Cross-medium aircraft system and control method thereof

Abstract

The invention provides a cross-medium aircraft system and a control method thereof. A task instruction sent by a ground station is received; the method comprises the following steps: acquiring sensing data of an aircraft in different medium environments in real time, and preprocessing the sensing data into sensing fusion data; receiving other sensing fusion data sent by other aircrafts in the cluster in real time, and generating a cluster collaborative trajectory according to the sensing fusion data and other sensing fusion data sent by other aircrafts in the cluster; determining a target task of the vehicle according to the cluster cooperation trajectory, the current operation state of the vehicle and the task instruction; and according to the target task, determining an optimal water access path parameter of the vehicle, and generating a driving signal based on the sensing fusion data or the optimal water access path parameter so as to continuously control the vehicle to reach a target state. The technical problems that a traditional aircraft is low in wing deformation efficiency, poor in water access control precision, lack of cluster cooperation capability and the like are solved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer processing, and in particular to a cross-medium vehicle system and a control method thereof. BACKGROUND

[0002] As an emerging technology, the cross-medium vehicle has the characteristics of strong adaptability and cross-domain operation, and has wide application prospects in the fields of ocean exploration, environmental monitoring, search and rescue operations, etc.

[0003] In the prior art, in order to realize continuous operation in underwater, water surface and air multi-medium environments, a single type of vehicle is usually used to switch modes for segmented operation in different medium environments. However, the existing cross-medium vehicle not only has poor control accuracy for entering and exiting water, which affects the continuity and safety of cross-medium operation, but also cannot realize large-scale and high-efficiency tasks, and the operation coverage is limited. SUMMARY

[0004] The present disclosure provides a cross-medium vehicle system and a control method thereof to at least solve the above technical problems existing in the prior art.

[0005] According to a first aspect of the present disclosure, a cross-medium vehicle system is provided, the system comprising: An autonomous decision and task planning system, configured to receive a task instruction sent by a ground station and send the task instruction to a cluster coordination system, wherein the task instruction is used to indicate a collective task that the cluster needs to complete; A multi-source information perception and state recognition system, configured to acquire perception data of the vehicle in different medium environments in real time, pre-process the perception data into perception fusion data, and send the perception fusion data to a multi-modal coordinated control and execution system and the cluster coordination system; The cluster coordination system is configured to receive the perception fusion data sent by the local multi-source information perception and state recognition system, other perception fusion data sent by other vehicles in the cluster, and the task instruction, generate a cluster coordination trajectory according to the perception fusion data and the other perception fusion data sent by the other vehicles in the cluster, and determine a target task of the vehicle according to the cluster coordination trajectory, a current running state of the vehicle and the task instruction, and send the target task to the autonomous decision and task planning system, wherein the current running state comprises current running parameters of the vehicle when driving in different media or crossing media, and the target task belongs to at least one subtask in the collective task; The autonomous decision and task planning system is further configured to receive the target task sent by the cluster coordination system, determine optimal water entry and exit path parameters of the vehicle according to the target task, and feed back the optimal water entry and exit path parameters to the multi-modal coordinated control and execution system; The multi-modal collaborative control and execution system is used for receiving the perception fusion data sent by the multi-source information perception and state recognition system and receiving the optimal in-out water path parameters sent by the autonomous decision and task planning system, generating a driving signal according to the perception fusion data or the optimal in-out water path parameters, and continuously controlling the vehicle to reach a target state.

[0006] In an optional embodiment, the perception data includes at least one of original depth data, original attitude data, original positioning data, and contact state data, and the multi-source information perception and state recognition system includes: a depth perception sensor module for measuring original depth data of the vehicle in real time when the vehicle is diving in water, and sending the original depth data to a data preprocessing module; an attitude and motion state detection module for obtaining original attitude data of the vehicle and sending the original attitude data to the data preprocessing module; a high-precision positioning navigation module for obtaining original positioning data of the vehicle and sending the original positioning data to the data preprocessing module; a medium environment detection module for detecting contact state data of each part of the vehicle and the water body in real time, and sending the contact state data to the data preprocessing module; a data preprocessing module for receiving and processing the original depth data, the original attitude data, the original positioning data, and the contact state data, and generating the perception fusion data; sending the perception fusion data to the multi-modal collaborative control and execution system and the cluster collaborative system.

[0007] In an optional embodiment, the cluster collaborative system includes: a cluster communication coordination module for receiving the perception fusion data sent by the data preprocessing module of each vehicle in the cluster, generating cluster shared data, and sending the cluster shared data to a cluster collaborative navigation control module; a cluster collaborative navigation control module for receiving the cluster shared data sent by the cluster communication coordination module and receiving other perception fusion data sent by other vehicles in the cluster, generating a cluster collaborative trajectory according to the cluster shared data and other perception fusion data, and sending the cluster collaborative trajectory to a collaborative task allocation module, a collaborative task allocation module for receiving the cluster collaborative trajectory sent by the cluster collaborative navigation control module, generating a target task based on the cluster collaborative trajectory, a current running state of the vehicle, and a task instruction, and sending the target task to a collaborative task execution strategy module through a distributed algorithm; The cooperative task execution strategy module receives the target task sent by the cooperative task allocation module, processes the target task based on a cooperative task allocation algorithm, and generates a cooperative operation control instruction and sends the instruction to the multi-modal cooperative control and execution system.

[0008] In an optional embodiment, the multi-modal cooperative control and execution system comprises: The control unit receives the perception fusion data sent by the data preprocessing module and the cooperative operation control instruction sent by the cooperative task execution strategy module, determines a first type of driving signal, a second type of driving signal, and a third type of driving signal according to the perception fusion data, the cooperative operation control instruction, and the current running state of the vehicle, sends the first type of driving signal to the bionic variable configuration wing control module, sends the second type of driving signal to the multi-modal propulsion control module, and sends the third type of driving signal to the attitude and heading control module, wherein the first type of driving signal is used to switch the wing configuration of the vehicle, the second type of driving signal is used to drive different propulsion units of the vehicle, and the third type of driving signal is used to drive the rudder surface control of the vehicle. The bionic variable configuration wing control module receives and responds to the first type of driving signal sent by the control unit to switch the wing configuration required by the vehicle in different media or in the case of cross-media, and generates a first feedback signal. The multi-modal propulsion control module receives and responds to the second type of driving signal sent by the control unit to drive the propulsion units required by the vehicle in different media or in the case of cross-media, and generates a second feedback signal. The attitude and heading control module receives and responds to the third type of driving signal sent by the control unit to maintain the attitude stability of the vehicle in different media or in the case of cross-media, and generates a third feedback signal. The first feedback signal, the second feedback signal, and the third feedback signal collected in real time are used as the current running parameters of the vehicle.

[0009] In an optional embodiment, the control unit comprises: The first control unit receives the perception fusion data sent by the data preprocessing module and the cooperative operation control instruction sent by the cooperative task execution strategy module, generates a coordinated control instruction according to the perception fusion data, the cooperative operation control instruction, and the current running state of the vehicle, and sends the instruction to the second control unit and the autonomous decision and task planning system. a second control unit configured to receive the coordinated control instruction of the first control unit, and in response to converting into a first type of driving signal, a second type of driving signal and a third type of driving signal, send the first type of driving signal to the bionic variable configuration wing control module, send the second type of driving signal to the multi-modal propulsion control module, and send the third type of driving signal to the attitude and heading control module.

[0010] In an optional embodiment, the autonomous decision-making and task planning system comprises: a modal switching decision module configured to, in the case of cross-medium modal switching, receive the coordinated control instruction of the first control unit, generate a modal switching control instruction based on an adaptive rule algorithm and the coordinated control instruction, and send the modal switching control instruction to the water-entry and exit path planning module; the water-entry and exit path planning module is configured to receive the modal switching control instruction sent by the modal switching decision module, calculate optimal water-entry and exit path parameters according to the modal switching control instruction, and send the optimal water-entry and exit path parameters to the second control unit to continuously control the vehicle to reach a target state; a communication and instruction management module configured to receive a task instruction sent by the ground station, send the task instruction to the collaborative task allocation module, and further configured to feed back task execution state information to the ground station.

[0011] In an optional embodiment, the second control unit is specifically configured to: in the case of cross-medium modal switching, receive the optimal water-entry and exit path parameters sent by the water-entry and exit path planning module, generate a new driving signal according to the optimal water-entry and exit path parameters, continuously update the first type of driving signal, the second type of driving signal and the third type of driving signal, and control the vehicle to reach a target state.

