CN-121985423-A - Method for planning physical observation task in heaven and earth integrated sensor network

CN121985423ACN 121985423 ACN121985423 ACN 121985423ACN-121985423-A

Abstract

The application provides a method for planning physical observation tasks in an heaven-earth integrated sensor network, which relates to the field of quantum computation and comprises the steps of obtaining available heaven-earth sensor capability data, discretizing a space-time range of a monitored scene, establishing a space, time and space-time mapping relation between the discretized space-time position and the sensor capability, carrying out quantum encoding on the space-time position, the sensor and the observation capability, carrying out quantum entanglement operation based on the three established mapping relations, establishing unified quantum state representation of the space-time observation capability of the heaven-earth sensor, establishing corresponding quantum operators aiming at specific computation requirements, carrying out efficient computation and measurement on the unified quantum state by using a quantum algorithm, so as to obtain a computation result, and realizing scheduling of the heaven-earth integrated sensor network. The technical scheme of the application fundamentally solves the inherent space-time observation capability characterization and calculation bottleneck under the classical calculation framework.

Inventors

LI JIE
HU CHULI
Ding xuan
WANG KE
CHEN NENGCHENG

Assignees

中国地质大学（武汉）

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (8)

1. A method of planning physical observation tasks in an integrated sensor network, the method being performed in a heterogeneous computing system comprising a processor and a quantum processor, comprising the steps of: Step S1, acquiring data of a space-time sensor and space-time observation capability data of the space-time sensor in a specific monitoring scene by the processor; step S2, discretizing a specific monitoring scene through the processor to obtain discrete space-time positions, and establishing three mapping relations among the discrete space-time positions, the space-earth sensor data and the space-time observation capability data; S3, quantum coding is carried out on the discrete space-time position, the space-time sensor data and the space-time observation capacity data based on three types of mapping relations through the quantum processor, and unified space-time observation capacity quantum state representation of the space-time sensor is constructed; And S4, constructing corresponding quantum operators and quantum circuits based on the acquired space-time observation capability calculation requirement through the quantum processor, calculating the space-time observation capability quantum state representation of the space-time sensor by using a quantum algorithm to generate a calculation result of the space-time observation capability, and dynamically adjusting an observation plan or parameter of at least one sensor in the space-time integrated sensor network based on the calculation result so as to execute collaborative physical observation on a target area.
2. The method for planning a physical observation task in an integrated sensor network as set forth in claim 1, wherein step S1 includes: Step S11, determining a space range and a time span of a specific monitoring scene, wherein the space range is defined by one or more polygons, and the time span is defined by a starting time point and an ending time point; and S12, acquiring an available space-earth sensor set in the space range and the time span, and determining space-time observation capability data of each space-earth sensor in the available space-earth sensor set, wherein the space-time observation capability data comprises observation start-stop time, earth observation coverage range, observation parameters and perception modes.
3. The method for planning a physical observation task in an integrated sensor network as set forth in claim 2, wherein step S2 includes: Step S21, discretizing the space range and the time span of the specific monitoring scene, wherein the space range is divided into one or more regular discrete grid units according to the specific spatial resolution, and the time span is divided into one or more discrete time periods according to the specific time resolution; Step S22, based on the observation start-stop time and the earth observation coverage of each earth sensor, establishing a space-time mapping relation between each discrete grid unit in each discrete time period and one or more sensors capable of observing the grid unit in the time period and space-time observation capability of the sensors; Establishing a time mapping relation between each discrete time period and one or more sensors capable of observing at least one discrete grid unit in the time period and space-time observation capacity of the sensors; Establishing a spatial mapping relation between each discrete grid unit and one or more sensors capable of observing the grid unit in at least one discrete time period and space-time observation capacity thereof; the three types of mapping relations comprise a space-time mapping relation, a time mapping relation and a space mapping relation.
4. A method of planning a physical observation task in an integrated sensor network as claimed in claim 3, wherein step S3 comprises: S31, quantum coding is executed, and a space-time position quantum state representing the discrete grid units and the discrete time period is constructed; The space-time position quantum state is used for identifying a mode control quantum state of the space, time and space-time mapping relation, a sensor quantum state representing the space-earth sensor set and identifying an observation capacity quantum state containing at least observation parameters and a perception mode; And S32, based on three types of mapping relations, using the mode control quantum state as core control, executing quantum entanglement operation, and establishing controllable association among the space-time position quantum state, the sensor quantum state and the observation capacity quantum state to form unified space-time observation capacity quantum state representation of the space-time sensor.
5. The method for planning a physical observation task in an integrated sensor network as set forth in claim 1, wherein step S4 includes: Step S41, inquiring the calculation requirement aiming at the specific space-time observation capability, converting the calculation requirement into one or more quantum operators suitable for quantum state representation based on the space-time observation capability quantum state representation of the space-time sensor, and constructing a corresponding quantum circuit; And S42, executing the quantum circuit by using a quantum algorithm, performing measurement operation on the executed quantum state, and decoding a measurement result to obtain a calculation result of space-time observation capability.
6. A method of planning a physical observation task in an integrated sensor network as defined in claim 1, further comprising: The quantum algorithm is a Grover search algorithm, the quantum operators comprise Oracle operators and Diffuser operators, and the calculation requirement is a spatial position and corresponding observation capability which can be observed by at least two sensors together in a specific time period.
7. A heterogeneous computing system comprising a processor and a quantum processor for implementing a method of planning physical observation tasks in an integrated sensor network according to any of claims 1-6, comprising a processor, a quantum processor, a memory for storing instructions, a user interface and a network interface for communicating to other devices, the processor and quantum processor for executing instructions stored in the memory.
8. A computer readable storage medium storing instructions which, when executed by a computer, perform the method of planning a physical observation task in an integrated sensor network according to any one of claims 1-6.

