Air-ground unmanned platform cooperative control method, device and medium for threat reconnaissance

Abstract

The present application belongs to threat detection technology, and specifically provides an air-ground unmanned platform cooperative control method, equipment and medium for threat reconnaissance, the control method is: the task is divided into multiple dynamic roles, each air-ground unmanned platform calculates the adaptation degree of each role based on the multi-modal state data of itself and broadcasts; through a distributed consistency algorithm, each platform reaches a consensus on role allocation, forming an initial cooperative networking; the unmanned platform obtains the information of the threat object by using analysis means according to the role it undertakes; according to the propagation characteristics of the threat object, predict suspicious threat sources and plan the inspection path; the platform executes the task along the path, adopts a task-driven hierarchical data security sharing mechanism in the process, and can trigger an emergency team based on dynamic trust to realize task connection when a node fails. The present application has the advantages of strong survivability, high traceability efficiency and the like.

Description

Technical Field

[0001] This invention relates to threat detection technology, and in particular to a collaborative control method, device and medium for air-to-ground unmanned platforms used for threat reconnaissance. Background Technology

[0002] Chemical warfare agents, due to their high toxicity, rapid action, wide diffusion range, and long-lasting effects, can generate rapid and sustained localized destructive power in a very short time. Developing on-site detection technologies capable of rapid response and accurate identification in complex battlefield environments is the core of chemical defense security.

[0003] Chemical warfare agent field detection technologies can be mainly divided into three categories: 1. While portable, technologies based on ion mobility spectroscopy and Raman spectroscopy rely heavily on pre-set databases and have a high false alarm rate.

[0004] 2. While technologies based on chemiluminescence and colorimetry possess a certain degree of sensitivity and selectivity, their accuracy and anti-interference capabilities are limited.

[0005] 3. Mass spectrometry, with its high sensitivity, high specificity, and broad-spectrum analytical capabilities, is gradually becoming the mainstream direction for the development of field testing equipment. Among them, miniaturized mass spectrometry technology, while maintaining excellent analytical performance, continues to overcome portability bottlenecks, leading the technological revolution in field testing equipment. However, all of these methods are limited to manual operation and are not suitable for direct detection in complex field environments.

[0006] With the rapid development of mobile devices such as drones and robotic dogs, mounting miniaturized mass spectrometers on mobile platforms is a novel application technology that can replace humans in rapidly detecting hazardous substances in chemical defense zones. However, it has several shortcomings, such as: 1. The use of a single platform (such as only drones or only ground robots) to carry detection equipment for reconnaissance has obvious limitations: drones and ground robots work independently, lack intelligent collaboration, and cannot form an efficient closed loop of "area scanning - point confirmation".

[0007] 2. Existing networks mostly adopt a preset master-slave mode, which cannot be dynamically reconfigured according to platform status, task progress and environmental changes, resulting in insufficient system resilience; full data sharing is often used between multiple platforms, which is inefficient and poses a risk of sensitive data leakage. In summary, the existing technologies suffer from the following systemic bottlenecks that urgently need to be addressed: (1) Lack of platform collaboration: UAVs and ground robots operate independently, lacking intelligent collaboration and dynamic task flow, and unable to form an efficient 'scan-confirm' closed loop; (2) Insufficient network resilience: The network mode is fixed, and it cannot be dynamically reconstructed when the platform fails or the environment changes; (3) Inefficient data sharing: Most of the data is shared in its entirety, which poses efficiency and security risks. Therefore, there is an urgent need for a collaborative control method for air-ground unmanned platforms that can achieve dynamic intelligent networking, adaptive task allocation, and safe and efficient collaboration. Summary of the Invention

[0008] To address the shortcomings of the existing technical solutions, this invention provides a collaborative control method for air-to-ground unmanned platforms used for threat reconnaissance.

