Perception decision communication integrated aircraft brain software system

Abstract

The invention relates to a perception, decision and communication integrated aircraft brain software system, which belongs to the field of artificial intelligence and comprises an intelligent master control scheduling module, a decision change module, a communication adaptive interaction module and a situation perception module. According to the method, a modular design idea is adopted, modules in the method are flexibly combined, functional modules are newly added, different tasks executed by an unmanned system can be achieved, and generalization of brain software is greatly improved. The intelligent capability and the online autonomous decision-making capability of the aircraft can be greatly improved, the method can be flexibly applied to various unmanned equipment, and the intelligent enabling capability of the unmanned equipment can be greatly improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of artificial intelligence and relates to a flight vehicle brain software system integrating perception, decision and communication. BACKGROUND

[0002] Deep learning, as an important technology in the field of artificial intelligence, has been widely applied in recent years in the fields of unmanned systems, intelligent medical treatment, intelligent transportation and intelligent security, and the application of intelligent perception technology and intelligent decision technology in unmanned flight vehicles is particularly remarkable. However, the current unmanned flight vehicles are not highly intelligent and have weak autonomy, and it is difficult to achieve rapid online perception, decision and communication integration, and it is necessary to simultaneously process perception, decision and communication information with the aid of the ground or other means, and it is impossible to achieve high autonomy.

[0003] At present, the decision-making processes such as task allocation and search of most unmanned flight vehicle systems excessively rely on preset rules or simple algorithms, and lack real-time dynamic decision-making ability according to actual conditions, and when facing unexpected situations or unknown challenges, it is impossible to make rapid and effective decisions. SUMMARY

[0004] The technical problem solved by the application is to overcome the shortcomings of the prior art and provide a flight vehicle brain software system integrating perception, decision and communication, which can greatly improve the intelligence and autonomy of the flight vehicle.

[0005] The technical solution of the application is as follows.

[0006] The flight vehicle brain software system integrating perception, decision and communication comprises an intelligent main control and scheduling module, a decision change module, a communication adaptive interaction module and a situation awareness module.

[0007] The intelligent main control and scheduling module obtains real-time situation information from a deduction platform during the execution of a task by the flight vehicle, performs real-time online scheduling on each module of the flight vehicle brain software system, and controls the distribution of information flow.

[0008] The decision change module comprises a target change model and a search allocation model. The target change model acts on the middle section of the flight vehicle, updates the final landing point of the flight vehicle according to the latest target situation information obtained from the deduction platform, and the search allocation model acts on the end section of the flight vehicle, and according to the latest situation information obtained from the deduction platform and other flight vehicle information obtained from the situation awareness module, according to the different state information of the flight vehicle and the target value, an optimal task execution scheme under multiple search allocation decision points is performed, and specific action instructions of the search allocation task are output.

[0009] The communication adaptive interaction module is divided into a sending end and a receiving end. The sending end is responsible for data compression and source / channel coding on the information to be transmitted. The receiving end is responsible for decoding and reconstruction of the received data and transmission to the corresponding module.

[0010] The situation awareness module acts on the terminal stage of the aircraft. It identifies the imaging information of the current aircraft brain software system itself and the reconstructed information from other aircrafts, obtains accurate position information and type information of the target, and provides them to the search and allocation model.

[0011] Preferably, in the communication adaptive interaction module, the sending end includes an information screening model, an information compression model and a communication coding model. The information screening model is responsible for priority sorting, screening and combination of the perception task interaction data and the decision task interaction data according to the current task demand, and determines the communication information. The information compression model selects a suitable compression model to compress the image data according to the current communication state. The communication coding model is responsible for encoding the data to be transmitted according to the communication protocol.

[0012] The receiving end includes a communication decoding model and an information reconstruction model. The communication decoding model decodes the data to be transmitted according to the communication protocol to obtain the interaction data information and the compressed image information. The information reconstruction model reconstructs the received compressed data according to the current communication state to obtain the original image information.

[0013] Preferably, the perception task interaction data includes target detection and identification results, and search area images.

[0014] Preferably, the decision task interaction data includes aircraft state and action instructions.

[0015] Preferably, the decision change module obtains the latest target situation information from the deduction platform,

[0016] Preferably, the running process of the aircraft brain software system is as follows:

[0017] 1) The aircraft brain software system is connected to the deduction platform, the deduction platform is run, the landing point of the aircraft cluster is designed, the aircraft is controlled to take off, and the aircraft brain software system is run.

