identify a selected number of groups among the destinations. For example, any partitioning clustering algorithm can be used, such as k-means, minimum spanning tree (MST), squared error, nearest neighbor, k-medoids, or expectation maximization (EM), or variations, alternatives, or derivatives of these.

[0219] In some embodiments, step 252 may obtain the number of transmit locations as a predefined value, for example from a memory. In some embodiments, the number of transmit locations may be provided to step 252 by an operator, for example as an input to the scheduling device 270. Figure 24 In FIG, the selected number of clusters is designated by #S. By increasing the number of clusters, the total distance traveled by the station 200′ increases, while the total flight path of the UAV decreases. Thus, the operator can adjust the number of clusters to achieve a desired balance between the distances traveled by the station 200′ and the UAV 1, respectively.

[0220] In some embodiments, step 252 runs a clustering algorithm on destination point 231 for different numbers of clusters, determines one or more evaluation parameters for the corresponding clustering results, and selects a number of launch locations based on the one or more evaluation parameters. Then, step 253 may determine launch locations 232 for the selected number of cluster sets generated in step 252. The evaluation parameters may be default parameters of the scheduling device 270 or selected by the operator. In some embodiments, the evaluation parameters may include a first parameter related to the UAV. For example, the first parameter may identify the maximum flight distance, maximum flight time, or maximum energy consumption among the UAVs when assuming the UAV travels from the geometric center of cluster 240 to the destination 231 within the cluster. Such UAV evaluation parameters can be used to ensure that cluster 240 is commensurate with the operational capabilities of the UAVs. If the maximum flight distance / flight time / energy consumption is too high for a cluster set, assigning a UAV to perform delivery according to that cluster set may be inappropriate or even impossible, even if stations in other clusters are used as a trampoline. In some embodiments, the evaluation parameters may include a second parameter related to station 200'. For example, the second parameter may identify the total energy consumption of the stations 200' during their mission, the total distance traveled by the stations 200' during their mission, or the time required for the stations 200' to complete their mission. The mission may begin when the stations 200' leave the distribution center and be completed when the stations return or when all UAVs have been launched. During the mission, one or more of the stations 200' may visit two or more launch locations.

[0221] Figure 23C is a flow chart of an example process for determining the number of transmit locations. As shown, the example process may be Figure 23B252 . In step 261, the current number of clusters (#C) is set to a starting value, such as a value #S entered by an operator or a default value. In one example, the starting value is equal to the maximum number of currently available stations 200 ′. Step 262 runs a clustering algorithm for destination 231 to define #C different clusters. Step 263 calculates the geometric centers of the clusters, and step 264 calculates at least one UAV evaluation parameter based on the geometric centers. If the at least one UAV evaluation parameter is deemed unacceptable, step 265 proceeds to step 266, which increases the number of clusters (#C). When the at least one UAV evaluation parameter is deemed acceptable, step 265 proceeds to step 267, which calculates at least one station evaluation parameter. If the at least one station evaluation parameter is deemed unacceptable, step 268 proceeds to step 266, which increases the number of clusters. When the at least one station evaluation parameter is deemed acceptable, step 268 proceeds to step 253, which may determine the transmit locations of the cluster set generated by the most recent execution of step 262. It can be recognized that Figure 23C The process in enables a balance to be struck between the respective travel parameters of the UAV and the station by setting appropriate evaluation criteria (such as thresholds) for the evaluation steps 265 and 268.

[0222] Below, several clauses are cited to summarize some aspects and embodiments disclosed above.

[0223] Clause 1: A system for launching an unmanned aerial vehicle (UAV) (1) to deliver cargo to a destination, the system comprising:

[0224] a lifting device (202A to 202D, 205, 207) physically separate from the UAV (1) and powered by a power source (204), the lifting device (202A to 202D, 205, 207) being operable to vertically raise the UAV (1) to a selected altitude (HL) relative to a reference horizontal plane prior to launch of the UAV (1), and

[0225] A control system (203) is configured to calculate a selected altitude (HL) and provide corresponding control signals (CS) to the lifting devices (202A to 202D, 205, 207).

[0226] Clause 2: The system of clause 1, wherein the control system (203) is configured to calculate the selected altitude (HL) based on parameter data representing at least one of the UAV (1), the lifting device (202A to 202D, 205, 207), or the ambient conditions between the lifting device (202A to 202D, 205, 207) and the destination.

[0227] Clause 3: A system according to clause 1 or 2, wherein the parameter data includes the location of the lifting device (202A to 202D, 205, 207) and the destination of the UAV (1).

[0228] Clause 4: A system according to any preceding clause, wherein the parameter data comprises one or more of the following: topographic data (TD) of the area including the lifting device (202A to 202D, 205, 207) and the destination, current environmental data (AD) of the area, current weight data (WD) of the UAV (1), current size data (SD) of the UAV (1), current energy content data (ED) of the local power source (1010) in the UAV (1), current energy content data (ES) of the power source (204), and ascent speed data (LS) of the lifting device (202A to 202D, 205, 207).

[0229] Clause 5: The system according to any preceding clause, wherein the control system (203) is configured to calculate the selected height (HL) by evaluating parameter data related to at least one predefined consumption parameter.

[0230] Clause 6: The system of clause 5, wherein the at least one predefined consumption parameter represents a consumption cost of performing the delivery at the destination.

[0231] Clause 7: A system according to clause 5 or 6, wherein the at least one predefined consumption parameter includes one or more of the following: energy consumption of the UAV (1) to deliver the cargo to the destination, total energy consumption including energy consumption of the UAV (1) and energy consumption of vertically raising the UAV (1) to a selected altitude (HL), a time period from a reference time point to the arrival of the UAV (1) at the destination, or a monetary cost of delivering the cargo at the destination.

[0232] Clause 8: The system according to any preceding clause, wherein the lifting device (202A to 202D, 205, 207) is provided on a mobile station (200').

[0233] Clause 9: A system according to any preceding clause, wherein the lifting device (202A, 202B, 207) is operable to vertically raise the platform (201) carrying the UAV (1).

[0234] Clause 10: The system of clause 9, wherein the lifting device (202A, 202B, 207) is configured to vertically raise the platform (201) when the UAV (1) is carried by the platform (201).

[0235] Clause 11: A system according to clause 9 or 10, wherein the control system (203) is further configured to operate the lifting device (202A, 202B, 207) to lower the platform (201) from the selected altitude (HL) to a predefined altitude (H0) after the UAV (1) is launched.

[0236] Clause 12: A system according to any one of clauses 9 to 11, wherein the lifting device includes a conveyor device (207) configured to lift the UAV (1) onto the platform (201) at least when the platform (201) is vertically raised.

[0237] Clause 13: The system of any preceding clause, wherein the lifting device comprises an extendable mechanical connection device (202A).

[0238] Clause 14: The system of any one of clauses 1 to 11, wherein the lifting device comprises an aircraft (202B).

[0239] Clause 15: The system of any preceding clause, wherein the lifting device comprises a flow generator (202D) configured to generate a vertical airflow towards the UAV (1).

[0240] Clause 16: A system according to any one of clauses 1 to 8, wherein the lifting device includes a cable (202C) that is releasably connected to the UAV (1) to supply energy from the power supply (204) to the UAV (1) when the control system (203) operates the propulsion system (1002) of the UAV (1) to vertically raise the UAV (1).

[0241] Clause 17: A system according to any preceding clause, the system being further configured to provide a transmission signal to the UAV (1) when the lifting device (202A to 202D, 205, 207) has raised the UAV (1) to a selected altitude (HL), the transmission signal causing the UAV (1) to be launched towards the destination.

[0242] Clause 18: An apparatus for configuring a delivery system, the delivery system comprising a plurality of systems according to any one of clauses 1 to 17, the plurality of systems being mobile, and the apparatus being configured to perform a method comprising the steps of:

[0243] obtaining (251) a plurality of destinations (231) for delivering goods via a delivery system;

[0244] Obtaining (252) the number of emission positions (232) to be determined;

[0245] determining (253) the number of said transmission locations (232) based on said plurality of destinations (231); and

[0246] One or more of the plurality of systems are directed (254) to the launch location (232).

[0247] Clause 19: An apparatus according to clause 18, wherein the step of determining (253) includes: operating a clustering algorithm for the plurality of destinations (231) to determine a number of clusters (240) equal to the number of the launch locations (232); calculating the geometric centers of the number of clusters (240); and determining the launch location (232) relative to the geometric center.

[0248] Clause 20: An apparatus according to clause 18, wherein the step of obtaining (252) the number of transmission locations comprises: operating a clustering algorithm on the plurality of destinations (231) in a plurality of repetitions, each repetition being configured to result in a different number of clusters (240); evaluating (264 to 268) the clusters (240) resulting in the plurality of repetitions with respect to at least one evaluation parameter to be optimized; and setting the number of transmission locations (232) based on the evaluation steps (264 to 268).