[0009] The objective of this invention is achieved through the following technical solution: A collaborative control method for air-to-ground unmanned platforms used for threat reconnaissance, wherein the control method is as follows: The task is decomposed into multiple dynamic roles. Each air-to-ground unmanned platform calculates its suitability for each role based on its own multimodal state data and broadcasts it. Through a distributed consensus algorithm, the platforms reach a consensus on role allocation and form an initial collaborative network. Based on their assigned roles, unmanned platforms utilize analytical methods to obtain information about threats. Based on the propagation characteristics of the threats, predict suspected threat sources and plan investigation routes; The unmanned platform carries out inspection tasks along the inspection path according to its assigned role; Different access attributes are set for different levels of analytical data. Each platform is granted corresponding decryption attributes according to its currently assigned task role, so that the platform can only decrypt data within its task permissions. When a failure or communication interruption is detected in an unmanned platform node, an emergency team formation process is triggered. The remaining platforms calculate the dynamic trust level between each other based on historical interaction data and form an emergency team with the partner node with the highest trust level to take over the critical task chain that the failed node has not completed.

[0010] Another objective of this invention is to provide an air-ground cooperative unmanned system for threat reconnaissance, which is achieved through the following technical solutions: The system includes: a command and control center, at least one aerial unmanned platform, and at least one ground unmanned platform; The aerial unmanned platform is suitable for undertaking the role of wide-area sniffing or communication relay, and is equipped with a gas chromatography-mass spectrometry system or detection equipment for large-area scanning. The ground-based unmanned platform is equipped with detection equipment for large-area scanning, making it suitable for frontline inspection or continuous monitoring. The command and control center is used to run collaborative control software to implement a collaborative control method for air-to-ground unmanned platforms used for threat reconnaissance.

[0011] Another objective of this invention is to provide an electronic device, which is achieved through the following technical solution: An electronic device includes a memory, a processor, and a computer program, wherein the processor executes the computer program in the memory to implement the control method of this application.

[0012] Another objective of this invention is to provide a computer-readable storage medium, which is achieved through the following technical solutions: A computer-readable storage medium stores a computer program that, when executed by a processor, implements the control method of this application.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0014] 1. Through dynamic role bidding based on distributed consensus and emergency team formation technology based on dynamic trust, the system has self-organization and self-healing capabilities, which greatly enhances the system's resilience and survivability in complex adversarial environments. 2. By fusing multi-source data to construct a spatial concentration gradient field and using a Bayesian optimization algorithm for predictive dynamic tracking, the reconnaissance behavior is transformed from "blind search" to "intelligent tracking," significantly improving the efficiency of threat source discovery and tracing. 3. Through a task-driven hierarchical data encryption and attribute-based decryption mechanism, while ensuring collaborative efficiency, "knowing secrets on demand" is achieved, minimizing the leakage of sensitive data and meeting the high security requirements of military applications; 4. This method is not only applicable to chemical threat reconnaissance, but its core networking and control logic can also be extended to diverse scenarios such as biological agent detection and radioactive monitoring. Attached Figure Description

[0015] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are merely illustrative of the technical solutions of this invention and are not intended to limit the scope of protection of this invention. In the drawings: Figure 1 This is a schematic diagram of a collaborative control method for air-to-ground unmanned platforms; Figure 2 This is a schematic diagram of the collaborative networking structure of air-to-ground unmanned platforms; Figure 3 This is a schematic diagram of the dynamic tracking sampling process; Figure 4 This is a logical block diagram of emergency team formation and secure data sharing. Detailed Implementation

[0016] Figures 1-4 The following description illustrates optional embodiments of the invention to teach those skilled in the art how to implement and reproduce the invention. Some conventional aspects have been simplified or omitted to teach the technical solutions of the invention. Those skilled in the art should understand that variations or substitutions derived from these embodiments will be within the scope of the invention. Those skilled in the art should understand that the following features can be combined in various ways to form multiple variations of the invention. Therefore, the invention is not limited to the optional embodiments described below, but is defined only by the claims and their equivalents. Example 1

[0017] This embodiment describes a collaborative control method for air-to-ground unmanned platforms used for threat reconnaissance, such as... Figure 1 As shown, specifically: The task is decomposed into multiple dynamic roles. Each air-to-ground unmanned platform calculates its suitability for each role based on its own multimodal state data and broadcasts it. Through a distributed consensus algorithm, the platforms reach a consensus on role allocation, forming an initial collaborative network. The dynamic roles include wide-area sniffing roles, forward inspection roles, and continuous monitoring roles. The multimodal state data includes the platform's remaining battery power, the type of onboard detection instruments, geographical location information, and distance to the mission target.