[0018] 2) The communication adaptive interaction module performs priority screening and decoding on the interaction information obtained by the aircraft brain software system from the deduction platform at the current time, judges whether the decision action needs to be performed according to the decoded information, and if no information related to the aircraft itself is received, goes to step 10), otherwise, goes to step 3).

[0019] 3) The intelligent master control scheduling module judges the decoded information. If the information is ground information, go to step 4), if the information is information of other aircrafts, go to step 5).

[0020] 4) Target change model in decision change module, according to the decoded information, obtaining the latest target allocation information of the current aircraft, updating the landing point of the aircraft, jumping to step 7);

[0021] 5) Information reconstruction model in communication adaptive interaction module, reconstructing the image compression information of other aircrafts according to the current communication state of the aircraft, entering step 6);

[0022] 6) The imaging simulation platform is called by the current search instruction information to image the specified area according to the current search instruction information, the reconstructed information and the current aircraft imaging information are detected and identified by the situation awareness module, the latest position and type information of the target are generated, the search allocation model in the decision change module generates the search instruction of the next moment of the aircraft according to the latest position and type information of the target, and enters step 8);

[0023] 7) Search allocation model in decision change module, real-time updating the tracking target according to the searched target information, real-time calculating the distance between the current position of the aircraft and the position of the tracking target, if the condition is met, no longer searching and target allocation, entering step 8); if the condition is not met, entering step 10);

[0024] 8) Screening the information to be sent, transmitting the information to be sent to the deduction platform by the information transmission model in the communication adaptive interaction module, entering step 9);

[0025] 9) If the information received by the deduction platform is instruction information, the deduction platform controls the corresponding aircraft model to act, entering step 10); if the information received by the deduction platform is image information, the deduction platform updates the situation information according to the image information, entering step 10);

[0026] 10) If the deduction simulation process is not ended, advancing one simulation step, repeating steps 2) to 9), otherwise, ending.

[0027] Preferably, in step 6), the search instruction includes which aircraft searches which area.

[0028] Preferably, in step 8), if the information to be sent is image information, it is compressed first and then transmitted to the deduction platform; if the information to be sent is search instruction and target allocation instruction, it is directly transmitted to the deduction platform.

[0029] Preferably, in step 1), when the aircraft brain software system is connected to the deduction platform, the information interaction between the brain software system and the deduction platform should be ensured.

[0030] Preferably, an aircraft model is run in the inference platform to simulate the motion state of the aircraft.

[0031] The present application has the following beneficial effects compared with the prior art:

[0032] (1) The perception decision communication integrated brain software system proposed in the present application can greatly improve the intelligent capability and online autonomous decision-making capability of the aircraft, and can be flexibly applied to various unmanned equipment, greatly improving the intelligent empowerment capability of the unmanned equipment.

[0033] (2) The perception decision communication integrated brain software system proposed in the present application adopts a modular design idea, and the modules can be flexibly combined and new functional modules can be added, so that different tasks performed by the unmanned system can be realized, and the generalization of the brain software is greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a composition schematic diagram of the present application;

[0035] Figure 2 is a hardware deployment schematic diagram;

[0036] Figure 3 is a flowchart of the present application. DETAILED DESCRIPTION

[0037] The present application will be further described below in conjunction with the drawings.

[0038] The present application proposes a highly autonomous architecture integrating intelligent perception, intelligent decision-making, and communication, which can online identify and perceive information during the execution of the task of the aircraft, and quickly execute decision-making actions according to the perceived information and the received information from the communication, so as to realize online autonomous perception and decision-making of the aircraft and improve the intelligent level of the aircraft. The present application solves the problems of poor autonomous capability and low intelligent level of the existing unmanned system.

[0039] As shown in Figure 1 , a perception decision communication integrated aircraft brain software system includes four functional modules, i.e., an intelligent main control scheduling module, a decision change module, a communication adaptive interaction module, and a situation awareness module, and can realize online autonomous perception and decision-making of the aircraft.

[0040] Intelligent main control scheduling module: The intelligent main control scheduling module is mainly responsible for the runtime sequence control and information flow control of each functional module in the online execution stage of the aircraft brain software system algorithm library, is responsible for realizing real-time online scheduling and runtime sequence control of each module of the aircraft brain software according to the real-time situation information obtained from the inference platform during the execution of the task of the aircraft, and controls the distribution of information flow, accurately distributes the perceived information and communication interaction information to each functional module.