[0249] Clause 21: An apparatus according to clause 20, wherein the at least one evaluation parameter includes: a first parameter representing the operation of one or more UAVs from a launch location (232) to the plurality of destinations (231); and / or a second parameter representing the operation of one or more systems that launch the one or more UAVs.

[0250] Clause 22: The apparatus of clause 21, wherein the first parameter represents a maximum flight distance, a maximum flight time, or a maximum energy consumption among the one or more UAVs when the one or more UAVs travel from a launch location (232) to the plurality of destinations (231).

[0251] Clause 23: The apparatus of clause 21 or 22, wherein the second parameter represents a total energy consumption of the one or more systems, a total distance traveled by the one or more systems, or a time required for the one or more systems to launch the one or more UAVs.

[0252] Clause 24: A method of launching an unmanned aerial vehicle (UAV) to deliver cargo to a destination, the method comprising the following steps:

[0253] calculating (221) a selected altitude relative to a reference horizontal plane for launching the UAV;

[0254] operating (222) the lifting device to vertically raise the UAV to a selected altitude; and

[0255] The (223) UAV is launched from a selected altitude.

[0256] Clause 25: The method of clause 24, wherein the selected altitude is calculated based on at least one of the UAV, the lifting device, or ambient conditions between the lifting device and the destination.

[0257] Clause 26: The method of clause 24 or 25, wherein the parameter data includes the location of the lifting device and the destination of the UAV.

[0258] Clause 27: A method according to any one of clauses 24 to 26, wherein the parameter data includes one or more of the following items: terrain data (TD) of the area including the lifting device and the destination, current environment data (AD) of the area, current weight data (WD) of the UAV, current size data (SD) of the UAV, current energy content data (ED) of the local power source in the UAV, current energy content data of the power source providing power to the lifting device, and ascent speed data (LS) of the lifting device.

[0259] Clause 28: The method of any of clauses 24 to 27, wherein the selected altitude is calculated by evaluating parameter data related to at least one predefined consumption parameter.

[0260] Clause 29: The method of clause 28, wherein the at least one predefined consumption parameter represents a consumption cost of performing the delivery at the destination.

[0261] Clause 30: A method according to clause 28 or 29, wherein the at least one predefined consumption parameter includes one or more of the following items: energy consumption of the UAV to deliver the goods to the destination, total energy consumption including energy consumption in the UAV and energy consumption for vertically raising the UAV to a selected altitude, a time period from a reference time point to the arrival of the UAV at the destination, or a monetary cost of delivering the goods at the destination.

[0262] Clause 31: The method of any one of clauses 24 to 30, wherein the lifting device is operated to vertically raise the platform carrying the UAV.

[0263] Clause 32: The method of clause 31, wherein the lifting device (202A, 202B, 207) is operated to vertically raise the platform while the UAV is carried by the platform.

[0264] Clause 33: The method of clause 31 or 32, further comprising the step of operating the lifting device after the UAV has been launched to lower the platform from the selected altitude to a predefined altitude.

[0265] Clause 34: The method of any one of clauses 31 to 33, wherein the step of operating (222) the lifting device includes operating the conveyor device to lift the UAV onto the platform at least when the platform is vertically raised.

[0266] Clause 35: The method of any one of clauses 24 to 34, wherein the step of operating (222) the lifting device comprises operating an extendable mechanical connection.

[0267] Clause 36: The method of any one of clauses 24 to 33, wherein the step of operating (222) the lifting device comprises operating an aircraft (202B) carrying a UAV.

[0268] Clause 37: The method of any one of clauses 24 to 36, wherein the step of operating (222) the lifting device comprises operating a flow generator to generate a vertical airflow toward the UAV.

[0269] Clause 38: A method according to any one of clauses 24 to 30, wherein the step of operating (222) the lifting device includes: supplying energy to the UAV from a ground-based power source via a cable connected to the UAV while the propulsion system of the UAV is operating to vertically raise the UAV; and wherein the step of causing (223) the UAV to be launched includes: releasing the cable from the UAV.

[0270] Clause 39: A method according to any one of clauses 24 to 38, wherein the step of causing (223) the UAV to be launched includes: when the lifting device has raised the UAV to a selected altitude, providing a transmission signal to the UAV, the transmission signal causing the UAV to be launched to the destination.

[0271] Clause 40: A computer readable medium comprising computer instructions (1102A) which, when executed by a processing system (1101), cause the processing system (1101) to perform the method of any one of clauses 24 to 39.

[0272] 4. Control a group of UAVs to deliver goods

[0273] This portion of the disclosure relates to improving energy efficiency when delivering cargo via UAVs.

[0274] A major challenge in operating UAVs for the purpose of delivering and / or collecting cargo is ensuring sufficient reach to complete their mission given the UAV's available power capacity and / or making the most efficient use of that power capacity. Conventionally, the range of UAV-based delivery systems can be extended through specialized long-range designs for the UAVs, through over-the-air charging, or through cargo transfers between UAVs. Each of these measures increases the cost and / or complexity of the delivery system.

[0275] Figure 25A An example delivery system is illustrated that includes a fleet of UAVs 1. The delivery system includes a control system 301 that is configured to communicate with the UAVs 1 via a wireless network 302. The control system 301 can be configured to assign tasks to the UAVs 1 in the fleet, where the tasks can include delivering and / or picking up cargo 4 at one or more destinations, as well as other operations involving repositioning of the UAVs (such as taking photos, performing measurements, etc.).

[0276] Energy efficiency of a single UAV or multiple UAVs can be improved by operating the UAVs in groups or "swarms." UAVs in a swarm fly in the same direction and at the same speed. The UAVs can be arranged into such a swarm for at least a portion of their respective missions, for example, as long as their respective destinations are approximately in the same direction. UAVs can leave and join the swarm during flight. Furthermore, a UAV swarm can be broken up into smaller groups, or multiple groups can be merged to form a larger group.

[0277] Figures 25B to 25C Two non-limiting examples of group formations ("flying formations") that can be implemented by a group of UAVs 1 are shown. Figure 25B In , the group has a linear formation, wherein the UAVs 1 are arranged one after another along the flight direction DIR of the group. Figure 25C In FIG, the group has a triangular formation, wherein the UAVs 1 are arranged to form an arrow shape along the flight direction DIR. Figure 25C It can be a plan view or an elevated side view. The group formation does not need to be two-dimensional (as shown), but can be three-dimensional.

[0278] Embodiments relate to techniques for configuring a swarm of UAVs, and more particularly, the relative positions of one or more UAVs in a swarm, to improve the performance of a single UAV in the swarm, the performance of a subset of UAVs in the swarm, or the overall performance of the swarm. The relative positions of the UAVs may be changed by changing the distance and / or direction between the UAV and one or more other UAVs in the swarm, and / or by interchanging the position of the UAV with another UAV in the swarm, and / or by changing the swarm formation.

[0279] Figure 26 is a flow chart of an example method 310 for controlling a group of UAVs that are performing one or more tasks of delivering and / or picking up cargo and are traveling in a common direction (see FIG. 25A to FIG. 25B The example method 310 may be performed by the control system 301 or as described below with reference to Figures 31A to 31C As described above, the monitoring is performed by one or more UAVs in the group. Step 311 obtains drag data indicating the air resistance of the set of UAVs in the group. Such a set of UAVs includes one or more UAVs and is hereinafter referred to as a "monitored set". Figures 29A to 29C As further described, the drag data may include a measured value or a predefined value or both. In some embodiments, step 311 may calculate a parameter value representing air resistance based on the drag data. Step 312 determines the relative position of each UAV in the monitored set based on the drag data or the parameter value (if calculated by step 311). In step 313, each UAV in the monitored set is controlled to reach its relative position.

[0280] Figure 30A The forces acting on an example UAV are illustrated, wherein the body of the UAV 1 includes at least two rotors or other spatially separated propulsion components 6, which are operated to generate respective thrusts FT1, FT2. In order for the UAV 1 to maintain its altitude, the vertical components of the thrusts FT1, FT2 need to be equal to the weight force FW acting on the UAV 1. The weight force FW corresponds to the total weight of the UAV, including any payload carried by the UAV. In order to move the UAV 1 in the flight direction DIR, the rotors 6 need to generate a horizontal component of thrust in that direction. As can be seen, this can be achieved by increasing the thrust FT2 relative to the thrust FT1, thereby causing the UAV 1 to tilt in the flight direction DIR. When the UAV 1 is moving at a constant speed, the horizontal component of the total thrust is equal to the force FD generated by the air resistance, commonly expressed as "drag" or "drag". The drag force FW is a function of the size and shape of the UAV 1 and the size and shape of the payload 4, if any (cf. Figure 25A ). The drag force FW is also a function of the current absolute travel speed of the UAV 1, as well as the current direction and absolute speed of the wind. As used herein, "absolute speed" is given relative to a fixed reference and can be ground speed. Furthermore, as used herein, the "airspeed" of the UAV refers to the sum of the absolute travel speed of the UAV in its current direction and the absolute wind speed in the direction of flight, where wind speeds opposite the direction of flight are negative.