[0012] In an optional embodiment, the bionic variable configuration wing control module is specifically configured to: receive the first type of driving signal sent by the control unit; when the current running mode of the vehicle meets a preset condition, adjust the wing configuration of the vehicle to a target wing configuration through the first type of driving signal; wherein the target wing configuration is selected according to the preset condition met by the current running mode of the vehicle as follows: when the current running mode is an underwater mode, adjust to a shark pectoral fin wing configuration; when the current running mode is a water-entry and exit mode, adjust to a flying fish pectoral fin wing configuration; when the current running mode is a ground effect flight mode, adjust to a ground effect flight wing configuration; adjusting to an albatross wing configuration when the current operating mode is an air flight mode.

[0013] In an optional embodiment, the adaptive rule algorithm is an algorithm for generating coordinated control instructions based on the output quantities of the sub-controllers in the vehicle and fuzzy rule weights.

[0014] According to a second aspect of the present disclosure, a control method of a cross-medium vehicle system is provided, characterized in that the method comprises: receiving a task instruction sent by a ground station, wherein the task instruction is used to indicate a collective task that a cluster needs to complete; real-time acquisition of perception data of the vehicle in different medium environments, and preprocessing the perception data into perception fusion data; real-time reception of other perception fusion data sent by other vehicles in the cluster, and generation of a cluster cooperative trajectory according to the perception fusion data and the other perception fusion data sent by the other vehicles in the cluster; determination of a target task of the vehicle according to the cluster cooperative trajectory, a current operating state of the vehicle, and the task instruction, wherein the current operating state comprises current operating parameters of the vehicle when driving in different media or across media, and the target task belongs to at least one sub-task in the collective task; determination of optimal water entry and exit path parameters of the vehicle according to the target task, and generation of a driving signal based on the perception fusion data or the optimal water entry and exit path parameters, so as to constantly control the vehicle to reach a target state.

[0015] According to a third aspect of the present disclosure, a control device of a cross-medium vehicle system is provided, characterized in that the device comprises: a task receiving module configured to receive a task instruction sent by a ground station, wherein the task instruction is used to indicate a collective task that a cluster needs to complete; a first data acquisition module configured to real-time acquisition of perception data of the vehicle in different medium environments, and preprocessing the perception data into perception fusion data; a second data acquisition module configured to real-time reception of other perception fusion data sent by other vehicles in the cluster, and generation of a cluster cooperative trajectory according to the perception fusion data and the other perception fusion data sent by the other vehicles in the cluster; a task determination module configured to determine a target task of the vehicle according to the cluster cooperative trajectory, a current operating state of the vehicle, and the task instruction, wherein the current operating state comprises current operating parameters of the vehicle when driving in different media or across media, and the target task belongs to at least one sub-task in the collective task; A state update module is used to determine the optimal entry and exit path parameters of the aircraft according to the target task, and then generate a drive signal based on the perception fusion data or the optimal entry and exit path parameters to continuously control the aircraft to reach the target state.

[0016] According to a fourth aspect of the present disclosure, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the present disclosure.

[0017] According to a fifth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the present disclosure.

[0018] The cross-media aircraft system and control method disclosed herein receive mission instructions sent by a ground station; obtain perception data of the aircraft in different media environments in real time, and preprocess the perception data into perception fusion data; receive other perception fusion data sent by other aircraft in the cluster in real time, and generate a cluster collaborative trajectory based on the perception fusion data and other perception fusion data sent by other aircraft in the cluster; determine the target mission of the aircraft based on the cluster collaborative trajectory, the current operating status of the aircraft and the mission instructions; determine the optimal entry and exit path parameters of the aircraft based on the target mission, and then generate a driving signal based on the perception fusion data or the optimal entry and exit path parameters to continuously control the aircraft to reach the target state, thereby solving technical problems such as low wing deformation efficiency, poor entry and exit control accuracy, and lack of cluster collaboration capability of traditional aircraft.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings, in which several embodiments of the present disclosure are shown by way of example and not limitation, wherein: In the drawings, the same or corresponding reference numerals denote the same or corresponding parts.

[0021] Figure 1A structural schematic diagram of a cross-medium vehicle system is shown. Figure 2 A structural schematic diagram of a multi-source information perception and state identification system in a cross-medium vehicle system is shown. Figure 3 A structural schematic diagram of a multi-modal collaborative control and execution system in a cross-medium vehicle system is shown. Figure 4 A structural schematic diagram of a cluster coordination system in a cross-medium vehicle system is shown. Figure 5 A structural schematic diagram of an autonomous decision-making and task planning system in a cross-medium vehicle system is shown. Figure 6 A complete structural schematic diagram of an exemplary cross-medium vehicle system is shown. Figure 7 A third perspective view of an exemplary cross-medium vehicle is shown. Figure 8 An exploded schematic diagram of a cross-medium vehicle is shown. Figure 9 A flowchart of a control method of a cross-medium vehicle system is shown. Figure 10 A structural schematic diagram of a control device of a cross-medium vehicle system is shown. DETAILED DESCRIPTION

[0022] In order to make the objectives, features and advantages of the present disclosure more obvious and easy to understand, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present disclosure.

[0023] Figure 1 A structural schematic diagram of a cross-medium vehicle system is provided, the system 10 comprising: an autonomous decision-making and task planning system 110, a multi-source information perception and state identification system 120, a multi-modal collaborative control and execution system 130, and a cluster coordination system 140, wherein, The autonomous decision-making and task planning system 110 is configured to receive a task instruction sent by a ground station and send the task instruction to the cluster coordination system 140, wherein the task instruction is used to indicate a collective task that the cluster needs to complete.

[0024] The cross-medium vehicle system provided by the embodiment is applied to a marine buoy monitoring task, needs to be floated from a preset depth under water to the water surface, and then is converted into a ground effect flight mode or air flight, so as to perform a task and transmit data in a required mode according to a task instruction sent by a ground station.

[0025] Specifically, the autonomous decision and task planning system 110 contains a module that can communicate with the ground station to realize bidirectional communication with the ground station, for example, can receive a task instruction sent by the ground station, the task instruction can be a collective task indicating that the cluster needs to complete, the collective task can be a task completed by all vehicles in the cluster, can be a task indicating that one or more vehicles in the cluster complete, or can be a task indicating that a specified vehicle in the cluster completes, and the embodiment does not limit it. After the autonomous decision and task planning system 110 receives the task instruction, the task instruction is sent to the cluster coordination system 140, and the task to be executed by the machine in this collective task is determined through the cluster coordination system 140.

[0026] The multi-source information perception and state recognition system 120 is used to acquire perception data of the vehicle in different medium environments in real time, pre-process the perception data into perception fusion data, and then send the perception fusion data to the multi-modal cooperative control and execution system 130 and the cluster coordination system 140.

[0027] The perception data includes at least one of original depth data, original attitude data, original positioning data, and contact state data.

[0028] The multi-source information perception and state recognition system 120 of the embodiment can be used to acquire depth data, attitude data, positioning data, and medium contact state data of the vehicle in different medium environments under water, on the water surface, and in the air, and through real-time monitoring of the above data, to provide data support for subsequent bionic wing deformation, rapid water entry and exit, and cluster coordination.

[0029] Specifically, the embodiment collects the perception data, and performs standardization, unification and other preprocessing operations to obtain perception fusion data. Then, the perception fusion data is sent to other systems in the vehicle for data support.

[0030] The cluster coordination system 140 is configured to receive the perception fusion data sent by the local multi-source information perception and state identification system 120, other perception fusion data sent by other vehicles in the cluster, and a task instruction, generate a cluster coordination trajectory according to the perception fusion data and the other perception fusion data sent by the other vehicles in the cluster, and determine a target task of the vehicle according to the cluster coordination trajectory, a current running state of the vehicle, and the task instruction, and send the target task to the autonomous decision and task planning system 110.

[0031] The current running state includes a current running parameter of the vehicle when the vehicle is running in different media or across media, and the target task belongs to at least one subtask in the collective task.

[0032] The target task can be a task allocation table and a priority list. The task allocation table records at least one task that needs to be completed by the vehicle, and the priority list records the priority of different tasks. The current running parameter can be an attitude feedback parameter, a propulsion feedback parameter, a wing feedback parameter, a speed parameter, and a position parameter of the vehicle when the vehicle is running. The cluster coordination trajectory is the formation position of the vehicle in the cluster and the route that needs to be traveled to cooperate with the cluster.

[0033] Specifically, the cluster coordination system 140 in the vehicle of the embodiment can clearly understand the running state of all vehicles in the cluster by receiving the perception fusion data of the vehicle and receiving the perception fusion data sent by other vehicles in the cluster. By analyzing the task instruction sent by the ground station and combining the current running parameter (feedback signal of part of the module) of the vehicle, the target task of the vehicle can be distributedly calculated and sent to the multi-modal cooperative control and execution system 130 for further planning and execution.