[0018] Depending on their role, drone platforms use analytical methods to obtain information about threats (such as chemical warfare agents, biological warfare agents, or radioactive materials). Based on the propagation characteristics of the threats, predict suspected threat sources and plan investigation routes.

[0019] The unmanned platform carries out inspection tasks along the inspection path according to its assigned role.

[0020] During the above process, when a failure of an unmanned platform node or a communication interruption is detected, the emergency team formation procedure is triggered: The remaining platforms calculate the dynamic trust level between each other based on historical interaction data, and form an emergency team with the partner node with the highest trust level; the emergency team takes over the critical task chain that the failed node has not completed through negotiation.

[0021] Different access attributes are set for different levels of analytical data. Each platform is granted corresponding decryption attributes based on its currently assigned task role, so that the platform can only decrypt data within its task permissions.

[0022] An electronic device according to this embodiment includes a memory, a processor, and a computer program. When the processor executes the computer program in the memory, it implements the control method of this embodiment.

[0023] This embodiment provides a computer-readable storage medium that stores a computer program, which, when executed by a processor, implements the control method of this embodiment. Example 2

[0024] Example of the application of the air-to-ground unmanned platform collaborative control method, equipment and medium in the reconnaissance of chemical threats according to Example 1.

[0025] like Figure 1 As shown, the collaborative control method for air-to-ground unmanned platforms used for threat reconnaissance is as follows: This embodiment establishes an air-to-ground collaborative networking system consisting of one command and control center, two UAVs (UAV-1 and UAV-2) equipped with portable GC-MS, and two quadruped robot dogs (UGV-1 and UGV-2) equipped with direct ionization mass spectrometers. All platforms are equipped with GPS modules and wireless self-organizing network communication modules. Collaborative control software runs at the command and control center, which includes role management, situational awareness fusion, dynamic tracking and decision-making, emergency team formation, and secure communication encryption modules.

[0026] Initial task allocation and network setup.

[0027] The command and control center received a mission to conduct chemical reconnaissance in a certain area. The mission was broken down into three dynamic roles: wide-area sniffing (responsible for rapidly scanning large areas), forward reconnaissance (responsible for accurately confirming suspicious locations), and communication relay (responsible for ensuring the data link in areas with poor communication conditions).

[0028] Each unmanned platform (UAV-1, UAV-2, UGV-1, UGV-2) immediately assessed its own status: UAV-1 had 95% battery power and was equipped with GC-MS, with a "wide-area sniffing" suitability of 0.92; UGV-1 had 80% battery power and was equipped with direct ionization mass spectrometry, with a "frontline inspection" suitability of 0.88; UAV-2 had 85% battery power and was in the best position, with a "communication relay" suitability of 0.90.

[0029] Each platform broadcasts its status and compatibility. Through a distributed consensus algorithm, the platforms quickly reach a consensus: UAV-1 acts as a wide-area sniffer, UGV-1 as a frontline surveillance unit, and UAV-2 as a communication relay. The initial collaborative network is thus established. Figure 2 As shown. The distributed consensus algorithm can employ a voting-based algorithm or a distributed hash table (DHT) protocol to achieve decentralized and rapid consensus.

[0030] Dynamic tracking and task redistribution.

[0031] During a wide-area scan, the UAV-1 detected trace amounts of soman mimic (DMMP) signal using GC-MS. The situation fusion module fused the DMMP mass spectrometry signal intensity (ion current intensity at m / z=124) with GPS coordinates and real-time wind direction data, and generated a real-time DMMP concentration spatial gradient distribution heatmap using a Gaussian process regression algorithm.

