Can you tell us a bit about your background and how you became interested in space and AI?
I studied computer engineering at the University of Castilla-La Mancha, where I developed a strong interest in AI and mathematical modelling, particularly optimisation algorithms. This passion led me to pursue a PhD focusing on these areas. After completing my PhD, I did a postdoc focused on computer vision on embedded systems, which afterwards led me to start working with Ubotica, initially aiming to integrate AI into systems for space applications. The biggest challenges include the harsh environment of space, hardware control, and designing for edge computing. It was an exciting time to work with AI in space, as there was not much being done with AI in nanosats and CubeSats. Ubotica provided me with the opportunity to apply my research to real-world applications.
As the Principal Engineer and Managing Director of the Spanish office, can you describe your primary responsibilities?
My primary responsibility is to manage the team and support them in achieving our project’s goals and milestones. This involves working closely with team members, assisting with specific tasks, and organising our efforts to ensure everything runs smoothly.
What are some of the most exciting projects you’ve worked on at Ubotica, particularly in the realm of Space?
One of the most exciting projects is CogniSAT-6, which represents the culmination of our work over the past few years, integrated into one satellite. Collaborating with our partners to advance AI in space has been incredibly rewarding.
How do you see AI transforming space exploration and satellite technology in the next decade?
AI is currently transforming the world, especially in space exploration. I believe it will enable decision-making processes without direct human intervention, significantly enhancing efficiency and capabilities. This is crucial because it addresses constraints like the impracticality of constant human interaction and maintaining a constant data connection with spacecraft or satellites.
How would you describe the work culture at Ubotica? What sets it apart from other places you’ve worked?
Ubotica’s work culture is rooted in freedom and flexibility. Employees have the freedom to work from anywhere, which offers significant advantages. The company strongly supports engaging in projects that align with personal interests and passions. Employees are encouraged to pursue work they truly enjoy.
How does Ubotica foster innovation within its teams, especially in the rapidly evolving field of AI for space?
Ubotica provides a flexible and supportive environment, encouraging innovative thinking. Employees are not limited to established methods but are also given opportunities to engage in research and explore new ideas. This commitment to developing new knowledge is evident in various initiatives within the company, allowing employees to work on projects they are genuinely interested in.
What are some of the biggest technical challenges you face when deploying AI into space systems?
One of the biggest challenges is the integration phase. Incorporating everything into a small satellite is complex; it’s not feasible to include any available system. We must carefully select the best options, considering constraints such as space, power consumption, and other parameters. Moreover, the current trend in AI is to develop huge systems with a lot of capabilities, whereas due to the limitations previously mentioned, those kinds of models can be used, and it is necessary to create smaller neural networks capable of doing practically the same.
What trends in the space and AI industries are you most excited about? How is Ubotica positioning itself to leverage these trends?
I am particularly interested in the development of AI for small CubeSats. The integration of these compact systems in space, the constant evolution of technology, and the challenges this presents are fascinating. Two trends I am currently following involve the use of foundation models, which can incorporate vast amounts of knowledge and be adapted to specific problems. This is especially intriguing due to the scarcity of labeled data needed for meaningful solutions and the possibility of using them to help other smaller models to learn from their already acquired “wisdom”. At Ubotica, we are developing quick methods to integrate these models into our ongoing development. Additionally, generating supplementary data using generative AI can help address the problem of lacking data from space.
We are working on generating data to improve the neural networks and AI models we are currently developing. This aspect is particularly exciting to me.
What is your long-term vision for Ubotica’s role in the space industry, and how do you see your contributions shaping that future?
I envision Ubotica as a leader in the integration of AI into every type of system that can benefit from it in space, whether for exploration, data collection, or other applications. We are already working on improving real-time communication within satellites, as demonstrated by our work on CogniSAT-6. There are trends that could make this process even faster in the near future.
What advice would you give to young engineers who are interested in pursuing a career in AI and space technology?
The space industry is a challenging yet rewarding field. Aspiring engineers need to master a significant amount of mathematics and gain a deep understanding of hardware. Despite its rapid growth and inherent difficulties, don’t be afraid. With dedication and perseverance, it is possible to adapt and succeed.
Recent Comments