The future of space

AIKO is a deep tech software company specialized in the design and development of Artificial Intelligence for the automation of space missions.

A picture of Sicily viewed from space, with the output of a Cloud Segmentation algorithm.

Automation will play a fundamental role in enabling two major benefits in space mission engineering:

  • reducing the costs of space missions
  • increasing the efficiency and performances of satellites

In orbit, our technology enables new capabilities on satellites, that become able to take the autonomous decisions on how to carry out their mission goals, even if not in contact with ground. On ground, our technology supports operators and engineering teams in optimizing how operations are carried out, taking more informed decisions, discovering patterns and novel behaviours, and focusing on critical activities.

Problems we focus on

A list of problems we are solving in the space industry. Several problems are there to be solved in how operations are managed in a space mission, and in AIKO we focus both on improving the flight-segment and ground-segment. For issues that we can solve on-board, we focus on enabling:
  • data understanding capabilities on the satellite, to prioritize and select data based on information content.
  • the recognition of observation opportunities, to enable new acquisition strategies and improve how data is collected during a mission
  • prediction and anticipation of failures, integrating predictive maintenance in how a satellite is operated
  • cooperation and coordination among satellites in a constellation
On ground, our technology is deployed in the Mission Control Centre. It helps in:
  • supporting operators in analysing data, in detecting patterns, novel behaviours and correlations, in order to take more informed decisions
  • managing constellations made of hundreds of satellites

Investor relations

SEED ROUND

MiRAGE: Autonomous Operations library

AIKO is currently negotiating the Seed Round, targeted at 2020. The objectives of the round are product finalization and space qualification, and business development.
Things we look for in an investor: smart money, strategy, network.
We put the dream team, the vision, the commitment.
Get in touch: drop us an email.

GEMINEA invests in AIKO

Geminea logo

GEMINEA (www.geminea.com), Milan-based VC focused on Artificial Intelligence startups, invests in AIKO and in its vision of using AI to enable autonomous space missions. The round is aimed at consolidating the team and the company to achieve a higher level in the technology development. We look forward to a shared path of growth and success!


Autonomous operations library

MiRAGE: Autonomous Operations library Co-founded by the European Commission

Enabling e4 autonomy

MiRAGE is an innovative software library to enable fully autonomous space missions, making use of state of the art Artificial Intelligence and is the result of years of research in Mission Autonomy.

The library has been co-funded by the H2020 program SME Instrument.

The Concept

MiRAGE: Autonomous Operations library

MiRAGE operates by analyzing the data made available during the mission (telemetry payload and external data) to generate and update the operations schedule. The satellites become aware, understanding their surroundings, the internal health, and eventual commands provided by the Mission Control Centre. MiRAGE library enables autonomous decisions during a mission, tailored to the needs of the end-users and the mission stakeholders.

Enabled benefits

Benefit: autonomous satellites

Planning and scheduling is a very resource-intensive activity during most of space missions. Performing schedule changes in reaction to unplanned mission events is even more challenging, due to poor predictability of these occurrences, to the necessity of having staff on-site 24/7, and to the quick reaction times requested in order to reduce eventual mission outages.
By shifting the decision-making on-board the satellite, MiRAGE library eliminates the decision-making loop in the Mission Control Centre for all the events that were accounted for during spacecraft development, including failures, events related to mission objectives, or events generated by other elements in a constellation.
Operators still maintain highest priority in commanding the spacecraft: they are now just focusing on more critical decisions that might still require ground intervention.

Benefit: aware satellites

We implement the latest advancements in Artificial Intelligence research to enable awareness on the satellite and to perform event detection. Our AIs are trained with datasets generated on purpose for each feature selected in MiRAGE: failure detection, payload analysis, pattern recognition including novely detection. Moreover, predictions are enabled during the mission: if certain trends are likely to result in a future Out Of Limit alert, our AIs are able to warn operators ahead of time.

Benefit: smart satellites

By using MiRAGE, more complex missions are at reach: tracking features on the surface, activating enhanced acquisition modes, selecting and prioritizing payload downlinked data. The result is increased efficiency, mission objectives are fulfilled sooner, and downlink bandwidth is occupied only with relevant data.

