A multi-mode switching system for a drone

Abstract

This invention relates to the field of unmanned aerial vehicle (UAV) control and communication technology, and discloses a multi-mode switching system for UAVs, including an airborne terminal, a ground control terminal, and a cloud-based dispatch center. The airborne terminal, ground control terminal, and cloud-based dispatch center interact via a 6G-WiFi 7 hybrid communication link. The airborne terminal integrates a flight control unit, an ARIS-NOMA fusion communication unit, a multimodal perception unit, and an energy management unit. The cloud-based dispatch center integrates an optimization-driven hierarchical deep reinforcement learning decision engine, achieving integrated coordinated switching of flight attitude, communication mode, and perception mode. This system offers advantages such as integrated coordinated switching, improved adaptability of UAVs in complex environments, and enhanced swarm collaboration capabilities, solving the problems of existing UAV multi-mode switching systems, including single-dimensional switching, weak anti-interference capabilities, high energy consumption, and poor swarm collaboration.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control and communication technology, specifically to a UAV multi-mode switching system. Background Technology

[0002] With the rapid development of the low-altitude economy, drones have been widely used in emergency rescue, high-altitude inspection, communication relay, topographic mapping, and other fields, placing higher demands on their flight adaptability, communication reliability, and perception accuracy. Multi-mode switching is a core technology for improving the environmental adaptability of drones. Currently, multi-mode switching technology for drones generally suffers from three major pain points: First, mode switching is mostly limited to flight attitude (such as multi-rotor / fixed-wing switching), failing to achieve coordinated linkage between communication and flight mode, leading to problems such as data transmission interruption and flight instability during the switching process; second, switching decisions rely on single sensor data, resulting in weak anti-interference capabilities, poor dynamic environmental adaptability, and difficulty in adapting to complex scenarios such as strong interference, terrain obstruction, and network load fluctuations; third, co-channel interference exists during multi-drone swarm switching, leading to low resource allocation efficiency and high energy consumption, which restricts the long-term collaborative operation capability of drone swarms.

[0003] In existing technologies, most UAV multi-mode switching systems focus on smooth switching of flight attitude, without integrating cutting-edge communication and intelligent decision-making technologies, and cannot achieve coordinated linkage of flight, communication and perception. At the same time, aerial reconfigurable smart surfaces (ARIS) and non-orthogonal multiple access (NOMA) technologies are mostly applied independently in the field of communication, without deep integration with UAV multi-mode switching, making it difficult to solve problems such as low channel utilization and severe interference in complex environments.

[0004] Therefore, we propose a multi-mode switching system for unmanned aerial vehicles (UAVs) to solve the above problems. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a multi-mode switching system for unmanned aerial vehicles (UAVs), which features three-in-one coordinated switching, enhancing the adaptability of UAVs in complex environments and their ability to work collaboratively in clusters. This system solves the problems of existing UAV multi-mode switching being single-dimensional, having weak anti-interference capabilities, high energy consumption, and poor cluster collaboration.

[0006] (II) Technical Solution To achieve the aforementioned three-in-one coordinated switching and enhance the adaptability and swarm collaborative operation capabilities of UAVs in complex environments, this invention provides the following technical solution: a multi-mode switching system for UAVs, comprising an airborne terminal, a ground control terminal, and a cloud-based dispatch center. The airborne terminal, ground control terminal, and cloud-based dispatch center interact with each other via a 6G-WiFi7 hybrid communication link. The airborne terminal integrates a flight control unit, an ARIS-NOMA fusion communication unit, a multimodal perception unit, and an energy management unit. The cloud-based dispatch center integrates an optimization-driven hierarchical deep reinforcement learning decision engine to achieve three-in-one coordinated switching of flight attitude, communication mode, and perception mode.

[0007] Furthermore, the ARIS-NOMA converged communication unit adopts an integrated design, including a PIN diode-based reconfigurable ARIS reflection unit, a NOMA RF transceiver module, a high-speed RF switching switch, and a synchronization control interface. The ARIS reflection unit and the NOMA RF transceiver module share the same RF front end. The high-speed RF switching switch adopts an SPDT structure and has a built-in isolator to achieve seamless switching between the ARIS reflection link and the NOMA transmission link.

[0008] Furthermore, the ARIS reflector operates in a frequency band adapted to the requirements of 6G low-altitude communication, supports continuous phase shift adjustment, and has preset adjustment accuracy; the NOMA radio frequency transceiver module adopts a multi-carrier non-orthogonal transmission architecture, integrates a power divider, a modem, and an interference suppression filter, and supports simultaneous access by multiple users.

