Multi-ranging positioning and collision avoidance method and system for unmanned aerial vehicle cluster

Abstract

The invention relates to the technical field of unmanned aerial vehicle perception and obstacle avoidance, in particular to an unmanned aerial vehicle cluster multi-ranging positioning and collision avoidance method and system, and in an unmanned aerial vehicle cluster, cluster cooperative information is transmitted between cluster groups formed by multiple unmanned aerial vehicles and between the cluster groups and a ground control center through a network. The method comprises the following steps: executing a first distance measurement process through a first wireless communication module, and obtaining first relative positioning information of other unmanned aerial vehicles in a first range; executing a second distance measurement process through the second wireless communication module, and obtaining second relative positioning information of other unmanned aerial vehicles in a second range; identifying the position of the unmanned aerial vehicle based on the first relative positioning information and / or the second relative positioning information, and generating a local environment perception graph; and executing a hierarchical collision avoidance decision based on the local environment perception graph. The scheme of the invention has the advantages of low cost, high reliability, no dependence on GNSS and suitability for cooperative positioning and collision avoidance of a large-scale unmanned aerial vehicle cluster.

Description

Technical Field

[0001] This invention relates to the field of drone perception and obstacle avoidance technology. More specifically, this invention relates to a method and system for multi-range positioning and collision avoidance of drone swarms. Background Technology

[0002] With the rapid development of drone swarm technology, the collision avoidance problem of multiple drones in dense formations and collaborative operations is becoming increasingly prominent. Existing technologies mainly utilize collision avoidance schemes based on GNSS, airborne sensors, and single wireless technologies; however, these schemes also have the following drawbacks: 1. GNSS (Global Navigation Satellite System) based collision avoidance systems (such as ADS-B, Automatic Dependent Surveillance Broadcast System): These systems are highly dependent on global satellite navigation systems and cannot function in environments where GNSS signals are denied or interfered with. Furthermore, they are costly and difficult to deploy on a large scale on low-cost UAV platforms.

[0003] 2. Obstacle avoidance based on airborne sensors (such as radar, lidar, and visual sensors): Due to limitations in the sensor's detection range, viewing angle, environmental adaptability (such as fog, rain, and dust), and computational complexity, there are problems such as blind spots, high latency, and high cost in high-speed, large-scale cluster applications.

[0004] 3. Collision avoidance systems based on a single wireless technology: such as those using only ad hoc networks or only Wi-Fi / Bluetooth for ranging. A single technology cannot simultaneously meet the requirements of long range, high accuracy, low latency, and interference resistance. Long-range technologies (such as ad hoc networks) lack accuracy, while short-range high-precision technologies (such as star-flash technology) have too short a warning time.

[0005] The international standard ISO 15964, "Unmanned Aerial Vehicle (UAV) perception and obstacle avoidance systems," focuses on the core systems for the safe operation of UAVs. It specifies the safety and quality requirements for UAV perception and obstacle avoidance systems, aiming to guide the development, production, testing, and use of UAV perception and obstacle avoidance systems and ensure UAV flight safety. While it defines the framework for UAV obstacle avoidance systems, it does not specify concrete technical implementation paths.

[0006] Therefore, there is an urgent need for a low-cost, highly reliable, GNSS-independent, and large-scale UAV swarm cooperative positioning and collision avoidance solution. Summary of the Invention

[0007] To address the technical problems of high cost, poor reliability, and poor applicability of the aforementioned drone swarm ranging, positioning, and obstacle avoidance solutions, this invention provides solutions in the following aspects.

[0008] In a first aspect, the present invention provides a method for multi-range positioning and collision avoidance in a drone swarm. In the drone swarm, multiple drones form a cluster, and the clusters, as well as the cluster and a ground control center, exchange swarm coordination information via a network. Multiple drones are equipped with a first wireless communication module and a second wireless communication module. The method includes: executing a first ranging procedure through the first wireless communication module to obtain first relative positioning information with other drones within a first range; executing a second ranging procedure through the second wireless communication module to obtain second relative positioning information with other drones within a second range, wherein the second range is smaller than the first range, and the accuracy of the second relative positioning information is higher than that of the first relative positioning information; identifying the drone positions based on the first and / or second relative positioning information to generate a local environmental perception map; and performing a layered collision avoidance decision based on the local environmental perception map, wherein the layered collision avoidance decision includes at least a local emergency collision avoidance decision based on the drone's own perception information and a collaborative collision avoidance decision based on swarm network coordination information.