[0032] The dynamic tracking decision module analyzes the heat map and predicts, using a Bayesian optimization algorithm, that the area with the highest concentration is located 500 meters downwind in an abandoned warehouse area (suspicious area A), and marks the coordinates of this area as a high-priority target.

[0033] At this point, the collaborative control software doesn't simply command UGV-1 to proceed. It initiates a new, smaller-scale task reassignment: UGV-1 and UGV-2 recalculate their suitability for the "inspecting Area A" task based on their own locations (UGV-1 is 200 meters from Area A, UGV-2 is 800 meters from Area A) and detection capabilities. Ultimately, UGV-1 wins with a higher suitability score and is dynamically assigned the forward inspection task, its path as follows... Figure 3 As shown. The fitness calculation can comprehensively consider distance factor, ability matching degree and remaining energy, and achieve quantitative evaluation through a weighted scoring model.

[0034] The UAV-1 unmanned platform converts the concentration signal from the mass spectrometer into specific motion control commands to achieve autonomous source locating, specifically: UAV-1 (wide-area sniffing) continuously acquires mass spectrometry data at a frequency of 2 Hz and extracts the ion current intensity of the target compound (DMMP, m / z 124) as the concentration signal c in real time. Each concentration sample has c... i With the corresponding GPS coordinates (x i ,y i and the current wind speed vector v w Binding. The situation fusion module uses the Gaussian process regression (GPR) algorithm to bind discrete concentration observation points (x... i ,y i , c i The algorithm fits the data to a continuous two-dimensional scalar field C(x, y), providing the predicted concentration value μ(x, y) at any location, and also giving the prediction uncertainty σ. 2 (x, y).

[0035] The dynamic pursuit decision module transforms the source-finding problem into an optimization problem, with its objective function being... Defined as: ,in α is the location coordinates to be evaluated, and α is a balance parameter (set to 1.5).

[0036] The decision module runs a Bayesian optimization algorithm, using a Gaussian process model to calculate the next optimal point (i.e., the point that enables...). The largest point This information is then sent to UGV-1. The local controller of UGV-1 receives the target point... The controller, based on the Model Predictive Control (MPC) framework, combines the robot dog's dynamic model, terrain information, and obstacle map to calculate a series of optimal control commands (wheel hub motor speed, joint servo angle, etc.). The robot dog executes these control commands and moves towards the target point.

[0037] As UGV-1 approaches the predicted source location, its onboard direct ionization mass spectrometer performs a touch-based detection of ground residues, obtaining more precise local concentration data. This data is transmitted back in real time, updating the global concentration field model C(x,y). Based on the updated model, the Bayesian optimizer may calculate a more accurate next target point and iterates until UGV-1 confirms the source location (detecting a concentration exceeding a set threshold and a gradual change in spatial gradient).

[0038] Suppose that while UGV-1 is en route to Area A, its communication system suddenly fails due to complex electromagnetic interference, causing it to lose contact with the command and control center and other platforms.

[0039] The emergency team-up module was triggered. UGV-2 and UAV-2, still within the communication network, detected that UGV-1 had lost contact. Based on historical interaction records (e.g., a 99% success rate in communication with UGV-1 over the past hour), they calculated that their dynamic trust level in UGV-1 remained "high." Based on this, UGV-2 proactively sent a team-up request to UAV-2 to form an emergency team. This emergency team, through shared situational information, confirmed that UGV-1's unfinished mission was "inspection of Area A." Therefore, with communication relay support from UAV-2, UGV-2 autonomously changed its original route and proceeded to Area A to relieve UGV-1 of its mission, ensuring the continuity of the reconnaissance mission. The process is as follows: Figure 4 As shown on the left. The dynamic trust level can be expressed by the formula T= w 1⋅ Rs + w 2⋅ Cq + w 3⋅ Ef Calculation, where Rs To improve the success rate of historical mission collaboration. Cq To score the quality of the communication link, Ef As a task efficiency factor, w 1, w 2, w3 represents an adjustable weight.