[0009] Furthermore, the ARIS-NOMA converged communication unit supports switching between active transmission mode and passive reflection mode. In active transmission mode, the ARIS reflection unit is turned off, the NOMA module operates at full power, and high-speed transmission is achieved using the IEEE 802.11be protocol and MU-MIMO technology. In passive reflection mode, the NOMA module retains only the channel detection unit, and the ARIS reflection unit is activated and dynamically adjusts the phase shift angle.

[0010] Furthermore, in the active transmission mode, the NOMA module adopts a dynamic power allocation algorithm based on channel conditions, combined with LDPC coding and modulation methods adapted to transmission requirements, to achieve multi-user power domain non-orthogonal multiplexing, which has low latency and low bit error rate characteristics; in the passive reflection mode, the ARIS reflection unit adjusts the phase shift based on the maximum channel gain criterion to form a directional reflection beam, which has strong interference suppression capability.

[0011] Furthermore, the mode switching of the ARIS-NOMA fusion communication unit adopts a smooth transition mechanism, including a preparation phase, a switching phase, and a stabilization phase, which has low switching latency and low signal attenuation characteristics, and is equipped with a redundancy backoff mechanism for switching failure.

[0012] Furthermore, the O-HDRL decision engine adopts a hierarchical architecture, with the upper layer being the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm and the lower layer being the semi-definite relaxation and alternation optimization method. The decision engine collects channel signal-to-noise ratio, network load, and UAV position parameters in real time, and constructs a weighted fusion switching trigger function to realize collaborative decision-making between communication mode and flight and perception modes.

[0013] Furthermore, the flight control unit is based on the improved INAV hybrid flight control system, integrating mixer configuration file technology and dynamic hybrid control algorithm, supporting smooth switching between multi-rotor and fixed-wing flight attitudes, and optimizing the seamless transition between motor output and control surface control; the multimodal perception unit integrates LiDAR, visible light camera and infrared thermal imager, supporting full perception fusion and single-modal focusing mode switching.

[0014] Furthermore, the energy management unit adopts a hydrogen-electric hybrid power system with an integrated mode adaptive power consumption adjustment mechanism, which can dynamically adjust power consumption according to different communication modes. It has a significant power consumption advantage over the separate design and supports cluster air recharging.

[0015] Furthermore, the ARIS-NOMA fused communication unit also integrates a cluster collaborative interference suppression mechanism. It identifies interference sources through an adaptive interference detection algorithm, dynamically allocates the ARIS reflection beam direction and NOMA transmission power by combining the O-HDRL decision engine, and is equipped with Turbo coding and continuous interference cancellation algorithm, thus possessing strong interference cancellation capabilities.

[0016] (III) Beneficial Effects Compared with the prior art, the present invention provides a multi-mode switching system for unmanned aerial vehicles (UAVs), which has the following beneficial effects: 1. This invention innovatively integrates ARIS and NOMA technologies to construct an integrated ARIS-NOMA fusion communication unit, realizing intelligent switching between active transmission and passive reflection communication modes. This breaks through the limitations of the fixed communication mode of traditional UAVs, improves channel utilization and anti-interference capability, and reduces hardware size and energy consumption through a shared radio frequency front-end design. 2. This invention constructs a hierarchical decision architecture through the O-HDRL decision engine, realizing the three-in-one coordinated switching of flight attitude, communication mode, and perception mode, solving the problems of unstable switching and data interruption caused by the independent operation of the three modes in the prior art, and improving the dynamic adaptability of UAVs in complex environments. 3. The overall design of the technical solution of the present invention is reasonable. The combination of core technologies and the synergistic mechanism are not obvious to those skilled in the art. It can be widely applied to high-end scenarios such as emergency rescue, 6G low-altitude communication, and energy inspection, and has significant practical value and industrialization prospects. Attached Figure Description

[0017] Figure 1 This is a structural block diagram of the airborne terminal of the present invention; Figure 2 This is a structural block diagram of the ARIS-NOMA fusion communication unit of the present invention; Figure 3 This is a flowchart illustrating the mode switching process of the ARIS-NOMA converged communication unit of the present invention. Figure 4 This is a diagram illustrating the hierarchical architecture of the O-HDRL decision engine of this invention; Figure 5 This is a flowchart of the system of the present invention. Detailed Implementation

[0018] 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 embodiments of the present invention, and not all embodiments. 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.

[0019] As attached Figure 1 As shown, this embodiment provides an ARIS-NOMA integrated UAV multi-mode intelligent switching system, including an airborne terminal, a ground control terminal, and a cloud-based dispatch center. The three are connected via a 6G-WiFi7 hybrid communication link to achieve high-speed, low-latency data interaction, ensuring the synchronous transmission of commands and data at each layer.

