Multi-Satellite In-Orbit Computing for Real-Time Remote Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current satellite systems transmit raw remote sensing images to the ground, resulting in high data volume and limited transmission efficiency, while in-orbit processing reduces data transmission burden but lacks efficient real-time analysis capabilities.
Innovation Solution
An in-orbit real-time task processing system utilizing multi-satellite collaborative computing, where a ground center determines a target satellite, sends instructions and location information, and the satellite performs data processing using a remote-sensing large model to generate result data, with optional semantic analysis and image processing tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If satellites transmit raw remote sensing images to the ground center, then the complete image data is available, but the data transmission burden is high and transmission efficiency is limited
Solution Approach 1:
The patent extracts and processes only the necessary information from raw remote sensing images on-board the satellite using AI models, rather than transmitting the complete image data. The satellite performs target detection, classification, and feature extraction to generate condensed result data that contains only essential information, thereby reducing transmission volume while maintaining analytical value.
Solution Approach 2:
The patent implements preliminary data processing actions on-board the satellite before transmission to the ground center. The satellite performs image preprocessing, target detection, and information extraction in-orbit, so that the data is already processed and condensed before being transmitted downlink, eliminating the need for extensive ground-based processing and reducing transmission requirements.
2Quantity of substance
If satellites process remote sensing images while in orbit, then the data transmission burden is reduced, but real-time analysis capability is insufficient
Solution Approach 1:
The patent enables satellites to perform self-processing of their own captured images using on-board AI models and computing resources. The satellite independently executes target detection, classification, and information extraction algorithms without requiring ground station intervention, achieving autonomous real-time analysis that reduces both transmission burden and processing time.
Solution Approach 2:
The patent implements dynamic task allocation and processing priorities based on real-time conditions. The system can dynamically adjust which satellites process which images, prioritize time-critical tasks, and adapt processing intensity based on transmission requirements, enabling flexible real-time analysis that responds to changing operational needs.
3Productivity
If multi-satellite interconnection and collaborative computing are implemented, then the utilization efficiency of communication links is improved, but the system complexity increases
Solution Approach 1:
The patent segments the remote sensing processing task across multiple satellites, where each satellite independently processes its own images and shares results through a coordinated network. This distributed segmentation allows parallel processing that improves communication efficiency while keeping individual satellite complexity manageable through independent but coordinated operation.
Solution Approach 2:
The patent introduces a ground-based task scheduling system and communication coordination module that acts as an intermediary between satellites. This intermediary manages the complex coordination by centralizing task allocation, optimizing communication routes, and synchronizing processing activities, thereby improving link utilization while abstracting the complexity from individual satellite operations.
Data Source
AI summary
Provided are an in-orbit real-time task processing system, method, and device based on multi-satellite collaborative computing. The system includes a ground center and a plurality of satellites. The ground center responds to instruction text of a user, determines a target satellite based on orbital trajectories of the satellites and the task location information, and sends the instruction text and the task location information to the target satellite. The target satellite collects a remote-sensing image in real-time, determines a data processing task corresponding to the instruction text through a remote-sensing large model, performs data processing on the remote-sensing image to obtain result data, and sends the result data to the ground center. The system realizes multi-satellite distributed computing and the forwarding of information to any location through the satellite interconnection network, and can obtain real-time remote data, which improves the convenience and efficiency of satellite remote sensing image applications.