[0281] It has been recognized that the drag force FD can have a significant impact on the energy consumption of the UAV 1. A UAV carrying a payload with a large cross-section and low weight can have the same total energy consumption as the same UAV carrying a payload with a small cross-section and high weight. In the context of a swarm, wind speed and possibly also wind direction depend on the position of the UAV relative to the other UAVs in the swarm. Therefore, since the drag force FW is a function of wind speed / direction, one or more UAVs can be selectively repositioned relative to the other UAVs in the swarm based on their actual or estimated drag force. In one example, the above-mentioned UAV carrying a payload with a large cross-section can be positioned to act as a "wind shield" for the other UAVs in the swarm, thereby significantly reducing the energy consumption of the other UAVs. In another example, in terms of energy consumption, the above-mentioned UAV carrying a payload with a large cross-section will significantly benefit from being positioned behind one or more other UAVs that provide a wind shield. It is thus recognized that, for example, according to the example method 310, significant improvements in energy consumption, range, flight time, etc. can be achieved by adjusting the relative positions of one or more UAVs in a group based on estimated drag.

[0282] like Figure 26 As shown, the determination in step 312 can be performed to achieve the efficiency target. In some embodiments, step 312 can involve: optimizing the respective relative positions of the UAVs in the monitored set to meet the efficiency target. The efficiency target of the group can change over time and can be set by the control system 301 or one of the UAVs in the group or the UAVs. In one example, the efficiency target is set to minimize the air drag of at least one UAV in the group, and the at least one UAV may or may not be included in the monitored set. In another example, the efficiency target is set to minimize the energy consumption of at least one UAV in the group. In another example, the efficiency target is set to minimize the flight time of at least one UAV in the group, for example, by increasing the speed of the group or by causing the at least one UAV to operate at an increased speed after leaving the group.

[0283] The example method 310 may include an optional step 311B of obtaining current state data of the group; and a step 312 of determining the respective relative positions of the UAVs in the monitored set based on the drag data and the current state data. For example, the current state data may indicate the number of UAVs in the group, the absolute speed of travel of the group, the total weight of one or more UAVs, the local power supply of one or more UAVs (see Figure 1The remaining energy level of the UAVs, the model identifier of the one or more UAVs, the destination of the one or more UAVs, or any combination of these. By including the current state data in the determination made in step 312, the group can be optimized more accurately.

[0284] FIG. 27A to FIG. 27B A simplified example of such use of current state data is shown. Figure 27A In FIG, a UAV 1 carrying a large payload 4 and having a large residual energy level (indicated by icon 315) is placed at the front of the group to act as a wind shield for the other UAVs. Figure 27B In the group, the UAV 1 carrying the large payload 4 has lower energy (as shown by icon 315) and is therefore placed at the tail of the group to benefit from the wind shield provided by the other UAVs 1.

[0285] In some embodiments, steps 312 and 313, and optionally steps 311 and / or 311B, may be performed continuously or intermittently for the group. This enables the group to take into account changes in external conditions (such as wind speed / direction), changes caused by one or more UAVs joining and / or leaving the group, and changes in the group's current state data mentioned above.

[0286] Figure 28 is a flow chart of an example process 320 for arranging UAVs into a swarm. The example process 320 may be performed intermittently during swarm travel (e.g., upon initial formation of the swarm) and whenever the primary structure of the swarm changes (e.g., when the number of UAVs in the swarm changes). In step 321, a flight formation for the swarm is determined. In some embodiments, the flight formation is determined to minimize air resistance for the swarm as a whole, taking into account wind direction at the location of the swarm. Alternatively or additionally, the flight formation may be determined based on the airspeed of one or more UAVs. Alternatively or additionally, the flight formation may be determined based on the number of UAVs in the swarm. In general, the flight formation may be determined based on drag data, current state data, or any combination thereof. In some embodiments, the flight formation is selected from a set of predefined flight formations, which includes, for example, linear formations ( Figure 25B ) and one or more arrow formations ( Figure 25C In a non-limiting example, if the group encounters a strong headwind, step 321 may select a linear formation, while if the group encounters a strong crosswind, an arrow formation may be selected.

[0287] In some embodiments, a predefined flight formation is associated with a corresponding flight template, which may include a predefined spatial arrangement of relative positions ("slots"), wherein each such slot is designated for a UAV. In some embodiments, the flight template may also associate an efficiency rating with the corresponding slot. The efficiency rating is a relative parameter that ranks the slots by air resistance under nominal conditions (e.g., assuming all UAVs are identical and carry no payload). A high efficiency rating may mean low air resistance. For example, a slot at the front of the flight formation is likely to have a low efficiency rating. Back Figure 28 , step 322 may obtain a flight template and associated efficiency rating for the flight formation selected in step 321. Step 323 may obtain travel capability data for one or more UAVs. For the respective UAVs, the travel capability data may include an estimate of the remaining flight range. In some embodiments, the remaining flight range may represent the flight distance (or equivalently, the flight time) that remains before the UAV's local power supply becomes empty or reaches a low energy limit. Below, the remaining flight range is designated as ERFD ("estimated remaining flight distance"). For example, the ERFD may be calculated continuously or intermittently by or for the respective UAV based on historical or current power consumption and the current energy level of the local power supply. The historical or current power consumption may be compensated for or replaced by a calculation based on drag data and status data (such as total frontal area, gross weight, airspeed, etc.). It should be noted that regardless of the calculation method, the ERFD is a composite parameter that inherently includes the drag data and status data of the UAV. Specifically, the instantaneous change in ERFD during UAV group flight is proportional to the change in air drag, since all other variables affecting ERFD can be assumed to be constant. In step 324, UAVs are assigned to slots so that the ERFD matches the efficiency class of the slot. For example, this matching step may involve assigning UAVs with higher ERFD to slots with lower efficiency classes.

[0288] It should be understood that other definitions of the remaining flight range are possible. In one example, the remaining flight range may represent the flight distance (or equivalently, flight time) that remains until the UAV has completed its mission (e.g., arrived at its destination to perform pickup / delivery of cargo). Such a remaining flight range can be used to optimize the monitored set so that the corresponding included UAVs can complete their missions.

[0289] Process 320 generates a basic group configuration that is optimized for flight distance or flight time based on a general estimate of the air drag of the respective UAVs. By applying this basic group configuration to the group, the magnitude and number of variations between the UAVs in the group during successive flights will be reduced (e.g., based on method 310). This, in turn, will reduce the complexity of controlling the UAVs during group flight.

[0290] In other embodiments, the efficiency rating and travel capability data may be omitted, and step 324 assigns the UAV to the vacant slot based on some other logic or randomly.

[0291] Back to Figure 26 In step 311 of the embodiment, the drag data may include predefined values ​​for respective UAVs in the measured set. The predefined values ​​may be pre-estimated air drag values ​​("drag values") for the UAV 1 and any payload 4 currently being carried. For example, UAVs may be classified into different UAV types based on air resistance, and payloads 4 may be classified into different payload types based on air resistance. Thus, based on the UAV type and payload type, a drag value may be estimated for the respective UAV.

[0292] Figure 29A Flowchart of an example process for determining a drag value for a UAV based on drag data including one or more predefined values. As shown, the example process may be part of step 311. Step 331 obtains an identifier ("cargo ID") for a payload carried by the UAV. The cargo ID may directly or indirectly indicate the type of payload mentioned above. The cargo ID may be obtained from a dispatch device in the delivery system, or may be read from the payload by the UAV, such as by a code reader mentioned above (see Figure 11020 in). Step 332 determines the frontal area of ​​the UAV in combination with its payload based on the cargo ID. Step 332 can be performed by using a lookup table or by an algorithm. It can be noted that the frontal area can be supplemented by data on the 2D or 3D shape of the payload 4 and / or other aerodynamic characteristics of the payload. Step 333 estimates the air speed of the UAV. The air speed can be given based on the absolute travel speed of the UAV (and therefore based on the group) and an estimate of the wind speed and direction. The wind speed and direction can be given based on the weather forecast for the location of the group, optionally modified based on the slot occupied by the UAV. Alternatively or in addition, the wind speed / direction can be wind data compiled based on wind measurements performed by a large number of UAVs, which may or may not be included in the fleet. Such a UAV can continuously measure local wind speed and direction and report the local wind speed and direction, along with 3D coordinates, to a wind server, which is configured to compile the reported data into a wind map, e.g., a 3D array layered on a map. The wind server can also generate wind maps for areas where the amount of reported data is insufficient by using a learning model (e.g., a convolutional neural network), which is trained based on wind maps compiled for different regions along with a 3D model of the region, as well as the wind and weather conditions of the region. The wind server can then use the trained learning model to estimate the wind map for any region based on the 3D model of the region and the wind and weather conditions of the region. Such a wind server can be queried to obtain wind data in any of the concepts, aspects, and embodiments described herein. Given the frontal area of ​​the payload and the estimated airspeed, and assuming that at least the frontal area of ​​the UAV is known, step 334 can determine the drag value of the UAV.