[0041] Decision change module: the decision change module is composed of a target change model and a search allocation model, the target change model mainly acts on the middle stage of the aircraft (the distance of half of the target), and the aircraft brain software system obtains the latest target situation information from the deduction platform and makes online change decision of the landing point of the aircraft. The search allocation model mainly acts on the final stage of the aircraft (very close to the target), and the aircraft brain software system combines the information received from other aircraft in the deduction platform and the information identified by the self-situation perception module to make optimal task decision under multiple decision points, and outputs specific action instructions of the corresponding task;

[0042] Communication adaptive interaction module: the communication adaptive interaction module is divided into a sending end and a receiving end, the sending end is composed of an information screening model, an information compression model and a communication coding model, the information screening model is responsible for prioritizing, screening and combining the perception task interaction data (target detection and identification results, search area image) and the decision task interaction data (aircraft state, action instruction, etc.) according to the current task demand, determining the communication information, the information compression model selects a suitable compression model to compress the image data according to the current communication state, and the communication coding model is responsible for encoding the data to be transmitted according to the communication protocol;

[0043] The receiving end is composed of a communication decoding model and an information reconstruction model, the communication decoding model is responsible for decoding the data to be transmitted according to the communication protocol to obtain the interaction data information and the compressed image information, and the information reconstruction model is responsible for reconstructing the received compressed data according to the current communication state to obtain the original image information;

[0044] Situation awareness module: the situation awareness module is composed of a detection and identification model, mainly acts on the final stage of the aircraft (very close to the target), as the support of the search allocation module, identifies the reconstructed image information of the aircraft and the current detected image information, obtains the accurate position information and type information of the target, and provides them to the search allocation model.

[0045] In the present application, when the aircraft cluster is about to approach the target, the target takes measures to limit the communication within a certain range, specifically, the bandwidth of transmission is reduced, the information obtained by the aircraft to be transmitted is compressed by the information compression model, and the compressed information is restored on the receiving aircraft brain according to the reconstruction model, achieving fast transmission of information under the condition of limited communication.

[0046] When the aircraft cluster starts searching for the target, the imaging information of the current aircraft brain itself and the imaging information transmitted from other aircraft are identified by the detection and identification model to obtain the accurate position information and type information of the target, guiding the subsequent aircraft brain to complete more target search and allocation.

[0047] As Figure 2 As shown, it is a schematic diagram of aircraft brain software system hardware deployment, including aircraft brain, CPU, GPU, embedded intelligent computing load, etc., the aircraft brain algorithm library includes intelligent master control scheduling module, communication adaptive interaction module, decision change module and situation awareness module, according to the task characteristics, each module of the aircraft brain algorithm library is deployed on the hardware unit with the best performance, such as the deep neural network contained in the situation awareness module is deployed on the GPU, and the mathematical calculation is deployed on the CPU.

[0048] As shown in Figure 3 The running process of the aircraft brain software system of the application is as follows:

[0049] 1) The aircraft brain software system is connected to the deduction platform to ensure that the brain software system and the deduction platform can interact with each other, the deduction platform is run, the landing point of the aircraft cluster is designed, the aircraft is controlled to take off, and the aircraft brain software system is run;

[0050] 2) The communication adaptive interaction module is run to decode the interaction information obtained by the aircraft brain software system from the deduction platform at the current time, and to determine whether the decision action needs to be executed according to the decoded information, if no information related to the aircraft itself is received, go to step 8), otherwise, go to step 3);

[0051] 3) The master control scheduling module is called to judge the decoded information, if the ground information is received, go to step 4), if the information of other aircrafts is received, go to step 5);

[0052] 4) The target change model in the decision change module is run to update the landing point of the aircraft according to the latest target allocation information in the ground information, and go to step 7);

[0053] 5) The aircraft brain software system calls the information reconstruction model in the communication adaptive interaction module, and completes the reconstruction of the information from other aircrafts according to the current communication state of the aircraft, and goes to step 6);

[0054] 6) The search allocation model in the decision change module is run, the aircraft brain software system images the specified area according to the current search instruction information, calls the target detection model in the situation awareness module to detect and identify the reconstructed information and the current aircraft imaging information, generates the position and type information of the target, and generates the search instruction of the next time of the aircraft, and goes to step 8);