[0009] In one embodiment, identifying the location of the UAV based on the first relative positioning information and / or the second relative positioning information includes: selecting the first relative positioning information for long-distance positioning, using the second relative positioning information for short-distance positioning, and determining the location of the UAV by combining the first and second relative positioning information for medium-distance positioning.

[0010] In one embodiment, after generating the local environmental perception map, the method further includes: marking the UAV position information in the perception map according to its ranging and positioning method; reporting the marked local environmental perception map to the group leader unit through the first wireless communication module; wherein the group leader unit generates a global environmental perception map, each group leader unit reports the global environmental perception map to the ground control center, the ground control center forms a spatial position envelope map based on multiple global environmental perception maps, and sends the spatial position envelope map to each group leader unit; and receiving a global environmental perception map sent by the group leader unit that exceeds the range of its own perception map.

[0011] In one embodiment, the local emergency collision avoidance decision includes a local collision avoidance strategy, the cooperative collision avoidance decision includes an intra-group collision avoidance strategy and an inter-group collision avoidance strategy, and the execution of the hierarchical collision avoidance decision includes: executing the local collision avoidance strategy, based on the second relative positioning information, performing high-priority collision avoidance actions on other UAVs entering the emergency distance; executing the intra-group collision avoidance strategy, based on the received global environment perception map, correcting the local perception map, and participating in intra-group cooperative path planning; and / or executing the inter-group collision avoidance strategy, whereby the group leader or ground control center forms a network space envelope map based on multiple global environment perception maps, and performs macro-path planning based on the network space envelope map to avoid the risk of group collisions.

[0012] In one embodiment, the method further includes: recording a timestamp of the positioning information and calculating the velocity vector of other drones; predicting the future positions of other drones based on the most recent positioning information and the velocity vector, in order to compensate for communication delay and calculation delay.

[0013] In one embodiment, the method further includes: when the first wireless communication module detects that another drone is approaching, sending a warning message to the second wireless communication module in advance; and the second wireless communication module pre-allocating connection and computing resources to the drone that is about to enter its communication range based on the warning message.

[0014] In one embodiment, the method further includes: if the first wireless communication module is detected to be malfunctioning, then relying primarily on the second wireless communication module for collision avoidance and adopting a conservative flight strategy; if the second wireless communication module is detected to be malfunctioning, then relying primarily on the first wireless communication module for collision avoidance and adjusting the aircraft's position within the formation.

[0015] In one embodiment, for drones in the cluster that do not have positioning capabilities, their positions are determined by at least four drones with positioning capabilities through a collaborative network positioning algorithm using the Time Difference of Arrival (TDoA) algorithm, and are marked in the global environment perception map.

[0016] In one embodiment, the first ranging process includes self-organizing network positioning or other methods with medium- and long-range positioning modules, and the second ranging process includes star-flash positioning or other methods with medium- and long-range positioning modules.

[0017] In a second aspect, the present invention also provides a multi-range positioning and collision avoidance system for unmanned aerial vehicle (UAV) swarms, comprising: a processor; and a memory storing computer program instructions, wherein when the processor executes the computer program instructions, it implements the multi-range positioning and collision avoidance method for UAV swarms as described in one or more of the preceding embodiments.

[0018] The beneficial effects of this invention are as follows: This invention obtains the relative positions of drone swarms through multiple wireless ranging methods, effectively enhancing ranging and collision avoidance performance and improving collision avoidance reliability. Utilizing a multi-collision avoidance strategy involving local emergency collision avoidance and cooperative collision avoidance decision-making, the safety of the drone swarm is effectively improved.