[0034] The embodiment realizes multi-vehicle formation control, cooperative task allocation, cluster communication coordination through the cluster coordination system 140, forms the advantage of cluster operation, and can effectively solve the problem that a single vehicle cannot realize large-scale and high-efficiency cluster cooperative tasks due to limited single-vehicle operation efficiency and limited operation coverage.

[0035] The autonomous decision and task planning system 110 is further configured to receive the target task sent by the cluster coordination system 140, determine an optimal water entry and exit path parameter of the vehicle according to the target task, and feed back the optimal water entry and exit path parameter to the multi-modal cooperative control and execution system 130.

[0036] The optimal water-entry and exit path parameters include the angle, speed, and time of water-entry and exit of the vehicle. Since the vehicle may need to repeatedly enter and exit water while performing a task and undergo different modal changes, the vehicle needs to feed back the optimal water-entry and exit path parameters to the multi-modal cooperative control and execution system 130 to determine the next modal operation time point and the desired operation state.

[0037] The autonomous decision-making and task planning system 110 in this embodiment determines the optimal water-entry and exit path parameters of the vehicle during the execution of the target task according to the perception fusion data and the target task, and implements intelligent cross-medium modal switching decisions, repeated and rapid water-entry and exit path planning, and cluster cooperative navigation control based on the optimal water-entry and exit path parameters. This new type of air-water cross-medium special vehicle effectively realizes the technical innovations of bionic wing deformation, rapid water-entry and exit, and cluster cooperation, improves the stability, rapidity, and cooperativeness during the modal switching process, and significantly improves the operation efficiency of a single platform and the cluster cooperation capability.

[0038] The multi-modal cooperative control and execution system 130 receives the perception fusion data sent by the multi-source information perception and state recognition system 120 and the optimal water-entry and exit path parameters sent by the autonomous decision-making and task planning system 110, generates driving signals according to the perception fusion data or the optimal water-entry and exit path parameters, and continuously controls the vehicle to reach the target state.

[0039] The target state is the operation state that the vehicle needs to reach when driving in different media or across media, which can be one of the shark pectoral fin wing configuration, the flying fish pectoral fin wing configuration, the ground effect flight wing configuration, or the albatross wing configuration. The target state can also be the position and speed required to be reached when driving in different media or across media according to the requirements of the target task.

[0040] For example, if the vehicle undergoes four stages during the execution of a task, the first stage is underwater diving, and the target state can be a state of diving at a first specific speed, a first specific trajectory, and a shark pectoral fin wing configuration. The second stage is flying out of the water, and the target state can be a state of flying at a second specific speed, a second specific trajectory, and a flying fish pectoral fin wing configuration. The third stage is water surface flight, and the target state can be a state of flying at a third specific speed, a third specific trajectory, and a ground effect flight wing configuration. The fourth stage is air flight, and the target state can be a state of flying at a fourth specific speed, a fourth specific trajectory, and an albatross wing configuration.

[0041] Specifically, the multi-modal cooperative control and execution system 130 in this embodiment can generate driving signals based on the perception fusion data to control the vehicle to reach the target state in different media, and can also generate driving signals through optimal in-out water path parameters to constantly control the vehicle to reach the target state when crossing media, so as to accurately cooperatively control the functions of wing control, multi-modal propulsion, and attitude and heading control of the vehicle, and realize the organic integration of various innovative functions.

[0042] Figure 2 A structural diagram of a multi-source information perception and state recognition system in a cross-medium vehicle system provided by the embodiments of the present disclosure is shown. As shown in Figure 2 The multi-source information perception and state recognition system 120 includes the multi-source information perception and state recognition system 120, which includes a depth perception sensor module 1210, an attitude and motion state detection module 1220, a high-precision positioning and navigation module 1230, a medium environment detection module 1240, and a data preprocessing module 1250. Among them, The depth perception sensor module 1210 is configured to measure the original depth data of the vehicle underwater in real time, and send the original depth data to the data preprocessing module 1250. The original depth data can be underwater depth information and vertical speed of ascent and descent of the vehicle.

[0043] Specifically, the depth perception sensor module 1210 in this embodiment adopts a high-precision pressure sensor array, which can accurately measure the original depth data such as underwater depth information and vertical speed of ascent and descent of the vehicle in real time, and send the original depth data to the data preprocessing module for filtering and calibration processing, so as to provide accurate depth reference data for subsequent modal switching decision, path planning and cluster cooperation. Alternatively, the embodiment can set a preset depth in advance, and when the vehicle reaches the preset depth, the remaining depth from the water surface is monitored in real time, and the original depth data is sent to the data preprocessing module 1250 to further achieve energy saving effect.

[0044] The attitude and motion state detection module 1220 is configured to obtain the original attitude data of the vehicle and send the original attitude data to the data preprocessing module. The original attitude data includes three-axis acceleration, three-axis angular velocity and three-axis magnetic field intensity of the vehicle.

[0045] Specifically, the attitude and motion state detection module 1220 in the embodiment adopts a nine-axis inertial measurement module (integrating a three-axis accelerometer, an angular velocity meter, and a magnetometer) to detect the attitude change and motion state of the aerial vehicle during wing deformation, to output three-axis acceleration, three-axis angular velocity, and three-axis magnetic field strength, etc. original attitude data, and send the original attitude data to the data preprocessing module 1250 for sensor fusion processing, to provide accurate real-time attitude feedback data for subsequent decision control, path planning, and cluster collaboration.

[0046] The high-precision positioning and navigation module 1230 is configured to obtain original positioning data of the aerial vehicle, and send the original positioning data to the data preprocessing module 1250; wherein the original positioning data is the accurate geographic position and motion trajectory of the aerial vehicle.

[0047] Specifically, the high-precision positioning and navigation module 1230 in the embodiment adopts a differential satellite navigation system to obtain the accurate geographic position and motion trajectory of the aerial vehicle, etc. original positioning data, and send the original positioning data to the data preprocessing module 1250 for processing, to provide accurate position reference data for subsequent formation control, task allocation, and collaborative navigation.

[0048] The medium environment detection module 1240 is configured to detect the contact state data of each part of the aerial vehicle with the water body in real time, and send the contact state data to the data preprocessing module 1250; wherein the contact state data is the contact state of each part of the aerial vehicle with the water body, including a completely immersed state, a partially contacted state, and a completely separated state.

[0049] Specifically, the medium environment detection module 1240 in the embodiment adopts a multi-point distributed conductivity sensor array to determine the medium environment of the aerial vehicle through conductivity change, that is, to detect the contact state of each part of the aerial vehicle with the water body in real time, and send the contact state data to the data preprocessing module 1250 for signal processing, to provide accurate environment state data for subsequent mode switching decision.

[0050] The data preprocessing module 1250 is configured to receive and process the original depth data, the original attitude data, the original positioning data, and the contact state data, to generate the perception fusion data; and send the perception fusion data to the multi-modal collaborative control and execution system, the cluster collaboration system, and the autonomous decision and task planning system.

[0051] Specifically, the data preprocessing module 1250 adopts an embedded processor to fuse the sensor data sent by each of the above modules, to perform filtering, calibration, time synchronization, and data format unification processing, to generate the perception fusion data, and send the perception fusion data to other modules to provide data support.

[0052] Figure 3 An embodiment of the present disclosure provides a structural diagram of a multi-modal collaborative control and execution system in a cross-medium vehicle system. As shown in Figure 3 The multi-modal collaborative control and execution system 130 includes a control unit 1310, a bionic variable configuration wing control module 1320, a multi-modal propulsion control module 1330, and an attitude and heading control module 1340. In addition, Figure 4 An embodiment of the present disclosure provides a structural diagram of a cluster collaborative system in a cross-medium vehicle system. As shown in Figure 4 The cluster collaborative system 140 includes a cluster communication coordination module 1410, a cluster collaborative navigation control module 1420, a collaborative task allocation module 1430, and a collaborative task execution strategy module 1440. Among them, The control unit 1310 is configured to receive the perception fusion data sent by the data preprocessing module 1250 and receive the collaborative task execution control instruction sent by the collaborative task execution strategy module 1440, determine a first type of driving signal, a second type of driving signal, and a third type of driving signal according to the perception fusion data, the collaborative task execution control instruction, and the current running state of the vehicle, send the first type of driving signal to the bionic variable configuration wing control module, send the second type of driving signal to the multi-modal propulsion control module, and send the third type of driving signal to the attitude and heading control module. The first type of driving signal is used to switch the wing configuration of the vehicle, the second type of driving signal is used to drive different propulsion units of the vehicle, and the third type of driving signal is used to drive the rudder surface control of the vehicle. The control unit 1310 can include a first control unit and a second control unit. The first control unit, i.e., the top-level decision control unit, adopts an embedded flight control computer, which can execute intelligent decision algorithms, cross-medium modal switching logic, cluster collaborative control, and exception handling decisions. The second control unit, i.e., the bottom-level execution control unit, adopts a real-time microcontroller, which is used to receive instructions from the first control unit and convert them into driving signals executable by the vehicle to achieve millisecond-level control response and precise execution without involving cross-medium modal switching.