[0055] 7) running the search assignment model in the decision change module, updating the tracking target in real time according to the target information searched by the search, calculating the distance between the current position of the current aircraft and the position of the tracked target in real time, and not searching and assigning targets again when the condition is met, entering step 8); if the condition is not met, entering step 10);

[0056] 8) screening the information to be sent, running the information transmission model in the communication adaptive interaction module to transmit the information to be sent; if the information to be sent is image information, it is compressed first and then transmitted to the deduction platform; if the information to be sent is search instructions and target assignment instructions, it is directly transmitted to the deduction platform, entering step 9);

[0057] 9) if the information received by the deduction platform is instruction information, the deduction platform controls the corresponding aircraft model to act, entering step 10); if the information received by the deduction platform is image information, the deduction platform updates the situation information according to the image information, entering step 10);

[0058] 10) if the deduction simulation process is not ended, steps 2) to 9) are cycled, otherwise, it is ended.

[0059] Embodiment:

[0060] In this embodiment, the running steps of the aircraft brain software system integrating perception, decision and communication are as follows:

[0061] Step 1: the aircraft brain software is placed on a desktop computer, the simulation deduction platform is run on another desktop computer, the aircraft model is in the simulation deduction platform, the motion state of the simulation aircraft is simulated, the imaging simulation system is run on a computer, responsible for imaging, the three computers are connected through a network cable, the communication protocol is TCP, the aircraft brain software sends corresponding control instructions to the platform to realize the motion control of the aircraft model;

[0062] Step 2: Run the deduction platform, select a typical simulation environment, and set the scene to track ten aircrafts with three targets. Two targets are imp_tar, indicating important targets, and one target is normal_tar, indicating a general target. The scene is mainly tracking important targets. A desktop computer equipped with aircraft brain software simulates ten aircraft brains. The simulation platform designs the initial landing points for the aircraft cluster, such as the initial landing points of the ten aircrafts being [[132, 16, 0], [132, 16, 0], [133, 16, 0], [133, 16, 0], [132, 15, 0], [132, 15, 0], [132, 14, 0], [132, 14, 0], [133, 15, 0], [133, 15, 0]]. The three values of [132, 16, 0] represent the longitude, latitude, and altitude in the geodetic coordinate system, respectively. Other values are similar. The aircrafts take off according to the designed landing points, run the aircraft brain software system, and the simulation platform updates the aircraft state information in real time;

[0063] Step 3: The aircraft brain software system obtains the current situation information from the simulation deduction platform, runs the communication adaptive interaction module in the aircraft brain software system, and the communication decoding model in the communication adaptive interaction module decodes the current situation information obtained at the current time. Process the decoded situation information, determine whether to execute decision actions, if no relevant information is received, jump to step 8, otherwise, run step 4;

[0064] Step 4: The aircraft brain software system calls the main control scheduling module to judge the decoded situation information. If the received information is related to the ground, run step 5, if the received information is other aircraft information, run step 6;

[0065] Step 5: Run the target change model in the decision change module, according to the latest target allocation information in the obtained ground information, such as the latest information obtained by aircraft 1 in the middle of the flight from the ground is {“fly1”:[132.5, 16, 0]}, indicating that the aircraft 1 executes the decision action, the latest landing point information [132.5, 16, 0], and the original landing point information of the aircraft 1 is [132, 16, 0]. Update the landing point of the aircraft, update the latest landing point information to the corresponding aircraft brain software system, and jump to step 8:

[0066] Step 6: Other aircraft information includes target image compression and target type, position information detected by other aircrafts. The aircraft brain software system calls the information reconstruction model in the communication adaptive interaction module, and selects the corresponding reconstruction model according to the current communication state of the aircraft, reconstructs the received image compression information, realizes the restoration of image information in the communication limited environment, and enters step 7;

[0067] Step 7: Run the search allocation model in the decision change module, and the aircraft brain software system obtains the search image information at the current time from the shared address, calls the target detection model yolov5 in the situational awareness module to detect and identify the reconstructed image and the current aircraft image, generates the latest position information (longitude, latitude and height information) of the target and type information, and generates the search instruction of the next time of the aircraft. The instruction includes which aircraft and which area to search;

[0068] Step 8: Run the search allocation model in the decision change module, and update the tracking target in real time according to the target information searched by the search, and calculate the distance between the current position of the current aircraft and the position of the tracked target in real time. If it is less than 20km, no search and target allocation will be performed;