[0020] As attached Figure 2 As shown, the airborne terminal serves as the core execution carrier for mode switching, integrating a flight control unit, an ARIS-NOMA fusion communication unit, a multimodal sensing unit, and an energy management unit. These units work together to achieve the execution and status feedback of multimodal switching.

[0021] (1) Flight Control Unit: Based on the improved INAV hybrid flight control system, it integrates hybrid configuration file technology and dynamic hybrid control algorithm, supports smooth switching between multi-rotor and fixed-wing flight attitudes, and avoids flight attitude fluctuations during the switching process by optimizing the seamless transition between motor output and control surface control, thus ensuring flight stability. At the same time, the flight control unit is linked with the ARIS-NOMA fusion communication unit and multimodal perception unit through the synchronous control interface to receive switching commands from the cloud scheduling center and synchronously adjust the flight attitude.

[0022] (2) ARIS-NOMA converged communication unit: as attached Figure 3As shown, it adopts an integrated design and is the core innovative component of this invention. Specifically, it includes a PIN diode-based reconfigurable ARIS reflector unit, a NOMA RF transceiver module, a high-speed RF switching switch, and a synchronization control interface. The ARIS reflector unit and the NOMA RF transceiver module share the same RF front-end, effectively reducing hardware size and power consumption. The high-speed RF switching switch adopts an SPDT structure with a built-in isolator, enabling seamless switching between the ARIS reflector link and the NOMA transmission link, avoiding signal crosstalk during the switching process.

[0023] The ARIS reflector operates in a frequency band suitable for 6G low-altitude communication requirements, supports continuous phase shift adjustment, has preset adjustment accuracy, and can dynamically adjust the reflection phase shift angle according to the channel status to achieve channel reconstruction and interference suppression. The NOMA RF transceiver module adopts a multi-carrier non-orthogonal transmission architecture, integrating a power divider, modem, and interference suppression filter, supporting simultaneous access by multiple users and improving spectrum utilization.

[0024] As attached Figure 4 As shown, this converged communication unit supports intelligent switching between active transmission mode (A-UAV mode) and passive reflection mode (P-UAV mode), with a clear and controllable switching process: In active transmission mode, the ARIS reflection unit is completely turned off, the NOMA module operates at full power, and high-speed data transmission is achieved using the IEEE 802.11be protocol and MU-MIMO technology. Combined with a dynamic power allocation algorithm based on channel conditions, LDPC coding, and an adapted modulation scheme, multi-user power domain non-orthogonal multiplexing is achieved, ensuring low latency and low bit error rate; In passive reflection mode, the NOMA module retains only the channel detection unit, the ARIS reflection unit is activated and adjusts the phase shift angle based on the maximum channel gain criterion to form a directional reflection beam, which has strong interference suppression capability.

[0025] The mode switching employs a smooth transition mechanism, divided into a preparation phase, a switching phase, and a stabilization phase. In the preparation phase, the current mode's core module enters a power attenuation phase, while the target mode module enters a warm-up phase and completes parameter initialization. In the switching phase, the high-speed RF switch completes the link switching, the current mode's core module is completely shut down, and the target mode module gradually increases its power to the rated value, achieving seamless signal connection through RF gain adjustment. In the stabilization phase, the target mode module enters a stable operating state, providing real-time feedback on its operating status, and the decision engine fine-tunes relevant parameters to ensure stable communication quality. A redundant fallback mechanism for switching failures is also implemented; if a signal interruption is detected during the switching process, the system automatically reverts to the original mode to prevent communication interruptions from affecting drone operations.

[0026] In addition, the ARIS-NOMA fusion communication unit also integrates a cluster cooperative interference suppression mechanism. It identifies interference sources and interference intensity within the cluster through an adaptive interference detection algorithm, uploads the interference information to the cloud scheduling center, and dynamically allocates the ARIS reflection beam direction and NOMA transmission power in conjunction with the O-HDRL decision engine. Combined with Turbo coding and continuous interference cancellation (SIC) algorithm, it effectively eliminates co-channel interference and improves the reliability of multi-UAV cluster cooperative communication.

[0027] (3) Multimodal sensing unit: Integrates LiDAR, visible light camera and infrared thermal imager, supporting switching between two sensing modes: full-sensory fusion and single-modal focusing. In full-sensory fusion mode, multi-source sensing data are fused with high precision through feature point registration algorithm, improving the accuracy of target recognition and terrain detection in complex environments; in single-modal focusing mode, only one sensing device is activated according to the needs of the operation scenario, reducing energy consumption. The sensing unit collects environmental data (terrain complexity, obstacle density, light intensity) and flight status data in real time and uploads them to the cloud dispatch center to provide data support for mode switching decisions.