[0053] The bionic variable configuration wing control module 1320 is configured to receive and respond to the first type of driving signal sent by the control unit 1310 to switch the wing configuration required by the vehicle in different medium driving and / or cross-medium conditions, and generate a first feedback signal.

[0054] The first feedback signal can be used to represent the wing feedback parameters of the current wing state of the vehicle.

[0055] Specifically, the bionic variable configuration wing control module 1320 is configured to receive the first type of driving signal from the control unit 1310 and generate a first feedback signal to feed back the current wing feedback parameter of the vehicle. For example, the first type of driving signal sent by the second control unit is received to simulate the contraction of the pectoral fin of a fish and the expansion mechanism of the wings of a bird, and the intelligent switching of the wing configuration is realized according to different medium environments.

[0056] In an implementation, the bionic variable configuration wing control module 1320 is specifically configured to: receive the first type of driving signal sent by the control unit; and adjust the wing configuration of the vehicle to a target wing configuration through the first type of driving signal when the current operation mode of the vehicle meets a preset condition; and the target wing configuration is selected from one of the following according to the preset condition met by the current operation mode of the vehicle: when the current operation mode is an underwater mode, the shark pectoral fin wing configuration is adjusted; when the current operation mode is an in-and-out-of-water mode, the flying fish pectoral fin wing configuration is adjusted; when the current operation mode is a ground effect flight mode, the ground effect flight wing configuration is adjusted; and when the current operation mode is an air flight mode, the albatross wing configuration is adjusted.

[0057] The bionic variable configuration wing control module 1320 adopts a steering engine driver and a deformation / flight integrated coordinated control technology, simulates the bending contraction mechanism of the pectoral fin of a fish and the expansion mechanism of the wings of a bird, and is configured to dynamically adjust the wing configuration according to different flight modes (underwater diving, in-and-out-of-water, ground effect flight, and air flight). The module receives a driving signal from a bottom-layer execution control unit, and realizes intelligent bionic wing deformation control through an adaptive fuzzy switching control algorithm. The core innovation of the bionic variable configuration wing control module lies in the engineering implementation of a multi-level bionic mechanism.

[0058] The shark pectoral fin wing configuration is also a shark pectoral fin bionic underwater configuration. When the depth data of the depth perception sensor module indicates that the vehicle is in an underwater environment, the contraction mechanism of the pectoral fin of a shark when swimming underwater is simulated, a fluid dynamics optimization algorithm based on deep learning is used to establish a mapping relationship between the wing angle and the hydrodynamic resistance, and the wing is optimized through an intelligent algorithm to tightly fit the fuselage to form a streamlined shape, reduce turbulence, and reduce hydrodynamic resistance.

[0059] The flying fish pectoral fin wing configuration is also a flying fish pectoral fin bionic out-of-water configuration. When the medium environment detection module detects a signal that the fuselage starts to separate from the water body, the rapid expansion mechanism of the pectoral fin of a flying fish when jumping out of the water is used for reference, a data-driven dynamic expansion control algorithm is used, and the wing is intelligently converted from a contraction state to an expansion state through a multi-joint hinge design to ensure that the best lift characteristics are obtained in the water exit moment; The wing configuration of ground effect flight, that is, the special configuration for ground effect flight, adopts an intelligent control algorithm based on ground effect bionics optimization when the vehicle is near the water surface based on the high-precision positioning and navigation module, the wing is adjusted to a downward reverse angle configuration, and a downward airflow is generated by the front symmetrical double propulsion units in the multi-modal propulsion control module, forming a ground effect lift augmentation zone between the lower surface of the wing and the water surface. The albatross wing configuration, that is, the albatross wing bionic flight configuration, adopts a data-driven optimization algorithm based on reinforcement learning when the vehicle is out of the ground effect area based on the high-precision positioning and navigation module, referring to the wing type adjustment mechanism of albatross gliding on the sea surface, the wing has active camber adjustment capability and can adjust the wing type geometric parameters in real time according to the flight environment to optimize the lift-drag ratio performance.

[0060] Specifically, the bionic wing deformation process in the embodiment is uniformly coordinated by a top-level decision control unit, and a self-adaptive fuzzy switching control algorithm is used to realize accurate modal switching control. The algorithm constructs a fuzzy reasoning system based on multi-sensor fusion data, automatically selects the optimal control strategy through a fuzzy rule base according to multi-dimensional input parameters such as depth information, attitude state, and medium environment, and realizes smooth switching between different modes.

[0061] The multi-modal propulsion control module 1330 is configured to receive and respond to the second type of driving signal sent by the control unit to drive the propulsion units required by the vehicle in different media and / or in the case of crossing media, and generate a second feedback signal.

[0062] The second feedback signal can be used to represent the propulsion feedback parameter of the current propulsion state of the vehicle.

[0063] Specifically, the multi-modal propulsion control module 1330 is configured to receive the control unit 1310, for example, receive the second type of driving signal sent by the second control unit, drive the front symmetrical double propulsion unit and the rear single propulsion module to adapt to the propulsion requirements of different media and rapid maneuvering, and feedback the propulsion feedback parameter of the propulsion state of the vehicle after driving as the second feedback signal.

[0064] The attitude and heading control module 1340 is configured to receive and respond to the third type of driving signal sent by the control unit to maintain the attitude stability of the vehicle in different media and / or in the case of crossing media, and generate a third feedback signal.

[0065] The third feedback signal can be used to represent the attitude feedback parameter of the current attitude state of the vehicle.

[0066] Specifically, the attitude and heading control module 1340 is configured to receive the control unit 1310, for example, receive the third type of driving signal sent by the second control unit, execute rudder control to ensure the attitude stability of the vehicle in different media and / or in the case of cross-media mode switching and cluster maneuvering process. And the attitude reached by the vehicle after driving is fed back as a third feedback signal.

[0067] In addition, the cluster communication coordination module 1410 in the cluster coordination system 140 is configured to receive the perception fusion data sent by the data preprocessing module of each vehicle in the cluster, generate cluster shared data, and send the cluster shared data to the cluster coordinated navigation control module 1420. The cluster shared data is a cluster state information data packet and an instruction synchronization queue. The cluster state information is the sharing of the cluster overall formation layout and the position and running state of each vehicle in the cluster. The instruction synchronization queue is used to ensure the synchronization of the instructions of each vehicle in the cluster.

[0068] Specifically, the cluster communication coordination module 1410 in the embodiment is used to realize the communication between the vehicles in the cluster, send the collected cluster shared data of each vehicle to the cluster coordinated navigation control module 1420, and ensure the real-time sharing of the state information of each vehicle and the accurate synchronization of the coordinated instructions.

[0069] The cluster coordinated navigation control module 1420 is configured to receive the cluster shared data sent by the cluster communication coordination module and receive other perception fusion data sent by other vehicles in the cluster, generate a cluster coordinated trajectory according to the cluster shared data and the other perception fusion data, and send the cluster coordinated trajectory to the coordinated task allocation module 1430.

[0070] The cluster coordinated trajectory can be a trajectory parameter referred to by the vehicle to complete the cluster formation requirement.

[0071] Specifically, the cluster coordinated navigation control module 1420 is configured to generate a cluster coordinated trajectory according to the formation requirement of the cluster coordination system 140 and the high-precision positioning and navigation data of each vehicle in the cluster, and distribute the cluster coordinated trajectory to the coordinated task allocation module 1430 in the cluster coordination system 140 of each vehicle, to ensure the coordinated navigation and formation control of each vehicle in the cluster.

[0072] The coordinated task allocation module 1430 is configured to receive the cluster coordinated trajectory sent by the local cluster coordinated navigation control module 1420, generate a target task based on the cluster coordinated trajectory, the current running state of the vehicle, and the task instruction, and send the target task to the coordinated task execution strategy module through a distributed algorithm.

[0073] The embodiment can know the running trajectory of other vehicles in the cluster through receiving the cluster cooperative trajectory, and can generate a target task through intelligent task allocation and load balancing by a distributed algorithm, in combination with the current running state of the local machine and the local task instruction. The embodiment can allocate vehicles with different modal advantages to perform corresponding tasks according to the bionic wing deformation capability and the cross-medium navigation performance of each vehicle: for example, underwater reconnaissance tasks are allocated to vehicles with shark pectoral fin configuration advantage, ground effect flight tasks are allocated to vehicles with ground effect flight configuration advantage, and high-altitude cruising tasks are allocated to vehicles with albatross wing configuration advantage.