[0019] Furthermore, this invention also effectively improves collision avoidance performance by performing predictive position correction when the ranging and positioning frequency is low.

[0020] Furthermore, this invention also uses a system network positioning method with at least four drones equipped with positioning functions to accurately locate drones that have failed ranging and positioning functions or do not have positioning functions, thereby effectively improving the collision avoidance effect and having high applicability. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to an embodiment of the present invention; Figure 2 This is a spatial location diagram of a drone according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the overall design of a three-layer collision avoidance scheme according to an embodiment of the present invention; Figure 4 This is a diagram showing the connection relationship of the main modules in a drone according to an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating inter-group collision avoidance control at the ground center according to an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating an application scenario of a three-layer collision avoidance strategy for drone swarms according to an embodiment of the present invention; Figure 7 This is a structural diagram illustrating a multi-range positioning and collision avoidance system for a drone swarm according to an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] Figure 1This is a flowchart illustrating a multi-range positioning and collision avoidance method 100 for a drone swarm according to an embodiment of the present invention. In the drone swarm, multiple drones form a cluster, and the clusters and the ground control center exchange cluster coordination information via a network. Multiple drones are equipped with a first wireless communication module and a second wireless communication module to achieve wireless communication positioning.

[0025] like Figure 1 As shown, in step S101, a first ranging procedure is executed to obtain first relative positioning information with other drones within a first range. Specifically, the first ranging procedure is executed through a first wireless communication module to obtain first relative positioning information with other drones within a first range.

[0026] In step S102, a second ranging procedure is executed to obtain second relative positioning information with other drones within a second range. Specifically, the second ranging procedure is executed through a second wireless communication module to obtain second relative positioning information with other drones within a second range, wherein the second range is smaller than the first range, and the accuracy of the second relative positioning information is higher than that of the first relative positioning information.

[0027] According to the context of this invention, the first ranging process includes self-organizing network positioning or other methods equipped with medium- and long-range positioning modules, and the second ranging process includes star-flash positioning or other methods equipped with medium- and long-range positioning modules. It should be noted that there are many other ways to achieve positioning in the art, and those skilled in the art can choose different positioning methods according to actual needs.

[0028] In step S103, a local environment perception map is generated. Specifically, the location of the UAV is identified based on the first relative positioning information and / or the second relative positioning information to generate the local environment perception map. In some embodiments, when generating the local environment perception map, the first relative positioning information is selected for long-range positioning, the second relative positioning information is used for short-range positioning, and the first and second relative positioning information are combined to determine medium-range positioning.

[0029] Furthermore, after generating a local environmental perception map, the UAV's position information in the perception map can be marked according to its ranging and positioning method. The marked local environmental perception map is then reported to the group leader unit via the first wireless communication module. The group leader unit generates a global environmental perception map, and each group leader unit reports its global environmental perception map to the ground control center. The ground control center forms a spatial position envelope map based on multiple global environmental perception maps and distributes the spatial position envelope map to each group leader unit. The system also receives global environmental perception maps from the group leader units that extend beyond its own local perception map range.

[0030] In step S104, a hierarchical collision avoidance decision is performed based on the local environment awareness map. The hierarchical collision avoidance decision includes at least a local emergency collision avoidance decision based on local awareness information and a collaborative collision avoidance decision based on cluster network coordination information. The local emergency collision avoidance decision includes a local collision avoidance strategy, and the collaborative collision avoidance decision includes intra-group collision avoidance strategies and inter-group collision avoidance strategies. The execution of the hierarchical collision avoidance decision involves three stages, mainly including: The system executes a collision avoidance strategy, and based on the second relative positioning information, performs high-priority collision avoidance actions on other drones that enter the emergency distance.

[0031] The system executes an intra-group collision avoidance strategy, corrects its local perception map based on the received global environment perception map, and participates in intra-group collaborative path planning.

[0032] To implement the inter-group collision avoidance strategy, the group leader or ground control center (hereinafter referred to as the ground center) forms a network space envelope map based on multiple global environmental perception maps, and performs macro-path planning based on the network space envelope map to avoid the risk of group collisions.

