The Valencian Community aims to position itself at the forefront of aerospace innovation with the NeuroAI4Space project, a strategic initiative focused on incorporating advanced artificial intelligence systems into small satellites, aircraft, and drones, with key applications in Earth Observation.
The project, coordinated by Intigia, includes the participation of EMXYS, a leading company in electronics for space instruments; the University of Alicante; and the ITACA Institute of the Universitat Politècnica de València (UPV), through its Fault-Tolerant Systems research group.
The objective is to design a lightweight, low-cost, and low-power processing board capable of running artificial intelligence directly on board the platforms, avoiding the transmission of large volumes of data to the ground and thus improving mission efficiency and autonomy.
The research is funded by the European Union through the FEDER Comunitat Valenciana 2021–2027 Programme, within the IVACE+i Innovation – Strategic Cooperative Projects calls (Reference INNEST/2025/319), and responds to a growing need in the space sector: equipping onboard systems with greater autonomy, energy efficiency, and real-time decision-making capability.
This demand is particularly relevant in the New Space sector, characterized by more agile and cost-effective missions, where it is essential to have technologies that reduce operating costs and energy consumption without compromising reliability.
In this context, NeuroAI4Space will, for example, enable a satellite or drone to process images directly in orbit or in flight, selecting only relevant information and significantly reducing energy and bandwidth consumption.
Technical details
The technological proposal of NeuroAI4Space is based on a low-cost, lightweight, and ultra-low-power processing board that combines RISC-V processors with a neuromorphic chip specifically designed for the space environment. The solution uses commercial-off-the-shelf (COTS) components and advanced radiation protection methodologies.
“Unlike solutions based on conventional deep learning, the project is committed to spiking neural networks (SNN), which enable much more energy-efficient processing and offer greater intrinsic radiation tolerance,” the consortium notes.
Inspired by the functioning of the human brain, neuromorphic artificial intelligence constitutes one of the project’s main distinguishing features. Compared to traditional AI systems, this technology enables complex calculations with significantly lower energy consumption and greater resistance to the extreme conditions of space.
“These neuromorphic networks represent a paradigm shift compared to traditional deep learning, combining energy efficiency and robustness—two critical factors in space applications,” the consortium emphasizes.
Contribution of ITACA’s Fault-Tolerant Systems group
Within the project, the Fault-Tolerant Systems group at the ITACA Institute will be responsible for developing a comprehensive radiation-hardening strategy for a neuromorphic IP core. This strategy will enable its implementation on commercial FPGAs and non-radiation-hardened ASICs, ensuring reliable operation in Low Earth Orbit (LEO).
The methodology will take into account the specific requirements of each mission—such as altitude, duration, or level of criticality—to define appropriate levels of reliability and availability, as well as the degree of protection required while remaining compatible with system constraints.
“The project’s main innovation lies in the combination of neuromorphic hardware, advanced fault-tolerance techniques in both hardware and software, and experimental validation through fault injection and radiation testing, with the aim of developing the first European neuromorphic platform for space applications based on open technologies and free from export restrictions”, explain the ITACA researchers.
Added value of the project
Beyond its technological impact, NeuroAI4Space will strengthen the Valencian and Spanish aerospace ecosystem by promoting collaboration between universities and companies, encouraging technology transfer, and generating new, highly qualified employment opportunities.
The project advances the development of an open European platform, free from export restrictions, with the potential to become a commercial product for future space missions, as well as for emergency management, environmental monitoring, and tracking extreme weather events.