[0293] In step 311, the drag data may include values ​​measured by or for the UAV in addition to or in place of the predefined values. Figure 29A In the example process, step 333 may estimate the airspeed of the UAV by obtaining measurement data from a wind sensor on the UAV or another UAV in the swarm.

[0294] Figure 29B Flowchart of another example process for determining the drag value of the UAV based on the measured value. As shown, the example process can be part of step 311. Assume that the UAV includes two or more spatially separated propulsion components (refer to Figure 30A The rotor 6 in the rotor 6), step 341 measures the distribution of propulsion energy between the propulsion components, and step 343 calculates the current air resistance based on the distribution. Figure 30AAs can be understood from the description of FIG, the difference in propulsion energy between the rotors 6 represents the energy spent on compensating for air resistance. When the acceleration is zero or close to zero, step 341 can be performed. Alternatively, the process can include step 342 of measuring the acceleration, and when calculating the air resistance, step 343 can compensate for the measured acceleration.

[0295] Figure 29C is a flow chart of another example process as part of step 311. Step 351 measures the orientation of the UAV relative to a reference plane. Figure 30B An example of such a measurement is shown, where the orientation is given by the inclination of the UAV 1 along the flight direction DIR, e.g., the inclination being represented by the angle θ between a reference direction RD of the UAV 1 and a vertical reference plane RP facing DIR. In general, the orientation may be represented by more than one angle (e.g., the conventional angles of pitch and roll of the UAV), such as those provided by one or more sensors on the UAV (referring to Figure 1 When the acceleration is zero or close to zero, or taking into account the current acceleration measured by step 352, step 353 calculates the current air resistance based on the orientation.

[0296] In some embodiments, the methods and processes described above are performed by Figure 25A The control system 301 in FIG. Figure 26 In the method 310 of FIG. 3 , the control system 301 may obtain (step 311) drag data and optionally obtain (step 311B) status data for respective UAVs in the monitored set, and send (step 313) control data to the swarm to cause the respective UAVs to reach their relative positions, as determined by the control system 301 (step 312). As previously understood, the control system 301 may or may not communicate with the swarm to obtain the drag data.

[0297] In some embodiments, at least a portion of the methods and processes are performed by respective UAVs in the monitored set. For example, Figures 29A to 29C One or more of the processes in can be performed by the corresponding UAV, and the final drag value can be sent to the control system 301.

[0298] In some embodiments, the methods and processes are performed by at least one of the UAVs in the swarm. For example, at least a portion of the methods and processes can be performed by a master UAV among the UAVs in the swarm. The master UAV can thus serve as the control system 301 and be configured to at least transmit control data to the UAVs in the swarm. The master UAV can also be configured to receive UAV data from the slave UAVs, such as drag data and / or status data. Figures 31A to 31BTwo examples of communication paths within a group are shown. Figure 31A In , the master UAV 1M communicates directly with the other UAVs in the group. Figure 31B In the embodiment of the present invention, the UAVs are configured to form a series of communication links between pairs of UAVs, wherein a corresponding UAV can pass at least a portion of the incoming data to the other UAV.

[0299] In some embodiments, a leader UAV is dynamically selected among the UAVs in the swarm, for example, by performing a voting process among the UAVs in the swarm. The voting process can be based on any conventional algorithm for leader election, such as those practiced in mesh networks. The voting process can be performed at the initial formation of the swarm and / or before the leader UAV leaves the swarm.

[0300] In other embodiments, the master UAV is pre-selected, for example, by the control system 301. The master UAV may be a UAV that is known to be included in the group from the start to the final destination.

[0301] In some embodiments, the methods and processes are collectively performed by UAVs in a swarm and operate the UAVs to interchange positions in the swarm in pairs based on UAV data exchanged between the UAVs. Figure 31C An example is shown in which two UAVs 1 exchange UAV data (as indicated by dashed arrows) and then swap positions (as indicated by solid arrow 360). It should be understood that UAV data can be exchanged between pairs of UAVs in other ways (e.g., via one or more other UAVs in the group). The UAV data can include drag data and optionally state data. In some embodiments, the UAV data can include travel capability data, which can be given by the remaining flight range mentioned above. Below, the use of travel capability data will be illustrated for ERFD. This description is also applicable to other definitions of travel capability data.

[0302] In some embodiments, UAV data can be used to update a group configured according to a flight template associated with efficiency levels. In one example, position swaps are determined based on the ERFD of the respective UAVs and the efficiency level of the slot currently occupied by the respective UAVs. This decision can be made by the paired UAVs, the master UAV, or the control system 301. Swapping can be performed to maximize the difference between the ERFD and the efficiency level of the respective slots. Therefore, if one of the UAVs has both a higher ERFD and a higher efficiency level than the other, the UAVs swap positions. In other words, when paired UAVs swap positions, if the UAVs' combined ERFD is deemed to be higher, a swap is performed. When such swaps are performed throughout the group, the UAV with the highest ERFD will be positioned to maximize air resistance, while the UAV with the lower ERFD will fly within a windshield. The result is increased efficiency, thereby extending the group's overall flight range. Additional information can be exchanged and used to determine swaps. For example, the additional information may include the planned flight distances of the respective UAVs, allowing the interchange to optimize the group to ensure that all UAVs can complete their planned flight distances. It should be noted that during flight, the ERFD of the respective UAVs may decrease, so it may be beneficial for the UAVs to repeatedly exchange UAV data and decide to interchange.

[0303] Interchanges based on UAV data exchange can be applied to achieve other efficiency goals. For example, an interchange can be determined to maximize the ERFD of one or more selected UAVs in the group, so that the selected UAV is placed in the slot with the highest efficiency level.

[0304] Back to Figure 26 In the example method 310 of FIG. 3 , step 312 may also consider upcoming events when determining the relative positions of the corresponding UAVs in the monitored set. In some embodiments, based on the presence of at least one UAV (e.g., Figure 32 1L in FIG) leaves the group, and / or at least one UAV (such as Figure 32 The relative positions can be determined based on the predicted time point when UAV 1L (shown as UAV 1A in the figure) arrives and joins the group. For example, UAV 1L and / or other UAVs in the group can be repositioned to facilitate UAV 1L's departure from the group. Similarly, UAVs in the group can be repositioned to provide a suitable position for the joining UAV 1A in the group, for example, based on the ERFD of the joining UAV 1A.

[0305] Below, several clauses are cited to summarize some aspects and embodiments disclosed above.

[0306] Clause 1: A method of controlling unmanned aerial vehicles (UAVs) (1), said UAVs being organized in groups to perform one or more tasks of delivering and / or collecting cargo (4), said method comprising the following steps:

[0307] obtaining ( 311 ) drag data indicative of air resistance of one or more UAVs in the group;

[0308] determining (312) respective relative positions of at least the one or more UAVs within the group based on the drag data; and

[0309] At least controlling (312) the one or more UAVs to reach respective relative positions.

[0310] Clause 2: The method of clause 1, wherein the drag data comprises at least one of: a predefined value of the one or more UAVs, or a value measured by or for the one or more UAVs.

[0311] Clause 3: The method according to clause 2, wherein the predefined value is obtained based on cargo carried by the one or more UAVs (1).

[0312] Clause 4: The method according to clause 2 or 3, wherein the predefined value indicates the frontal area of ​​the one or more UAVs (1) when carrying its cargo (4).

[0313] Clause 5: A method according to any one of clauses 2 to 4, wherein the measured value is at least one of: the travel speed of the one or more UAVs, the airspeed at the one or more UAVs, the wind speed and / or wind direction, the acceleration of the one or more UAVs, the orientation of the one or more UAVs relative to a reference plane (RP), or the distribution of propulsion energy to spatially separated propulsion components (6) on the one or more UAVs.

[0314] Clause 6: A method according to any preceding clause, wherein the step of determining (312) the respective relative positions and the step of controlling (313) are performed continuously or intermittently when the UAVs (1) are organized in groups.