[0033] The above-mentioned triple collision avoidance decision can be executed in its entirety, or an independent collision avoidance strategy can be selected based on the actual situation. Those skilled in the art can choose according to actual needs. For example, when there is only one drone swarm, only the local collision avoidance strategy and the intra-swarm collision avoidance strategy can be executed.

[0034] When implementing collision avoidance strategies, due to the large amount of information and the lag in ground-based calculations, timestamps of positioning information can be recorded and velocity vectors of other UAVs can be calculated. Based on the most recent positioning information and velocity vectors, the future positions of other UAVs can be predicted to compensate for communication and computation delays.

[0035] In some embodiments, the solution of the present invention also effectively ensures the reliability of the collision avoidance strategy by rationally allocating computing resources to different modules. Specifically, when the first wireless communication module detects that another drone is approaching, it sends a warning message to the second wireless communication module in advance. Based on the warning message, the second wireless communication module pre-allocates connection and computing resources to the drone that is about to enter its communication range.

[0036] In addition, a fault response mechanism is set up to ensure that the UAV can complete its mission in an orderly manner under emergency conditions. Specifically, if the first wireless communication module fails, the second wireless communication module will be relied upon primarily for collision avoidance, and a conservative flight strategy will be adopted. If the second wireless communication module fails, the first wireless communication module will be relied upon primarily for collision avoidance, and the UAV's position within the formation will be adjusted.

[0037] In addition, for drones in the cluster that do not have positioning capabilities, their positions are determined by at least four drones with positioning capabilities through a collaborative network positioning algorithm using the Time Difference of Arrival (TDoA) algorithm, and are marked in the global environment perception map.

[0038] The solution of this invention will now be described in detail in conjunction with its practical application.

[0039] Collision avoidance in a drone swarm essentially involves real-time perception of the relative positions of each drone within the swarm in three-dimensional space, forming a real-time three-dimensional spatial position map, and then determining which drones are likely to collide, calling upon the flight control module to adjust the drones' spatial positions, and thus avoiding collisions.

[0040] This invention achieves dual positioning of the relative positions of the drones by setting a first wireless communication module and a second wireless communication module in the drone. For example... Figure 2 The diagram shows the spatial positions of some drones in a drone swarm. The length of the line segments represents the distance between the drones, and the direction and distance together determine the relative positions of the drones.

[0041] This invention employs a combined approach of self-organizing network ranging and positioning with satellite-based ranging. For both self-organizing networks and satellite-based positioning, the most suitable ranging and positioning scheme among their various methods can be used, and this invention does not impose any limitations on this approach. The positioning valve can be broadly categorized into three directions: ranging-based positioning, angle-based positioning, and fingerprint-based positioning. This embodiment primarily utilizes ranging-based positioning methods, such as ToA, TDoA, RToF, and RSSI.

[0042] Furthermore, this invention also employs a network-based collaborative positioning technology among multiple drones to achieve a more robust and reliable collision avoidance effect.

[0043] like Figure 3 As shown, the drone swarm collision avoidance scheme uses two ranging and positioning schemes simultaneously: self-organizing network and star-flash. The self-organizing network focuses on providing ranging and positioning at medium to long distances, while the star-flash focuses on providing ranging and positioning at medium to short distances. The combination of the two ranging and positioning schemes provides a more reliable ranging and positioning effect.

[0044] The UAV swarm collision avoidance design in this invention has a three-layer design: local collision avoidance design, intra-swarm collision avoidance design, and inter-swarm collision avoidance design. The local collision avoidance design focuses on local emergency collision avoidance. The main control unit selects an appropriate local collision avoidance algorithm and decision based on the self-organizing network and star-flash positioning results, and sends collision avoidance decisions and action commands to the UAV flight controller, thereby controlling the flight controller to complete the collision avoidance implementation.

[0045] The collision avoidance design within the group emphasizes organized, coordinated collision avoidance. The group leader communicates with the drones within the group via an ad hoc network, and the group leader's main control unit controls the drones within the group to maintain distance.