[0069] Step 9: Screen the information to be sent. For target update information, the instruction form is as follows:

[0070] Cmd={“unit”:”fly1”,”target_lla”:[132.5,16,0]}, for the search instruction of the search allocation model, it mainly includes two parts. One part is to compress the image information obtained by the current aircraft, and transmit the compressed information. The instruction form is as follows: Cmd={“send”:”fly1”,”rec”:”fly2”,”img_data”:img_bin,”det_data”:[131,15,0,1,0.8]}, where send represents the sender, rec represents the receiver, img_data represents the compressed image information transmitted, and det_dat represents the target information and type information identified. Where [131,15,0,”imp_tar”,0.8] represents 131 as the longitude information, 15 as the latitude information, 0 as the height information, imp_tar as the important target information, and 0.8 as the confidence of identifying the important target information. The other part is the search instruction of the next time of the current aircraft, which is as follows: Cmd={“unit”:”fly1”,”search”:[131,15,0]}, where unit is the intelligent agent that performs the search action, which is aircraft 1, and the center point of the search area is [131,15,0]. There is also a target allocation instruction as follows: Cmd={”fly1”:[132.5,16,0],”fly2”:[132.7,16,0],”fly3”:[132.7,16,0]....}, which represents the final landing point information of each aircraft. Run the information transmission model in the communication adaptive interaction module, and select the instruction information for transmission through the TCP protocol according to the specific situation;

[0071] Step 9: The deduction platform controls the corresponding aircraft model to act according to the control instruction sent by the aircraft brain software system, if the received is target update instruction information, the simulation platform controls the aircraft to change the landing point according to the field, and re-calculates the flight trajectory, if the received is search decision instruction, the simulation platform controls the search point to be transmitted to the imaging simulation platform according to the unit field, the imaging simulation platform images the corresponding search point, and places the imaging result under the specified shared folder, if the received is compressed image information and identification result information, the simulation platform updates the corresponding information in the situation, and the specified aircraft brain software system reads the situation information to obtain, and the current simulation step is ended:

[0072] Step 10: If the deduction simulation process is not ended, the simulation end condition is that all aircrafts are within 1km from the assigned landing point, the steps 3 to 9 are cycled, otherwise, the simulation is ended.

[0073] The application discloses a kind of aircraft brain software systems of perception decision communication integration.It includes scheduling model in intelligent main control scheduling module, and intelligent main control scheduling module is mainly responsible for the running time sequence control and information flow control of each functional module in aircraft brain algorithm library online execution phase.Decision change module includes target change model and search allocation model, and is responsible for aircraft online goal update, optimal decision change task execution scheme under multiple decision point conditions.Communication adaptive interaction module includes information compression model, information reconstruction model and information transmission model, and communication adaptive interaction module is divided into sending end and receiving end, and is responsible for information reception and transmission.Situation awareness module includes target detection model, and is responsible for detecting and identifying image information received by aircraft.Aircraft brain software system adopts hardware dynamic adaptation, software uses the working mode of modular design scheme, can be configured according to scene change and task demand change Required module.Through information interaction with simulation deduction platform, aircraft can be quickly applied in different scenes.

[0074] Parts of the application not described in detail belong to the known technology of those skilled in the art.

Claims

1. An aircraft brain software system integrating perception, decision-making and communication, characterized by: It includes intelligent master control scheduling module, decision change module, communication adaptive interaction module and situation awareness module; Intelligent master control scheduling module: During the aircraft's mission execution, it obtains real-time situation information from the simulation platform, performs real-time online scheduling of various modules of the aircraft brain software system, and controls the distribution of information flow; The decision-making change module includes a target change model and a search allocation model. The target change model acts on the mid-stage of the aircraft and updates the aircraft's final landing point based on the latest target situation information obtained from the deduction platform. The search allocation model acts on the terminal stage of the aircraft and, based on the latest situation information obtained from the deduction platform and other aircraft information obtained from the situational awareness module, determines the optimal task execution plan under multiple search allocation decision points according to the aircraft's own state information and target value, and outputs specific action instructions for the search allocation task. Communication adaptive interaction module: It is divided into a transmitter and a receiver. The transmitter is responsible for data compression and source / channel coding of the transmitted information, while the receiver is responsible for decoding and reconstructing the received data and transmitting it to the corresponding module. Situational awareness module: acts on the terminal section of the aircraft, identifies the imaging information of the current aircraft brain software system itself and the reconstructed information from other aircraft, obtains the accurate location information and type information of the target, and provides it to the search allocation model.