[0315] Clause 7: A method according to any preceding clause, the method further comprising: obtaining (311B) current status data of the group, wherein the respective relative positions are determined based on the drag data and the current status data, and wherein the current status data includes one or more of the following items: the total weight of the corresponding UAV, the count of UAVs in the group, the speed of the corresponding UAV, the remaining energy level of the power supply (1010) of the corresponding UAV, the model identifier of the corresponding UAV, or the destination of the corresponding UAV.

[0316] Clause 8: A method according to any preceding clause, wherein the step of determining (312) the respective relative positions comprises optimizing the respective relative positions of at least the one or more UAVs within the group based on the drag data so as to meet an efficiency target.

[0317] Clause 9: The method of clause 8, wherein the efficiency goal comprises minimizing one or more of air drag, energy consumption, or flight time for at least one UAV in the group.

[0318] Clause 10: The method according to any preceding clause, further comprising: determining (321, 322) a flight formation of the group.

[0319] Clause 11: The method of clause 10, wherein the flight formation is selected from a set of predefined flight formations.

[0320] Clause 12: A method according to clause 10 or 11, wherein the flight formation is associated with a template of predefined locations, wherein the method further comprises the step of: selectively assigning (324) at least a subset of the predefined locations to at least the one or more UAVs.

[0321] Clause 13: The method according to clause 12, further comprising the steps of obtaining (323) travel capability data for at least the one or more UAVs, wherein the predefined locations have corresponding efficiency levels, and wherein at least the one or more UAVs are selectively assigned to predefined locations so that the travel capability data matches the efficiency levels of at least a subset of the predefined locations.

[0322] Clause 14: A method according to any one of clauses 1 to 12, wherein the step of obtaining (311) drag data includes: obtaining travel capability data of at least the one or more UAVs; wherein the corresponding relative positions are determined based on the travel capability data.

[0323] Clause 15: The method according to clause 14 further comprises the following steps: repeatedly updating the travel capability data; wherein the step of determining (312) the corresponding relative positions and the step of controlling (313) include: selectively causing one or more pairs of UAVs to interchange positions in the group based on the updated travel capability data.

[0324] Clause 16: A method according to clause 14 or 15, wherein the step of obtaining (311) drag data includes: causing each UAV among the one or more UAVs to determine its travel capability data, and causing the one or more UAVs to exchange travel capability data.

[0325] Clause 17: A method as described in any of clauses 13 to 16, wherein the travel capability data indicates a remaining flight range of the one or more UAVs at a defined point in time.

[0326] Clause 18: A method according to any one of clauses 1 to 16, wherein the step of obtaining (311) drag data, the step of determining (312) the corresponding relative position and the step of controlling (313) are performed by the master UAV (1M) in the group.

[0327] Clause 19: A method according to any preceding clause, wherein the respective relative positions are further determined based on a predicted point in time for at least one UAV (1A, 1L) to leave or join the group.

[0328] Clause 20: A control system configured to perform the method of any one of clauses 1 to 19 and comprising communication means (1103) for communicating with at least one UAV in the group.

[0329] Clause 21: An unmanned aerial vehicle (UAV) configured to be organized into a group with at least one other UAV and to perform the method according to any one of clauses 1 to 19, the UAV comprising a communication device (1013) for communicating with the at least one other UAV.

[0330] Clause 22: A computer readable medium comprising computer instructions (1102A) which, when executed by a processing system (1101), cause the processing system (1101) to perform the method of any one of clauses 1 to 19.

[0331] 5. Develop a system for delivering goods via UAVs

[0332] This portion of the disclosure relates to a technique for configuring a delivery system to improve the resource efficiency and / or flexibility of the delivery system.

[0333] One problem that arises in systems that deliver goods via UAVs is that the recipient may not be present to receive the goods. Often, leaving the goods at the recipient's location (e.g., outside their house) is not feasible. Conventionally, deliveries must be planned so that the UAV has sufficient energy to return the goods to the point of origin, such as a warehouse. Furthermore, UAV deliveries involve complex route / resource planning issues to optimize the utilization of the UAV fleet.

[0334] This part of the disclosure is based on the following insight: utilization and flexibility can be improved by introducing mobile storage points (also referred to herein as "mobile storage cabinets (nest)"), which can be used as intermediate storage facilities, and goods can be stored in these intermediate storage facilities for a shorter period of time. The introduction of mobile storage cabinets provides better resource efficiency, because the mobile storage cabinets can be positioned to facilitate the operation of the delivery system. For example, a UAV can return goods to a mobile storage cabinet instead of a warehouse, and then perform another delivery, thereby allowing the final delivery of the goods to be performed by the UAV or other UAVs at a later time. In addition, the provision of storage cabinets can extend the distance over which UAV-based delivery goods are delivered, because one UAV can transport goods to the storage cabinet, and another UAV can collect the goods for transportation to the recipient or another storage cabinet. Storage cabinets can also be used as interchange points between different delivery systems, for example, for handing over goods from one provider to another.

[0335] Figures 33A to 33C An example delivery system configured according to an embodiment is illustrated. The delivery system comprises: one or more central storage facilities 400 (one central storage facility is shown), for example, a warehouse; and one or more local storage facilities ("fixed storage cabinets") 401 (one fixed storage cabinet is shown). The storage facilities 400, 401 are fixed and therefore have fixed positions 400', 401' in a common coordinate system (for example, given by (x, y, z) coordinates as shown). The delivery system also includes a mobile storage cabinet 402 (one mobile storage cabinet is shown) with a variable position in the common coordinate system. In general, the delivery system can be regarded as comprising a "first storage point" 400, 401 with a fixed position, and a "second storage point" 402 with a variable position. The delivery system can also include a fleet of UAVs 1.

[0336] Figure 34A 4 is a flow chart of an exemplary method 410 for transporting goods ("items") in this type of delivery system. In this example, it is assumed that at the beginning of the method 410, the items 4 to be delivered are located at the warehouse 400, such as Figure 33A In step 411, one or more mobile storage cabinets 402 are moved to corresponding selected static positions 402' ( Figure 33A In step 412, the ordered items for delivery are transported from the warehouse 400 to the fixed storage cabinet 401 and the mobile storage cabinet 402 at one or more time points, such as Figure 33B Any suitable vehicle 405 may be used for such transportation, for example, a truck, a van (as shown), a car, a ship, a UAV, etc. In step 413, the item 4 is delivered to the recipient from the fixed storage cabinet 401 and the mobile storage cabinet 402 by the UAV 1, and may be delivered to the recipient from the warehouse 400, as shown. Figure 33C It is recognized that the resting position 402' of the corresponding mobile storage cabinet 402 can be changed to achieve various purposes, such as increasing the capacity of the delivery system, extending the range of the delivery system, reducing energy consumption, increasing flexibility, etc.

[0337] In this example method 410, first control described one or more mobile storage cabinets 402 to move to corresponding static position 402 ', then the article to be delivered is transported to the mobile storage cabinet 402 and the fixed storage cabinet 401 that are in static position 402 '. Therefore, for example, based on the estimated capacity demand, mobile storage cabinets 402 can be well distributed in advance. The estimation can be made based on historical data and / or by preliminary requests for delivery or collection by users using the delivery system. An advantage of the time separation between storage cabinet distribution and goods distribution is that goods can be delivered to storage cabinets during adverse conditions (such as major events, disasters or traffic jams). For example, if goods are transported to storage cabinets 401, 402 by freight drones from warehouse 400, then even if, for example, there is traffic jams during peak hours, goods can be efficiently delivered to storage cabinets 401, 402. In high-risk areas for disasters, such as earthquake zones or quarantine areas, mobile storage cabinets 402 can be distributed and used as intermediate storage points to enable UAV-based delivery of supplies and medicines that would be complicated or reach the at-risk areas by ground transportation.

[0338] In an alternative embodiment of the example method 410 , at least some of the items to be delivered are transported by the mobile storage cabinet 402 to a rest position 402 ′.

[0339] The fixed storage cabinet 401 and the mobile storage cabinet 402 can be used as an intermediate storage point for items near the final delivery point of the items, where the items will be received / retrieved by the consignee. For example, the storage cabinets 401 and 402 can be placed in an unused location, such as on the roof of a building or in an unused parking space.

[0340] Items may be transported to lockers 401, 402 using an optimal delivery method, for example based on one or more of the following: distance, weight of items, number of items, value of items, urgency, cost, speed of delivery, and available resources for transportation (including vehicles and staff).

[0341] The storage cabinets 401, 402 may be configured to ensure tamper-proof storage of items both for items delivered by the vehicle 105 and for items that may be dropped off by the UAV for intermediate storage. Tamper-proofing may be provided by physical barriers or by a monitoring system that alerts the operator to unauthorized access. The storage cabinets may also include a charging station that allows one or more UAVs to land and charge the UAV's local power source (see Figure 1 1010) for recharging.