[0046] Collision avoidance design between groups focuses on collision avoidance flight path planning at the swarm intelligence and command and control levels. Group leaders communicate via a network to maintain overall group distance. Distance maintenance is typically controlled by the command and control center.

[0047] The aforementioned dual ranging and positioning + three-layer collision avoidance scheme is a customizable design. It can be appropriately tailored to different drones and mission types. For example, only one ranging and positioning scheme can be used, or only the onboard collision avoidance design can be employed to reduce costs, improve reliability, and enhance swarm intelligence. For instance, for high-speed drone swarms, dual ranging and positioning + three-layer collision avoidance achieves high collision avoidance reliability and swarm intelligence capabilities; for slow-speed drones, only star-flash ranging and positioning + onboard collision avoidance can be used to reduce system complexity; for medium-to-large drone swarms with large distances between drones, only self-organizing network ranging and positioning + onboard collision avoidance + network collaborative positioning and collision avoidance can be used to reduce costs. Furthermore, a small number of drones without ranging and positioning capabilities can be mixed in to achieve greater engineering practicality and reduce costs.

[0048] The solution of this invention will be described below in conjunction with the connection relationship and functions of the main modules on the drone.

[0049] like Figure 4 As shown, the main control unit in the UAV is connected to the pod module, flight control module, ad hoc network module, and star-flash module to achieve positioning and collision avoidance control. The star-flash module includes G-nodes and T-nodes. Based on Figure 4 The connection relationships shown are illustrated, and the single-machine positioning and collision avoidance technology solution is described as follows: 1. The StarShock module continuously runs a precise ranging algorithm at close range and performs positioning, and reports the positioning data to the local main control unit; 2. The self-organizing network module continuously runs the medium and long distance coarse ranging algorithm to perform positioning, and reports the positioning data to the local main control unit; 3. The main control unit of this machine performs the following actions: a) The main control unit combines historical ranging and positioning data (which can be used to calculate the approximate position, flight direction, and flight speed) and distance to determine whether to accept data from the star-flash module or the self-organizing network module (in principle, the self-organizing network ranging results are accepted for long distances, the star-flash ranging results are accepted for short distances, and a comprehensive judgment is made for medium distances), and triggers the flight control module to perform collision avoidance actions as needed. b) The main control unit runs the positioning algorithm to update the relative spatial position information of all UAVs within its sensing range. The relative spatial position information will indicate whether it is ad hoc network ranging and positioning data or star-flash positioning data (i.e., "coloring"), and transmit this relative spatial position information to the group leader through the ad hoc network module.

[0050] c) The main control unit feeds back the colored "key UAVs" numbers to the ad hoc network module and the star-flash module respectively. The ad hoc network module receives the UAV numbers for medium to long range (ad hoc networks are not good at short-range high-precision positioning), while the star-flash module receives the UAV numbers for medium to short range (long-range UAVs are beyond the star-flash detection range).

[0051] like Figure 5 and Figure 6 As shown, UAV network cooperative positioning and collision avoidance technology includes intra-group collision avoidance and inter-group collision avoidance. The key design for intra-group collision avoidance is that the group leader calculates a global spatial position map in real time based on the local spatial position information reported by each UAV and broadcasts it to each UAV. The key design for inter-group collision avoidance is that the ground center forms a spatial envelope for each group based on the spatial position maps and BeiDou position information transmitted back by each group leader, continuously correcting the flight trajectories of each group to maintain a certain distance between them (the spatial envelopes do not conflict and have a certain interval). Due to the large amount of information and the lag in ground center calculations, the ground center does not perform specific collision avoidance control for individual UAVs, but only for UAV swarms as a group.

[0052] Specifically, the UAV network cooperative localization and collision avoidance process mainly includes the following: First, each UAV within the group independently performs star-flash ranging and positioning, as well as self-organizing network ranging and positioning, and the main control unit independently calculates and generates a local spatial position map. Figure 5 The spatial positions of each UAV in self-organizing network ranging and each UAV in star-flash ranging are respectively colored (labeled). Figure 6 Within Group B, UAVs B-2 and B-1 can each perform collision avoidance autonomously through ranging.