[0040] When a node in the system fails, a newly formed emergency team securely and automatically obtains the key needed to decrypt the data, such as... Figure 4 As shown on the right, the specific process is as follows: Upon joining the network, all unmanned platforms (UAVs, UGVs) are assigned a unique digital identity certificate by the Certification Authority (CA). The Command and Control Center, acting as the Attribute Authorization Center (AA), manages a list of attributes (e.g., Role: Frontline Inspection, Permission Level: High, Mission Area: Area A).

[0041] 1) Before the mission begins, AA pre-generates and securely distributes corresponding attribute private keys for each platform based on its initial role. The private key held by UGV-1, which acts as "Frontline Inspector," contains the attributes of role - Frontline Inspector and permission level - High.

[0042] 2) When UGV-1 loses contact while en route to Area A, the system confirms its status as "failed".

[0043] 3) Based on the historical communication success rate (98% for both UGV-2 and UAV-2), the dynamic trust level between them is calculated to be "high," and an emergency response team is automatically formed. UGV-2 is recommended as the new inspection unit.

[0044] 4) UGV-2 sends an attribute key update request to the Attribute Authorization Center (AA). This request is signed with its digital certificate to ensure authenticity and states: "Request to take over task T-A001 (Inspection Area A), requesting the corresponding attribute private key." Upon receiving the request, AA first verifies UGV-2's certificate and signature. Subsequently, AA's policy engine dynamically generates a new attribute private key based on the current task status (UGV-1 is invalid, task T-A001 needs to be taken over) and UGV-2's new role (temporary frontline inspector). This new private key contains attributes such as role - frontline inspector, permission level - high, and task ID: T-A001.

[0045] 5) AA uses UGV-2's public key (obtained from its certificate) to encrypt the newly generated attribute private key. The encrypted attribute private key is securely sent to UGV-2 via the remaining communication link (relayed through UAV-2). UGV-2 decrypts the key using its own private key to obtain the new attribute private key.

[0046] 6) At this point, the detailed reconnaissance data for Area A (the precise mass spectrometry characteristic data of DMMP) is encrypted using an attribute-based encryption (ABE) scheme, with the access policy: Role-Frontline Surveillance AND Task ID: T-A001. UGV-2, using its newly acquired attribute private key, perfectly satisfies this access policy, thus successfully decrypting the data and seamlessly taking over the reconnaissance mission. Other platforms in the system that do not meet this policy (UAV-1, still performing wide-area scanning) are unable to decrypt the data. The attribute-based encryption (ABE) scheme can employ ciphertext policy attribute-based encryption (CP-ABE), with the attribute authorization center (AA) managing attributes and distributing keys to achieve fine-grained access control.

[0047] Throughout the process, the data security sharing module continues to operate.

[0048] As a "wide-area sniffer," the raw mass spectra generated by the UAV-1 are marked as high-secret and encrypted using ABE. Only the command and control center and platforms assigned the "data analysis" role (which is not available in this example) have the authority to decrypt them.

[0049] The UAV-1 broadcasts characteristic data, namely, "DMMP detected, concentration level: medium, coordinates (X,Y)", which is a medium-density level.

[0050] The command and control center issued the instruction to UGV-1 to "proceed to coordinates (X,Y) to confirm the DMMP threat." This was a mission instruction and was classified as low-secret.

[0051] When UGV-2 took over the mission, it was given the role of "forward surveillance," granting it access to decrypt and receive more detailed DMMP signature data about Area A from the command and control center, but it still could not access UAV-1's original full spectrum. This mechanism ensures the security of data during the sharing process, with the logic as follows: Figure 4 As shown on the right.

[0052] An electronic device includes a memory and a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the control method of this embodiment.

[0053] A computer-readable storage medium stores a computer program that, when executed by a processor, implements the control method of this embodiment. The storage medium may be an optical disc or a magnetic disk.

