Summary
Nowadays, uncrewed spacecraft missions not only push the boundaries of our scientific understanding of the Universe, but are an indispensable tool for monitoring and increasing our understanding of Earth. However, deploying and operating spacecraft comes with many challenges usually not encountered in ground-based systems. Two major challenges are the construction, repair and modification of structures in space (e.g. telescopes like JWST) and their operation under harsh constraints like limited power and transmission bandwidths, restricting especially the computing power available onboard spacecraft.
BASE (Biologically-inspired Autonomous Systems for Space Exploration) envisions a future of independent and autonomous spacecraft that can adapt to unexpected situations and reconfigure their structure accordingly without human guidance. To achieve this bold goal, this action will lay the conceptual foundation for such innovative technologies by developing novel concepts combining techniques from AI, neuromorphic computing and mathematics. More specifically, I will (i) for the first time thoroughly analyze the potential benefits (i.e. energy demands, performance) of spiking neural networks to enable onboard AI and (ii) develop methods for the autonomous self-assembly and self-modification of space structures from individual components, focussing on scenarios that are compatible with currently envisioned designs.
The ambitious applications considered in BASE will drive highly innovative progress in the considered disciplines, e.g., by deriving the first mathematical theory for characterising computational properties of spiking neural networks. By tightly binding theory and algorithms to concrete applications, the transferability of the outcome of this action to impactful technologies is guaranteed. Finally, the objectives of BASE are complementary, providing both the computing platform and methodology for spacecraft autonomy, yielding a highly unique research program.
BASE (Biologically-inspired Autonomous Systems for Space Exploration) envisions a future of independent and autonomous spacecraft that can adapt to unexpected situations and reconfigure their structure accordingly without human guidance. To achieve this bold goal, this action will lay the conceptual foundation for such innovative technologies by developing novel concepts combining techniques from AI, neuromorphic computing and mathematics. More specifically, I will (i) for the first time thoroughly analyze the potential benefits (i.e. energy demands, performance) of spiking neural networks to enable onboard AI and (ii) develop methods for the autonomous self-assembly and self-modification of space structures from individual components, focussing on scenarios that are compatible with currently envisioned designs.
The ambitious applications considered in BASE will drive highly innovative progress in the considered disciplines, e.g., by deriving the first mathematical theory for characterising computational properties of spiking neural networks. By tightly binding theory and algorithms to concrete applications, the transferability of the outcome of this action to impactful technologies is guaranteed. Finally, the objectives of BASE are complementary, providing both the computing platform and methodology for spacecraft autonomy, yielding a highly unique research program.
Unfold all
/
Fold all
More information & hyperlinks
| Web resources: | https://cordis.europa.eu/project/id/101103062 |
| Start date: | 01-09-2024 |
| End date: | 31-08-2026 |
| Total budget - Public funding: | - 199 440,00 Euro |
Cordis data
Original description
Nowadays, uncrewed spacecraft missions not only push the boundaries of our scientific understanding of the Universe, but are an indispensable tool for monitoring and increasing our understanding of Earth. However, deploying and operating spacecraft comes with many challenges usually not encountered in ground-based systems. Two major challenges are the construction, repair and modification of structures in space (e.g. telescopes like JWST) and their operation under harsh constraints like limited power and transmission bandwidths, restricting especially the computing power available onboard spacecraft.BASE (Biologically-inspired Autonomous Systems for Space Exploration) envisions a future of independent and autonomous spacecraft that can adapt to unexpected situations and reconfigure their structure accordingly without human guidance. To achieve this bold goal, this action will lay the conceptual foundation for such innovative technologies by developing novel concepts combining techniques from AI, neuromorphic computing and mathematics. More specifically, I will (i) for the first time thoroughly analyze the potential benefits (i.e. energy demands, performance) of spiking neural networks to enable onboard AI and (ii) develop methods for the autonomous self-assembly and self-modification of space structures from individual components, focussing on scenarios that are compatible with currently envisioned designs.
The ambitious applications considered in BASE will drive highly innovative progress in the considered disciplines, e.g., by deriving the first mathematical theory for characterising computational properties of spiking neural networks. By tightly binding theory and algorithms to concrete applications, the transferability of the outcome of this action to impactful technologies is guaranteed. Finally, the objectives of BASE are complementary, providing both the computing platform and methodology for spacecraft autonomy, yielding a highly unique research program.
Status
SIGNEDCall topic
HORIZON-MSCA-2022-PF-01-01Update Date
31-07-2023
Geographical location(s)