[0053] Second, each UAV main control unit within the group reports its spatial location map and timestamp to the group leader, for example... Figure 5 The drones A1, A2, A3, A4 and A5 shown (simplified as A-1, A-2, A-3, A-4 and A-5, and so on) report their spatial location maps and timestamps to the A-group leader drone. Figure 6The B-1 and B-2 drones in the middle group will report their respective spatial position maps and timestamps to the group leader drone of group N. The group leader drone's main control unit will aggregate the spatial position information maps reported by each drone in the group, update the historical global spatial position information map, and form a global spatial position information map that includes all drones in the group. Figure 5 There will also be two colors, representing star-flash ranging and self-organizing network ranging respectively.

[0054] Third, the leader drone transmits the spatial location information map, timestamp, and GNSS location information back to the ground center. The ground center then forms the spatial location envelope of each group of UAVs and plans the next flight path for each group (including collision avoidance considerations, i.e., maintaining a certain safe distance between groups), and sends the flight path to the leader drone.

[0055] IV. The group leader's main control unit calculates and broadcasts the spatial position information map of all UAVs in the group to all UAVs. The group leader's main control unit also adjusts the flight trajectories of all UAVs in the group based on the flight path plan issued by the ground center. Each UAV in the group performs the following operations: (a) The main control unit compares the spatial position map of the UAV in the group broadcast by the group leader with the spatial position map obtained by the local ranging and positioning. If the timestamps are very different, the local ranging and positioning error is corrected; otherwise, the local ranging and positioning result shall prevail.

[0056] (b) The main control unit determines whether obstacle avoidance is required based on the corrected ranging and positioning results, and sends instructions to the flight controller. (c) The main control unit of the machine issues instructions to the flight controller to adjust the flight path according to the newly issued flight path planning requirements.

[0057] In addition to the sudden trajectory changes caused by collision avoidance that trigger collision avoidance maneuvers by other drones, sudden trajectory changes may also occur due to some drones actively avoiding obstacles. In this case, a network-based collaborative collision avoidance mechanism can be established, following the process described above.

[0058] The position of a drone swarm is constantly changing during the ranging and positioning process. Therefore, it's crucial to avoid situations where the actual position has significantly changed by the time the ranging and positioning results are available, leading to outdated positioning data. The solution is to add a precise timestamp (e.g., accurate to milliseconds) to the ranging and positioning process. Dividing the spatial vector difference between the current and previous positions by the time difference yields the velocity vector. Therefore, by adding the product of the velocity vector and the time difference to the most recent ranging and positioning position, the estimated current spatial position of the drone can be corrected to obtain the corrected position.

[0059] Each time ranging and positioning is performed, the velocity vector is updated with a timestamp. When the spatial location information is reported to the head unit and the ground center, the timestamp and velocity vector are also added. This allows for the calculation of a more accurate estimate of the current position.

[0060] The frequency of ranging and positioning can be significantly reduced using the methods described above. Within the limits of the system's capabilities, increasing the frequency of ranging and positioning can still effectively improve positioning accuracy.

[0061] This invention uses a combination of self-organizing network ranging and positioning and star-flash ranging and positioning: self-organizing network can achieve ranging and positioning in the medium to long range, while star-flash can achieve ranging and positioning in the medium to short range. Therefore, the combination of the two can achieve better ranging and positioning results.

[0062] The main implementation methods of cooperative ranging and positioning are as follows: (1) Starburst and self-organizing network each perform their own ranging and positioning; (2) The star flash and the self-organizing network respectively report their ranging and positioning results (spatial position information and timestamp of each UAV) to the main control unit in their respective UAVs; (3) The main control unit combines the ranging results of the star flash and the self-organizing network to form a complete spatial location map within the local detection range; (4) The main control unit selects other UAVs at medium distances detected by the self-organizing network, forms a spatial location map of them, and sends it to the star-flash ranging and positioning module; (5) The main control unit selects the UAVs with medium distances (close to the farthest limit of star flash positioning) for star flash detection, forms a spatial position map of them, and sends it to the self-organizing network ranging and positioning module; (6) The Star Flash ranging and positioning module performs ranging and positioning on the one hand for connected UAVs within the detectable range, and on the other hand, estimates how many UAVs will enter the Star Flash detection range based on the spatial location map, reserving sufficient computing power resources to quickly connect newly approaching UAVs and perform ranging and positioning as soon as possible. (7) The self-organizing network ranging and positioning module eliminates nearby UAVs in order to reduce the consumption of computing resources for ranging and positioning.