[0074] The cooperative task execution strategy module 1440 is configured to receive the target task sent by the cooperative task allocation module 1430, process the target task based on a cooperative task allocation algorithm, and generate a cooperative operation control instruction and send the cooperative operation control instruction to the control unit 1310 in the multi-modal cooperative control and execution system 130.

[0075] The cooperative operation control instruction is a cluster team control instruction obtained by analyzing the target task, and is used to control the vehicles to form a formation and change the formation according to requirements.

[0076] Specifically, the cooperative task execution strategy module 1440 in the embodiment is configured to receive the task allocation table and the priority list of the cooperative task allocation module 1430, generate a cooperative operation control instruction according to a cooperative task allocation algorithm, and realize dynamic formation and formation change of multiple vehicles.

[0077] In addition, the cooperative task execution strategy module 1440 in the embodiment adopts a bionic swarm intelligence algorithm to coordinate the modal switching and timing control of each vehicle. Underwater formation control based on fish swimming behavior: when multiple vehicles are underwater, the vehicles are coordinated to maintain a shark pectoral fin bionic underwater configuration, so as to realize a close formation and flexible obstacle avoidance; Air formation control based on bird flocking mode: when the vehicle group enters an air flight mode, the vehicles are coordinated to switch to an albatross wing bionic flight configuration, so as to maintain a stable formation and optimal energy consumption; Entry and exit water formation coordination based on dolphin group water jumping behavior: the timing of repeated and rapid entry and exit of water of each vehicle is coordinated to ensure that the cluster synchronously enters and exits water without collision, and the overall cross-medium conversion efficiency is optimized.

[0078] The embodiment can realize multi-directional, multi-level cooperative operation and efficient task execution by sending the cooperative operation control instruction to the top-level decision control unit in the multi-modal cooperative control and execution system 130 of each vehicle.

[0079] In an implementable manner, the control unit 1310 includes: a first control unit configured to receive the perception fusion data sent by the data preprocessing module 1250 and the cooperative task execution strategy module 1440, and generate a coordinated control instruction according to the perception fusion data, the cooperative task execution control instruction, and the current running state of the aerial vehicle, and send the coordinated control instruction to a second control unit and the autonomous decision and task planning system 110; a second control unit configured to receive the coordinated control instruction of the first control unit, and in response to converting into a first type of driving signal, a second type of driving signal, and a third type of driving signal, send the first type of driving signal to the bionic variable-configuration wing control module 1320, send the second type of driving signal to the multi-modal propulsion control module 1330, and send the third type of driving signal to the attitude and heading control module 1340.

[0080] The coordinated control instruction can be a control instruction for controlling the running track and the mode switching of the aerial vehicle.

[0081] Specifically, the first control unit in the embodiment is a top-level decision control unit, and the second control unit is a bottom-level execution control unit. The top-level decision control unit and the bottom-level execution control unit adopt a hierarchical cooperative control architecture. The top-level decision control unit runs an adaptive control firmware and is responsible for executing high-level control algorithms such as adaptive fuzzy switching control, track tracking, and mode switching decision. The bottom-level execution control unit focuses on receiving the control instruction from the top-level decision control unit and the optimal water entry and exit path parameters from the water entry and exit path planning module and converting them into pulse width modulation signals for output, so as to realize the precise driving of the actuators of the bionic variable-configuration wing control module, the multi-modal propulsion control module, and the attitude and heading control module. Real-time communication between high-level and low-level control modules is realized through a standard communication bus, and the modularity design, reliability, and functional expandability of the control system are improved.

[0082] For example, the second control unit receives the coordinated control instruction of the first control unit, and in response to converting into a first type of driving signal, a second type of driving signal, and a third type of driving signal, sends the first type of driving signal to the bionic variable-configuration wing control module 1320, sends the second type of driving signal to the multi-modal propulsion control module 1330, and sends the third type of driving signal to the attitude and heading control module 1340, so as to adjust the aerial vehicle to reach the target state.

[0083] Figure 5 A structure diagram of an autonomous decision and task planning system in a cross-medium aerial vehicle system is shown. Figure 5As shown, the autonomous decision and task planning system 110 includes a mode switching decision module 1110, an in-out water path planning module 1120, and a communication and instruction management module 1130. Among them, The mode switching decision module 1110 is configured to receive the coordinated control instruction of the first control unit in the case of cross-medium mode switching of the vehicle, generate a mode switching control instruction based on an adaptive rule algorithm and the coordinated control instruction, and send the mode switching control instruction to the in-out water path planning module 1120.

[0084] The mode switching control instruction can be an instruction for controlling the vehicle to switch modes in a cross-medium situation.

[0085] Specifically, the mode switching decision module 1110 is configured to receive the coordinated control instruction of the first control unit, process the multi-sensor data uncertainty of the multi-source information perception and state recognition system in combination with an adaptive fuzzy control rule base, realize intelligent mode switching decision based on the bionic wing deformation state and medium environment characteristics, generate a mode switching control instruction using a finite state machine and a fuzzy logic control algorithm, and send the mode switching control instruction to the in-out water path planning module 1120.

[0086] The in-out water path planning module 1120 receives the mode switching control instruction sent by the mode switching decision module 1110, calculates optimal in-out water path parameters according to the mode switching control instruction, and sends the optimal in-out water path parameters to the second control unit to continuously control the vehicle to reach a target state.

[0087] The optimal in-out water path parameters can be the in-out water angle, speed sequence, and timing control parameters of the vehicle when entering and leaving the water.

[0088] Specifically, the in-out water path planning module 1120 is configured to receive the mode switching control instruction of the mode switching decision module 1110, process the depth reference data, real-time attitude feedback data, and environmental state data of the multi-source information perception and state recognition system based on a neural network algorithm, calculate optimal in-out water path parameters, send the optimal in-out water path parameters to the second control unit in the multi-mode collaborative control and execution system 130, and realize trajectory planning and attitude optimization for repeated and rapid in-out water of the vehicle. The communication and instruction management module 1130 is configured to receive the task instruction sent by the ground station, send the task instruction to the collaborative task allocation module, and further configured to feed back the task execution state information to the ground station.

[0089] Specifically, the communication and instruction management module 1130 is configured to perform bidirectional data interaction with the ground station, receive task instructions and distribute them to the collaborative task allocation module 1430 in the cluster coordination system 140, support online updating and remote monitoring of control parameters, and feed back system running status and task execution results to the ground station to realize remote command and control.

[0090] In an implementation, the second control unit is further configured to: In the case of cross-medium modal switching, the optimal in-out water path parameters sent by the in-out water path planning module 1120 are received, and the first, second and third driving signals are updated according to the optimal in-out water path parameters and the cluster coordination trajectory, to control the vehicle to reach the target state.

[0091] Since the cluster vehicle not only needs to perform cluster coordination during task execution, but also needs to perform corresponding position transformation for different modes and tasks, and needs to complete the task target of the local vehicle according to the target task, the vehicle often needs to repeatedly enter and exit the water multiple times during task execution. The embodiment can parse and generate corresponding driving signals by sending optimal in-out water path parameters to the second control unit, update the first, second and third driving signals in time, and thus complete the requirements of repeated in-out water and formation.

[0092] Figure 6 A complete structure schematic diagram of an exemplary cross-medium vehicle system provided by the embodiment of the present disclosure is shown. As shown in Figure 6 The embodiment takes the first control unit and the second control unit as the core and exemplarily shows a clearer and more complete vehicle system structure. The interaction between different modules has been described in detail above and will not be repeated here. It should be noted that since the vehicle needs to complete the target task given by the cluster according to its position and characteristics in the cluster during task execution, and often repeatedly enters and exits the water during the process of completing the target task, the mechanism for controlling the vehicle body is different in different situations. For example, in the case of different media, the first control unit in the vehicle of the embodiment can generate coordination control instructions to the second control unit by sending the perception fusion data sent by the data preprocessing module, the collaborative work control instructions sent by the collaborative task execution strategy module, and the current running state of the vehicle itself, to reach the target state. At the same time, the first control unit also sends the coordination control instructions to the modal switching decision module, so that the vehicle can generate modal switching control instructions based on the adaptive rule algorithm in the modal switching decision module to send them to the in-out water path planning module, to constantly control the vehicle to reach the target state by calculating the optimal in-out water path parameters from the in-out water path planning module and sending them to the second control unit.

