The Phoenix Project
FET-Open Horizon2020 project dedicated to developing
novel techniques to exploring unknown environments
Humans have been exploring the world from the depths of the oceans to the edges of the universe. Yet many environments remain inaccessible, even to modern cutting-edge technology. For example, exploring the status of waste water under the Fukushima reactor or discovering suitable sites for underground CO2 storage.
The Phoenix Project aims to investigate a new line of technology that will enable the exploration of difficult-to-access environments through the inspectors of the future: smart sensing agents with learning abilities and the autonomous capacity to take decisions.
The project uses a pipeline inspection test case to validate the results. The resulting technology however, could be broadly applied in contexts such as water, oil, waste, and heat distribution systems, chemical reactor vessels, space exploration, and eventually even inside the human body.
Exploring inaccessible environments
As a project our aim is to investigate a new line of technology that will enable the exploration of difficult-to-access environments exploiting a risky, highly-novel approach called Phoenix. Phoenix will explore inaccessible environments with physical agents that are extremely limited in size and resources, and can operate without direct control over software and hardware.
Phoenix is a radically new, high risk/high reward project. It also holds the promise to shed light on emergent properties of self-organization, local adaptation and division of labour in autonomous systems. The high societal benefits, foundational character and long-term focus make Phoenix a perfect fit for the Future and Emerging Technologies (FET) programme of the European Commission
The objective of the Phoenix is to develop a method to explore unknown and inaccessible environments. This involves three issues:
- Development of a co-evolutionary framework;
- Design of versatile agent technology and;
- Techniques to formalize different kinds of expert knowledge, even uncertain knowledge to influence the design of agents and the evolutionary algorithm which influences their evolution and “rebirth” in the Phoenix system.
Achieving these objectives will shed light on emergent properties of self-organization, local adaptation and division of labour in autonomous systems.
Evolution is achieved through injecting successive instantiations of swarms of miniaturized sensor agents into the inaccessible environment: at each generation, part of the agents are recovered, and the collected data are then used to jointly optimize, by means of evolutionary algorithms, a model of the environment, as well as the hardware and behavioural parameters of the agents themselves, before starting the next generation.
Phoenix will explore inaccessible environments with physical agents that are extremely limited in size and resources, and can operate without direct control over software and hardware.
How does it work?
- Phoenix starts with processing a user question through a GUI.
- It then assesses available knowledge from the Phoenix Knowledge Base and generates the first generation of motes to explore the environment.
- Next, Phoenix initiates an evolutionary process involving two nested generational loops:
- In the outer loop Phoenix develops, deploys and retrieves physical agents capable of penetrating the inaccessible environment and gathering information. Based on this knowledge, a model of the unknown environment is developed and evaluated.
- This model is refined in the inner loop, where environmental models and abstract representations of the physical agents (virtual agents) co-evolve in a virtual world until an improved generation of physical agents is ready for deployment. The goal of this co-evolution is to maximize the information captured about the unknown environment by progressively optimized agents.
Since the dawn of humankind, humans have explored the Earth directly or through measurement devices. Countless discoveries arose from that activity. The Phoenix project tries to find methods for evolving generations of small sensor agents that are increasingly adept at exploring an inaccessible environment.
The value of exploration cannot be underestimated and the outcomes of Phoenix are poised to make substantive social impacts:
- Mapping pipelines to find obstructions, leaks or faults. This could be part of a strategy to more efficiently deliver drinking water or to prevent contamination.
- Exploring underground channels, which cannot be otherwise accessed without damaging them. This could be part of a strategy related to more efficient extraction of oil or natural gas or to the search of natural CO2 storage locations.
- Measuring from the depths of glaciers or inside volcanos. This could be part of a strategy to better model climate change.
For now, we focus on the first use-case. We believe that the outcome from this line of research will be part of the foundation we need to elaborate more complex use-cases like the second and third one above.