[0063] While the main control unit within the UAV makes a collision avoidance decision and instructs the flight control unit to execute the collision avoidance maneuver, it must also inform friendly aircraft of the decision / maneuver. This serves two purposes: firstly, it enables potentially colliding friendly aircraft to take counter-collision maneuvers, making the collision avoidance actions of both sides more efficient; secondly, it allows unrelated friendly aircraft to coordinate collision avoidance based on this information. Furthermore, the lead aircraft and ground control can proactively correct the velocity vector and spatial position information of relevant UAVs based on this information.

[0064] The collision avoidance maneuver message will also be sent by the "involved drone" to the lead drone as a high-priority message. The lead drone will then forward the message to the ground control center.

[0065] Furthermore, the main control unit can determine the failure status of the ad hoc network by the frequency of heartbeats or reports from the ad hoc network modules, and activate a short-range collision avoidance scheme. This means that collision avoidance no longer relies on the ranging and positioning function of the ad hoc network, but instead relies on the ranging and positioning data from the satellite flash. If the ad hoc network communication function is still functioning, but only the positioning function is failing, the main control unit can also cross-reference the spatial position map of each unit within the group broadcast by the group leader with the short-range ranging and positioning results, and select the most reasonable collision avoidance strategy through network collaborative positioning.

[0066] In addition, for individual drones experiencing starlight malfunctions, the starlight module G node on the drone continuously monitors the survival status of the T node (where multiple starlight modules exist on the drone) and reports this information to the main control unit. The main control unit determines the failure status of the starlight module by its heartbeat or the frequency of its reporting and activates a remote collision avoidance scheme. This means that collision avoidance no longer relies on the starlight's ranging and positioning function, but instead on the ranging and positioning data from the self-organizing network. The main control unit can also cross-reference its own ranging and positioning data with the remote ranging and positioning data based on the spatial position map of each drone in the group broadcast from the group leader, and select the most appropriate collision avoidance strategy through network-coordinated positioning.

[0067] For drones that lack positioning capabilities but possess communication capabilities, this invention relies on network cooperative positioning technology for collision avoidance (as described above, drones with positioning capabilities mutually locate each other, and the leader drone calculates a spatial position map, which is then distributed to all drones within the group). Furthermore, drones without positioning capabilities autonomously perform all collision avoidance maneuvers. It is important to note that a certain number of drones with ranging and positioning capabilities must be present, for example, at least four, to meet the minimum requirements for TDoA ranging and positioning.

[0068] This invention achieves the design of a collision avoidance scheme for UAV ranging and positioning by using a dual ranging and positioning method combined with a three-layer collision avoidance method. This reduces costs, improves reliability, and enhances swarm intelligence, effectively improving the reliability of UAV swarm operations.

[0069] Figure 7 This is a structural diagram illustrating a multi-range positioning and collision avoidance system for a drone swarm according to an embodiment of the present invention.

[0070] This invention also provides a multi-range positioning and collision avoidance system for unmanned aerial vehicle (UAV) swarms, including a processor and a memory. The memory stores computer program instructions, and when the processor executes the computer program instructions, it implements the multi-range positioning and collision avoidance method for UAV swarms as described above.

[0071] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and therefore will not be described in detail here.