[0342] In some embodiments, storage cabinets 401, 402 can be used for intermediate storage of items until one of the following events occurs: the delivery system detects that a consignee (e.g., a recipient) is ready to receive the item, the UAV on the storage cabinet 401, 402 is recharged and operable to deliver the item, or a new UAV arrives at the storage cabinet 401, 402 to continue the delivery of the item.

[0343] In some embodiments, once an item is stored in a locker 401, 402 within range of the final delivery point, the consignee can be notified about the item and allowed to schedule delivery, depending on the availability of a UAV to perform the final delivery.

[0344] Below, an embodiment of determining the stationary position ("second storage point") of the corresponding mobile storage cabinet 402 will be described. For clarity of presentation, these embodiments are illustrated with respect to delivering items to a delivery point, which may represent the final delivery point selected by the consignee for delivery, or the location of another first storage point or second storage point in the delivery system. However, it can be noted that the embodiments are also applicable to collecting items at a collection point, which may represent the location selected by the consignor (for example, a user who wants to collect a small package) for collection, or the location of another first storage point or second storage point in the delivery system. Generally, regardless of the direction in which the UAV transports the items relative to the first storage point and the second storage point, the delivery point and the collection point are both "destination points" of the UAV.

[0345] The second storage point 402 can be a road-based vehicle, such as Figures 33A to 33CAs shown. Typically, the second storage point 402 is any storage facility that can be freely positioned and repositioned at any stationary position within the common coordinate system. Thus, the second storage point 402 can be a container that is unloaded from a road-based vehicle or a trailer that is disconnected from the road-based vehicle at a selected stationary position and is later retrieved by this or another road-based vehicle. The first storage points 400, 401 can be any storage facility that cannot be freely positioned and has a predefined position at least at one or more predefined points in time. Thus, the corresponding first storage point can be any type of permanent structure (such as a building), or a structure with limited mobility located on a vehicle (such as a train or bus). For example, a first storage point can be formed based on the stopping of a train or bus at a predefined point in time. Typically, the first storage points 400, 401 are associated with a first position known to the delivery system at any given point in time.

[0346] Figure 34B is a flow chart of an example method 420 for configuring a delivery system including the first and second storage points mentioned above. Step 421 obtains location data for a plurality of delivery points to which items are to be delivered. Step 422 runs a clustering algorithm on the location data to determine clusters of delivery points. Any density-based clustering algorithm or partitioning clustering algorithm may be used. In some embodiments, the clustering algorithm may be configured to divide the delivery points into any number of clusters given by the clusters themselves. In other embodiments, the number of clusters may be predefined for the clustering algorithm. In some embodiments, such a predefined number of clusters may be set relative to the number of second storage points currently available. Step 423 evaluates the clusters relative to the first position of the first storage point to determine a set of stationary positions ("stop positions") of at least one second storage point.

[0347] The method 420 provides a technique for adapting the capacity of a delivery system to the current demand for delivering shipments in a resource efficient manner by considering the first location of a first storage point when determining the set of stationary locations of a second storage point.

[0348] In some embodiments, the set of stationary locations is determined based on an evaluation function that evaluates resource efficiency of assigning corresponding destination points in a cluster to a reference point for the cluster relative to assigning corresponding destination points to one of the first locations. Such an evaluation function can evaluate resource efficiency based on energy consumption, travel distance, delivery time, etc.

[0349] Figure 35AAn example delivery system is shown having six first storage points 400, 401, which are composed of a warehouse 400 and five fixed storage cabinets 401 distributed in an area around the warehouse 400. The delivery system also includes four second storage points 402, which are stationed at the warehouse 400 when not in use. The black dots represent multiple delivery points 403 to which items are to be delivered by UAVs. Figure 35A Also illustrated are four clusters C1 to C4 that have been generated by step 422. It is recognized that step 423 can be performed in different ways to evaluate the clusters C1 to C4 relative to the first storage point 400, 401 for determining the rest position. Figures 34C to 34D as well as Figures 35B to 35E Some non-limiting examples are given.

[0350] Back to Figure 34B , the example method 420 may include the following step 424: assigning one or more of the first storage points 400, 401 to a corresponding first subset of delivery points, and assigning one or more stationary locations of the second storage points 402 to a corresponding second subset of delivery points. All first subsets and second subsets may be mutually exclusive. In this context, "assignment" infers that the first storage point / second storage point is associated with a delivery point of the corresponding first subset / second subset and will be used as a starting point for delivering goods to the delivery points via one or more UAVs. The example method 420 may also include the following step 425: assigning the UAV 1 to perform delivery of goods from the corresponding first storage point 400, 401 to its associated delivery point 403, and to deliver goods from the corresponding stationary location of the second storage point 402 to its associated delivery point 403. Based on the assignment in step 424, the item can be transported to the one or more second storage points 402 at the stationary locations determined in step 423 via the vehicle 105. Based on the assignments made in step 425, UAVs in the UAV fleet may be scheduled to perform the corresponding deliveries. Some of the UAVs may be transported with the items to the second storage location 402. Other UAVs may be scheduled to fly to a stationary location at a specific point in time to perform one or more deliveries.

[0351] Figure 34C 4 is a flow chart of an example process for evaluating the clusters generated by the clustering in step 422. As shown, this example process can be part of step 423. Step 430 determines the cost value of performing a delivery to each delivery point. The cost value is determined by assuming that the delivery is performed via a reference point of the cluster to which the delivery point belongs. Figure 35AIn FIG, the reference point is represented as C1′ to C4′ in the clusters C1 to C4. The reference point can be set at or near the center point of the cluster. The center point can be calculated relative to the included delivery points, for example as an average of the locations of these delivery points. Therefore, the cost value of the delivery takes into account at least the cost of the UAV transporting the item from the reference points C1′ to C4 to the delivery point 403, as shown in FIG. Figure 35A As shown by the arrows in . The cost value may also take into account, for example, the transport of items and / or the second station 402 from the central warehouse 400 to the reference points C1' to C4', and / or the transport of the UAV 1 to the respective reference points C1' to C4'. The cost value does not have to be a monetary value, but may be given in any type of unit. In some embodiments, the cost value may be calculated based on the distance in a common coordinate system and / or based on the estimated energy consumption. In this context, at least when transport by road is involved, the distance may be determined to take into account the availability of roads and the speed limits of these roads, for example to meet the time limit for reaching the respective reference points C1' to C4'. The cost value may also be calculated to take into account "total costs", such as the costs of the driver at the second storage point, parking fees at the second storage point, etc. Such total costs may be specific to each cluster and may be divided across all delivery points associated with the cluster.

[0352] The process can then use the cost value to determine the second storage point 402 to be used for delivering the goods, as well as the resting location of the first storage points 400, 401. In some cases, this determination can result in only the first storage point 400, 401 or only the second storage point 402 being used. Figure 34C In the example of , the determination involves the following step 431: based on the cost value, identifying and selecting a delivery point that is not suitable for delivery via the second storage point 402, or equivalently, a delivery point that is more suitable for delivery via the first storage points 400, 401. For example, step 431 can select all delivery points whose cost value exceeds a cost threshold CTH. These selected delivery points are hereinafter referred to as "first storage delivery points" (FSDP). CTH can be set to represent the nominal cost of performing delivery via a first storage point in the first storage points 400, 401. It is conceivable that CTH is specific to each cluster C1 to C4, or even specific to each delivery point. In one embodiment, in order to set the CTH of each delivery point, step 431 can determine a "candidate cost value" for delivery to the delivery point via each first storage point in the first storage points 400, 401, and set CTH according to the candidate cost value. For example, CTH can be set to be associated with the minimum value among the candidate cost values. Thus, step 431 will identify all delivery points that are more cost-effective for delivering the item via the first storage point 400, 401.

[0353] Then, after removing the FSDP from the cluster, step 432 determines the set of remaining clusters in the cluster, wherein each such remaining cluster includes at least one delivery point. Then, step 433 determines the resting position based on the remaining clusters, for example, by setting the resting position at the reference point of the corresponding remaining cluster. By comparing Figure 35A and Figure 35B , we can see the effect of steps 431 to 433, Figure 35B Illustrated in the Figure 35A The remaining clusters C2 and C3 after removing FSDP from the clusters C1 to C4 in the image. The stationary position can be given by the reference points C2' and C3' in the remaining clusters C2 and C3. It should be noted that the reference points C2' and C3' are Figure 35A and Figure 35B is different because the included delivery points have changed.

