Onboard AI Cloud Detection with PUS for Satellite Power Savings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing onboard cloud detection systems for complex space missions face challenges such as hardware incompatibility with space-grade standards, high costs, increased complexity due to parallel AI processing, and excessive power consumption, which are not optimized for efficient resource management.
Innovation Solution
A dedicated onboard cloud detection system utilizing a packet utilization standard (PUS) that includes a camera, data processing unit, and AI algorithms to detect clouds, enabling or disabling main payload equipment based on cloud coverage, thereby optimizing memory and power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a dedicated AI image processing unit is implemented for each instrument or a centralized DPU processes multiple data streams in parallel, then cloud detection capability is improved, but device complexity increases
Solution Approach 1:
The patent implements a centralized DPU that serves multiple instruments simultaneously, making the cloud detection system universal rather than dedicated to each payload. The DPU processes data streams from multiple sources through a unified architecture, reducing overall system complexity while maintaining detection capabilities across all instruments.
Solution Approach 2:
The patent merges cloud detection functionality into the existing DPU architecture rather than adding separate processing units. By combining cloud detection with the centralized data processing function, the system avoids parallel hardware implementations and reduces device complexity while maintaining reliable cloud detection across multiple instruments.
2Reliability
If the payload operates continuously to ensure operational readiness, then detection capability is improved, but power consumption increases
Solution Approach 1:
The patent implements periodic cloud detection at predetermined intervals rather than continuous payload operation. The DPU acquires images from the camera at specific time intervals and processes them for cloud detection, allowing the payload to enter low-power states between detection cycles while maintaining operational readiness when needed.
Solution Approach 2:
The system uses a dedicated camera and DPU for cloud detection that operates independently from the main payload processing. This self-service detection capability monitors cloud conditions without requiring the main payload to remain fully operational, enabling power savings while maintaining detection functionality.
3Loss of information
If cloud coverage data is continuously stored on-board for analysis, then data availability is improved, but memory space consumption increases
Solution Approach 1:
The patent extracts only the essential cloud detection results (cloud presence/absence at detection intervals) rather than storing all raw images and processing data. The DPU processes images and stores only the relevant cloud coverage information, significantly reducing memory space requirements while maintaining data availability for mission analysis.
Solution Approach 2:
The system stores cloud detection data with varying detail levels based on local requirements. Full-resolution processing occurs only when needed, while routine cloud monitoring uses compressed or summarized data storage, optimizing memory usage according to the specific quality requirements of different mission phases.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An onboard cloud detection system comprising: a camera (1000) configured to acquire images of the Earth at predetermined acquisition intervals; and a data processing unit (2000) comprising: a cloud detection unit (2210) configured to use artificial intelligence, AI, algorithms to detect clouds; a packet utilization standard, PUS, application layer (2230) configured to issue telemetry and/or telecommands corresponding to a predetermined parameter of the output of the cloud detection unit (2210); and an interface configured to distribute the telemetry and/or telecommands to an external hardware and/or an external software terminal (3000, 4000).