-
StatusOngoing
-
Status date2024-07-24
-
Activity Code1A.127
The primary goal of this project is to enhance our collective
understanding of the advantages and challenges associated with employing generative AI in satellite communication (SatCom) design. It seeks to pinpoint the most advanced tools and methods presently utilised in generative AI across various industries, determining their relevance to the SatCom sector and the necessary modifications to tailor these tools for SatCom applications.
The task also includes the delivery of three proof-of-concept generative design tools.
Ultimately, the project outlines a detailed plan for integrating these tools into SatCom design, complete with projected timelines and development expenses.
The primary challenge of this project lies in the untested nature and complexity of generative AI techniques within the SatCom domain. These AI technologies, while promising, present a steep learning curve and uncertainty regarding their direct applicability and effectiveness in SatCom design.
It is essential to ensure that the AI-generated solutions are not only innovative but also practical, manufacturable, and compliant with industry standards. Additionally, the success of generative AI heavily relies on the availability and quality of data.
Generative AI has experienced a massive explosion of advancement and interest in the past few years. Exploring the capability of generative AI in design processes offers potential for significant improvements in design methodologies and the refinement of final designs.
By enabling the rapid generation of numerous design alternatives that meet specific criteria, generative design significantly reduces the time from concept to completion. And, unlike previous generative methods, generative AI learns from existing data, leading to innovative and optimised solutions by identifying patterns and possibilities that may not be immediately apparent to human designers. An AI-driven approach also supports better decision-making by providing a broader array of design options and potential outcomes.
Furthermore, it significantly reduces costs by automating parts of the design process that have traditionally required human input. It also optimises designs for material use and energy efficiency.
This project advances our understanding of how recent generative AI developments apply to SatCom design.
Focus on Generative Design: The project centres around exploring and utilising generative design methodologies, particularly in the context of SatCom systems.
State-of-the-Art Exploration: A significant portion of the project involves conducting literature reviews on the latest advancements in generative design and tools specific to SatCom.
Phased Approach: The project follows a phased structure, starting with a kickoff, followed by a literature review on generative design, and then moving on to explore state-of-the-art (SOTA) tools in SatCom design applications.
Innovation and Optimisation: By focusing on generative design and SOTA tools, the project aims to innovate and optimise SatCom design processes. This could lead to more efficient, cost-effective, and high-performing SatCom systems.
Progressive Development: The project progresses from a feasibility study to a demo phase, and potentially to an operational phase, reflecting the ambition not just to explore and understand but to apply and implement findings in practical, operational settings.
Using state-of-the-art AI techniques, such as diffusion models and transformers, we aim to embed generative AI within the design processes. We use cloud computing and GPUs to process and analyse vast amounts of data efficiently.
The project covers the delivery of software proof-of-concepts of the following key components:
-
A generative model capable of delivering 3D designs for SatCom parts.
-
Large Language Models (LLMs) as design support, offering an innovative way to interact with design software.
-
A diffusion-based satellite image generator.
To ensure these tools are accessible and user-friendly, a user interface is provided that allows people to experiment with and explore the capabilities of our software.
This project spans 9 months, starting from the 1st of July 2024.
The project tasks include:
-
Survey and Assessment of Current Generative AI Techniques
-
Identification of Potential SatCom Applications for Generative Design
-
Development of Generative AI Models
-
Hardware Assessment
-
Programmatic Legal and Development Gaps
Milestone 1 is achieved upon completing tasks 1 and 2, followed by Milestone 2 with the completion of task 3. The final milestone is reached at the project's conclusion.
Additionally, a workshop with industry experts takes place at the end of Task 2.
The project, started on 1st July 2024, is progressing well. Currently, a literature review on the state-of-the-art (SOTA) in generative design is underway, laying the foundational understanding necessary for the project's success. The next phase involves a comprehensive review of SOTA tools in satellite communication (SatCom) design, set to begin shortly. This sequential approach ensures a thorough exploration of existing knowledge and technologies, moving the project from its initial feasibility study towards a more developed and demonstrative phase.