[0354] Step 433 may also comprise assigning the FSDP among the first storage points 400, 401 according to any suitable algorithm. It is also envisaged that step 433 performs a further refinement of the remaining clusters before determining the rest position. Figure 34C is a flow chart of an example process for assigning an FSDP according to one embodiment. As shown, the example process can be part of step 433.

[0355] In step 434, the corresponding delivery point in the FSDP is associated with its nearest first deposit point. Figure 35C The effect of step 434 is visualized, wherein the arrows from the first deposit points 400, 401 to the delivery points (black dots) represent the resulting associations.

[0356] In step 435, the first storage points 400, 401 are evaluated based on a cost criterion. In one example, step 435 may evaluate the first storage points 400, 401 to detect a negative cost balance. Such a negative cost balance may be identified if the cost of performing a delivery via the first storage point 400, 401 to its associated delivery point exceeds a cost limit. The cost limit may represent a setup cost for the first storage point 400, 401, or any other limit. Step 435 may also perform a corresponding evaluation of the second storage point 402 located at the reference point C2', C3' in the remaining clusters C2, C3. Thus, step 435 may generate a set of first storage points and / or second storage points 400, 401, 402 having a negative cost balance, hereinafter referred to as a "candidate storage point set" (CSS).

[0357] Step 436 runs another clustering algorithm on the locations of the storage points in the CSS. The other clustering algorithm can be a density-based algorithm such as DBSCAN, OPTICS, mean shift, etc. Step 436 generates one or more storage point clusters. Figure 35D The effect of step 436 is visualized, where both subsets of the first storage point 401 are included in the corresponding storage point clusters N1, N2. Figure 35D In the example, the warehouse 400, the top fixed storage cabinet 401, and the second storage points at reference points C2' and C3' are not included in the candidate storage point set, so they are not found to have a negative cost equilibrium in step 435. These storage points will retain their assigned delivery points, which were assigned by steps 432 to 433 for the second storage point 402 (see Figure 35B ) and step 434 for the first storage point 400, 401 (refer to Figure 35C ) given.

[0358] In step 437, one storage point is selected from each storage point cluster N1, N2, ultimately obtaining a "selected storage point" (SS), or equivalently a "selected storage point location". In some embodiments, step 437 excludes any second storage point (if any), and selects the SS among the first storage points in the corresponding storage point cluster. In some embodiments, step 437 determines the center point of the corresponding storage point cluster, determines the distance from each storage point in the storage point cluster to the center point, and selects the SS based on the distance. In some embodiments, the SS can be selected based on the distance and the number of delivery points associated with each storage point in the storage point cluster. For example, the distance of each storage point can be reduced based on the number of delivery points associated with the storage point. Figure 35D In the example, the center points of the storage point clusters N1 and N2 are specified by N1' and N2'.

[0359] In step 438, all delivery points associated with the storage points in the corresponding storage point cluster are assigned to the SS of the storage point cluster. Figure 35D The effect of step 438 is visualized, where the delivery point designated by arrow 401X has been added to the selected storage points in the storage point clusters N1, N2 ( Figure 35D ).

[0360] Finally, in step 439, if steps 435 to 437 have changed the content of the delivery points in the remaining clusters, the resting position of the second deposit point may be updated. The updated resting position may be set relative to the center point of the corresponding remaining cluster. Figure 35D In the example of , step 439 is not required.

[0361] Figure 35E The resulting configuration of the delivery system is illustrated after completing step 439. It should be understood that, regardless of the implementation, method 420 determines a first location 401' of a first storage point to be used for delivery to a corresponding subset of delivery points, and a second location (a resting location) to be reached via a second storage point for delivery to the corresponding subset of delivery points.

[0362] The above reference can be executed by the control system in the delivery system Figures 34A to 35E Describe the methods and procedures. Figure 36 An example of such a control system 440 is shown, which includes a calculation module 441 and a distribution module 442. The calculation module 441 can execute Figure 34B Method 420 (optionally based on Figure 34C and / or Figure 34D ), to calculate a set of first locations [401'] and a set of second locations [402'] to be used for delivering goods to multiple delivery points. The delivery point

[403] is set as the input of the calculation module 441. As shown in the figure, the input of the calculation module 441 may also include: the number of available second storage points 402 (#M), the locations (FP) of all available first storage points 400, 401, or a predefined cost parameter (CD) that can be used for calculation, such as the total cost mentioned above, a predefined threshold or limit (if used), etc. The distribution module 442 can be configured to receive the first location set

[401] and the second location set

[402] , and transmit the corresponding configuration data to the second storage point 402, the UAV 1, and the transport vehicle 405. Thus, the distribution module 442 can perform by providing the configuration data to the second storage point 402 in step 411, providing the configuration data to the transport vehicle 405 in step 412, and providing the configuration data to the UAV 1 in step 413. Figure 34A Transport method 410.

[0363] Below, several clauses are cited to summarize some aspects and embodiments disclosed above.

[0364] Clause 1: A method of configuring a delivery system, the delivery system comprising a first storage point (400, 401) having a predefined location, a second storage point (402) having a variable location, and an unmanned aerial vehicle (UAV) (1), the UAV being operable to transport cargo between a destination point (403) and the first and second storage points (400 to 402), the destination point being a location where the UAV picks up or delivers the cargo, the method comprising the following steps:

[0365] Obtaining (421) location data of the destination point;

[0366] running (422) a clustering algorithm on the location data to determine clusters (C1 to C4) of destination points; and

[0367] The cluster is evaluated (423) relative to predefined positions of the first storage points to determine a set of stopping positions (402') of at least one of the second storage points (402).

[0368] Clause 2. A method according to clause 1, wherein the set of stopping locations (402') is determined by evaluating the resource efficiency of assigning the destination point to the reference points (C1' to C4') in the cluster (C1 to C4) relative to assigning the destination point to the first storage point (400, 401).

[0369] Clause 3. The method according to clause 1 or 2 further comprises the following steps: associating one or more first storage points (400, 401) with a corresponding first subset of destination points (424); and associating each stop position in the set of stop positions (402') with a corresponding second subset of destination points (424).

[0370] Clause 4. The method according to clause 3 further comprises the following steps: assigning (425) a subset of UAVs (1) for delivery from the one or more first storage points in the first storage points (400, 401) to a corresponding first subset of destination points, and for delivery from the respective stopping locations to a corresponding second subset of destination points.

[0371] Clause 5. A method according to any preceding clause, wherein the evaluation step (423) comprises: determining (430) for a corresponding destination point in the corresponding cluster a cost value of transporting to or from the corresponding destination point via a reference position (C1' to C4') within the corresponding cluster, wherein the set of stop locations (402') is determined based on the cost value.

[0372] Clause 6. A method according to clause 5, wherein the evaluation step (423) further includes: identifying (431) a selected set of destination points among the destination points whose cost values ​​exceed a cost threshold; determining (432) a set of remaining clusters among the clusters after removing the selected set of destination points from the clusters (C1 to C4), the corresponding remaining clusters including at least one destination point; and determining (433) a set of stop locations based on the set of remaining clusters.

[0373] Clause 7. The method according to clause 6, wherein the step of determining (433) a set of stop positions based on the set of remaining clusters comprises: for the respective remaining cluster, setting the stop position (402') relative to a reference position of the respective remaining cluster.

[0374] Clause 8. A method according to clause 6 or 7, wherein the step of evaluating (423) further comprises: associating (434) the selected set of destination points with a corresponding set of nearest predefined locations among predefined locations; identifying (435) a set of candidate storage point locations among the set of nearest predefined locations based on a cost criterion; running (436) another clustering algorithm on the set of candidate storage point locations to generate one or more storage point clusters (N1, N2); and determining (437) a selected storage point location (401') for a corresponding storage point cluster among the one or more storage point clusters (N1, N2).

[0375] Clause 9. The method of clause 8, wherein the step of evaluating (423) further comprises assigning (438) the selected storage point location (401') to all destination points associated with the corresponding storage point cluster (N1, N2).

[0376] Clause 10. A method according to clause 8 or 9, wherein, for the corresponding storage point cluster (N1, N2), the selected storage point location (401') is determined based on at least one of the following items: the distance from the corresponding candidate storage point location in the corresponding storage point cluster (N1, N2) to the reference point (N1', N2') of the corresponding storage point cluster (N1, N2), or the number of destination locations associated with the corresponding candidate storage point location in the corresponding storage point cluster (N1, N2).

[0377] Clause 11. The method according to any one of clauses 6 to 10 further comprises the following steps: determining a cost threshold for a corresponding destination point among the destination points, wherein the step of determining the cost threshold comprises: determining a candidate cost value for transporting between the corresponding destination point and a predefined location of the first storage point (400, 401); and setting the cost threshold based on the candidate cost value.

[0378] Clause 12. The method of any one of clauses 5 to 11, wherein the cost value corresponds to estimated energy consumption.