Reducing operations costs

The real, long-term benefit of fully autonomous mission is the reduction of the operations costs. This is particularly important for small satellite missions, where operations costs have been shown not to scale with spacecraft size, and for constellations, where the high number of operators per spacecraft simply cannot be afforded with hundreds, thousands of satellites.

TRL and Compatibility

MiRAGE, currently at TRL 6, is scheduled to reach TRL 7 by end of 2018.

MiRAGE TRL plan


The on-board version is compatible with Linux-based OS, while the ground-based one supports Win/Unix/Mac OS architectures.

Two MiRAGE versions
We believe future space missions will be fully autonomous, and our technology is the enabler.

Interested in seeing MiRAGE live?
Request information or a demonstration tailored on your mission concept.

Technology

AIKO provides state of the art expertise and technology in three distinct areas related to the development and operations of space missions.

State of the art AI-enabled automation

AIKO is an Artificial Intelligence and Automation company. For this reason, our core company values revolve around providing the most advanced, state of the art technology for the problems we set out to solve. Our pipeline for designing, developing and deploying an AI algorithm is lean and efficient, and our technology references are updated at least twice per year.

AI on the edge

A NVIDIA Jetson Nano board.

AIKO’s unique expertise involves developing and deploying state of art AI to edge devices, including recent releases of AI accelerators such as the Intel Myriad X, the NVIDIA Jetson, and the Google Coral. We have successfully deployed AI algorithms to commercial processors (ARM and desktop architectures), to mobile GPUs, and to Accelerators. From 2020, we started working on FPGAs too, to further extend our knowledge and capabilities in the embedded world.

AI for safety-critical applications

A picture of a spacecraft being tested.

In addition to the expertise in quickly and reliably integrating state of art technologies upstream, AIKO expertise is growing in the field of verification and validation of AI for safety-critical applications, both in the field of supervised AI and in the field of robust and explainable AI.

Expertise

Advanced Flight Software

Mission specific Command and Data Handling algorithms for E4-level autonomy.
Artificial Intelligence for Mission Replanning, Failure Detection, Isolation and Reconfiguration, Event Detection, Data Selection.

Big Data applications for space missions

Analysis algorithms for Mission Data. Feature Extraction, Image Classification, Pattern Recognition.

Expert Systems for Space Engineering

Employing Case-Based Reasoning to perform System Diagnosis, Design and Decision-Making, Project Planning.

Artificial Intelligence

Artificial Intelligence on embedded platforms

We make of Artificial Intelligence our flagship technology:
- Deep Learning
- Expert Systems
adapting the technology to Space Systems (C&DH, AOCS, GNC, EPS) and to Space Missions. Out team has the perfect intersection between Space Systems Engineering knowledge and Artificial Intelligence state of the art. We know exactly which algortihm to use depending on what data is analyzed and where this data comes from.

Autonomous Operations

Leaders on autonomous operations

Years of research on autonomous operations for space missions ensure our work reflects most recent approaches to autonomy. Our team has research published in important peer-reviewed international journals (Mechanical Systems and Signal Processing, Robotics and Autonomous Systems, Journal of Guidance, Control, and Dynamics), and several participations in international aerospace congresses (IAC, AIAA Space, IEEE Aerospace, ESA 4S, European CubeSat Symposium).

On-Board Software

Experts in integrating Artificial Intelligence to embedded

At AIKO, we develop embedded software for most typical satellite COTS processors:
- ARM family, MSP family
Development heritage on the flown e-st@r I and II satellites.