[0093] The embodiment realizes comprehensive perception of the vehicle in different medium environments through the multi-source information perception and state recognition system, provides accurate data support for the deformation of the bionic wing, rapid water entry and exit, and cluster cooperation; realizes accurate cooperation of the bionic variable-configuration wing, rapid water entry and exit trajectory control, and multi-modal propulsion through the multi-modal cooperative control and execution system, and each module cooperates to form a unified control system; realizes formation control, task allocation, and cooperative operation of multiple vehicles through the cluster cooperation system, and fully develops the cluster advantage; realizes intelligent modal switching decision, path planning, and cooperative control through the autonomous decision and task planning system, and overall plans and optimizes the global. Application of the system effectively realizes intelligent control of the deformation of the bionic wing, accurate control of repeated rapid water entry and exit, and cooperative operation of the multi-vehicle cluster, cross-medium modal conversion and cooperative control of a single platform in underwater diving, surface ground effect flight, and air flight, and significantly improves the stability, continuity, rapidity, and autonomy of cross-medium operation, and solves the technical problems that the traditional vehicle cannot cross the border, the wing deformation efficiency is low, the water entry and exit control precision is poor, and the cluster cooperation capability is poor.

[0094] In an implementable manner, the adaptive rule algorithm is an algorithm for generating a coordinated control instruction based on an output quantity of an inner controller of the vehicle and a fuzzy rule weight.

[0095] The inner controller can be a special controller designed for different navigation modes of the vehicle. For example, it can include: an underwater diving mode controller for controlling the bionic underwater configuration of the shark pectoral fin; a water exit mode controller for controlling the bionic water exit configuration of the flying fish pectoral fin; a ground effect flight mode controller for controlling the ground effect flight special configuration; and an air flight mode controller for controlling the bionic flight configuration of the albatross wing.

[0096] Specifically, the calculation process of the adaptive rule algorithm can refer to the following: The attitude quaternion solution formula (1) is expressed as: (1) wherein, is an attitude quaternion describing the current attitude of the vehicle, is an angular velocity vector, represents quaternion multiplication; is a time derivative of the attitude quaternion; is the current time of the vehicle.

[0097] The embodiment can determine real-time attitude information of the vehicle according to the perception fusion data collected by the vehicle, construct a state vector based on the real-time attitude information, and determine the control instruction of the vehicle through the adaptive rule algorithm. The state vector is constructed as , for example, it can include depth value, attitude angle, medium contact state, speed, etc. Enter the fuzzy weight calculation formula (2).

[0098] Among them, the fuzzy weight calculation formula (2) is expressed as: (2) in, For the The activation weight of each fuzzy rule; For the The membership function value of a fuzzy rule; For the The membership function value of a fuzzy rule; is the total number of fuzzy rules; is the current system state vector; the denominator ensures that the sum of all weights is 1 to achieve normalization.

[0099] Specifically, this embodiment can be based on the current state of the aircraft Calculate the activation weight of each mode to determine how to fuse different sub-controllers and obtain the fuzzy weight , and then the fuzzy weight Enter the adaptive fuzzy switching control algorithm formula (3). For example, when the vehicle switches from underwater to air, the weight of the underwater sub-controller gradually decreases, and the weight of the out-of-water mode sub-controller increases, achieving smooth mode switching.

[0100] Among them, the adaptive fuzzy switching control algorithm formula (3) is expressed as: (3) in, For the The activation weight of the fuzzy rules, For the The output of the sub-controller, n is the total number of fuzzy rules, corresponding to four navigation modes, n=4; is the total control output.

[0101] Specifically, the total control output is the final control instruction sent to the second control unit, The output of the dedicated controller corresponding to the four modes of underwater diving, water exit, ground effect flight, and air flight, Indicates the The activation degree of each mode. , which can fuse all sub-controller outputs to generate the final control instruction. Since the spacecraft may have execution deviations during the mission execution, this embodiment also provides an adaptive parameter update rate formula (4) The adaptive parameter update rate formula (4) is: (4) in, For the Adaptive parameter vector of each sub-controller; is the adaptive gain, , e(t) is the tracking error; is a positive definite weight matrix, For the The basis function vector of the modes; Specifically, Adjust the speed for the control parameter; e(t) can be the difference between the desired trajectory and the actual trajectory; The update weights of different parameters can be adjusted. The characteristic function of a certain mode of navigation of an aircraft can be described. In this embodiment, the execution effect of the aircraft can be learned by collecting data, thereby updating the sub-controller parameters of the aircraft in a certain mode.

[0102] The adaptive rule algorithm of this embodiment can dynamically adjust control parameters according to environmental conditions and mission requirements, ensuring intelligent switching and precise control between different modes such as underwater diving, rapid water exit, ground effect flight, and aerial cruising. Precise drive is achieved through the second control unit, that is, the underlying execution control unit, and the entire deformation process can be completed quickly to ensure the continuity of cross-media conversion.

[0103] Figure 7 A third perspective diagram of an exemplary cross-media vehicle provided in an embodiment of the present disclosure is provided. Figure 8 An exploded schematic diagram of an exemplary cross-media vehicle provided in an embodiment of the present disclosure.

[0104] like Figure 8 As shown, the cross-medium vehicle includes but is not limited to: wings 810, front propellers 821, rear propellers 822, fuselage 830, and tail 840. Wings 810 are integrated with the bionic variable configuration wing control module 1320 of the above-mentioned embodiment, enabling the bionic variable configuration wing control module 1320 to adjust the wing configuration; front propellers 821 and rear propellers 822 are integrated with the multi-modal propulsion control module 1330, enabling the multi-modal propulsion control module 1330 to achieve a propulsion state; and tail 840 is integrated with the attitude and heading control module 1340 of the above-mentioned embodiment, enabling the attitude and heading control module 1340 to maintain attitude stability, thereby completing the mission issued by the ground station.

[0105] Figure 9A flowchart of a control method of a cross-medium vehicle system is provided for the embodiments of the present disclosure. The method can be executed by a control device of a cross-medium vehicle system provided by the embodiments of the present disclosure, and the device can be implemented in the form of software and / or hardware. The method specifically includes the following steps: S910, receiving a task instruction sent by a ground station.

[0106] The task instruction is used to indicate a collective task that the cluster needs to complete.

[0107] Specifically, the task instruction is a task instruction and operation requirement for switching from underwater to air mode, which is received by a communication and instruction management module in each vehicle of the vehicle cluster and sent by the ground station, and is sent to a cooperative task allocation module.

[0108] S920, acquiring perception data of the vehicle in different medium environments in real time, and preprocessing the perception data into perception fusion data.

[0109] The perception data includes at least one of original depth data, original attitude data, original positioning data, and contact state data.

[0110] Specifically, the vehicle receives the above-mentioned original data of each sensor, performs filtering, calibration, time synchronization, and data format unification processing, and takes the processed environmental perception information as the perception fusion data.

[0111] S930, receiving other perception fusion data sent by other vehicles in the cluster in real time, and generating a cluster cooperative trajectory according to the perception fusion data and the other perception fusion data sent by the other vehicles in the cluster.

[0112] Specifically, the vehicle of the present embodiment generates cluster shared data according to the local perception fusion data, and generates a cluster cooperative trajectory in combination with the perception fusion data sent by the other vehicles in the cluster, so as to obtain the running trajectory of the vehicle in the cluster.

[0113] S940, determining a target task of the vehicle according to the cluster cooperative trajectory, a current running state of the vehicle, and the task instruction.

[0114] The current running state includes current running parameters of the vehicle when driving in different media or across media, and the target task belongs to at least one subtask in the collective task.

[0115] Specifically, the vehicle of the present embodiment generates a target task belonging to the vehicle based on the cluster cooperative trajectory, the current running state of the vehicle, and the task instruction through a distributed algorithm, and finally processes the target task based on a cooperative task allocation algorithm.

[0116] S950, determining optimal in-out water path parameters of the vehicle according to the target task, and generating driving signals based on the perception fusion data or the optimal in-out water path parameters to continuously control the vehicle to reach a target state.

[0117] Specifically, in the case of not involving cross-medium, the embodiment can generate a coordination control instruction executable by the vehicle itself according to the perception fusion data, and convert the coordination control instruction into the first type of driving signal, the second type of driving signal and the third type of driving signal to control the vehicle to reach the target state; in the case of involving cross-medium, the embodiment can generate a coordination control instruction executable by the vehicle itself according to the optimal in-out water path parameters, and convert the coordination control instruction into the first type of driving signal, the second type of driving signal and the third type of driving signal to control the vehicle to reach the target state.

[0118] The first type of driving signal is used to switch the wing configuration of the vehicle, the second type of driving signal is used to drive different propulsion units of the vehicle, and the third type of driving signal is used to drive the rudder control of the vehicle.