[0379] Clause 13. The method according to any one of clauses 5 to 12, further comprising the steps of: determining a center point of the respective cluster (C1 to C4); and setting a reference position (C1' to C4') relative to the center point.

[0380] Clause 14. The method of any preceding clause, wherein the clustering algorithm is configured to generate a predefined number of clusters.

[0381] Clause 15. The method according to any preceding clause, further comprising the steps of: providing (411) an instruction to arrange at least one second storage point in the set of stop locations within a first time interval; and providing (412) an instruction to transport a subset of the goods to the at least one second storage point in the set of stop locations within a second time interval after the first time interval.

[0382] Clause 16. A computer-readable medium comprising computer instructions (1102A) that, when executed by a processing system (1101), cause the processing system (1101) to perform the method of any one of clauses 1 to 15.

[0383] Clause 17. A control system configured to perform the method according to any one of clauses 1 to 15, and comprising a communication device (1103) for communicating with a second storage point (402).

[0384] The structures and methods disclosed herein can be implemented by hardware or a combination of software and hardware. In some embodiments, such hardware includes one or more software-controlled computer systems. Figure 37 A computer system 1100 is schematically depicted that includes a processing system 1101, a computer memory 1102, and a communication interface or device 1103 for inputting and / or outputting data. Depending on the implementation, the computer system 1100 may be included in the UAV, in any external computing resource that communicates with the UAV, or in any other computing resource. The communication interface 1103 may be configured for wired and / or wireless communication. The processing system 1101 may, for example, include one or more of the following: a CPU ("Central Processing Unit"), a DSP ("Digital Signal Processor"), a microprocessor, a microcontroller, an ASIC ("Application Specific Integrated Circuit"), a combination of discrete analog and / or digital components, or some other programmable logic device such as an FPGA ("Field Programmable Gate Array"). A control program 1102A comprising computer instructions is stored in the memory 1102 and executed by the processing system 1101 to perform any of the methods, operations, functions, or steps illustrated above. As shown in FIG. Figure 37As shown, the memory 1102 may also store control data 102B for use by the processing system 1102. The control program 1102A may be provided to the computer system 1100 via a computer-readable medium 1110, which may be a tangible (non-transitory) product (e.g., magnetic media, optical disk, read-only memory, flash memory, etc.) or a propagated signal.

[0385] While the disclosed subject matter has been described in connection with what are presently considered to be the most practical embodiments, it is to be understood that the disclosed subject matter is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

[0386] Furthermore, while multiple operations are depicted in a particular order in the drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desired results. In certain circumstances, parallel processing may be advantageous.

[0387] As mentioned above, any and all combinations of the concepts and embodiments described above are possible and can provide synergies. Several non-limiting examples are presented below. The embodiments presented in Section 1 can be combined with any of the embodiments presented in Sections 2 to 5 to ensure secure delivery of cargo. The embodiments for improving utilization presented in Section 2 can be combined with the embodiments for secure delivery presented in Section 1 to minimize the amount of lost / damaged cargo, for example, by including in the utilization cost a predicted risk of loss / damage of cargo in a certain geographic location or sub-region. The embodiments presented in Section 2 can also be combined with the embodiments for controlling a group of UAVs presented in Section 4, for example by using the group position / direction as an input to the calculation of aggregate cost data. The embodiments presented in Section 2 can also be combined with the embodiments for improving the launch of UAVs presented in Section 3, for example by including in the calculation of aggregate cost data the launch location and altitude of the respective UAVs being launched. The embodiments for controlling a group of UAVs presented in Section 4 may be combined with the embodiments for improving the launch of UAVs presented in Section 3 and / or the embodiments for configuring a delivery system presented in Section 5, for example by launching a group of UAVs from a launch station and / or a storage locker to further increase the number of flights. The embodiments for configuring a delivery system presented in Section 5 may be combined with the embodiments for improving the launch of UAVs presented in Section 3, for example by including the corresponding launch location as the location of the second storage point (mobile storage locker). The embodiments presented in Section 5 may also be combined with the embodiments for secure delivery presented in Section 1, for example by selecting a specific location of the second storage point at the stopping position, for example by considering the second storage point as the payload for secure delivery.

Claims

1. A method for delivering a payload (4) by an unmanned aerial vehicle (UAV) (1), the method comprising the following steps: obtaining (21) a designated location for delivering said payload (4), obtaining (22) at least one payload vulnerability index of said payload (4), obtaining (23) a collection time indicating the point in time when the payload (4) is expected to be retrieved by the recipient, selecting (24) a final delivery range at or near the designated location based on the at least one payload vulnerability index and the pickup time, and The UAV (1) is operated (25) to deliver the payload (4) to the final delivery range, wherein the step of selecting (24) comprises: predicting (24A) environmental conditions in a set of candidate delivery ranges located in the vicinity of the designated location, the environmental conditions being predicted for a dwell period extending from an estimated delivery time to the pickup time; and selecting (204) the final delivery range from the set of candidate delivery ranges based on the predicted environmental conditions and the at least one payload vulnerability index, The environmental conditions include at least one of the following: sunlight, moisture, and wind.

2. The method according to claim 1, wherein The at least one payload vulnerability index represents one or more of: monetary value, temperature sensitivity, an acceptable storage temperature range, moisture sensitivity, or wind sensitivity.

3. The method according to claim 1, wherein The environmental conditions are predicted based on a three-dimensional model of the area surrounding the delivery location and weather data for the area during the dwell period.

4. The method according to any one of claims 1 to 3, wherein: The environmental conditions are predicted based on sensor data from a sensor device (5; 1018; 1020) on the UAV (1) or on another UAV.

5. The method according to claim 4, wherein The sensor data includes at least one two-dimensional representation taken in the vicinity of a delivery location.

6. The method according to claim 1, wherein The final delivery range is also selected based on input data (39) provided by the recipient and indicating at least one candidate delivery range from the set of candidate delivery ranges.

7. The method according to claim 6, wherein: The input data (39) includes one or more of a photographic copy and a textual description.

8. The method according to claim 1, further comprising the steps of: An imaging device (5; 1018) on the UAV (1) is operated (26; 60) to capture at least one image of the final delivery range according to at least one image criterion.

9. The method according to claim 8, wherein The at least one image criterion indicates one or more objectively identifiable features to be included in the at least one image.

10. The method according to claim 8 or 9, further comprising the following steps: The at least one image is provided (63) to the recipient for display in real time or substantially real time.

11. The method according to claim 8 or 9, further comprising the steps of: While providing the at least one image for display in real time, enabling the recipient to (66) externally control at least one of the imaging device (5; 1018) and flight controller (1005) of the UAV (1).

12. The method according to claim 8 or 9, further comprising the following steps: The UAV (1) is operated (60-63) to capture images of the final delivery range while at least one of the UAV (1) and the imaging device (5; 1018) is moved until the at least one image is deemed to match a predefined image.

13. The method according to claim 8 or 9, wherein: The at least one image is taken after the payload (4) is delivered to the final delivery range to include the payload (4).

14. A control system comprising logic (1101, 1102) configured to perform the method according to any one of claims 1 to 13.

15. An unmanned aerial vehicle comprising the control system according to claim 14.

16. A computer-readable medium comprising computer instructions (1102A) which, when executed by a processing system (1101), cause the processing system (1101) to perform the method according to any one of claims 1 to 13.

17. A drone configured to: obtain a designated location for delivering a payload (4); obtain at least one payload vulnerability index for the payload (4); obtain a collection time indicating a point in time when the payload (4) is expected to be retrieved by a recipient; selecting a final delivery range at or near the designated location based on the at least one payload vulnerability index and the pickup time; and delivering the payload to the final delivery range (4), The drone is further configured to: predict environmental conditions within a set of candidate delivery ranges located within the vicinity of the designated location, the environmental conditions being predicted for a dwell period extending from an estimated delivery time to a pickup time; and selecting the final delivery range from the set of candidate delivery ranges based on the predicted environmental conditions and the at least one payload vulnerability index, The environmental conditions include at least one of the following: sunlight, moisture, and wind.

18. The drone of claim 17, further configured to predict the environmental condition based on sensor data from a sensor device (5; 1018; 1020) on the drone or on another drone.

19. The drone according to claim 17 or 18, further configured to: operate an imaging device (5; 1018) on the drone to capture at least one image of the final delivery range according to at least one image standard.

20. The drone of claim 19, further configured to enable the recipient to manipulate at least one of the imaging device (5; 1018) and flight controller (1005) of the drone while the drone (1) provides the at least one image for display in real time.

21. The drone of claim 19, further configured to compare the at least one image with a predefined image and capture images of the final delivery range until the at least one image is deemed to match the predefined image.

Images