[0028] (4) Energy Management Unit: Employs a hydrogen-electric hybrid power system with an integrated adaptive power consumption adjustment mechanism, which can dynamically adjust the power consumption of each unit according to the working mode of the ARIS-NOMA fusion communication unit. In active transmission mode, priority is given to ensuring the power supply of the NOMA module; in passive reflection mode, the overall power consumption is reduced, which has a significant power consumption advantage over the traditional separate design. At the same time, the energy management unit supports aerial recharging of the cluster and can dock with the "aerial power bank" drones in the cluster to achieve rapid recharging and extend the drone's endurance.

[0029] The ground control unit is responsible for the local monitoring, parameter configuration, and emergency intervention of a single UAV, integrating a mode switching monitoring panel, a data decoding module, and a local decision backup unit. The mode switching monitoring panel displays the UAV's current flight attitude, communication mode, perception mode, energy consumption status, and switching logs in real time, and supports manual issuance of switching commands by staff (suitable for emergency scenarios). The data decoding module performs real-time decoding and preprocessing of multimodal data and channel status data transmitted back from the airborne terminal, removing noisy data to ensure data accuracy. The local decision backup unit is used to deal with communication interruptions at the cloud dispatch center. When communication is interrupted, it automatically takes over the switching decision, adopts preset strategies to ensure the normal operation of the UAV, and seamlessly switches to cloud dispatch after communication is restored.

[0030] The cloud-based dispatch center is the core of the system's global decision-making, integrating the O-HDRL decision engine, big data analysis module, and trajectory planning module. (See attached...) Figure 5As shown, the O-HDRL decision engine adopts a hierarchical architecture. The upper layer is the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, which treats each UAV as an independent agent, learning its own state (flight, communication, perception, energy consumption) and the cluster state (position of other UAVs, interference situation) in real time, and generating the optimal switching strategy and trajectory planning scheme. The lower layer uses the Semi-Deterministic Relaxation (SDR) and Alternating Optimization (AO) methods to transform the switching strategy generated by the reinforcement learning in the upper layer into specific execution parameters (such as ARIS reflection phase shift, NOMA power allocation ratio), improving the accuracy and feasibility of the decision.

[0031] The big data analytics module analyzes the location information, channel status, task load, and energy consumption data of the UAV cluster in real time, and builds a dynamic environment model to provide data support for the switching decision of the O-HDRL decision engine. The trajectory planning module generates the optimal flight trajectory based on the location information and environment model of multiple UAVs, avoids collisions between UAVs in the cluster, and optimizes the switching timing to reduce energy consumption and interference during the switching process.

[0032] As attached Figure 5 As shown, the specific workflow of this system is as follows: 1. Initialization: After the system starts, the airborne terminal, ground control terminal, and cloud dispatch center complete self-checks and connections. The ARIS-NOMA fusion communication unit enters active transmission mode by default, the flight control unit enters multi-rotor flight attitude by default, and the multimodal perception unit enters full perception fusion mode by default. 2. Data Acquisition: The multimodal sensing unit collects environmental data and flight status data in real time, the ARIS-NOMA fusion communication unit collects channel status data and network load data, and the energy management unit collects energy consumption data. All data is uploaded to the cloud dispatch center through a 6G-WiFi7 hybrid communication link. 3. Decision generation: The O-HDRL decision engine in the cloud-based scheduling center constructs a weighted and fused switching trigger function based on the collected data to determine whether a mode switch is required. If so, it generates a coordinated switching command for flight attitude, communication mode, and perception mode. 4. Mode switching: After receiving the switching command, the flight control unit, ARIS-NOMA fusion communication unit, and multimodal perception unit simultaneously perform the switching operation. The ARIS-NOMA fusion communication unit completes the link switching through a high-speed radio frequency switching switch, the flight control unit completes the flight attitude switching through a dynamic hybrid control algorithm, and the multimodal perception unit completes the perception mode switching. The switching process adopts a smooth transition mechanism. 5. Status Feedback and Optimization: After the handover is completed, the airborne terminal will feed back the handover result and the current operating status to the cloud dispatch center. The O-HDRL decision engine will fine-tune the relevant parameters and optimize the handover strategy based on the feedback data. At the same time, the ground control terminal will monitor the operating status after the handover in real time and intervene manually when necessary. 6. Cluster Collaboration: When multiple drones operate in a cluster, each drone communicates with the others via a 6G link. The cloud-based dispatch center uses the O-HDRL decision engine and cluster collaboration interference suppression mechanism to achieve coordinated switching and interference suppression of drones within the cluster, ensuring the overall operational performance of the cluster. 7. Emergency Handling: When communication between the cloud dispatch center and the airborne terminal is interrupted, the local decision backup unit of the ground control terminal automatically takes over to ensure the normal operation of the UAV; when the ARIS-NOMA fusion communication unit fails to switch, the redundancy fallback mechanism is activated to fall back to the original mode to avoid communication interruption.