[0119] In the case of cross-medium modal switching, the vehicle of the embodiment receives the coordination control instruction of the first control unit, generates a modal switching control instruction based on an adaptive rule algorithm and the coordination control instruction, calculates the optimal in-out water path parameters according to the modal switching control instruction, and continuously adjusts the running state of the vehicle to reach the target state according to the perception fusion data or the optimal in-out water path parameters to complete the target task. Specifically, the first type of driving signal, the second type of driving signal and the third type of driving signal are updated by the optimal in-out water path parameters to realize continuous updating of the running state of the vehicle. In addition, the embodiment is also used to feed back the task execution state information to the ground station.

[0120] In addition, after receiving the first type of driving signal, the embodiment receives the perception fusion data of the vehicle in real time; the current state of the vehicle is monitored in real time based on the perception fusion data; when the current state meets a preset condition, the wing configuration of the vehicle is adjusted to a target wing configuration by the first type of driving signal; wherein the target wing configuration is selected according to the preset condition met by the current state as follows: when the current state is an underwater mode, the shark pectoral fin wing configuration is adjusted; when the current state is an in-out water mode, the flying fish pectoral fin wing configuration is adjusted; when the current state is a ground effect flight mode, the ground effect flight wing configuration is adjusted; and when the current state is an air flight mode, the albatross wing configuration is adjusted.

[0121] For the convenience of understanding, the embodiment gives an exemplary more detailed control flow method: Specifically, the communication and instruction management module receives the task instructions and operation requirements for the mode switching from underwater to air mode issued by the ground station, and sends the task requirements to the collaborative task allocation module. The collaborative task allocation module receives the cluster coordination trajectory from the cluster communication coordination module, the task requirements from the communication and instruction management module, and the current operating parameters of the vehicle, and performs intelligent task allocation and load balancing to generate target tasks, i.e., task allocation table and priority list. The collaborative task execution strategy module receives the task allocation table and priority list, generates collaborative operation control instructions according to the collaborative task allocation algorithm, and sends them to the top-level decision control unit.

[0122] In the underwater ascent phase without cross-medium mode switching, the top-level decision control unit receives the fused sensor data of the multi-source information perception and state recognition system, the collaborative operation control instructions of the cluster coordination system, and the current operating parameters of the vehicle. Based on the intelligent decision algorithm, the coordination control instructions are generated and directly sent to the bottom-level execution control unit. The bottom-level execution control unit converts the coordination control instructions into pulse width modulation driving signals, which are sent to the rear single-propulsion module in the multi-mode propulsion control module to provide stable vertical ascent thrust, the rudder control module in the attitude and heading control module to ensure the attitude stability during the ascent process, and the bionic variable configuration wing control module to maintain the shark pectoral fin bionic underwater configuration. At the same time, the cluster communication coordination module continuously updates the position information of each vehicle, and the collaborative task execution strategy module ensures the formation coordination of each vehicle during the ascent process.

[0123] When the depth perception sensor module detects that the vehicle has risen to a preset cross-medium switching depth threshold, or when the medium environment detection module detects that the vehicle body has started to separate from the water (partial contact state), the data preprocessing module sends the depth reference data and other environmental state data to the top-level decision control unit. The top-level decision control unit determines that cross-medium mode switching is needed, generates coordination control instructions, and sends them to the mode switching decision module.

[0124] The mode switching decision module processes the uncertainty of multi-sensor data based on the adaptive fuzzy control rule base, realizes intelligent mode switching decision based on the bionic wing deformation state and medium environment characteristics, generates mode switching control instructions, and sends them to the repeated rapid water entry and exit path planning module. The repeated rapid water entry and exit path planning module processes multi-source sensor fusion data based on neural network algorithm, calculates the optimal water exit path parameters (including water exit angle, speed sequence and timing control parameters), and sends the optimal water entry and exit path parameters to the bottom-level execution control unit. The bottom-level execution control unit converts the optimal water entry and exit path parameters into precise driving signals, and simultaneously drives the bionic variable configuration wing control module, the mode propulsion control module, and the mode propulsion control module.

[0125] Specifically, the bionic variable configuration wing control module quickly converts from the shark pectoral fin bionic underwater configuration to the flying fish pectoral fin bionic water-emulating configuration, simulating the rapid deployment mechanism of the flying fish pectoral fin to ensure the best lift characteristics at the moment of emerging from the water. The front and rear propulsion units of the modal propulsion control module work together to provide precise cross-medium conversion thrust to overcome water surface resistance; the attitude and heading control module performs precise rudder control to maintain the attitude stability according to the planned path during the water-emergence process.

[0126] Once the environmental monitoring module confirms the vehicle has completely exited the water and entered the air, it sends environmental status data to the data preprocessing module. Simultaneously, the high-precision positioning and navigation module initiates GPS positioning and sends positioning data to the data preprocessing module. The data preprocessing module transmits environmental status and positioning information to the top-level decision control unit, which executes the intelligent decision-making algorithm to generate coordinated control instructions that are sent to the bottom-level execution control unit.

[0127] The underlying execution control unit then drives the bionic variable configuration wing control module, adjusting the wing to a configuration specifically for ground effect flight, using an intelligent control algorithm based on ground effect bionic optimization. Simultaneously, the multi-modal propulsion control module activates the symmetrical dual propulsion units at the front, generating downward airflow and forming a ground effect lift zone between the wing's underside and the water surface. Furthermore, the swarm collaborative navigation control module updates the swarm's collaborative trajectory, and the collaborative mission execution strategy module coordinates the various vehicles to establish a formation at a preset ground effect flight altitude.

[0128] After each vehicle has flown a predetermined distance in ground-effect flight, the top-level decision-making control unit completes an attitude stability check and self-checks each subsystem's status. It then generates coordinated control instructions and sends them to the bottom-level execution control unit. This unit then activates the bionic variable-configuration wing control module to adjust the wing to a bionic flight configuration similar to an albatross wing, simulating the efficient gliding mechanism of the albatross. This improves aerodynamic performance through a data-driven optimization algorithm based on reinforcement learning. Simultaneously, it activates the single propulsion module in the rear of the multimodal propulsion control module to increase thrust.

[0129] The cluster collaborative navigation control module generates the cluster collaborative trajectory for air cruising, and the collaborative task execution strategy module coordinates each aircraft to climb to the preset cruising altitude and enter the air cruising formation mode.

[0130] During airborne cruise, the CCM establishes a high-quality wireless communication link with the ground station to transmit ocean buoy monitoring data. After completing the data transmission mission, the CCM can coordinate returning underwater or continuing other airborne missions based on new instructions from the ground station via the CCM.

[0131] The embodiment effectively realizes intelligent deformation of the bionic wing, accurate control of repeated and rapid water entry and exit, and cooperative operation of the multi-vehicle cluster, significantly improves the stability and autonomy of cross-medium operation, and solves the technical problems of low deformation efficiency, poor water entry and exit control accuracy, and lack of cluster cooperation capability of the traditional vehicle wing.

[0132] Meanwhile, the embodiment can also flexibly adjust the setting of the bionic wing deformation parameter, the rapid water entry and exit control parameter, and the cluster cooperation strategy according to the actual sea conditions and task requirements, to ensure the best overall performance and safety. When the sea conditions are poor, the triggering condition of the bionic wing deformation and the trajectory parameter of the rapid water entry and exit can be adjusted; when rapid cluster maneuvering is required, the bionic wing deformation speed and cluster cooperation timing can be optimized.

[0133] Figure 10 is a structural schematic diagram of a control device of a cross-medium vehicle system provided by the embodiment of the disclosure, and the device specifically comprises: The task receiving module 1010 is configured to receive a task instruction sent by a ground station, wherein the task instruction is used to instruct a collective task that needs to be completed by the cluster. The first data acquisition module 1020 is configured to acquire perception data of the vehicle in different medium environments in real time, and pre-process the perception data into perception fusion data. The second data acquisition module 1030 is configured to receive perception fusion data sent by other vehicles in the cluster in real time, and generate a cluster cooperation trajectory according to the perception fusion data and the perception fusion data sent by other vehicles in the cluster. The task determining module 1040 is configured to determine a target task of the vehicle according to the cluster cooperation trajectory, a current running state of the vehicle, and the task instruction, wherein the current running state comprises a current running parameter of the vehicle when driving in different media or crossing media, and the target task belongs to at least one subtask in the collective task. The state updating module 1050 is configured to determine optimal water entry and exit path parameters of the vehicle according to the target task, and generate a driving signal based on the perception fusion data or the optimal water entry and exit path parameters, to constantly control the vehicle to reach a target state.

[0134] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, the steps described in the disclosure can be executed in parallel, in sequence, or in a different order, as long as the desired results of the technical solutions of the disclosure can be achieved, which is not limited herein.