2. The aircraft brain software system integrating perception, decision-making and communication according to claim 1, characterized in that: In the communication adaptive interaction module, the sending end includes an information screening model, an information compression model, and a communication coding model. The information screening model is responsible for prioritizing, screening, and combining the perception task interaction data and the decision task interaction data according to the current task requirements to determine the communication information; the information compression model selects an appropriate compression model to compress the image data according to the current communication status; and the communication coding model is responsible for encoding the data to be transmitted according to the communication protocol. The receiving end includes a communication decoding model and an information reconstruction model. The communication decoding model decodes the transmitted data according to the communication protocol to obtain interactive data information and compressed image information. The information reconstruction model reconstructs the received compressed data according to the current communication status to obtain the original image information.

3. The aircraft brain software system integrating perception, decision-making and communication according to claim 2, characterized in that: The perception task interaction data includes target detection and recognition results and search area images.

4. The aircraft brain software system integrating perception, decision-making and communication according to claim 2, characterized in that: The decision-making task interaction data includes aircraft status and action instructions.

5. The aircraft brain software system integrating perception, decision-making and communication according to claim 1, characterized in that: The decision change module obtains the latest target situation information from the deduction platform.

6. The aircraft brain software system integrating perception, decision-making and communication according to claim 2, characterized in that: The operation process of the aircraft brain software system is as follows: 1) Connect the aircraft brain software system to the simulation platform, run the simulation platform, design the landing point for the aircraft cluster, control the aircraft takeoff, and run the aircraft brain software system; 2) The communication adaptive interaction module prioritizes and decodes the interaction information from the simulation platform obtained by the aircraft brain software system at the current moment, and determines whether a decision action needs to be executed based on the decoded information. If no information related to the aircraft itself is received, it jumps to step 10); otherwise, it goes to step 3); 3) The intelligent master control scheduling module determines the decoded information. If the information is ground information, it proceeds to step 4); if the information is information from another aircraft, it proceeds to step 5); 4) The target change model in the decision change module obtains the latest target allocation information of the current aircraft based on the decoded information, updates the landing point of the aircraft, and jumps to step 7); 5) The information reconstruction model in the communication adaptive interaction module reconstructs the image compression information of other aircraft according to the current communication state of the aircraft, and then proceeds to step 6); 6) The deduction platform calls the imaging simulation platform to image the designated area based on the current search instruction information. The situational awareness module detects and identifies the reconstructed information and the current aircraft imaging information to generate the latest position and type information of the target. The search allocation model in the decision change module generates the aircraft's search instruction at the next moment based on the latest position and type information of the target, and then proceeds to step 8); 7) The search allocation model in the decision change module updates the tracking target in real time based on the searched target information, and calculates the distance between the current position of the aircraft and the position of the tracking target in real time. If the conditions are met, the search and target allocation will no longer be performed and the process will proceed to step 8). If the conditions are not met, the process will proceed to step 10). 8) Filter the information to be sent, run the information transmission model in the communication adaptive interaction module to transmit the information to be sent to the deduction platform, and proceed to step 9); 9) If the information received by the deduction platform is instruction information, the deduction platform controls the corresponding aircraft model to perform an action, and the process proceeds to step 10); if the information received by the deduction platform is image information, the deduction platform updates the situation information according to the image information, and the process proceeds to step 10); 10) If the simulation process is not finished, advance one simulation step forward and loop from step 2) to step 9). Otherwise, end.

7. The aircraft brain software system integrating perception, decision-making and communication according to claim 6, characterized in that: In step 6), the search instruction includes which aircraft searches which area.

8. The aircraft brain software system integrating perception, decision-making and communication according to claim 6, characterized in that: In step 8), if the information to be sent is image information, it is first compressed and then transmitted to the deduction platform; if the information to be sent is a search instruction and a target allocation instruction, it is directly transmitted to the deduction platform.

9. The aircraft brain software system integrating perception, decision-making and communication according to claim 6, characterized in that: In the step 1), when the aircraft brain software system is connected to the deduction platform, it should be ensured that information exchange can be carried out between the brain software system and the deduction platform.

10. The aircraft brain software system integrating perception, decision-making and communication according to claim 6, characterized in that: An aircraft model runs on the simulation platform to simulate the aircraft's motion state.

Images