Claims

1. A method for cooperative control of air-to-ground unmanned platforms for threat reconnaissance, characterized in that, The method for collaborative control of air-to-ground unmanned platforms used for threat reconnaissance is as follows: The task is decomposed into multiple dynamic roles. Each air-to-ground unmanned platform calculates its suitability for each role based on its own multimodal state data and broadcasts it. Through a distributed consensus algorithm, the platforms reach a consensus on role allocation and form an initial collaborative network. The unmanned platform uses analytical methods to obtain information about threats based on its assigned role; Based on the propagation characteristics of the threats, predict suspected threat sources and plan investigation routes; The unmanned platform carries out inspection tasks along the inspection path according to its assigned role; Different access attributes are set for different levels of analytical data. Each platform is granted corresponding decryption attributes according to its currently assigned task role, so that the platform can only decrypt data within its task permissions. When a failure or communication interruption is detected in an unmanned platform node, an emergency team formation process is triggered. The remaining platforms calculate the dynamic trust level between each other based on historical interaction data and form an emergency team with the partner node with the highest trust level to take over the critical task chain that the failed node has not completed.

2. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 1, characterized in that, The dynamic roles include wide-area sniffing roles, front-line inspection roles, and continuous monitoring roles; the multimodal status data includes the platform's remaining battery power, the type of detection instruments on board, geographical location information, and distance from the mission target.

3. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 2, characterized in that, The drone platform, which plays the role of wide-area sniffing, collects ambient air data in real time during flight and performs mass spectrometry analysis to obtain information on the concentration of threat substances; By combining geographical location and meteorological data, a spatial concentration gradient field of threat substances is constructed and updated; Based on the concentration gradient field, the control center uses an algorithm to predict the suspicious area with the highest concentration of threat substances and generates new reconnaissance waypoints or routes. The control center dynamically assigns forward inspection tasks to the ground unmanned platforms that are closest to the suspected area and have the corresponding capabilities.

4. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 3, characterized in that, The concentration gradient field is constructed and updated using a Gaussian process regression algorithm; the control center predicts the suspicious area with the highest concentration of threat substances using a Bayesian optimization algorithm; and the ground unmanned platform generates local motion control commands using a model predictive control algorithm to autonomously move to the target point for precise authentication.

5. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 1, characterized in that, The analytical method employed was mass spectrometry. The raw mass spectrum data was classified as high-density, the compound identification result as medium-density, and the threat level assessment result as low-density.

6. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 1, characterized in that, The calculation of the dynamic trust level is based on historical interaction data, including at least one of the following: historical task collaboration success rate, communication link quality history record, and task completion efficiency data.

7. The air-to-ground unmanned platform cooperative control method for threat reconnaissance according to claim 1, characterized in that, During the collaborative network execution of tasks, when the wide-area sniffing role platform detects that the concentration of threats exceeds a preset threshold, or when the credibility of the predicted suspicious threat source reaches a certain level, a new round of task allocation process is triggered for the suspicious threat source, and the currently idle or most suitable platform competes to assume the front-line inspection role.

8. An air-ground cooperative unmanned system for threat reconnaissance, characterized in that, include: One command and control center, at least one unmanned aerial platform, and at least one unmanned ground platform; The aerial unmanned platform is suitable for undertaking the role of wide-area sniffing or communication relay, and is equipped with a gas chromatograph-mass spectrometer or detection equipment for large-area scanning. The ground-based unmanned platform is equipped with detection equipment for large-area scanning, making it suitable for frontline inspection or continuous monitoring. The command and control center is used to run collaborative control software to implement the control method as described in any one of claims 1 to 7.

9. An electronic device comprising a memory, a processor, and a computer program, characterized in that, When the processor executes the computer program in the memory, it implements the air-to-ground unmanned platform collaborative control method for threat reconnaissance as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the air-to-ground unmanned platform cooperative control method for threat reconnaissance as described in any one of claims 1 to 7.

Images