[0135] In addition, the terms "first", "second", etc. are used only for the purpose of description, and should not be understood as indicating or implying relative importance or an indicated number of technical features. Therefore, the features defined as "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0136] The above description is merely a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present disclosure, which should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A cross-media aircraft system, characterized in that: The system comprises: An autonomous decision-making and mission planning system, configured to receive mission instructions from a ground station and send them to the cluster collaboration system, wherein the mission instructions are used to indicate a collective mission that the cluster needs to complete; A multi-source information perception and state recognition system is used to obtain the perception data of the aircraft in different media environments in real time, pre-process the perception data into perception fusion data, and then send the perception fusion data to the multi-modal collaborative control and execution system and the cluster collaborative system; a swarm collaboration system configured to receive perception fusion data sent by the aircraft's multi-source information perception and state recognition system, other perception fusion data sent by other aircraft in the swarm, and mission instructions, generate a swarm collaboration trajectory based on the perception fusion data and other perception fusion data sent by other aircraft in the swarm, and determine a target mission for the aircraft based on the swarm collaboration trajectory, the aircraft's current operating state, and the mission instructions, and transmit the mission to an autonomous decision-making and mission planning system, wherein the current operating state includes current operating parameters of the aircraft when traveling in different media or crossing media, and the target mission belongs to at least one subtask of the collective mission; The autonomous decision-making and mission planning system is further configured to receive the target mission sent by the cluster collaboration system, determine the optimal entry and exit path parameters of the vehicle based on the target mission, and then feed the optimal entry and exit path parameters back to the multimodal collaborative control and execution system; A multimodal collaborative control and execution system is used to receive the perception fusion data sent by the multi-source information perception and state recognition system and the optimal water entry and exit path parameters sent by the autonomous decision-making and mission planning system, and generate a driving signal based on the perception fusion data or the optimal water entry and exit path parameters to continuously control the vehicle to reach the target state.

2. The system according to claim 1, wherein: The perception data includes at least one of original depth data, original posture data, original positioning data, and contact state data. The multi-source information perception and state recognition system includes: A depth perception sensor module is used to measure the original depth data of the vehicle diving in the water in real time and send the original depth data to the data preprocessing module; An attitude and motion state detection module is used to obtain the original attitude data of the aircraft and send the original attitude data to the data preprocessing module; a high-precision positioning and navigation module, configured to obtain raw positioning data of the aircraft and send the raw positioning data to the data preprocessing module; a medium environment detection module, configured to detect in real time the contact status data between various parts of the vehicle and the water body, and send the contact status data to the data preprocessing module; A data preprocessing module is used to receive and process the original depth data, the original posture data, the original positioning data and the contact state data to generate the perception fusion data; and send the perception fusion data to the multimodal collaborative control and execution system and the cluster collaborative system.

3. The system according to claim 2, characterized in that The cluster collaboration system includes: A cluster communication coordination module is used to receive the perception fusion data sent by the data pre-processing module of each aircraft in the cluster, generate cluster shared data, and send the cluster shared data to the cluster collaborative navigation control module; a cluster collaborative navigation control module, configured to receive cluster shared data sent by the cluster communication coordination module, as well as other perception fusion data sent by other vehicles in the cluster, generate cluster collaborative trajectories based on the cluster shared data and other perception fusion data, and send the generated trajectories to the collaborative task allocation module; a collaborative task allocation module, configured to receive the cluster collaborative trajectory sent by the cluster collaborative navigation control module, generate a target task through a distributed algorithm based on the cluster collaborative trajectory, the current operating state of the vehicle, and the task instruction, and send the target task to the collaborative task execution strategy module; The collaborative task execution strategy module is used to receive the target task sent by the collaborative task allocation module, process the target task based on the collaborative task allocation algorithm, generate collaborative operation control instructions and send them to the multimodal collaborative control and execution system.

4. The system according to claim 3, characterized in that The multimodal collaborative control and execution system includes: a control unit, configured to receive the perception fusion data sent by the data preprocessing module and the collaborative operation control instruction sent by the collaborative task execution strategy module, determine a first type of drive signal, a second type of drive signal, and a third type of drive signal based on the perception fusion data, the collaborative operation control instruction, and the current operating state of the aircraft, send the first type of drive signal to the bionic variable configuration wing control module, send the second type of drive signal to the multimodal propulsion control module, and send the third type of drive signal to the attitude and heading control module, wherein the first type of drive signal is used to switch the wing configuration of the aircraft, the second type of drive signal is used to drive different propulsion units of the aircraft, and the third type of drive signal is used to drive the control surface of the aircraft; a bionic variable configuration wing control module, configured to receive and respond to a first type of drive signal sent by the control unit to switch to a wing configuration required for the aircraft to travel on or across different media, and to generate a first feedback signal; a multi-modal propulsion control module, configured to receive and respond to a second type of driving signal sent by the control unit to drive a propulsion unit required by the vehicle for traveling in different media or across different media, and to generate a second feedback signal; an attitude and heading control module, configured to receive and respond to a third type of driving signal sent by the control unit to maintain attitude stability of the vehicle when traveling on or across different media, and to generate a third feedback signal; The first feedback signal, the second feedback signal and the third feedback signal collected in real time are used as the current operating parameters of the aircraft.

5. The system according to claim 4, characterized in that The control unit comprises: a first control unit, configured to receive the perception fusion data sent by the data preprocessing module and the collaborative operation control instructions sent by the collaborative task execution strategy module, and generate a coordination control instruction based on the perception fusion data, the collaborative operation control instructions, and the current operating state of the aircraft, and send the coordination control instruction to the second control unit and the autonomous decision-making and mission planning system; The second control unit is used to receive the coordinated control instructions of the first control unit, and in response convert them into a first type of drive signal, a second type of drive signal and a third type of drive signal, send the first type of drive signal to the bionic variable configuration wing control module, send the second type of drive signal to the multimodal propulsion control module, and send the third type of drive signal to the attitude and heading control module.

6. The system according to claim 5, characterized in that The autonomous decision-making and mission planning system includes: a mode switching decision module, configured to receive the coordinated control instruction from the first control unit when the vehicle switches between media modes, and generate a mode switching control instruction based on an adaptive rule algorithm and the coordinated control instruction, and send the generated mode switching control instruction to the water entry and exit path planning module; a water entry and exit path planning module, configured to receive the mode switching control instruction sent by the mode switching decision module, calculate optimal water entry and exit path parameters according to the mode switching control instruction, and send the optimal water entry and exit path parameters to the second control unit to continuously control the vehicle to reach a target state; The communication and instruction management module is used to receive the task instructions sent by the ground station and send the task instructions to the collaborative task allocation module; and is also used to feed back task execution status information to the ground station.

7. The system according to claim 6, wherein: The second control unit is further configured to: In the case of cross-medium mode switching, the optimal water entry and exit path parameters sent by the water entry and exit path planning module are received, and the first type of drive signal, the second type of drive signal and the third type of drive signal are updated according to the optimal water entry and exit path parameters and the cluster collaborative trajectory to control the aircraft to reach the target state.

8. The system according to claim 4, wherein: The bionic variable configuration wing control module is specifically used for: receiving a first type of driving signal sent by the control unit; When the current operating mode of the aircraft meets a preset condition, adjusting the wing configuration of the aircraft to a target wing configuration through the first type of driving signal; The target wing configuration is selected from the following according to the preset conditions satisfied by the current operating mode of the aircraft: When the current operating mode is the underwater mode, adjusting to a shark pectoral fin wing configuration; When the current operating mode is the water entry and exit mode, adjusting to a flying fish pectoral fin wing configuration; When the current operating mode is the ground effect flight mode, adjusting to a ground effect flight wing configuration; When the current operating mode is the aerial flight mode, the configuration is adjusted to the albatross wing configuration.

9. The system according to claim 6, wherein: The adaptive rule algorithm is an algorithm for generating coordinated control instructions based on the output of the sub-controllers in the aircraft and the fuzzy rule weights.

10. A control method for a cross-media aircraft system, characterized in that: The method comprises: Receiving a task instruction sent by a ground station, wherein the task instruction is used to indicate a collective task that the cluster needs to complete; Acquire the perception data of the aircraft in different media environments in real time, and preprocess the perception data into perception fusion data; receiving other perception fusion data sent by other aircraft in the cluster in real time, and generating a cluster collaborative trajectory based on the perception fusion data and the other perception fusion data sent by other aircraft in the cluster; determining a target task for the aircraft based on the cluster collaborative trajectory, the current operating state of the aircraft, and the task instruction, wherein the current operating state includes current operating parameters of the aircraft when traveling in different media or crossing media, and the target task belongs to at least one subtask of the collective task; The optimal water entry and exit path parameters of the aircraft are determined according to the target task, and then a driving signal is generated based on the perception fusion data or the optimal water entry and exit path parameters to continuously control the aircraft to reach the target state.

Images