[0072] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0073] In the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

[0074] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A method for multi-range positioning and collision avoidance in a drone swarm, characterized in that, In a drone swarm, multiple drones form a cluster, and the clusters, as well as the cluster and the ground control center, exchange swarm coordination information via a network. Multiple drones are equipped with a first wireless communication module and a second wireless communication module. The method includes: The first wireless communication module executes a first ranging procedure to obtain first relative positioning information with other drones within a first range; The second ranging process is executed through the second wireless communication module to obtain the second relative positioning information with other drones within the second range, wherein the second range is smaller than the first range, and the accuracy of the second relative positioning information is higher than that of the first relative positioning information; The location of the UAV is identified based on the first relative positioning information and / or the second relative positioning information to generate a local environmental perception map; based on the local environmental perception map, a hierarchical collision avoidance decision is executed, which includes at least a local emergency collision avoidance decision based on the local perception information and a cooperative collision avoidance decision based on the cluster network cooperation information.

2. The method for multi-range positioning and collision avoidance of UAV swarms according to claim 1, characterized in that, Identifying the location of the UAV based on the first relative positioning information and / or the second relative positioning information includes: For long-range positioning, the first relative positioning information is selected; for short-range positioning, the second relative positioning information is used; and for medium-range positioning, the first and second relative positioning information are combined to determine the positioning.

3. The method for multi-range positioning and collision avoidance of UAV swarms according to claim 1, characterized in that, After generating the local environment-aware map, the process further includes: The location information of the UAV in the perception map is marked according to its ranging and positioning method; The identified local environmental perception map is reported to the group head unit through the first wireless communication module; wherein the group head unit generates a global environmental perception map, each group head unit reports the global environmental perception map to the ground control center, the ground control center forms a spatial location envelope map based on multiple global environmental perception maps, and sends the spatial location envelope map to each group head unit; Receive a global environment perception map that exceeds the range of the local perception map, sent by the group head unit.

4. The UAV swarm multi-range positioning and collision avoidance method according to claim 3, characterized in that, The local emergency collision avoidance decision includes a local collision avoidance strategy; the cooperative collision avoidance decision includes an intra-group collision avoidance strategy and an inter-group collision avoidance strategy; and the execution of the hierarchical collision avoidance decision includes: The system executes a collision avoidance strategy, and based on the second relative positioning information, performs high-priority collision avoidance actions on other drones that enter the emergency distance. Execute intra-group collision avoidance strategy, correct local perception map based on the received global environment perception map, and participate in intra-group cooperative path planning; and / or To implement inter-group collision avoidance strategies, the lead aircraft or ground control center forms a network space envelope map based on multiple global environmental perception maps, and performs macro-path planning based on the network space envelope map to avoid the risk of group collisions.

5. The method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, Also includes: Record the timestamp of the positioning information and calculate the velocity vector of other drones; Based on the most recent positioning information and the velocity vector, the future positions of other drones are predicted to compensate for communication and computation delays.

6. The method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, Also includes: When the first wireless communication module detects that other drones are approaching, it sends an early warning message to the second wireless communication module in advance. Based on the warning information, the second wireless communication module pre-allocates connection and computing resources for the drone that is about to enter its communication range.

7. The method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, Also includes: If the first wireless communication module is detected to be malfunctioning, the system will primarily rely on the second wireless communication module for collision avoidance and will adopt a conservative flight strategy. If the second wireless communication module is detected to be malfunctioning, the system will primarily rely on the first wireless communication module for collision avoidance and adjust its position within the formation.

8. The method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to claim 1 or 3, characterized in that, For drones in the cluster that do not have positioning capabilities, their positions are determined by at least four drones with positioning capabilities through a collaborative network positioning algorithm using the Time Difference of Arrival (TDoA) algorithm, and are marked in the global environment perception map.

9. The method for multi-range positioning and collision avoidance of unmanned aerial vehicle (UAV) swarms according to claim 1, characterized in that, The first ranging process includes self-organizing network positioning or other methods with medium- and long-distance positioning modules, and the second ranging process includes star-flash positioning or other methods with medium- and long-distance positioning modules.

10. A multi-range positioning and collision avoidance system for unmanned aerial vehicle (UAV) swarms, characterized in that, include: processor; The memory stores computer program instructions, which, when executed by the processor, implement the UAV swarm multi-range positioning and collision avoidance method as described in any one of claims 1-9.

Images