Who we are

Engineering

Lorenzo

CEO
Mission Autonomy

Loris

Systems Engineering
Knowledge Management

Christian

Software
Embedded

Alessandro

Autonomous reasoning
Intelligent agents

Ilaria

Deep Learning
Data Science

Mattia

Deep Learning
Embedded AI

Legal and Business

Giorgio

Co-founder
Doctor in Law
Legal

Enrico

Senior Consultant
@I3P
AIKO tutor

01/02/2018
Space Science Institute

Machine Learning algorithms for Mars Event Detection on a SmallSat mission

SSI Logo

AIKO signs a collaboration with the Space Science Institute, Colorado, US.
The collaboration will be focused on evaluating the implementation of Machine Learning algorithms for a Mars SmallSat mission concept.
The Space Science Institute (SSI) is a nonprofit research and education corporation based in Boulder (CO, USA) with a vision to expand humankind's understanding and appreciation of planet Earth, our Solar System, and the universe beyond. SSI is home to more than 60 scientists carrying out cutting edge research in astrophysics, planetary science, and space physics. It is also a leader in translating science into education and outreach projects, and it was home to the Cassini Imaging Science Subsystem Ops team, which provided the best pictures of Saturn, its rings, and moons until the end of NASA's Cassini mission.

Mars gif cloud detection Image source: NASA. Processing: AIKO.

The project opens an important commercial direction for AIKO, with the first collaboration aimed at the implementation of AI-based algorithms for interplanetary missions. Small Satellites are characterized by severe limitations, that make their utilization in interplanetary missions challenging. AIKO algorithms are an enabling technology that will make future Small Satellites missions to other planets possible.



01/09/2018
ESA Advanced Concepts Team

Deep Learning for spacecraft applications

AIKO starts a collaboration with ESA Advanced Concepts Team, ESTEC, The Netherlands.
The collaboration will focus on developing a tool for evaluating the computational requirements for deep learning state of the art algorithms for embedded applications. ACT Logo
The outcome of the project will allow an user to estimate, given a specific or custom architecture of Deep Neural Network, the computational requirements of the algorithm. In particular, information such as operation counts, memory utilization and inference time for each processor will be displayed. The tool represents an initial step towards promoting the diffusion and utilization of Artificial Intelligence in space applications and will be provided as a freely available web application on AIKO website.

Deep Learning on Embedded


The Advanced Concepts Team (ACT) is part of the ESA’s Directorate of Technical and Quality Management (TEC-SF). The team is essentially a channel for the study of technologies and ideas that are of strategical importance in the long term planning of the Agency. It serves the function of a think tank providing decision makers the support of a highly multidisciplinary research group. Based at ESTEC, they carry out research work on advanced topics and emerging technologies and perform highly skilled analysis on a wide range of topics.



PRIVACY

AIKO does not handle personal data.
For additional enquiries, please contact privacy@aikospace.com

30/04/2019
In-orbit demonstration agreement

Deep Learning for Earth Observation applications

AIKO signs an agreement with Satellogic (satellogic.com), an Earth Observation satellite company to validate Deep Learning technology. The activity will be carried out within Satellogic's Software and Hardware Hosted Payloads Program during 2019.

Why this is important for us

In-orbit validation is a key step in order to improve AIKO heritage. The Technology Readiness Level of AIKO technology will reach the status of TRL 9: demonstrated in real mission scenario.
AIKO's technology enables more ambitious commercial and scientific goals for Earth Observation missions, as an increased degree of on-board autonomy helps in implementing more effective mission operations. In particular, functionalities like event detection, autonomous planning and goal management, if implemented on-board, introduce several benefits to the way operations are managed. The deep learning opens new possibilities and horizons, such as:

  • Prioritization and payload data selection
  • Onboard data enhancing

In addition to the activities carried out during the program, AIKO will further improve its internal procedures for implementing and optimizing Deep Learning software in the context of space systems on-board execution. Given this opportunity, AIKO further consolidates its role in the Space industry, as provider of automation and Artificial Intelligence solutions, dramatically expanding its know-how and expertise.

A sincere thank you to Satellogic for the opportunity!

Earth from ISS at night

Facilities

AIKO premises are located in the I3P incubator of Politecnico di Torino, Italy, ranked in 2014 among the top 25 best incubators in the world. AIKO offices are equipped with the needed infrastructure to carry out modern engineering activities related to AIKO background. This includes, Artificial Intelligence design and training, software R&D, teleconferences and concurrent engineering. AIKO makes use of state of the art cloud computing infrastructure provided by Google, NVIDIA and Microsoft, in addition to local GPU-enabled resources available to each AIKO engineer.

I3P offices from outside


AIKO offices are located in Torino, in a central location fully connected to train and bus station, and within walking distance from the Politecnico di Torino.

Torino city