[0033] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An ARIS-NOMA integrated unmanned aerial vehicle (UAV) multi-mode intelligent switching system, characterized in that, It includes an airborne terminal, a ground control terminal, and a cloud-based dispatch center. The airborne terminal, ground control terminal, and cloud-based dispatch center achieve data interaction through a 6G-WiFi7 hybrid communication link. The airborne terminal integrates a flight control unit, an ARIS-NOMA fusion communication unit, a multimodal perception unit, and an energy management unit. The cloud-based dispatch center integrates an optimization-driven hierarchical deep reinforcement learning decision engine to perform three-in-one coordinated switching of flight attitude, communication mode, and perception mode.

2. The system according to claim 1, characterized in that, The ARIS-NOMA converged communication unit adopts an integrated design, including a PIN diode-based reconfigurable ARIS reflection unit, a NOMA RF transceiver module, a high-speed RF switching switch, and a synchronization control interface. The ARIS reflection unit and the NOMA RF transceiver module share the same RF front end. The high-speed RF switching switch adopts an SPDT structure and has a built-in isolator to achieve seamless switching between the ARIS reflection link and the NOMA transmission link.

3. The system according to claim 2, characterized in that, The ARIS reflector operates in a frequency band adapted to the requirements of 6G low-altitude communication, and the adjustment accuracy is preset through continuous phase shift adjustment; the NOMA radio frequency transceiver module adopts a multi-carrier non-orthogonal transmission architecture and integrates a power divider, a modem and an interference suppression filter.

4. The system according to claim 2, characterized in that, The ARIS-NOMA fused communication unit supports switching between active transmission mode and passive reflection mode. In active transmission mode, the ARIS reflection unit is turned off, the NOMA module operates at full power, and transmission is achieved using the IEEE 802.11be protocol and MU-MIMO technology. In passive reflection mode, the NOMA module retains only the channel detection unit, and the ARIS reflection unit is activated and dynamically adjusts the phase shift angle.

5. The system according to claim 4, characterized in that, In the active transmission mode, the NOMA module employs a dynamic power allocation algorithm based on channel conditions, combined with LDPC coding and modulation methods adapted to transmission requirements, to perform multi-user power domain non-orthogonal multiplexing; in the passive reflection mode, the ARIS reflection unit adjusts the phase shift based on the maximum channel gain criterion to form a directional reflection beam for interference suppression.

6. The system according to claim 4, characterized in that, The ARIS-NOMA fusion communication unit employs a smooth transition mechanism for mode switching, including a preparation phase, a switching phase, and a stabilization phase. It is capable of low switching latency and low signal attenuation, and is equipped with a redundancy fallback mechanism for switching failures.

7. The system according to claim 1, characterized in that, The O-HDRL decision engine adopts a layered architecture. The upper layer is a multi-agent deep deterministic policy gradient algorithm, and the lower layer is a semi-definite relaxation and alternation optimization method. The decision engine collects channel signal-to-noise ratio, network load, and UAV position parameters in real time, constructs a weighted fusion switching trigger function, and performs collaborative decision-making on communication mode, flight mode, and perception mode.

8. The system according to claim 1, characterized in that, The flight control unit is based on the improved INAV hybrid flight control system, integrating mixer configuration file technology and dynamic hybrid control algorithm to enable smooth switching between multi-rotor and fixed-wing flight attitudes and seamless transition between motor output and control surface control; the multimodal perception unit integrates LiDAR, visible light camera and infrared thermal imager to perform full perception fusion and single-modal focusing mode switching.

9. The system according to claim 1, characterized in that, The energy management unit adopts a hydrogen-electric hybrid power system and an integrated mode adaptive power consumption adjustment mechanism, which can dynamically adjust power consumption according to different communication modes and enable the cluster to replenish power in the air.

10. The system according to any one of claims 1-9, characterized in that, The ARIS-NOMA fusion communication unit also integrates a cluster collaborative interference suppression mechanism. It identifies interference sources through an adaptive interference detection algorithm, dynamically allocates the ARIS reflection beam direction and NOMA transmission power by combining the O-HDRL decision engine, and performs interference cancellation by combining Turbo coding and continuous interference cancellation algorithm.

Images