Description

Method for planning physical observation task in heaven and earth integrated sensor network Technical Field The application relates to the field of quantum computation, in particular to a method for planning a physical observation task in an heaven-earth integrated sensor network. Background In many critical application fields, such as disaster emergency response, environmental dynamic monitoring and comprehensive perception of smart cities, timely and comprehensive space-time information is the basis for realizing scientific decisions and efficient actions. The world sensors (e.g., satellites, aerial remote sensing, ground monitoring stations, etc.) are the core infrastructure for obtaining such information. However, to achieve efficient planning and scheduling of these massive, heterogeneous, ubiquitous sensor resources, a necessary premise is comprehensive, dynamic knowledge of the "spatiotemporal observability" of the sensor. The space-time observation capability of the sensor not only comprises static attributes such as space-time resolution, observation parameters and the like, but also covers dynamic states such as earth observation coverage, residual storage space and the like, and is a core index for measuring the observation efficiency of the sensor. In particular, with massive deployment of the world sensors, the converged space-time observation capability of the world sensors is forming a novel complex data resource which needs to be efficiently managed. This challenge is particularly pronounced in sudden disaster scenarios such as urban inland inundation. Such scenarios place extremely high demands on real-time and comprehensive disaster monitoring, which is highly dependent on comprehensive fine knowledge of available sensors and their capabilities and efficient planning and scheduling. Therefore, how to efficiently and comprehensively characterize and calculate the complex space-time observation capability, and apply the complex space-time observation capability to planning of physical observation tasks in an heaven-earth integrated sensor network has become a key technical problem with general significance to be solved urgently. Currently, modeling of space-time observability of world sensors is based mainly on "object", "field" and "object field" models in the geographic information science. The object model has the characteristics of comprehensive capability representation but complex query, the field model has the characteristics of incomplete representation but easy query, and the object field model aims to have the advantages of both. Specifically, modeling based on object fields firstly models space-time positions as fields, then models sensors and the capabilities thereof as objects, then builds an association mapping of the fields and the objects, and finally realizes multi-sensor capability characterization and calculation based on the positions. However, in the classical computing framework, this approach suffers from inherent characterization and computing drawbacks. As the number of world sensors, the target scene spatiotemporal scale and resolution increase, the complexity of characterizing and computing the spatiotemporal observation capability will increase explosively. For example, when the spatial and temporal resolutions are respectively increasedThe spatial complexity of the capability characterization and the temporal complexity of the capability calculation will both correspondingly increase by at least a factorMultiple times. The method can severely restrict the cognitive accuracy and efficiency of the time-space observation capability, and further prevent efficient and reliable sensor inquiry discovery and planning scheduling. Therefore, how to break through the space-time observation capability characterization and calculation bottleneck under the classical calculation framework is a core technical problem which needs to be solved currently. In 1982, feynman proposed the concept of "quantum computers". This is a novel computational paradigm based on quantum mechanics (e.g., quantum superposition and entanglement), which exploits the inherent quantum parallelism, and is recognized as having far beyond the computational power of classical computers in dealing with specific problems. Over forty years of development, quantum computing has made significant progress in both hardware (a general purpose quantum computer with hundreds of bits implemented) and theory (e.g., the shell and Grover algorithms). Notably, quantum computing has been successfully applied in recent years to image characterization and computation, particularly when processing raster images resembling "field" models, achieving more efficient characterization and computation than classical. Although the structure of the object field model is more complicated, so that the quantum field model algorithm cannot be directly applied, the method provides key implications for breaking through the characterization